1 \input texinfo @c -*-texinfo-*-
5 @c oooooooooooooooooooooooooooooooooooooooooooooooooooooooooooooooooooooooooooo
7 @c GNAT DOCUMENTATION o
11 @c Copyright (C) 1995-2004 Free Software Foundation o
14 @c GNAT is maintained by Ada Core Technologies Inc (http://www.gnat.com). o
16 @c oooooooooooooooooooooooooooooooooooooooooooooooooooooooooooooooooooooooooooo
18 @setfilename gnat_rm.info
19 @settitle GNAT Reference Manual
20 @setchapternewpage odd
23 @include gcc-common.texi
25 @dircategory GNU Ada tools
27 * GNAT Reference Manual: (gnat_rm). Reference Manual for GNU Ada tools.
31 Copyright @copyright{} 1995-2004, Free Software Foundation
33 Permission is granted to copy, distribute and/or modify this document
34 under the terms of the GNU Free Documentation License, Version 1.2
35 or any later version published by the Free Software Foundation;
36 with the Invariant Sections being ``GNU Free Documentation License'',
37 with the Front-Cover Texts being ``GNAT Reference Manual'', and with
38 no Back-Cover Texts. A copy of the license is included in the section
39 entitled ``GNU Free Documentation License''.
44 @title GNAT Reference Manual
45 @subtitle GNAT, The GNU Ada 95 Compiler
46 @subtitle GCC version @value{version-GCC}
47 @author Ada Core Technologies, Inc.
50 @vskip 0pt plus 1filll
57 @node Top, About This Guide, (dir), (dir)
58 @top GNAT Reference Manual
64 GNAT, The GNU Ada 95 Compiler@*
65 GCC version @value{version-GCC}@*
68 Ada Core Technologies, Inc.
72 * Implementation Defined Pragmas::
73 * Implementation Defined Attributes::
74 * Implementation Advice::
75 * Implementation Defined Characteristics::
76 * Intrinsic Subprograms::
77 * Representation Clauses and Pragmas::
78 * Standard Library Routines::
79 * The Implementation of Standard I/O::
81 * Interfacing to Other Languages::
82 * Specialized Needs Annexes::
83 * Implementation of Specific Ada Features::
84 * Project File Reference::
85 * GNU Free Documentation License::
88 --- The Detailed Node Listing ---
92 * What This Reference Manual Contains::
93 * Related Information::
95 Implementation Defined Pragmas
97 * Pragma Abort_Defer::
103 * Pragma C_Pass_By_Copy::
105 * Pragma Common_Object::
106 * Pragma Compile_Time_Warning::
107 * Pragma Complex_Representation::
108 * Pragma Component_Alignment::
109 * Pragma Convention_Identifier::
111 * Pragma CPP_Constructor::
112 * Pragma CPP_Virtual::
113 * Pragma CPP_Vtable::
115 * Pragma Elaboration_Checks::
117 * Pragma Export_Exception::
118 * Pragma Export_Function::
119 * Pragma Export_Object::
120 * Pragma Export_Procedure::
121 * Pragma Export_Value::
122 * Pragma Export_Valued_Procedure::
123 * Pragma Extend_System::
125 * Pragma External_Name_Casing::
126 * Pragma Finalize_Storage_Only::
127 * Pragma Float_Representation::
129 * Pragma Import_Exception::
130 * Pragma Import_Function::
131 * Pragma Import_Object::
132 * Pragma Import_Procedure::
133 * Pragma Import_Valued_Procedure::
134 * Pragma Initialize_Scalars::
135 * Pragma Inline_Always::
136 * Pragma Inline_Generic::
138 * Pragma Interface_Name::
139 * Pragma Interrupt_Handler::
140 * Pragma Interrupt_State::
141 * Pragma Keep_Names::
144 * Pragma Linker_Alias::
145 * Pragma Linker_Section::
146 * Pragma Long_Float::
147 * Pragma Machine_Attribute::
148 * Pragma Main_Storage::
150 * Pragma Normalize_Scalars::
151 * Pragma Obsolescent::
154 * Pragma Propagate_Exceptions::
155 * Pragma Psect_Object::
156 * Pragma Pure_Function::
158 * Pragma Restricted_Run_Time::
159 * Pragma Restriction_Warnings::
160 * Pragma Source_File_Name::
161 * Pragma Source_File_Name_Project::
162 * Pragma Source_Reference::
163 * Pragma Stream_Convert::
164 * Pragma Style_Checks::
166 * Pragma Suppress_All::
167 * Pragma Suppress_Exception_Locations::
168 * Pragma Suppress_Initialization::
171 * Pragma Task_Storage::
172 * Pragma Thread_Body::
173 * Pragma Time_Slice::
175 * Pragma Unchecked_Union::
176 * Pragma Unimplemented_Unit::
177 * Pragma Universal_Data::
178 * Pragma Unreferenced::
179 * Pragma Unreserve_All_Interrupts::
180 * Pragma Unsuppress::
181 * Pragma Use_VADS_Size::
182 * Pragma Validity_Checks::
185 * Pragma Weak_External::
187 Implementation Defined Attributes
197 * Default_Bit_Order::
205 * Has_Discriminants::
211 * Max_Interrupt_Priority::
213 * Maximum_Alignment::
217 * Passed_By_Reference::
228 * Unconstrained_Array::
229 * Universal_Literal_String::
230 * Unrestricted_Access::
236 The Implementation of Standard I/O
238 * Standard I/O Packages::
247 * Operations on C Streams::
248 * Interfacing to C Streams::
252 * Ada.Characters.Latin_9 (a-chlat9.ads)::
253 * Ada.Characters.Wide_Latin_1 (a-cwila1.ads)::
254 * Ada.Characters.Wide_Latin_9 (a-cwila9.ads)::
255 * Ada.Command_Line.Remove (a-colire.ads)::
256 * Ada.Command_Line.Environment (a-colien.ads)::
257 * Ada.Direct_IO.C_Streams (a-diocst.ads)::
258 * Ada.Exceptions.Is_Null_Occurrence (a-einuoc.ads)::
259 * Ada.Exceptions.Traceback (a-exctra.ads)::
260 * Ada.Sequential_IO.C_Streams (a-siocst.ads)::
261 * Ada.Streams.Stream_IO.C_Streams (a-ssicst.ads)::
262 * Ada.Strings.Unbounded.Text_IO (a-suteio.ads)::
263 * Ada.Strings.Wide_Unbounded.Wide_Text_IO (a-swuwti.ads)::
264 * Ada.Text_IO.C_Streams (a-tiocst.ads)::
265 * Ada.Wide_Text_IO.C_Streams (a-wtcstr.ads)::
266 * GNAT.Array_Split (g-arrspl.ads)::
267 * GNAT.AWK (g-awk.ads)::
268 * GNAT.Bounded_Buffers (g-boubuf.ads)::
269 * GNAT.Bounded_Mailboxes (g-boumai.ads)::
270 * GNAT.Bubble_Sort (g-bubsor.ads)::
271 * GNAT.Bubble_Sort_A (g-busora.ads)::
272 * GNAT.Bubble_Sort_G (g-busorg.ads)::
273 * GNAT.Calendar (g-calend.ads)::
274 * GNAT.Calendar.Time_IO (g-catiio.ads)::
275 * GNAT.Case_Util (g-casuti.ads)::
276 * GNAT.CGI (g-cgi.ads)::
277 * GNAT.CGI.Cookie (g-cgicoo.ads)::
278 * GNAT.CGI.Debug (g-cgideb.ads)::
279 * GNAT.Command_Line (g-comlin.ads)::
280 * GNAT.Compiler_Version (g-comver.ads)::
281 * GNAT.Ctrl_C (g-ctrl_c.ads)::
282 * GNAT.CRC32 (g-crc32.ads)::
283 * GNAT.Current_Exception (g-curexc.ads)::
284 * GNAT.Debug_Pools (g-debpoo.ads)::
285 * GNAT.Debug_Utilities (g-debuti.ads)::
286 * GNAT.Directory_Operations (g-dirope.ads)::
287 * GNAT.Dynamic_HTables (g-dynhta.ads)::
288 * GNAT.Dynamic_Tables (g-dyntab.ads)::
289 * GNAT.Exception_Actions (g-excact.ads)::
290 * GNAT.Exception_Traces (g-exctra.ads)::
291 * GNAT.Exceptions (g-except.ads)::
292 * GNAT.Expect (g-expect.ads)::
293 * GNAT.Float_Control (g-flocon.ads)::
294 * GNAT.Heap_Sort (g-heasor.ads)::
295 * GNAT.Heap_Sort_A (g-hesora.ads)::
296 * GNAT.Heap_Sort_G (g-hesorg.ads)::
297 * GNAT.HTable (g-htable.ads)::
298 * GNAT.IO (g-io.ads)::
299 * GNAT.IO_Aux (g-io_aux.ads)::
300 * GNAT.Lock_Files (g-locfil.ads)::
301 * GNAT.MD5 (g-md5.ads)::
302 * GNAT.Memory_Dump (g-memdum.ads)::
303 * GNAT.Most_Recent_Exception (g-moreex.ads)::
304 * GNAT.OS_Lib (g-os_lib.ads)::
305 * GNAT.Perfect_Hash.Generators (g-pehage.ads)::
306 * GNAT.Regexp (g-regexp.ads)::
307 * GNAT.Registry (g-regist.ads)::
308 * GNAT.Regpat (g-regpat.ads)::
309 * GNAT.Secondary_Stack_Info (g-sestin.ads)::
310 * GNAT.Semaphores (g-semaph.ads)::
311 * GNAT.Signals (g-signal.ads)::
312 * GNAT.Sockets (g-socket.ads)::
313 * GNAT.Source_Info (g-souinf.ads)::
314 * GNAT.Spell_Checker (g-speche.ads)::
315 * GNAT.Spitbol.Patterns (g-spipat.ads)::
316 * GNAT.Spitbol (g-spitbo.ads)::
317 * GNAT.Spitbol.Table_Boolean (g-sptabo.ads)::
318 * GNAT.Spitbol.Table_Integer (g-sptain.ads)::
319 * GNAT.Spitbol.Table_VString (g-sptavs.ads)::
320 * GNAT.Strings (g-string.ads)::
321 * GNAT.String_Split (g-strspl.ads)::
322 * GNAT.Table (g-table.ads)::
323 * GNAT.Task_Lock (g-tasloc.ads)::
324 * GNAT.Threads (g-thread.ads)::
325 * GNAT.Traceback (g-traceb.ads)::
326 * GNAT.Traceback.Symbolic (g-trasym.ads)::
327 * GNAT.Wide_String_Split (g-wistsp.ads)::
328 * Interfaces.C.Extensions (i-cexten.ads)::
329 * Interfaces.C.Streams (i-cstrea.ads)::
330 * Interfaces.CPP (i-cpp.ads)::
331 * Interfaces.Os2lib (i-os2lib.ads)::
332 * Interfaces.Os2lib.Errors (i-os2err.ads)::
333 * Interfaces.Os2lib.Synchronization (i-os2syn.ads)::
334 * Interfaces.Os2lib.Threads (i-os2thr.ads)::
335 * Interfaces.Packed_Decimal (i-pacdec.ads)::
336 * Interfaces.VxWorks (i-vxwork.ads)::
337 * Interfaces.VxWorks.IO (i-vxwoio.ads)::
338 * System.Address_Image (s-addima.ads)::
339 * System.Assertions (s-assert.ads)::
340 * System.Memory (s-memory.ads)::
341 * System.Partition_Interface (s-parint.ads)::
342 * System.Restrictions (s-restri.ads)::
343 * System.Rident (s-rident.ads)::
344 * System.Task_Info (s-tasinf.ads)::
345 * System.Wch_Cnv (s-wchcnv.ads)::
346 * System.Wch_Con (s-wchcon.ads)::
350 * Text_IO Stream Pointer Positioning::
351 * Text_IO Reading and Writing Non-Regular Files::
353 * Treating Text_IO Files as Streams::
354 * Text_IO Extensions::
355 * Text_IO Facilities for Unbounded Strings::
359 * Wide_Text_IO Stream Pointer Positioning::
360 * Wide_Text_IO Reading and Writing Non-Regular Files::
362 Interfacing to Other Languages
365 * Interfacing to C++::
366 * Interfacing to COBOL::
367 * Interfacing to Fortran::
368 * Interfacing to non-GNAT Ada code::
370 Specialized Needs Annexes
372 Implementation of Specific Ada Features
373 * Machine Code Insertions::
374 * GNAT Implementation of Tasking::
375 * GNAT Implementation of Shared Passive Packages::
376 * Code Generation for Array Aggregates::
378 Project File Reference
380 GNU Free Documentation License
387 @node About This Guide
388 @unnumbered About This Guide
391 This manual contains useful information in writing programs using the
392 GNAT compiler. It includes information on implementation dependent
393 characteristics of GNAT, including all the information required by Annex
396 Ada 95 is designed to be highly portable.
397 In general, a program will have the same effect even when compiled by
398 different compilers on different platforms.
399 However, since Ada 95 is designed to be used in a
400 wide variety of applications, it also contains a number of system
401 dependent features to be used in interfacing to the external world.
402 @cindex Implementation-dependent features
405 Note: Any program that makes use of implementation-dependent features
406 may be non-portable. You should follow good programming practice and
407 isolate and clearly document any sections of your program that make use
408 of these features in a non-portable manner.
411 * What This Reference Manual Contains::
413 * Related Information::
416 @node What This Reference Manual Contains
417 @unnumberedsec What This Reference Manual Contains
420 This reference manual contains the following chapters:
424 @ref{Implementation Defined Pragmas}, lists GNAT implementation-dependent
425 pragmas, which can be used to extend and enhance the functionality of the
429 @ref{Implementation Defined Attributes}, lists GNAT
430 implementation-dependent attributes which can be used to extend and
431 enhance the functionality of the compiler.
434 @ref{Implementation Advice}, provides information on generally
435 desirable behavior which are not requirements that all compilers must
436 follow since it cannot be provided on all systems, or which may be
437 undesirable on some systems.
440 @ref{Implementation Defined Characteristics}, provides a guide to
441 minimizing implementation dependent features.
444 @ref{Intrinsic Subprograms}, describes the intrinsic subprograms
445 implemented by GNAT, and how they can be imported into user
446 application programs.
449 @ref{Representation Clauses and Pragmas}, describes in detail the
450 way that GNAT represents data, and in particular the exact set
451 of representation clauses and pragmas that is accepted.
454 @ref{Standard Library Routines}, provides a listing of packages and a
455 brief description of the functionality that is provided by Ada's
456 extensive set of standard library routines as implemented by GNAT@.
459 @ref{The Implementation of Standard I/O}, details how the GNAT
460 implementation of the input-output facilities.
463 @ref{The GNAT Library}, is a catalog of packages that complement
464 the Ada predefined library.
467 @ref{Interfacing to Other Languages}, describes how programs
468 written in Ada using GNAT can be interfaced to other programming
471 @ref{Specialized Needs Annexes}, describes the GNAT implementation of all
472 of the specialized needs annexes.
475 @ref{Implementation of Specific Ada Features}, discusses issues related
476 to GNAT's implementation of machine code insertions, tasking, and several
480 @ref{Project File Reference}, presents the syntax and semantics
485 @cindex Ada 95 ISO/ANSI Standard
487 This reference manual assumes that you are familiar with Ada 95
488 language, as described in the International Standard
489 ANSI/ISO/IEC-8652:1995, Jan 1995.
492 @unnumberedsec Conventions
493 @cindex Conventions, typographical
494 @cindex Typographical conventions
497 Following are examples of the typographical and graphic conventions used
502 @code{Functions}, @code{utility program names}, @code{standard names},
509 @file{File Names}, @samp{button names}, and @samp{field names}.
518 [optional information or parameters]
521 Examples are described by text
523 and then shown this way.
528 Commands that are entered by the user are preceded in this manual by the
529 characters @samp{$ } (dollar sign followed by space). If your system uses this
530 sequence as a prompt, then the commands will appear exactly as you see them
531 in the manual. If your system uses some other prompt, then the command will
532 appear with the @samp{$} replaced by whatever prompt character you are using.
534 @node Related Information
535 @unnumberedsec Related Information
537 See the following documents for further information on GNAT:
541 @cite{GNAT User's Guide}, which provides information on how to use
542 the GNAT compiler system.
545 @cite{Ada 95 Reference Manual}, which contains all reference
546 material for the Ada 95 programming language.
549 @cite{Ada 95 Annotated Reference Manual}, which is an annotated version
550 of the standard reference manual cited above. The annotations describe
551 detailed aspects of the design decision, and in particular contain useful
552 sections on Ada 83 compatibility.
555 @cite{DEC Ada, Technical Overview and Comparison on DIGITAL Platforms},
556 which contains specific information on compatibility between GNAT and
560 @cite{DEC Ada, Language Reference Manual, part number AA-PYZAB-TK} which
561 describes in detail the pragmas and attributes provided by the DEC Ada 83
566 @node Implementation Defined Pragmas
567 @chapter Implementation Defined Pragmas
570 Ada 95 defines a set of pragmas that can be used to supply additional
571 information to the compiler. These language defined pragmas are
572 implemented in GNAT and work as described in the Ada 95 Reference
575 In addition, Ada 95 allows implementations to define additional pragmas
576 whose meaning is defined by the implementation. GNAT provides a number
577 of these implementation-dependent pragmas which can be used to extend
578 and enhance the functionality of the compiler. This section of the GNAT
579 Reference Manual describes these additional pragmas.
581 Note that any program using these pragmas may not be portable to other
582 compilers (although GNAT implements this set of pragmas on all
583 platforms). Therefore if portability to other compilers is an important
584 consideration, the use of these pragmas should be minimized.
587 * Pragma Abort_Defer::
593 * Pragma C_Pass_By_Copy::
595 * Pragma Common_Object::
596 * Pragma Compile_Time_Warning::
597 * Pragma Complex_Representation::
598 * Pragma Component_Alignment::
599 * Pragma Convention_Identifier::
601 * Pragma CPP_Constructor::
602 * Pragma CPP_Virtual::
603 * Pragma CPP_Vtable::
605 * Pragma Elaboration_Checks::
607 * Pragma Export_Exception::
608 * Pragma Export_Function::
609 * Pragma Export_Object::
610 * Pragma Export_Procedure::
611 * Pragma Export_Value::
612 * Pragma Export_Valued_Procedure::
613 * Pragma Extend_System::
615 * Pragma External_Name_Casing::
616 * Pragma Finalize_Storage_Only::
617 * Pragma Float_Representation::
619 * Pragma Import_Exception::
620 * Pragma Import_Function::
621 * Pragma Import_Object::
622 * Pragma Import_Procedure::
623 * Pragma Import_Valued_Procedure::
624 * Pragma Initialize_Scalars::
625 * Pragma Inline_Always::
626 * Pragma Inline_Generic::
628 * Pragma Interface_Name::
629 * Pragma Interrupt_Handler::
630 * Pragma Interrupt_State::
631 * Pragma Keep_Names::
634 * Pragma Linker_Alias::
635 * Pragma Linker_Section::
636 * Pragma Long_Float::
637 * Pragma Machine_Attribute::
638 * Pragma Main_Storage::
640 * Pragma Normalize_Scalars::
641 * Pragma Obsolescent::
644 * Pragma Propagate_Exceptions::
645 * Pragma Psect_Object::
646 * Pragma Pure_Function::
648 * Pragma Restricted_Run_Time::
649 * Pragma Restriction_Warnings::
650 * Pragma Source_File_Name::
651 * Pragma Source_File_Name_Project::
652 * Pragma Source_Reference::
653 * Pragma Stream_Convert::
654 * Pragma Style_Checks::
656 * Pragma Suppress_All::
657 * Pragma Suppress_Exception_Locations::
658 * Pragma Suppress_Initialization::
661 * Pragma Task_Storage::
662 * Pragma Thread_Body::
663 * Pragma Time_Slice::
665 * Pragma Unchecked_Union::
666 * Pragma Unimplemented_Unit::
667 * Pragma Universal_Data::
668 * Pragma Unreferenced::
669 * Pragma Unreserve_All_Interrupts::
670 * Pragma Unsuppress::
671 * Pragma Use_VADS_Size::
672 * Pragma Validity_Checks::
675 * Pragma Weak_External::
678 @node Pragma Abort_Defer
679 @unnumberedsec Pragma Abort_Defer
681 @cindex Deferring aborts
689 This pragma must appear at the start of the statement sequence of a
690 handled sequence of statements (right after the @code{begin}). It has
691 the effect of deferring aborts for the sequence of statements (but not
692 for the declarations or handlers, if any, associated with this statement
696 @unnumberedsec Pragma Ada_83
705 A configuration pragma that establishes Ada 83 mode for the unit to
706 which it applies, regardless of the mode set by the command line
707 switches. In Ada 83 mode, GNAT attempts to be as compatible with
708 the syntax and semantics of Ada 83, as defined in the original Ada
709 83 Reference Manual as possible. In particular, the new Ada 95
710 keywords are not recognized, optional package bodies are allowed,
711 and generics may name types with unknown discriminants without using
712 the @code{(<>)} notation. In addition, some but not all of the additional
713 restrictions of Ada 83 are enforced.
715 Ada 83 mode is intended for two purposes. Firstly, it allows existing
716 legacy Ada 83 code to be compiled and adapted to GNAT with less effort.
717 Secondly, it aids in keeping code backwards compatible with Ada 83.
718 However, there is no guarantee that code that is processed correctly
719 by GNAT in Ada 83 mode will in fact compile and execute with an Ada
720 83 compiler, since GNAT does not enforce all the additional checks
724 @unnumberedsec Pragma Ada_95
733 A configuration pragma that establishes Ada 95 mode for the unit to which
734 it applies, regardless of the mode set by the command line switches.
735 This mode is set automatically for the @code{Ada} and @code{System}
736 packages and their children, so you need not specify it in these
737 contexts. This pragma is useful when writing a reusable component that
738 itself uses Ada 95 features, but which is intended to be usable from
739 either Ada 83 or Ada 95 programs.
741 @node Pragma Annotate
742 @unnumberedsec Pragma Annotate
747 pragma Annotate (IDENTIFIER @{, ARG@});
749 ARG ::= NAME | EXPRESSION
753 This pragma is used to annotate programs. @var{identifier} identifies
754 the type of annotation. GNAT verifies this is an identifier, but does
755 not otherwise analyze it. The @var{arg} argument
756 can be either a string literal or an
757 expression. String literals are assumed to be of type
758 @code{Standard.String}. Names of entities are simply analyzed as entity
759 names. All other expressions are analyzed as expressions, and must be
762 The analyzed pragma is retained in the tree, but not otherwise processed
763 by any part of the GNAT compiler. This pragma is intended for use by
764 external tools, including ASIS@.
767 @unnumberedsec Pragma Assert
774 [, static_string_EXPRESSION]);
778 The effect of this pragma depends on whether the corresponding command
779 line switch is set to activate assertions. The pragma expands into code
780 equivalent to the following:
783 if assertions-enabled then
784 if not boolean_EXPRESSION then
785 System.Assertions.Raise_Assert_Failure
792 The string argument, if given, is the message that will be associated
793 with the exception occurrence if the exception is raised. If no second
794 argument is given, the default message is @samp{@var{file}:@var{nnn}},
795 where @var{file} is the name of the source file containing the assert,
796 and @var{nnn} is the line number of the assert. A pragma is not a
797 statement, so if a statement sequence contains nothing but a pragma
798 assert, then a null statement is required in addition, as in:
803 pragma Assert (K > 3, "Bad value for K");
809 Note that, as with the @code{if} statement to which it is equivalent, the
810 type of the expression is either @code{Standard.Boolean}, or any type derived
811 from this standard type.
813 If assertions are disabled (switch @code{-gnata} not used), then there
814 is no effect (and in particular, any side effects from the expression
815 are suppressed). More precisely it is not quite true that the pragma
816 has no effect, since the expression is analyzed, and may cause types
817 to be frozen if they are mentioned here for the first time.
819 If assertions are enabled, then the given expression is tested, and if
820 it is @code{False} then @code{System.Assertions.Raise_Assert_Failure} is called
821 which results in the raising of @code{Assert_Failure} with the given message.
823 If the boolean expression has side effects, these side effects will turn
824 on and off with the setting of the assertions mode, resulting in
825 assertions that have an effect on the program. You should generally
826 avoid side effects in the expression arguments of this pragma. However,
827 the expressions are analyzed for semantic correctness whether or not
828 assertions are enabled, so turning assertions on and off cannot affect
829 the legality of a program.
831 @node Pragma Ast_Entry
832 @unnumberedsec Pragma Ast_Entry
838 pragma AST_Entry (entry_IDENTIFIER);
842 This pragma is implemented only in the OpenVMS implementation of GNAT@. The
843 argument is the simple name of a single entry; at most one @code{AST_Entry}
844 pragma is allowed for any given entry. This pragma must be used in
845 conjunction with the @code{AST_Entry} attribute, and is only allowed after
846 the entry declaration and in the same task type specification or single task
847 as the entry to which it applies. This pragma specifies that the given entry
848 may be used to handle an OpenVMS asynchronous system trap (@code{AST})
849 resulting from an OpenVMS system service call. The pragma does not affect
850 normal use of the entry. For further details on this pragma, see the
851 DEC Ada Language Reference Manual, section 9.12a.
853 @node Pragma C_Pass_By_Copy
854 @unnumberedsec Pragma C_Pass_By_Copy
855 @cindex Passing by copy
856 @findex C_Pass_By_Copy
860 pragma C_Pass_By_Copy
861 ([Max_Size =>] static_integer_EXPRESSION);
865 Normally the default mechanism for passing C convention records to C
866 convention subprograms is to pass them by reference, as suggested by RM
867 B.3(69). Use the configuration pragma @code{C_Pass_By_Copy} to change
868 this default, by requiring that record formal parameters be passed by
869 copy if all of the following conditions are met:
873 The size of the record type does not exceed@*@var{static_integer_expression}.
875 The record type has @code{Convention C}.
877 The formal parameter has this record type, and the subprogram has a
878 foreign (non-Ada) convention.
882 If these conditions are met the argument is passed by copy, i.e.@: in a
883 manner consistent with what C expects if the corresponding formal in the
884 C prototype is a struct (rather than a pointer to a struct).
886 You can also pass records by copy by specifying the convention
887 @code{C_Pass_By_Copy} for the record type, or by using the extended
888 @code{Import} and @code{Export} pragmas, which allow specification of
889 passing mechanisms on a parameter by parameter basis.
892 @unnumberedsec Pragma Comment
898 pragma Comment (static_string_EXPRESSION);
902 This is almost identical in effect to pragma @code{Ident}. It allows the
903 placement of a comment into the object file and hence into the
904 executable file if the operating system permits such usage. The
905 difference is that @code{Comment}, unlike @code{Ident}, has
906 no limitations on placement of the pragma (it can be placed
907 anywhere in the main source unit), and if more than one pragma
908 is used, all comments are retained.
910 @node Pragma Common_Object
911 @unnumberedsec Pragma Common_Object
912 @findex Common_Object
917 pragma Common_Object (
918 [Internal =>] LOCAL_NAME,
919 [, [External =>] EXTERNAL_SYMBOL]
920 [, [Size =>] EXTERNAL_SYMBOL] );
924 | static_string_EXPRESSION
928 This pragma enables the shared use of variables stored in overlaid
929 linker areas corresponding to the use of @code{COMMON}
930 in Fortran. The single
931 object @var{local_name} is assigned to the area designated by
932 the @var{External} argument.
933 You may define a record to correspond to a series
934 of fields. The @var{size} argument
935 is syntax checked in GNAT, but otherwise ignored.
937 @code{Common_Object} is not supported on all platforms. If no
938 support is available, then the code generator will issue a message
939 indicating that the necessary attribute for implementation of this
940 pragma is not available.
942 @node Pragma Compile_Time_Warning
943 @unnumberedsec Pragma Compile_Time_Warning
944 @findex Compile_Time_Warning
949 pragma Compile_Time_Warning
950 (boolean_EXPRESSION, static_string_EXPRESSION);
954 This pragma can be used to generate additional compile time warnings. It
955 is particularly useful in generics, where warnings can be issued for
956 specific problematic instantiations. The first parameter is a boolean
957 expression. The pragma is effective only if the value of this expression
958 is known at compile time, and has the value True. The set of expressions
959 whose values are known at compile time includes all static boolean
960 expressions, and also other values which the compiler can determine
961 at compile time (e.g. the size of a record type set by an explicit
962 size representation clause, or the value of a variable which was
963 initialized to a constant and is known not to have been modified).
964 If these conditions are met, a warning message is generated using
965 the value given as the second argument. This string value may contain
966 embedded ASCII.LF characters to break the message into multiple lines.
968 @node Pragma Complex_Representation
969 @unnumberedsec Pragma Complex_Representation
970 @findex Complex_Representation
975 pragma Complex_Representation
976 ([Entity =>] LOCAL_NAME);
980 The @var{Entity} argument must be the name of a record type which has
981 two fields of the same floating-point type. The effect of this pragma is
982 to force gcc to use the special internal complex representation form for
983 this record, which may be more efficient. Note that this may result in
984 the code for this type not conforming to standard ABI (application
985 binary interface) requirements for the handling of record types. For
986 example, in some environments, there is a requirement for passing
987 records by pointer, and the use of this pragma may result in passing
988 this type in floating-point registers.
990 @node Pragma Component_Alignment
991 @unnumberedsec Pragma Component_Alignment
992 @cindex Alignments of components
993 @findex Component_Alignment
998 pragma Component_Alignment (
999 [Form =>] ALIGNMENT_CHOICE
1000 [, [Name =>] type_LOCAL_NAME]);
1002 ALIGNMENT_CHOICE ::=
1010 Specifies the alignment of components in array or record types.
1011 The meaning of the @var{Form} argument is as follows:
1014 @findex Component_Size
1015 @item Component_Size
1016 Aligns scalar components and subcomponents of the array or record type
1017 on boundaries appropriate to their inherent size (naturally
1018 aligned). For example, 1-byte components are aligned on byte boundaries,
1019 2-byte integer components are aligned on 2-byte boundaries, 4-byte
1020 integer components are aligned on 4-byte boundaries and so on. These
1021 alignment rules correspond to the normal rules for C compilers on all
1022 machines except the VAX@.
1024 @findex Component_Size_4
1025 @item Component_Size_4
1026 Naturally aligns components with a size of four or fewer
1027 bytes. Components that are larger than 4 bytes are placed on the next
1030 @findex Storage_Unit
1032 Specifies that array or record components are byte aligned, i.e.@:
1033 aligned on boundaries determined by the value of the constant
1034 @code{System.Storage_Unit}.
1038 Specifies that array or record components are aligned on default
1039 boundaries, appropriate to the underlying hardware or operating system or
1040 both. For OpenVMS VAX systems, the @code{Default} choice is the same as
1041 the @code{Storage_Unit} choice (byte alignment). For all other systems,
1042 the @code{Default} choice is the same as @code{Component_Size} (natural
1047 If the @code{Name} parameter is present, @var{type_local_name} must
1048 refer to a local record or array type, and the specified alignment
1049 choice applies to the specified type. The use of
1050 @code{Component_Alignment} together with a pragma @code{Pack} causes the
1051 @code{Component_Alignment} pragma to be ignored. The use of
1052 @code{Component_Alignment} together with a record representation clause
1053 is only effective for fields not specified by the representation clause.
1055 If the @code{Name} parameter is absent, the pragma can be used as either
1056 a configuration pragma, in which case it applies to one or more units in
1057 accordance with the normal rules for configuration pragmas, or it can be
1058 used within a declarative part, in which case it applies to types that
1059 are declared within this declarative part, or within any nested scope
1060 within this declarative part. In either case it specifies the alignment
1061 to be applied to any record or array type which has otherwise standard
1064 If the alignment for a record or array type is not specified (using
1065 pragma @code{Pack}, pragma @code{Component_Alignment}, or a record rep
1066 clause), the GNAT uses the default alignment as described previously.
1068 @node Pragma Convention_Identifier
1069 @unnumberedsec Pragma Convention_Identifier
1070 @findex Convention_Identifier
1071 @cindex Conventions, synonyms
1075 @smallexample @c ada
1076 pragma Convention_Identifier (
1077 [Name =>] IDENTIFIER,
1078 [Convention =>] convention_IDENTIFIER);
1082 This pragma provides a mechanism for supplying synonyms for existing
1083 convention identifiers. The @code{Name} identifier can subsequently
1084 be used as a synonym for the given convention in other pragmas (including
1085 for example pragma @code{Import} or another @code{Convention_Identifier}
1086 pragma). As an example of the use of this, suppose you had legacy code
1087 which used Fortran77 as the identifier for Fortran. Then the pragma:
1089 @smallexample @c ada
1090 pragma Convention_Identifier (Fortran77, Fortran);
1094 would allow the use of the convention identifier @code{Fortran77} in
1095 subsequent code, avoiding the need to modify the sources. As another
1096 example, you could use this to parametrize convention requirements
1097 according to systems. Suppose you needed to use @code{Stdcall} on
1098 windows systems, and @code{C} on some other system, then you could
1099 define a convention identifier @code{Library} and use a single
1100 @code{Convention_Identifier} pragma to specify which convention
1101 would be used system-wide.
1103 @node Pragma CPP_Class
1104 @unnumberedsec Pragma CPP_Class
1106 @cindex Interfacing with C++
1110 @smallexample @c ada
1111 pragma CPP_Class ([Entity =>] LOCAL_NAME);
1115 The argument denotes an entity in the current declarative region
1116 that is declared as a tagged or untagged record type. It indicates that
1117 the type corresponds to an externally declared C++ class type, and is to
1118 be laid out the same way that C++ would lay out the type.
1120 If (and only if) the type is tagged, at least one component in the
1121 record must be of type @code{Interfaces.CPP.Vtable_Ptr}, corresponding
1122 to the C++ Vtable (or Vtables in the case of multiple inheritance) used
1125 Types for which @code{CPP_Class} is specified do not have assignment or
1126 equality operators defined (such operations can be imported or declared
1127 as subprograms as required). Initialization is allowed only by
1128 constructor functions (see pragma @code{CPP_Constructor}).
1130 Pragma @code{CPP_Class} is intended primarily for automatic generation
1131 using an automatic binding generator tool.
1132 See @ref{Interfacing to C++} for related information.
1134 @node Pragma CPP_Constructor
1135 @unnumberedsec Pragma CPP_Constructor
1136 @cindex Interfacing with C++
1137 @findex CPP_Constructor
1141 @smallexample @c ada
1142 pragma CPP_Constructor ([Entity =>] LOCAL_NAME);
1146 This pragma identifies an imported function (imported in the usual way
1147 with pragma @code{Import}) as corresponding to a C++
1148 constructor. The argument is a name that must have been
1149 previously mentioned in a pragma @code{Import}
1150 with @code{Convention} = @code{CPP}, and must be of one of the following
1155 @code{function @var{Fname} return @var{T}'Class}
1158 @code{function @var{Fname} (@dots{}) return @var{T}'Class}
1162 where @var{T} is a tagged type to which the pragma @code{CPP_Class} applies.
1164 The first form is the default constructor, used when an object of type
1165 @var{T} is created on the Ada side with no explicit constructor. Other
1166 constructors (including the copy constructor, which is simply a special
1167 case of the second form in which the one and only argument is of type
1168 @var{T}), can only appear in two contexts:
1172 On the right side of an initialization of an object of type @var{T}.
1174 In an extension aggregate for an object of a type derived from @var{T}.
1178 Although the constructor is described as a function that returns a value
1179 on the Ada side, it is typically a procedure with an extra implicit
1180 argument (the object being initialized) at the implementation
1181 level. GNAT issues the appropriate call, whatever it is, to get the
1182 object properly initialized.
1184 In the case of derived objects, you may use one of two possible forms
1185 for declaring and creating an object:
1188 @item @code{New_Object : Derived_T}
1189 @item @code{New_Object : Derived_T := (@var{constructor-call with} @dots{})}
1193 In the first case the default constructor is called and extension fields
1194 if any are initialized according to the default initialization
1195 expressions in the Ada declaration. In the second case, the given
1196 constructor is called and the extension aggregate indicates the explicit
1197 values of the extension fields.
1199 If no constructors are imported, it is impossible to create any objects
1200 on the Ada side. If no default constructor is imported, only the
1201 initialization forms using an explicit call to a constructor are
1204 Pragma @code{CPP_Constructor} is intended primarily for automatic generation
1205 using an automatic binding generator tool.
1206 See @ref{Interfacing to C++} for more related information.
1208 @node Pragma CPP_Virtual
1209 @unnumberedsec Pragma CPP_Virtual
1210 @cindex Interfacing to C++
1215 @smallexample @c ada
1218 [, [Vtable_Ptr =>] vtable_ENTITY,]
1219 [, [Position =>] static_integer_EXPRESSION]);
1223 This pragma serves the same function as pragma @code{Import} in that
1224 case of a virtual function imported from C++. The @var{Entity} argument
1226 primitive subprogram of a tagged type to which pragma @code{CPP_Class}
1227 applies. The @var{Vtable_Ptr} argument specifies
1228 the Vtable_Ptr component which contains the
1229 entry for this virtual function. The @var{Position} argument
1230 is the sequential number
1231 counting virtual functions for this Vtable starting at 1.
1233 The @code{Vtable_Ptr} and @code{Position} arguments may be omitted if
1234 there is one Vtable_Ptr present (single inheritance case) and all
1235 virtual functions are imported. In that case the compiler can deduce both
1238 No @code{External_Name} or @code{Link_Name} arguments are required for a
1239 virtual function, since it is always accessed indirectly via the
1240 appropriate Vtable entry.
1242 Pragma @code{CPP_Virtual} is intended primarily for automatic generation
1243 using an automatic binding generator tool.
1244 See @ref{Interfacing to C++} for related information.
1246 @node Pragma CPP_Vtable
1247 @unnumberedsec Pragma CPP_Vtable
1248 @cindex Interfacing with C++
1253 @smallexample @c ada
1256 [Vtable_Ptr =>] vtable_ENTITY,
1257 [Entry_Count =>] static_integer_EXPRESSION);
1261 Given a record to which the pragma @code{CPP_Class} applies,
1262 this pragma can be specified for each component of type
1263 @code{CPP.Interfaces.Vtable_Ptr}.
1264 @var{Entity} is the tagged type, @var{Vtable_Ptr}
1265 is the record field of type @code{Vtable_Ptr}, and @var{Entry_Count} is
1266 the number of virtual functions on the C++ side. Not all of these
1267 functions need to be imported on the Ada side.
1269 You may omit the @code{CPP_Vtable} pragma if there is only one
1270 @code{Vtable_Ptr} component in the record and all virtual functions are
1271 imported on the Ada side (the default value for the entry count in this
1272 case is simply the total number of virtual functions).
1274 Pragma @code{CPP_Vtable} is intended primarily for automatic generation
1275 using an automatic binding generator tool.
1276 See @ref{Interfacing to C++} for related information.
1279 @unnumberedsec Pragma Debug
1284 @smallexample @c ada
1285 pragma Debug (PROCEDURE_CALL_WITHOUT_SEMICOLON);
1287 PROCEDURE_CALL_WITHOUT_SEMICOLON ::=
1289 | PROCEDURE_PREFIX ACTUAL_PARAMETER_PART
1293 The argument has the syntactic form of an expression, meeting the
1294 syntactic requirements for pragmas.
1296 If assertions are not enabled on the command line, this pragma has no
1297 effect. If asserts are enabled, the semantics of the pragma is exactly
1298 equivalent to the procedure call statement corresponding to the argument
1299 with a terminating semicolon. Pragmas are permitted in sequences of
1300 declarations, so you can use pragma @code{Debug} to intersperse calls to
1301 debug procedures in the middle of declarations.
1303 @node Pragma Elaboration_Checks
1304 @unnumberedsec Pragma Elaboration_Checks
1305 @cindex Elaboration control
1306 @findex Elaboration_Checks
1310 @smallexample @c ada
1311 pragma Elaboration_Checks (RM | Static);
1315 This is a configuration pragma that provides control over the
1316 elaboration model used by the compilation affected by the
1317 pragma. If the parameter is RM, then the dynamic elaboration
1318 model described in the Ada Reference Manual is used, as though
1319 the @code{-gnatE} switch had been specified on the command
1320 line. If the parameter is Static, then the default GNAT static
1321 model is used. This configuration pragma overrides the setting
1322 of the command line. For full details on the elaboration models
1323 used by the GNAT compiler, see section ``Elaboration Order
1324 Handling in GNAT'' in the @cite{GNAT User's Guide}.
1326 @node Pragma Eliminate
1327 @unnumberedsec Pragma Eliminate
1328 @cindex Elimination of unused subprograms
1333 @smallexample @c ada
1335 [Unit_Name =>] IDENTIFIER |
1336 SELECTED_COMPONENT);
1339 [Unit_Name =>] IDENTIFIER |
1341 [Entity =>] IDENTIFIER |
1342 SELECTED_COMPONENT |
1344 [,[Parameter_Types =>] PARAMETER_TYPES]
1345 [,[Result_Type =>] result_SUBTYPE_NAME]
1346 [,[Homonym_Number =>] INTEGER_LITERAL]);
1348 PARAMETER_TYPES ::= (SUBTYPE_NAME @{, SUBTYPE_NAME@})
1349 SUBTYPE_NAME ::= STRING_LITERAL
1353 This pragma indicates that the given entity is not used outside the
1354 compilation unit it is defined in. The entity may be either a subprogram
1357 If the entity to be eliminated is a library level subprogram, then
1358 the first form of pragma @code{Eliminate} is used with only a single argument.
1359 In this form, the @code{Unit_Name} argument specifies the name of the
1360 library level unit to be eliminated.
1362 In all other cases, both @code{Unit_Name} and @code{Entity} arguments
1363 are required. If item is an entity of a library package, then the first
1364 argument specifies the unit name, and the second argument specifies
1365 the particular entity. If the second argument is in string form, it must
1366 correspond to the internal manner in which GNAT stores entity names (see
1367 compilation unit Namet in the compiler sources for details).
1369 The remaining parameters are optionally used to distinguish
1370 between overloaded subprograms. There are two ways of doing this.
1372 Use @code{Parameter_Types} and @code{Result_Type} to specify the
1373 profile of the subprogram to be eliminated in a manner similar to that
1375 the extended @code{Import} and @code{Export} pragmas, except that the
1376 subtype names are always given as string literals, again corresponding
1377 to the internal manner in which GNAT stores entity names.
1379 Alternatively, the @code{Homonym_Number} parameter is used to specify
1380 which overloaded alternative is to be eliminated. A value of 1 indicates
1381 the first subprogram (in lexical order), 2 indicates the second etc.
1383 The effect of the pragma is to allow the compiler to eliminate
1384 the code or data associated with the named entity. Any reference to
1385 an eliminated entity outside the compilation unit it is defined in,
1386 causes a compile time or link time error.
1388 The parameters of this pragma may be given in any order, as long as
1389 the usual rules for use of named parameters and position parameters
1392 The intention of pragma @code{Eliminate} is to allow a program to be compiled
1393 in a system independent manner, with unused entities eliminated, without
1394 the requirement of modifying the source text. Normally the required set
1395 of @code{Eliminate} pragmas is constructed automatically using the gnatelim
1396 tool. Elimination of unused entities local to a compilation unit is
1397 automatic, without requiring the use of pragma @code{Eliminate}.
1399 Note that the reason this pragma takes string literals where names might
1400 be expected is that a pragma @code{Eliminate} can appear in a context where the
1401 relevant names are not visible.
1403 @node Pragma Export_Exception
1404 @unnumberedsec Pragma Export_Exception
1406 @findex Export_Exception
1410 @smallexample @c ada
1411 pragma Export_Exception (
1412 [Internal =>] LOCAL_NAME,
1413 [, [External =>] EXTERNAL_SYMBOL,]
1414 [, [Form =>] Ada | VMS]
1415 [, [Code =>] static_integer_EXPRESSION]);
1419 | static_string_EXPRESSION
1423 This pragma is implemented only in the OpenVMS implementation of GNAT@. It
1424 causes the specified exception to be propagated outside of the Ada program,
1425 so that it can be handled by programs written in other OpenVMS languages.
1426 This pragma establishes an external name for an Ada exception and makes the
1427 name available to the OpenVMS Linker as a global symbol. For further details
1428 on this pragma, see the
1429 DEC Ada Language Reference Manual, section 13.9a3.2.
1431 @node Pragma Export_Function
1432 @unnumberedsec Pragma Export_Function
1433 @cindex Argument passing mechanisms
1434 @findex Export_Function
1439 @smallexample @c ada
1440 pragma Export_Function (
1441 [Internal =>] LOCAL_NAME,
1442 [, [External =>] EXTERNAL_SYMBOL]
1443 [, [Parameter_Types =>] PARAMETER_TYPES]
1444 [, [Result_Type =>] result_SUBTYPE_MARK]
1445 [, [Mechanism =>] MECHANISM]
1446 [, [Result_Mechanism =>] MECHANISM_NAME]);
1450 | static_string_EXPRESSION
1455 | TYPE_DESIGNATOR @{, TYPE_DESIGNATOR@}
1459 | subtype_Name ' Access
1463 | (MECHANISM_ASSOCIATION @{, MECHANISM_ASSOCIATION@})
1465 MECHANISM_ASSOCIATION ::=
1466 [formal_parameter_NAME =>] MECHANISM_NAME
1474 Use this pragma to make a function externally callable and optionally
1475 provide information on mechanisms to be used for passing parameter and
1476 result values. We recommend, for the purposes of improving portability,
1477 this pragma always be used in conjunction with a separate pragma
1478 @code{Export}, which must precede the pragma @code{Export_Function}.
1479 GNAT does not require a separate pragma @code{Export}, but if none is
1480 present, @code{Convention Ada} is assumed, which is usually
1481 not what is wanted, so it is usually appropriate to use this
1482 pragma in conjunction with a @code{Export} or @code{Convention}
1483 pragma that specifies the desired foreign convention.
1484 Pragma @code{Export_Function}
1485 (and @code{Export}, if present) must appear in the same declarative
1486 region as the function to which they apply.
1488 @var{internal_name} must uniquely designate the function to which the
1489 pragma applies. If more than one function name exists of this name in
1490 the declarative part you must use the @code{Parameter_Types} and
1491 @code{Result_Type} parameters is mandatory to achieve the required
1492 unique designation. @var{subtype_ mark}s in these parameters must
1493 exactly match the subtypes in the corresponding function specification,
1494 using positional notation to match parameters with subtype marks.
1495 The form with an @code{'Access} attribute can be used to match an
1496 anonymous access parameter.
1499 @cindex Passing by descriptor
1500 Note that passing by descriptor is not supported, even on the OpenVMS
1503 @cindex Suppressing external name
1504 Special treatment is given if the EXTERNAL is an explicit null
1505 string or a static string expressions that evaluates to the null
1506 string. In this case, no external name is generated. This form
1507 still allows the specification of parameter mechanisms.
1509 @node Pragma Export_Object
1510 @unnumberedsec Pragma Export_Object
1511 @findex Export_Object
1515 @smallexample @c ada
1516 pragma Export_Object
1517 [Internal =>] LOCAL_NAME,
1518 [, [External =>] EXTERNAL_SYMBOL]
1519 [, [Size =>] EXTERNAL_SYMBOL]
1523 | static_string_EXPRESSION
1527 This pragma designates an object as exported, and apart from the
1528 extended rules for external symbols, is identical in effect to the use of
1529 the normal @code{Export} pragma applied to an object. You may use a
1530 separate Export pragma (and you probably should from the point of view
1531 of portability), but it is not required. @var{Size} is syntax checked,
1532 but otherwise ignored by GNAT@.
1534 @node Pragma Export_Procedure
1535 @unnumberedsec Pragma Export_Procedure
1536 @findex Export_Procedure
1540 @smallexample @c ada
1541 pragma Export_Procedure (
1542 [Internal =>] LOCAL_NAME
1543 [, [External =>] EXTERNAL_SYMBOL]
1544 [, [Parameter_Types =>] PARAMETER_TYPES]
1545 [, [Mechanism =>] MECHANISM]);
1549 | static_string_EXPRESSION
1554 | TYPE_DESIGNATOR @{, TYPE_DESIGNATOR@}
1558 | subtype_Name ' Access
1562 | (MECHANISM_ASSOCIATION @{, MECHANISM_ASSOCIATION@})
1564 MECHANISM_ASSOCIATION ::=
1565 [formal_parameter_NAME =>] MECHANISM_NAME
1573 This pragma is identical to @code{Export_Function} except that it
1574 applies to a procedure rather than a function and the parameters
1575 @code{Result_Type} and @code{Result_Mechanism} are not permitted.
1576 GNAT does not require a separate pragma @code{Export}, but if none is
1577 present, @code{Convention Ada} is assumed, which is usually
1578 not what is wanted, so it is usually appropriate to use this
1579 pragma in conjunction with a @code{Export} or @code{Convention}
1580 pragma that specifies the desired foreign convention.
1583 @cindex Passing by descriptor
1584 Note that passing by descriptor is not supported, even on the OpenVMS
1587 @cindex Suppressing external name
1588 Special treatment is given if the EXTERNAL is an explicit null
1589 string or a static string expressions that evaluates to the null
1590 string. In this case, no external name is generated. This form
1591 still allows the specification of parameter mechanisms.
1593 @node Pragma Export_Value
1594 @unnumberedsec Pragma Export_Value
1595 @findex Export_Value
1599 @smallexample @c ada
1600 pragma Export_Value (
1601 [Value =>] static_integer_EXPRESSION,
1602 [Link_Name =>] static_string_EXPRESSION);
1606 This pragma serves to export a static integer value for external use.
1607 The first argument specifies the value to be exported. The Link_Name
1608 argument specifies the symbolic name to be associated with the integer
1609 value. This pragma is useful for defining a named static value in Ada
1610 that can be referenced in assembly language units to be linked with
1611 the application. This pragma is currently supported only for the
1612 AAMP target and is ignored for other targets.
1614 @node Pragma Export_Valued_Procedure
1615 @unnumberedsec Pragma Export_Valued_Procedure
1616 @findex Export_Valued_Procedure
1620 @smallexample @c ada
1621 pragma Export_Valued_Procedure (
1622 [Internal =>] LOCAL_NAME
1623 [, [External =>] EXTERNAL_SYMBOL]
1624 [, [Parameter_Types =>] PARAMETER_TYPES]
1625 [, [Mechanism =>] MECHANISM]);
1629 | static_string_EXPRESSION
1634 | TYPE_DESIGNATOR @{, TYPE_DESIGNATOR@}
1638 | subtype_Name ' Access
1642 | (MECHANISM_ASSOCIATION @{, MECHANISM_ASSOCIATION@})
1644 MECHANISM_ASSOCIATION ::=
1645 [formal_parameter_NAME =>] MECHANISM_NAME
1653 This pragma is identical to @code{Export_Procedure} except that the
1654 first parameter of @var{local_name}, which must be present, must be of
1655 mode @code{OUT}, and externally the subprogram is treated as a function
1656 with this parameter as the result of the function. GNAT provides for
1657 this capability to allow the use of @code{OUT} and @code{IN OUT}
1658 parameters in interfacing to external functions (which are not permitted
1660 GNAT does not require a separate pragma @code{Export}, but if none is
1661 present, @code{Convention Ada} is assumed, which is almost certainly
1662 not what is wanted since the whole point of this pragma is to interface
1663 with foreign language functions, so it is usually appropriate to use this
1664 pragma in conjunction with a @code{Export} or @code{Convention}
1665 pragma that specifies the desired foreign convention.
1668 @cindex Passing by descriptor
1669 Note that passing by descriptor is not supported, even on the OpenVMS
1672 @cindex Suppressing external name
1673 Special treatment is given if the EXTERNAL is an explicit null
1674 string or a static string expressions that evaluates to the null
1675 string. In this case, no external name is generated. This form
1676 still allows the specification of parameter mechanisms.
1678 @node Pragma Extend_System
1679 @unnumberedsec Pragma Extend_System
1680 @cindex @code{system}, extending
1682 @findex Extend_System
1686 @smallexample @c ada
1687 pragma Extend_System ([Name =>] IDENTIFIER);
1691 This pragma is used to provide backwards compatibility with other
1692 implementations that extend the facilities of package @code{System}. In
1693 GNAT, @code{System} contains only the definitions that are present in
1694 the Ada 95 RM@. However, other implementations, notably the DEC Ada 83
1695 implementation, provide many extensions to package @code{System}.
1697 For each such implementation accommodated by this pragma, GNAT provides a
1698 package @code{Aux_@var{xxx}}, e.g.@: @code{Aux_DEC} for the DEC Ada 83
1699 implementation, which provides the required additional definitions. You
1700 can use this package in two ways. You can @code{with} it in the normal
1701 way and access entities either by selection or using a @code{use}
1702 clause. In this case no special processing is required.
1704 However, if existing code contains references such as
1705 @code{System.@var{xxx}} where @var{xxx} is an entity in the extended
1706 definitions provided in package @code{System}, you may use this pragma
1707 to extend visibility in @code{System} in a non-standard way that
1708 provides greater compatibility with the existing code. Pragma
1709 @code{Extend_System} is a configuration pragma whose single argument is
1710 the name of the package containing the extended definition
1711 (e.g.@: @code{Aux_DEC} for the DEC Ada case). A unit compiled under
1712 control of this pragma will be processed using special visibility
1713 processing that looks in package @code{System.Aux_@var{xxx}} where
1714 @code{Aux_@var{xxx}} is the pragma argument for any entity referenced in
1715 package @code{System}, but not found in package @code{System}.
1717 You can use this pragma either to access a predefined @code{System}
1718 extension supplied with the compiler, for example @code{Aux_DEC} or
1719 you can construct your own extension unit following the above
1720 definition. Note that such a package is a child of @code{System}
1721 and thus is considered part of the implementation. To compile
1722 it you will have to use the appropriate switch for compiling
1723 system units. See the GNAT User's Guide for details.
1725 @node Pragma External
1726 @unnumberedsec Pragma External
1731 @smallexample @c ada
1733 [ Convention =>] convention_IDENTIFIER,
1734 [ Entity =>] local_NAME
1735 [, [External_Name =>] static_string_EXPRESSION ]
1736 [, [Link_Name =>] static_string_EXPRESSION ]);
1740 This pragma is identical in syntax and semantics to pragma
1741 @code{Export} as defined in the Ada Reference Manual. It is
1742 provided for compatibility with some Ada 83 compilers that
1743 used this pragma for exactly the same purposes as pragma
1744 @code{Export} before the latter was standardized.
1746 @node Pragma External_Name_Casing
1747 @unnumberedsec Pragma External_Name_Casing
1748 @cindex Dec Ada 83 casing compatibility
1749 @cindex External Names, casing
1750 @cindex Casing of External names
1751 @findex External_Name_Casing
1755 @smallexample @c ada
1756 pragma External_Name_Casing (
1757 Uppercase | Lowercase
1758 [, Uppercase | Lowercase | As_Is]);
1762 This pragma provides control over the casing of external names associated
1763 with Import and Export pragmas. There are two cases to consider:
1766 @item Implicit external names
1767 Implicit external names are derived from identifiers. The most common case
1768 arises when a standard Ada 95 Import or Export pragma is used with only two
1771 @smallexample @c ada
1772 pragma Import (C, C_Routine);
1776 Since Ada is a case insensitive language, the spelling of the identifier in
1777 the Ada source program does not provide any information on the desired
1778 casing of the external name, and so a convention is needed. In GNAT the
1779 default treatment is that such names are converted to all lower case
1780 letters. This corresponds to the normal C style in many environments.
1781 The first argument of pragma @code{External_Name_Casing} can be used to
1782 control this treatment. If @code{Uppercase} is specified, then the name
1783 will be forced to all uppercase letters. If @code{Lowercase} is specified,
1784 then the normal default of all lower case letters will be used.
1786 This same implicit treatment is also used in the case of extended DEC Ada 83
1787 compatible Import and Export pragmas where an external name is explicitly
1788 specified using an identifier rather than a string.
1790 @item Explicit external names
1791 Explicit external names are given as string literals. The most common case
1792 arises when a standard Ada 95 Import or Export pragma is used with three
1795 @smallexample @c ada
1796 pragma Import (C, C_Routine, "C_routine");
1800 In this case, the string literal normally provides the exact casing required
1801 for the external name. The second argument of pragma
1802 @code{External_Name_Casing} may be used to modify this behavior.
1803 If @code{Uppercase} is specified, then the name
1804 will be forced to all uppercase letters. If @code{Lowercase} is specified,
1805 then the name will be forced to all lowercase letters. A specification of
1806 @code{As_Is} provides the normal default behavior in which the casing is
1807 taken from the string provided.
1811 This pragma may appear anywhere that a pragma is valid. In particular, it
1812 can be used as a configuration pragma in the @file{gnat.adc} file, in which
1813 case it applies to all subsequent compilations, or it can be used as a program
1814 unit pragma, in which case it only applies to the current unit, or it can
1815 be used more locally to control individual Import/Export pragmas.
1817 It is primarily intended for use with OpenVMS systems, where many
1818 compilers convert all symbols to upper case by default. For interfacing to
1819 such compilers (e.g.@: the DEC C compiler), it may be convenient to use
1822 @smallexample @c ada
1823 pragma External_Name_Casing (Uppercase, Uppercase);
1827 to enforce the upper casing of all external symbols.
1829 @node Pragma Finalize_Storage_Only
1830 @unnumberedsec Pragma Finalize_Storage_Only
1831 @findex Finalize_Storage_Only
1835 @smallexample @c ada
1836 pragma Finalize_Storage_Only (first_subtype_LOCAL_NAME);
1840 This pragma allows the compiler not to emit a Finalize call for objects
1841 defined at the library level. This is mostly useful for types where
1842 finalization is only used to deal with storage reclamation since in most
1843 environments it is not necessary to reclaim memory just before terminating
1844 execution, hence the name.
1846 @node Pragma Float_Representation
1847 @unnumberedsec Pragma Float_Representation
1849 @findex Float_Representation
1853 @smallexample @c ada
1854 pragma Float_Representation (FLOAT_REP);
1856 FLOAT_REP ::= VAX_Float | IEEE_Float
1861 allows control over the internal representation chosen for the predefined
1862 floating point types declared in the packages @code{Standard} and
1863 @code{System}. On all systems other than OpenVMS, the argument must
1864 be @code{IEEE_Float} and the pragma has no effect. On OpenVMS, the
1865 argument may be @code{VAX_Float} to specify the use of the VAX float
1866 format for the floating-point types in Standard. This requires that
1867 the standard runtime libraries be recompiled. See the
1868 description of the @code{GNAT LIBRARY} command in the OpenVMS version
1869 of the GNAT Users Guide for details on the use of this command.
1872 @unnumberedsec Pragma Ident
1877 @smallexample @c ada
1878 pragma Ident (static_string_EXPRESSION);
1882 This pragma provides a string identification in the generated object file,
1883 if the system supports the concept of this kind of identification string.
1884 This pragma is allowed only in the outermost declarative part or
1885 declarative items of a compilation unit. If more than one @code{Ident}
1886 pragma is given, only the last one processed is effective.
1888 On OpenVMS systems, the effect of the pragma is identical to the effect of
1889 the DEC Ada 83 pragma of the same name. Note that in DEC Ada 83, the
1890 maximum allowed length is 31 characters, so if it is important to
1891 maintain compatibility with this compiler, you should obey this length
1894 @node Pragma Import_Exception
1895 @unnumberedsec Pragma Import_Exception
1897 @findex Import_Exception
1901 @smallexample @c ada
1902 pragma Import_Exception (
1903 [Internal =>] LOCAL_NAME,
1904 [, [External =>] EXTERNAL_SYMBOL,]
1905 [, [Form =>] Ada | VMS]
1906 [, [Code =>] static_integer_EXPRESSION]);
1910 | static_string_EXPRESSION
1914 This pragma is implemented only in the OpenVMS implementation of GNAT@.
1915 It allows OpenVMS conditions (for example, from OpenVMS system services or
1916 other OpenVMS languages) to be propagated to Ada programs as Ada exceptions.
1917 The pragma specifies that the exception associated with an exception
1918 declaration in an Ada program be defined externally (in non-Ada code).
1919 For further details on this pragma, see the
1920 DEC Ada Language Reference Manual, section 13.9a.3.1.
1922 @node Pragma Import_Function
1923 @unnumberedsec Pragma Import_Function
1924 @findex Import_Function
1928 @smallexample @c ada
1929 pragma Import_Function (
1930 [Internal =>] LOCAL_NAME,
1931 [, [External =>] EXTERNAL_SYMBOL]
1932 [, [Parameter_Types =>] PARAMETER_TYPES]
1933 [, [Result_Type =>] SUBTYPE_MARK]
1934 [, [Mechanism =>] MECHANISM]
1935 [, [Result_Mechanism =>] MECHANISM_NAME]
1936 [, [First_Optional_Parameter =>] IDENTIFIER]);
1940 | static_string_EXPRESSION
1944 | TYPE_DESIGNATOR @{, TYPE_DESIGNATOR@}
1948 | subtype_Name ' Access
1952 | (MECHANISM_ASSOCIATION @{, MECHANISM_ASSOCIATION@})
1954 MECHANISM_ASSOCIATION ::=
1955 [formal_parameter_NAME =>] MECHANISM_NAME
1960 | Descriptor [([Class =>] CLASS_NAME)]
1962 CLASS_NAME ::= ubs | ubsb | uba | s | sb | a | nca
1966 This pragma is used in conjunction with a pragma @code{Import} to
1967 specify additional information for an imported function. The pragma
1968 @code{Import} (or equivalent pragma @code{Interface}) must precede the
1969 @code{Import_Function} pragma and both must appear in the same
1970 declarative part as the function specification.
1972 The @var{Internal} argument must uniquely designate
1973 the function to which the
1974 pragma applies. If more than one function name exists of this name in
1975 the declarative part you must use the @code{Parameter_Types} and
1976 @var{Result_Type} parameters to achieve the required unique
1977 designation. Subtype marks in these parameters must exactly match the
1978 subtypes in the corresponding function specification, using positional
1979 notation to match parameters with subtype marks.
1980 The form with an @code{'Access} attribute can be used to match an
1981 anonymous access parameter.
1983 You may optionally use the @var{Mechanism} and @var{Result_Mechanism}
1984 parameters to specify passing mechanisms for the
1985 parameters and result. If you specify a single mechanism name, it
1986 applies to all parameters. Otherwise you may specify a mechanism on a
1987 parameter by parameter basis using either positional or named
1988 notation. If the mechanism is not specified, the default mechanism
1992 @cindex Passing by descriptor
1993 Passing by descriptor is supported only on the OpenVMS ports of GNAT@.
1995 @code{First_Optional_Parameter} applies only to OpenVMS ports of GNAT@.
1996 It specifies that the designated parameter and all following parameters
1997 are optional, meaning that they are not passed at the generated code
1998 level (this is distinct from the notion of optional parameters in Ada
1999 where the parameters are passed anyway with the designated optional
2000 parameters). All optional parameters must be of mode @code{IN} and have
2001 default parameter values that are either known at compile time
2002 expressions, or uses of the @code{'Null_Parameter} attribute.
2004 @node Pragma Import_Object
2005 @unnumberedsec Pragma Import_Object
2006 @findex Import_Object
2010 @smallexample @c ada
2011 pragma Import_Object
2012 [Internal =>] LOCAL_NAME,
2013 [, [External =>] EXTERNAL_SYMBOL],
2014 [, [Size =>] EXTERNAL_SYMBOL]);
2018 | static_string_EXPRESSION
2022 This pragma designates an object as imported, and apart from the
2023 extended rules for external symbols, is identical in effect to the use of
2024 the normal @code{Import} pragma applied to an object. Unlike the
2025 subprogram case, you need not use a separate @code{Import} pragma,
2026 although you may do so (and probably should do so from a portability
2027 point of view). @var{size} is syntax checked, but otherwise ignored by
2030 @node Pragma Import_Procedure
2031 @unnumberedsec Pragma Import_Procedure
2032 @findex Import_Procedure
2036 @smallexample @c ada
2037 pragma Import_Procedure (
2038 [Internal =>] LOCAL_NAME,
2039 [, [External =>] EXTERNAL_SYMBOL]
2040 [, [Parameter_Types =>] PARAMETER_TYPES]
2041 [, [Mechanism =>] MECHANISM]
2042 [, [First_Optional_Parameter =>] IDENTIFIER]);
2046 | static_string_EXPRESSION
2050 | TYPE_DESIGNATOR @{, TYPE_DESIGNATOR@}
2054 | subtype_Name ' Access
2058 | (MECHANISM_ASSOCIATION @{, MECHANISM_ASSOCIATION@})
2060 MECHANISM_ASSOCIATION ::=
2061 [formal_parameter_NAME =>] MECHANISM_NAME
2066 | Descriptor [([Class =>] CLASS_NAME)]
2068 CLASS_NAME ::= ubs | ubsb | uba | s | sb | a | nca
2072 This pragma is identical to @code{Import_Function} except that it
2073 applies to a procedure rather than a function and the parameters
2074 @code{Result_Type} and @code{Result_Mechanism} are not permitted.
2076 @node Pragma Import_Valued_Procedure
2077 @unnumberedsec Pragma Import_Valued_Procedure
2078 @findex Import_Valued_Procedure
2082 @smallexample @c ada
2083 pragma Import_Valued_Procedure (
2084 [Internal =>] LOCAL_NAME,
2085 [, [External =>] EXTERNAL_SYMBOL]
2086 [, [Parameter_Types =>] PARAMETER_TYPES]
2087 [, [Mechanism =>] MECHANISM]
2088 [, [First_Optional_Parameter =>] IDENTIFIER]);
2092 | static_string_EXPRESSION
2096 | TYPE_DESIGNATOR @{, TYPE_DESIGNATOR@}
2100 | subtype_Name ' Access
2104 | (MECHANISM_ASSOCIATION @{, MECHANISM_ASSOCIATION@})
2106 MECHANISM_ASSOCIATION ::=
2107 [formal_parameter_NAME =>] MECHANISM_NAME
2112 | Descriptor [([Class =>] CLASS_NAME)]
2114 CLASS_NAME ::= ubs | ubsb | uba | s | sb | a | nca
2118 This pragma is identical to @code{Import_Procedure} except that the
2119 first parameter of @var{local_name}, which must be present, must be of
2120 mode @code{OUT}, and externally the subprogram is treated as a function
2121 with this parameter as the result of the function. The purpose of this
2122 capability is to allow the use of @code{OUT} and @code{IN OUT}
2123 parameters in interfacing to external functions (which are not permitted
2124 in Ada functions). You may optionally use the @code{Mechanism}
2125 parameters to specify passing mechanisms for the parameters.
2126 If you specify a single mechanism name, it applies to all parameters.
2127 Otherwise you may specify a mechanism on a parameter by parameter
2128 basis using either positional or named notation. If the mechanism is not
2129 specified, the default mechanism is used.
2131 Note that it is important to use this pragma in conjunction with a separate
2132 pragma Import that specifies the desired convention, since otherwise the
2133 default convention is Ada, which is almost certainly not what is required.
2135 @node Pragma Initialize_Scalars
2136 @unnumberedsec Pragma Initialize_Scalars
2137 @findex Initialize_Scalars
2138 @cindex debugging with Initialize_Scalars
2142 @smallexample @c ada
2143 pragma Initialize_Scalars;
2147 This pragma is similar to @code{Normalize_Scalars} conceptually but has
2148 two important differences. First, there is no requirement for the pragma
2149 to be used uniformly in all units of a partition, in particular, it is fine
2150 to use this just for some or all of the application units of a partition,
2151 without needing to recompile the run-time library.
2153 In the case where some units are compiled with the pragma, and some without,
2154 then a declaration of a variable where the type is defined in package
2155 Standard or is locally declared will always be subject to initialization,
2156 as will any declaration of a scalar variable. For composite variables,
2157 whether the variable is initialized may also depend on whether the package
2158 in which the type of the variable is declared is compiled with the pragma.
2160 The other important difference is that there is control over the value used
2161 for initializing scalar objects. At bind time, you can select whether to
2162 initialize with invalid values (like Normalize_Scalars), or with high or
2163 low values, or with a specified bit pattern. See the users guide for binder
2164 options for specifying these cases.
2166 This means that you can compile a program, and then without having to
2167 recompile the program, you can run it with different values being used
2168 for initializing otherwise uninitialized values, to test if your program
2169 behavior depends on the choice. Of course the behavior should not change,
2170 and if it does, then most likely you have an erroneous reference to an
2171 uninitialized value.
2173 Note that pragma @code{Initialize_Scalars} is particularly useful in
2174 conjunction with the enhanced validity checking that is now provided
2175 in GNAT, which checks for invalid values under more conditions.
2176 Using this feature (see description of the @code{-gnatV} flag in the
2177 users guide) in conjunction with pragma @code{Initialize_Scalars}
2178 provides a powerful new tool to assist in the detection of problems
2179 caused by uninitialized variables.
2181 @node Pragma Inline_Always
2182 @unnumberedsec Pragma Inline_Always
2183 @findex Inline_Always
2187 @smallexample @c ada
2188 pragma Inline_Always (NAME [, NAME]);
2192 Similar to pragma @code{Inline} except that inlining is not subject to
2193 the use of option @code{-gnatn} and the inlining happens regardless of
2194 whether this option is used.
2196 @node Pragma Inline_Generic
2197 @unnumberedsec Pragma Inline_Generic
2198 @findex Inline_Generic
2202 @smallexample @c ada
2203 pragma Inline_Generic (generic_package_NAME);
2207 This is implemented for compatibility with DEC Ada 83 and is recognized,
2208 but otherwise ignored, by GNAT@. All generic instantiations are inlined
2209 by default when using GNAT@.
2211 @node Pragma Interface
2212 @unnumberedsec Pragma Interface
2217 @smallexample @c ada
2219 [Convention =>] convention_identifier,
2220 [Entity =>] local_name
2221 [, [External_Name =>] static_string_expression],
2222 [, [Link_Name =>] static_string_expression]);
2226 This pragma is identical in syntax and semantics to
2227 the standard Ada 95 pragma @code{Import}. It is provided for compatibility
2228 with Ada 83. The definition is upwards compatible both with pragma
2229 @code{Interface} as defined in the Ada 83 Reference Manual, and also
2230 with some extended implementations of this pragma in certain Ada 83
2233 @node Pragma Interface_Name
2234 @unnumberedsec Pragma Interface_Name
2235 @findex Interface_Name
2239 @smallexample @c ada
2240 pragma Interface_Name (
2241 [Entity =>] LOCAL_NAME
2242 [, [External_Name =>] static_string_EXPRESSION]
2243 [, [Link_Name =>] static_string_EXPRESSION]);
2247 This pragma provides an alternative way of specifying the interface name
2248 for an interfaced subprogram, and is provided for compatibility with Ada
2249 83 compilers that use the pragma for this purpose. You must provide at
2250 least one of @var{External_Name} or @var{Link_Name}.
2252 @node Pragma Interrupt_Handler
2253 @unnumberedsec Pragma Interrupt_Handler
2254 @findex Interrupt_Handler
2258 @smallexample @c ada
2259 pragma Interrupt_Handler (procedure_LOCAL_NAME);
2263 This program unit pragma is supported for parameterless protected procedures
2264 as described in Annex C of the Ada Reference Manual. On the AAMP target
2265 the pragma can also be specified for nonprotected parameterless procedures
2266 that are declared at the library level (which includes procedures
2267 declared at the top level of a library package). In the case of AAMP,
2268 when this pragma is applied to a nonprotected procedure, the instruction
2269 @code{IERET} is generated for returns from the procedure, enabling
2270 maskable interrupts, in place of the normal return instruction.
2272 @node Pragma Interrupt_State
2273 @unnumberedsec Pragma Interrupt_State
2274 @findex Interrupt_State
2278 @smallexample @c ada
2279 pragma Interrupt_State (Name => value, State => SYSTEM | RUNTIME | USER);
2283 Normally certain interrupts are reserved to the implementation. Any attempt
2284 to attach an interrupt causes Program_Error to be raised, as described in
2285 RM C.3.2(22). A typical example is the @code{SIGINT} interrupt used in
2286 many systems for an @kbd{Ctrl-C} interrupt. Normally this interrupt is
2287 reserved to the implementation, so that @kbd{Ctrl-C} can be used to
2288 interrupt execution. Additionally, signals such as @code{SIGSEGV},
2289 @code{SIGABRT}, @code{SIGFPE} and @code{SIGILL} are often mapped to specific
2290 Ada exceptions, or used to implement run-time functions such as the
2291 @code{abort} statement and stack overflow checking.
2293 Pragma @code{Interrupt_State} provides a general mechanism for overriding
2294 such uses of interrupts. It subsumes the functionality of pragma
2295 @code{Unreserve_All_Interrupts}. Pragma @code{Interrupt_State} is not
2296 available on OS/2, Windows or VMS. On all other platforms than VxWorks,
2297 it applies to signals; on VxWorks, it applies to vectored hardware interrupts
2298 and may be used to mark interrupts required by the board support package
2301 Interrupts can be in one of three states:
2305 The interrupt is reserved (no Ada handler can be installed), and the
2306 Ada run-time may not install a handler. As a result you are guaranteed
2307 standard system default action if this interrupt is raised.
2311 The interrupt is reserved (no Ada handler can be installed). The run time
2312 is allowed to install a handler for internal control purposes, but is
2313 not required to do so.
2317 The interrupt is unreserved. The user may install a handler to provide
2322 These states are the allowed values of the @code{State} parameter of the
2323 pragma. The @code{Name} parameter is a value of the type
2324 @code{Ada.Interrupts.Interrupt_ID}. Typically, it is a name declared in
2325 @code{Ada.Interrupts.Names}.
2327 This is a configuration pragma, and the binder will check that there
2328 are no inconsistencies between different units in a partition in how a
2329 given interrupt is specified. It may appear anywhere a pragma is legal.
2331 The effect is to move the interrupt to the specified state.
2333 By declaring interrupts to be SYSTEM, you guarantee the standard system
2334 action, such as a core dump.
2336 By declaring interrupts to be USER, you guarantee that you can install
2339 Note that certain signals on many operating systems cannot be caught and
2340 handled by applications. In such cases, the pragma is ignored. See the
2341 operating system documentation, or the value of the array @code{Reserved}
2342 declared in the specification of package @code{System.OS_Interface}.
2344 Overriding the default state of signals used by the Ada runtime may interfere
2345 with an application's runtime behavior in the cases of the synchronous signals,
2346 and in the case of the signal used to implement the @code{abort} statement.
2348 @node Pragma Keep_Names
2349 @unnumberedsec Pragma Keep_Names
2354 @smallexample @c ada
2355 pragma Keep_Names ([On =>] enumeration_first_subtype_LOCAL_NAME);
2359 The @var{LOCAL_NAME} argument
2360 must refer to an enumeration first subtype
2361 in the current declarative part. The effect is to retain the enumeration
2362 literal names for use by @code{Image} and @code{Value} even if a global
2363 @code{Discard_Names} pragma applies. This is useful when you want to
2364 generally suppress enumeration literal names and for example you therefore
2365 use a @code{Discard_Names} pragma in the @file{gnat.adc} file, but you
2366 want to retain the names for specific enumeration types.
2368 @node Pragma License
2369 @unnumberedsec Pragma License
2371 @cindex License checking
2375 @smallexample @c ada
2376 pragma License (Unrestricted | GPL | Modified_GPL | Restricted);
2380 This pragma is provided to allow automated checking for appropriate license
2381 conditions with respect to the standard and modified GPL@. A pragma
2382 @code{License}, which is a configuration pragma that typically appears at
2383 the start of a source file or in a separate @file{gnat.adc} file, specifies
2384 the licensing conditions of a unit as follows:
2388 This is used for a unit that can be freely used with no license restrictions.
2389 Examples of such units are public domain units, and units from the Ada
2393 This is used for a unit that is licensed under the unmodified GPL, and which
2394 therefore cannot be @code{with}'ed by a restricted unit.
2397 This is used for a unit licensed under the GNAT modified GPL that includes
2398 a special exception paragraph that specifically permits the inclusion of
2399 the unit in programs without requiring the entire program to be released
2400 under the GPL@. This is the license used for the GNAT run-time which ensures
2401 that the run-time can be used freely in any program without GPL concerns.
2404 This is used for a unit that is restricted in that it is not permitted to
2405 depend on units that are licensed under the GPL@. Typical examples are
2406 proprietary code that is to be released under more restrictive license
2407 conditions. Note that restricted units are permitted to @code{with} units
2408 which are licensed under the modified GPL (this is the whole point of the
2414 Normally a unit with no @code{License} pragma is considered to have an
2415 unknown license, and no checking is done. However, standard GNAT headers
2416 are recognized, and license information is derived from them as follows.
2420 A GNAT license header starts with a line containing 78 hyphens. The following
2421 comment text is searched for the appearance of any of the following strings.
2423 If the string ``GNU General Public License'' is found, then the unit is assumed
2424 to have GPL license, unless the string ``As a special exception'' follows, in
2425 which case the license is assumed to be modified GPL@.
2427 If one of the strings
2428 ``This specification is adapted from the Ada Semantic Interface'' or
2429 ``This specification is derived from the Ada Reference Manual'' is found
2430 then the unit is assumed to be unrestricted.
2434 These default actions means that a program with a restricted license pragma
2435 will automatically get warnings if a GPL unit is inappropriately
2436 @code{with}'ed. For example, the program:
2438 @smallexample @c ada
2441 procedure Secret_Stuff is
2447 if compiled with pragma @code{License} (@code{Restricted}) in a
2448 @file{gnat.adc} file will generate the warning:
2453 >>> license of withed unit "Sem_Ch3" is incompatible
2455 2. with GNAT.Sockets;
2456 3. procedure Secret_Stuff is
2460 Here we get a warning on @code{Sem_Ch3} since it is part of the GNAT
2461 compiler and is licensed under the
2462 GPL, but no warning for @code{GNAT.Sockets} which is part of the GNAT
2463 run time, and is therefore licensed under the modified GPL@.
2465 @node Pragma Link_With
2466 @unnumberedsec Pragma Link_With
2471 @smallexample @c ada
2472 pragma Link_With (static_string_EXPRESSION @{,static_string_EXPRESSION@});
2476 This pragma is provided for compatibility with certain Ada 83 compilers.
2477 It has exactly the same effect as pragma @code{Linker_Options} except
2478 that spaces occurring within one of the string expressions are treated
2479 as separators. For example, in the following case:
2481 @smallexample @c ada
2482 pragma Link_With ("-labc -ldef");
2486 results in passing the strings @code{-labc} and @code{-ldef} as two
2487 separate arguments to the linker. In addition pragma Link_With allows
2488 multiple arguments, with the same effect as successive pragmas.
2490 @node Pragma Linker_Alias
2491 @unnumberedsec Pragma Linker_Alias
2492 @findex Linker_Alias
2496 @smallexample @c ada
2497 pragma Linker_Alias (
2498 [Entity =>] LOCAL_NAME
2499 [Alias =>] static_string_EXPRESSION);
2503 This pragma establishes a linker alias for the given named entity. For
2504 further details on the exact effect, consult the GCC manual.
2506 @node Pragma Linker_Section
2507 @unnumberedsec Pragma Linker_Section
2508 @findex Linker_Section
2512 @smallexample @c ada
2513 pragma Linker_Section (
2514 [Entity =>] LOCAL_NAME
2515 [Section =>] static_string_EXPRESSION);
2519 This pragma specifies the name of the linker section for the given entity.
2520 For further details on the exact effect, consult the GCC manual.
2522 @node Pragma Long_Float
2523 @unnumberedsec Pragma Long_Float
2529 @smallexample @c ada
2530 pragma Long_Float (FLOAT_FORMAT);
2532 FLOAT_FORMAT ::= D_Float | G_Float
2536 This pragma is implemented only in the OpenVMS implementation of GNAT@.
2537 It allows control over the internal representation chosen for the predefined
2538 type @code{Long_Float} and for floating point type representations with
2539 @code{digits} specified in the range 7 through 15.
2540 For further details on this pragma, see the
2541 @cite{DEC Ada Language Reference Manual}, section 3.5.7b. Note that to use
2542 this pragma, the standard runtime libraries must be recompiled. See the
2543 description of the @code{GNAT LIBRARY} command in the OpenVMS version
2544 of the GNAT User's Guide for details on the use of this command.
2546 @node Pragma Machine_Attribute
2547 @unnumberedsec Pragma Machine_Attribute
2548 @findex Machine_Attribute
2552 @smallexample @c ada
2553 pragma Machine_Attribute (
2554 [Attribute_Name =>] string_EXPRESSION,
2555 [Entity =>] LOCAL_NAME);
2559 Machine dependent attributes can be specified for types and/or
2560 declarations. Currently only subprogram entities are supported. This
2561 pragma is semantically equivalent to
2562 @code{__attribute__((@var{string_expression}))} in GNU C,
2563 where @code{@var{string_expression}} is
2564 recognized by the GNU C macros @code{VALID_MACHINE_TYPE_ATTRIBUTE} and
2565 @code{VALID_MACHINE_DECL_ATTRIBUTE} which are defined in the
2566 configuration header file @file{tm.h} for each machine. See the GCC
2567 manual for further information.
2569 @node Pragma Main_Storage
2570 @unnumberedsec Pragma Main_Storage
2572 @findex Main_Storage
2576 @smallexample @c ada
2578 (MAIN_STORAGE_OPTION [, MAIN_STORAGE_OPTION]);
2580 MAIN_STORAGE_OPTION ::=
2581 [WORKING_STORAGE =>] static_SIMPLE_EXPRESSION
2582 | [TOP_GUARD =>] static_SIMPLE_EXPRESSION
2587 This pragma is provided for compatibility with OpenVMS VAX Systems. It has
2588 no effect in GNAT, other than being syntax checked. Note that the pragma
2589 also has no effect in DEC Ada 83 for OpenVMS Alpha Systems.
2591 @node Pragma No_Return
2592 @unnumberedsec Pragma No_Return
2597 @smallexample @c ada
2598 pragma No_Return (procedure_LOCAL_NAME);
2602 @var{procedure_local_NAME} must refer to one or more procedure
2603 declarations in the current declarative part. A procedure to which this
2604 pragma is applied may not contain any explicit @code{return} statements,
2605 and also may not contain any implicit return statements from falling off
2606 the end of a statement sequence. One use of this pragma is to identify
2607 procedures whose only purpose is to raise an exception.
2609 Another use of this pragma is to suppress incorrect warnings about
2610 missing returns in functions, where the last statement of a function
2611 statement sequence is a call to such a procedure.
2613 @node Pragma Normalize_Scalars
2614 @unnumberedsec Pragma Normalize_Scalars
2615 @findex Normalize_Scalars
2619 @smallexample @c ada
2620 pragma Normalize_Scalars;
2624 This is a language defined pragma which is fully implemented in GNAT@. The
2625 effect is to cause all scalar objects that are not otherwise initialized
2626 to be initialized. The initial values are implementation dependent and
2630 @item Standard.Character
2632 Objects whose root type is Standard.Character are initialized to
2633 Character'Last. This will be out of range of the subtype only if
2634 the subtype range excludes this value.
2636 @item Standard.Wide_Character
2638 Objects whose root type is Standard.Wide_Character are initialized to
2639 Wide_Character'Last. This will be out of range of the subtype only if
2640 the subtype range excludes this value.
2644 Objects of an integer type are initialized to base_type'First, where
2645 base_type is the base type of the object type. This will be out of range
2646 of the subtype only if the subtype range excludes this value. For example,
2647 if you declare the subtype:
2649 @smallexample @c ada
2650 subtype Ityp is integer range 1 .. 10;
2654 then objects of type x will be initialized to Integer'First, a negative
2655 number that is certainly outside the range of subtype @code{Ityp}.
2658 Objects of all real types (fixed and floating) are initialized to
2659 base_type'First, where base_Type is the base type of the object type.
2660 This will be out of range of the subtype only if the subtype range
2661 excludes this value.
2664 Objects of a modular type are initialized to typ'Last. This will be out
2665 of range of the subtype only if the subtype excludes this value.
2667 @item Enumeration types
2668 Objects of an enumeration type are initialized to all one-bits, i.e.@: to
2669 the value @code{2 ** typ'Size - 1}. This will be out of range of the
2670 enumeration subtype in all cases except where the subtype contains
2671 exactly 2**8, 2**16, or 2**32 elements.
2675 @node Pragma Obsolescent
2676 @unnumberedsec Pragma Obsolescent
2681 @smallexample @c ada
2682 pragma Obsolescent [(static_string_EXPRESSION)];
2686 This pragma must occur immediately following a subprogram
2687 declaration. It indicates that the associated function or procedure
2688 is considered obsolescent and should not be used. Typically this is
2689 used when an API must be modified by eventually removing or modifying
2690 existing subprograms. The pragma can be used at an intermediate stage
2691 when the subprogram is still present, but will be removed later.
2693 The effect of this pragma is to output a warning message that the
2694 subprogram is obsolescent if the appropriate warning option in the
2695 compiler is activated. If a parameter is present, then a second
2696 warning message is given containing this text.
2698 @node Pragma Passive
2699 @unnumberedsec Pragma Passive
2704 @smallexample @c ada
2705 pragma Passive ([Semaphore | No]);
2709 Syntax checked, but otherwise ignored by GNAT@. This is recognized for
2710 compatibility with DEC Ada 83 implementations, where it is used within a
2711 task definition to request that a task be made passive. If the argument
2712 @code{Semaphore} is present, or the argument is omitted, then DEC Ada 83
2713 treats the pragma as an assertion that the containing task is passive
2714 and that optimization of context switch with this task is permitted and
2715 desired. If the argument @code{No} is present, the task must not be
2716 optimized. GNAT does not attempt to optimize any tasks in this manner
2717 (since protected objects are available in place of passive tasks).
2719 @node Pragma Polling
2720 @unnumberedsec Pragma Polling
2725 @smallexample @c ada
2726 pragma Polling (ON | OFF);
2730 This pragma controls the generation of polling code. This is normally off.
2731 If @code{pragma Polling (ON)} is used then periodic calls are generated to
2732 the routine @code{Ada.Exceptions.Poll}. This routine is a separate unit in the
2733 runtime library, and can be found in file @file{a-excpol.adb}.
2735 Pragma @code{Polling} can appear as a configuration pragma (for example it
2736 can be placed in the @file{gnat.adc} file) to enable polling globally, or it
2737 can be used in the statement or declaration sequence to control polling
2740 A call to the polling routine is generated at the start of every loop and
2741 at the start of every subprogram call. This guarantees that the @code{Poll}
2742 routine is called frequently, and places an upper bound (determined by
2743 the complexity of the code) on the period between two @code{Poll} calls.
2745 The primary purpose of the polling interface is to enable asynchronous
2746 aborts on targets that cannot otherwise support it (for example Windows
2747 NT), but it may be used for any other purpose requiring periodic polling.
2748 The standard version is null, and can be replaced by a user program. This
2749 will require re-compilation of the @code{Ada.Exceptions} package that can
2750 be found in files @file{a-except.ads} and @file{a-except.adb}.
2752 A standard alternative unit (in file @file{4wexcpol.adb} in the standard GNAT
2753 distribution) is used to enable the asynchronous abort capability on
2754 targets that do not normally support the capability. The version of
2755 @code{Poll} in this file makes a call to the appropriate runtime routine
2756 to test for an abort condition.
2758 Note that polling can also be enabled by use of the @code{-gnatP} switch. See
2759 the @cite{GNAT User's Guide} for details.
2761 @node Pragma Propagate_Exceptions
2762 @unnumberedsec Pragma Propagate_Exceptions
2763 @findex Propagate_Exceptions
2764 @cindex Zero Cost Exceptions
2768 @smallexample @c ada
2769 pragma Propagate_Exceptions (subprogram_LOCAL_NAME);
2773 This pragma indicates that the given entity, which is the name of an
2774 imported foreign-language subprogram may receive an Ada exception,
2775 and that the exception should be propagated. It is relevant only if
2776 zero cost exception handling is in use, and is thus never needed if
2777 the alternative @code{longjmp} / @code{setjmp} implementation of
2778 exceptions is used (although it is harmless to use it in such cases).
2780 The implementation of fast exceptions always properly propagates
2781 exceptions through Ada code, as described in the Ada Reference Manual.
2782 However, this manual is silent about the propagation of exceptions
2783 through foreign code. For example, consider the
2784 situation where @code{P1} calls
2785 @code{P2}, and @code{P2} calls @code{P3}, where
2786 @code{P1} and @code{P3} are in Ada, but @code{P2} is in C@.
2787 @code{P3} raises an Ada exception. The question is whether or not
2788 it will be propagated through @code{P2} and can be handled in
2791 For the @code{longjmp} / @code{setjmp} implementation of exceptions,
2792 the answer is always yes. For some targets on which zero cost exception
2793 handling is implemented, the answer is also always yes. However, there
2794 are some targets, notably in the current version all x86 architecture
2795 targets, in which the answer is that such propagation does not
2796 happen automatically. If such propagation is required on these
2797 targets, it is mandatory to use @code{Propagate_Exceptions} to
2798 name all foreign language routines through which Ada exceptions
2801 @node Pragma Psect_Object
2802 @unnumberedsec Pragma Psect_Object
2803 @findex Psect_Object
2807 @smallexample @c ada
2808 pragma Psect_Object (
2809 [Internal =>] LOCAL_NAME,
2810 [, [External =>] EXTERNAL_SYMBOL]
2811 [, [Size =>] EXTERNAL_SYMBOL]);
2815 | static_string_EXPRESSION
2819 This pragma is identical in effect to pragma @code{Common_Object}.
2821 @node Pragma Pure_Function
2822 @unnumberedsec Pragma Pure_Function
2823 @findex Pure_Function
2827 @smallexample @c ada
2828 pragma Pure_Function ([Entity =>] function_LOCAL_NAME);
2832 This pragma appears in the same declarative part as a function
2833 declaration (or a set of function declarations if more than one
2834 overloaded declaration exists, in which case the pragma applies
2835 to all entities). It specifies that the function @code{Entity} is
2836 to be considered pure for the purposes of code generation. This means
2837 that the compiler can assume that there are no side effects, and
2838 in particular that two calls with identical arguments produce the
2839 same result. It also means that the function can be used in an
2842 Note that, quite deliberately, there are no static checks to try
2843 to ensure that this promise is met, so @code{Pure_Function} can be used
2844 with functions that are conceptually pure, even if they do modify
2845 global variables. For example, a square root function that is
2846 instrumented to count the number of times it is called is still
2847 conceptually pure, and can still be optimized, even though it
2848 modifies a global variable (the count). Memo functions are another
2849 example (where a table of previous calls is kept and consulted to
2850 avoid re-computation).
2853 Note: Most functions in a @code{Pure} package are automatically pure, and
2854 there is no need to use pragma @code{Pure_Function} for such functions. One
2855 exception is any function that has at least one formal of type
2856 @code{System.Address} or a type derived from it. Such functions are not
2857 considered pure by default, since the compiler assumes that the
2858 @code{Address} parameter may be functioning as a pointer and that the
2859 referenced data may change even if the address value does not.
2860 Similarly, imported functions are not considered to be pure by default,
2861 since there is no way of checking that they are in fact pure. The use
2862 of pragma @code{Pure_Function} for such a function will override these default
2863 assumption, and cause the compiler to treat a designated subprogram as pure
2866 Note: If pragma @code{Pure_Function} is applied to a renamed function, it
2867 applies to the underlying renamed function. This can be used to
2868 disambiguate cases of overloading where some but not all functions
2869 in a set of overloaded functions are to be designated as pure.
2871 @node Pragma Ravenscar
2872 @unnumberedsec Pragma Ravenscar
2877 @smallexample @c ada
2882 A configuration pragma that establishes the following set of restrictions:
2885 @item No_Abort_Statements
2886 [RM D.7] There are no abort_statements, and there are
2887 no calls to Task_Identification.Abort_Task.
2889 @item No_Select_Statements
2890 There are no select_statements.
2892 @item No_Task_Hierarchy
2893 [RM D.7] All (non-environment) tasks depend
2894 directly on the environment task of the partition.
2896 @item No_Task_Allocators
2897 [RM D.7] There are no allocators for task types
2898 or types containing task subcomponents.
2900 @item No_Dynamic_Priorities
2901 [RM D.7] There are no semantic dependencies on the package Dynamic_Priorities.
2903 @item No_Terminate_Alternatives
2904 [RM D.7] There are no selective_accepts with terminate_alternatives
2906 @item No_Dynamic_Interrupts
2907 There are no semantic dependencies on Ada.Interrupts.
2909 @item No_Implicit_Heap_Allocations
2910 [RM D.7] No constructs are allowed to cause implicit heap allocation
2912 @item No_Protected_Type_Allocators
2913 There are no allocators for protected types or
2914 types containing protected subcomponents.
2916 @item No_Local_Protected_Objects
2917 Protected objects and access types that designate
2918 such objects shall be declared only at library level.
2920 @item No_Requeue_Statements
2921 Requeue statements are not allowed.
2924 There are no semantic dependencies on the package Ada.Calendar.
2926 @item No_Relative_Delay
2927 There are no delay_relative_statements.
2929 @item No_Task_Attributes
2930 There are no semantic dependencies on the Ada.Task_Attributes package and
2931 there are no references to the attributes Callable and Terminated [RM 9.9].
2933 @item Boolean_Entry_Barriers
2934 Entry barrier condition expressions shall be boolean
2935 objects which are declared in the protected type
2936 which contains the entry.
2938 @item Max_Asynchronous_Select_Nesting = 0
2939 [RM D.7] Specifies the maximum dynamic nesting level of asynchronous_selects.
2940 A value of zero prevents the use of any asynchronous_select.
2942 @item Max_Task_Entries = 0
2943 [RM D.7] Specifies the maximum number of entries
2944 per task. The bounds of every entry family
2945 of a task unit shall be static, or shall be
2946 defined by a discriminant of a subtype whose
2947 corresponding bound is static. A value of zero
2948 indicates that no rendezvous are possible. For
2949 the Ravenscar pragma, the value of Max_Task_Entries is always
2952 @item Max_Protected_Entries = 1
2953 [RM D.7] Specifies the maximum number of entries per
2954 protected type. The bounds of every entry family of
2955 a protected unit shall be static, or shall be defined
2956 by a discriminant of a subtype whose corresponding
2957 bound is static. For the Ravenscar pragma the value of
2958 Max_Protected_Entries is always 1.
2960 @item Max_Select_Alternatives = 0
2961 [RM D.7] Specifies the maximum number of alternatives in a selective_accept.
2962 For the Ravenscar pragma the value is always 0.
2964 @item No_Task_Termination
2965 Tasks which terminate are erroneous.
2967 @item No_Entry_Queue
2968 No task can be queued on a protected entry. Note that this restrictions is
2969 checked at run time. The violation of this restriction generates a
2970 Program_Error exception.
2974 This set of restrictions corresponds to the definition of the ``Ravenscar
2975 Profile'' for limited tasking, devised and published by the
2976 @cite{International Real-Time Ada Workshop}, 1997,
2977 and whose most recent description is available at
2978 @url{ftp://ftp.openravenscar.org/openravenscar/ravenscar00.pdf}.
2980 The above set is a superset of the restrictions provided by pragma
2981 @code{Restricted_Run_Time}, it includes five additional restrictions
2982 (@code{Boolean_Entry_Barriers}, @code{No_Select_Statements},
2984 @code{No_Relative_Delay} and @code{No_Task_Termination}). This means
2985 that pragma @code{Ravenscar}, like the pragma @code{Restricted_Run_Time},
2986 automatically causes the use of a simplified, more efficient version
2987 of the tasking run-time system.
2989 @node Pragma Restricted_Run_Time
2990 @unnumberedsec Pragma Restricted_Run_Time
2991 @findex Restricted_Run_Time
2995 @smallexample @c ada
2996 pragma Restricted_Run_Time;
3000 A configuration pragma that establishes the following set of restrictions:
3003 @item No_Abort_Statements
3004 @item No_Entry_Queue
3005 @item No_Task_Hierarchy
3006 @item No_Task_Allocators
3007 @item No_Dynamic_Priorities
3008 @item No_Terminate_Alternatives
3009 @item No_Dynamic_Interrupts
3010 @item No_Protected_Type_Allocators
3011 @item No_Local_Protected_Objects
3012 @item No_Requeue_Statements
3013 @item No_Task_Attributes
3014 @item Max_Asynchronous_Select_Nesting = 0
3015 @item Max_Task_Entries = 0
3016 @item Max_Protected_Entries = 1
3017 @item Max_Select_Alternatives = 0
3021 This set of restrictions causes the automatic selection of a simplified
3022 version of the run time that provides improved performance for the
3023 limited set of tasking functionality permitted by this set of restrictions.
3025 @node Pragma Restriction_Warnings
3026 @unnumberedsec Pragma Restriction_Warnings
3027 @findex Restriction_Warnings
3031 @smallexample @c ada
3032 pragma Restriction_Warnings
3033 (restriction_IDENTIFIER @{, restriction_IDENTIFIER@});
3037 This pragma allows a series of restriction identifiers to be
3038 specified (the list of allowed identifiers is the same as for
3039 pragma @code{Restrictions}). For each of these identifiers
3040 the compiler checks for violations of the restriction, but
3041 generates a warning message rather than an error message
3042 if the restriction is violated.
3044 @node Pragma Source_File_Name
3045 @unnumberedsec Pragma Source_File_Name
3046 @findex Source_File_Name
3050 @smallexample @c ada
3051 pragma Source_File_Name (
3052 [Unit_Name =>] unit_NAME,
3053 Spec_File_Name => STRING_LITERAL);
3055 pragma Source_File_Name (
3056 [Unit_Name =>] unit_NAME,
3057 Body_File_Name => STRING_LITERAL);
3061 Use this to override the normal naming convention. It is a configuration
3062 pragma, and so has the usual applicability of configuration pragmas
3063 (i.e.@: it applies to either an entire partition, or to all units in a
3064 compilation, or to a single unit, depending on how it is used.
3065 @var{unit_name} is mapped to @var{file_name_literal}. The identifier for
3066 the second argument is required, and indicates whether this is the file
3067 name for the spec or for the body.
3069 Another form of the @code{Source_File_Name} pragma allows
3070 the specification of patterns defining alternative file naming schemes
3071 to apply to all files.
3073 @smallexample @c ada
3074 pragma Source_File_Name
3075 (Spec_File_Name => STRING_LITERAL
3076 [,Casing => CASING_SPEC]
3077 [,Dot_Replacement => STRING_LITERAL]);
3079 pragma Source_File_Name
3080 (Body_File_Name => STRING_LITERAL
3081 [,Casing => CASING_SPEC]
3082 [,Dot_Replacement => STRING_LITERAL]);
3084 pragma Source_File_Name
3085 (Subunit_File_Name => STRING_LITERAL
3086 [,Casing => CASING_SPEC]
3087 [,Dot_Replacement => STRING_LITERAL]);
3089 CASING_SPEC ::= Lowercase | Uppercase | Mixedcase
3093 The first argument is a pattern that contains a single asterisk indicating
3094 the point at which the unit name is to be inserted in the pattern string
3095 to form the file name. The second argument is optional. If present it
3096 specifies the casing of the unit name in the resulting file name string.
3097 The default is lower case. Finally the third argument allows for systematic
3098 replacement of any dots in the unit name by the specified string literal.
3100 A pragma Source_File_Name cannot appear after a
3101 @ref{Pragma Source_File_Name_Project}.
3103 For more details on the use of the @code{Source_File_Name} pragma,
3104 see the sections ``Using Other File Names'' and
3105 ``Alternative File Naming Schemes'' in the @cite{GNAT User's Guide}.
3107 @node Pragma Source_File_Name_Project
3108 @unnumberedsec Pragma Source_File_Name_Project
3109 @findex Source_File_Name_Project
3112 This pragma has the same syntax and semantics as pragma Source_File_Name.
3113 It is only allowed as a stand alone configuration pragma.
3114 It cannot appear after a @ref{Pragma Source_File_Name}, and
3115 most importantly, once pragma Source_File_Name_Project appears,
3116 no further Source_File_Name pragmas are allowed.
3118 The intention is that Source_File_Name_Project pragmas are always
3119 generated by the Project Manager in a manner consistent with the naming
3120 specified in a project file, and when naming is controlled in this manner,
3121 it is not permissible to attempt to modify this naming scheme using
3122 Source_File_Name pragmas (which would not be known to the project manager).
3124 @node Pragma Source_Reference
3125 @unnumberedsec Pragma Source_Reference
3126 @findex Source_Reference
3130 @smallexample @c ada
3131 pragma Source_Reference (INTEGER_LITERAL, STRING_LITERAL);
3135 This pragma must appear as the first line of a source file.
3136 @var{integer_literal} is the logical line number of the line following
3137 the pragma line (for use in error messages and debugging
3138 information). @var{string_literal} is a static string constant that
3139 specifies the file name to be used in error messages and debugging
3140 information. This is most notably used for the output of @code{gnatchop}
3141 with the @code{-r} switch, to make sure that the original unchopped
3142 source file is the one referred to.
3144 The second argument must be a string literal, it cannot be a static
3145 string expression other than a string literal. This is because its value
3146 is needed for error messages issued by all phases of the compiler.
3148 @node Pragma Stream_Convert
3149 @unnumberedsec Pragma Stream_Convert
3150 @findex Stream_Convert
3154 @smallexample @c ada
3155 pragma Stream_Convert (
3156 [Entity =>] type_LOCAL_NAME,
3157 [Read =>] function_NAME,
3158 [Write =>] function_NAME);
3162 This pragma provides an efficient way of providing stream functions for
3163 types defined in packages. Not only is it simpler to use than declaring
3164 the necessary functions with attribute representation clauses, but more
3165 significantly, it allows the declaration to made in such a way that the
3166 stream packages are not loaded unless they are needed. The use of
3167 the Stream_Convert pragma adds no overhead at all, unless the stream
3168 attributes are actually used on the designated type.
3170 The first argument specifies the type for which stream functions are
3171 provided. The second parameter provides a function used to read values
3172 of this type. It must name a function whose argument type may be any
3173 subtype, and whose returned type must be the type given as the first
3174 argument to the pragma.
3176 The meaning of the @var{Read}
3177 parameter is that if a stream attribute directly
3178 or indirectly specifies reading of the type given as the first parameter,
3179 then a value of the type given as the argument to the Read function is
3180 read from the stream, and then the Read function is used to convert this
3181 to the required target type.
3183 Similarly the @var{Write} parameter specifies how to treat write attributes
3184 that directly or indirectly apply to the type given as the first parameter.
3185 It must have an input parameter of the type specified by the first parameter,
3186 and the return type must be the same as the input type of the Read function.
3187 The effect is to first call the Write function to convert to the given stream
3188 type, and then write the result type to the stream.
3190 The Read and Write functions must not be overloaded subprograms. If necessary
3191 renamings can be supplied to meet this requirement.
3192 The usage of this attribute is best illustrated by a simple example, taken
3193 from the GNAT implementation of package Ada.Strings.Unbounded:
3195 @smallexample @c ada
3196 function To_Unbounded (S : String)
3197 return Unbounded_String
3198 renames To_Unbounded_String;
3200 pragma Stream_Convert
3201 (Unbounded_String, To_Unbounded, To_String);
3205 The specifications of the referenced functions, as given in the Ada 95
3206 Reference Manual are:
3208 @smallexample @c ada
3209 function To_Unbounded_String (Source : String)
3210 return Unbounded_String;
3212 function To_String (Source : Unbounded_String)
3217 The effect is that if the value of an unbounded string is written to a
3218 stream, then the representation of the item in the stream is in the same
3219 format used for @code{Standard.String}, and this same representation is
3220 expected when a value of this type is read from the stream.
3222 @node Pragma Style_Checks
3223 @unnumberedsec Pragma Style_Checks
3224 @findex Style_Checks
3228 @smallexample @c ada
3229 pragma Style_Checks (string_LITERAL | ALL_CHECKS |
3230 On | Off [, LOCAL_NAME]);
3234 This pragma is used in conjunction with compiler switches to control the
3235 built in style checking provided by GNAT@. The compiler switches, if set,
3236 provide an initial setting for the switches, and this pragma may be used
3237 to modify these settings, or the settings may be provided entirely by
3238 the use of the pragma. This pragma can be used anywhere that a pragma
3239 is legal, including use as a configuration pragma (including use in
3240 the @file{gnat.adc} file).
3242 The form with a string literal specifies which style options are to be
3243 activated. These are additive, so they apply in addition to any previously
3244 set style check options. The codes for the options are the same as those
3245 used in the @code{-gnaty} switch to @code{gcc} or @code{gnatmake}.
3246 For example the following two methods can be used to enable
3251 @smallexample @c ada
3252 pragma Style_Checks ("l");
3257 gcc -c -gnatyl @dots{}
3262 The form ALL_CHECKS activates all standard checks (its use is equivalent
3263 to the use of the @code{gnaty} switch with no options. See GNAT User's
3266 The forms with @code{Off} and @code{On}
3267 can be used to temporarily disable style checks
3268 as shown in the following example:
3270 @smallexample @c ada
3274 pragma Style_Checks ("k"); -- requires keywords in lower case
3275 pragma Style_Checks (Off); -- turn off style checks
3276 NULL; -- this will not generate an error message
3277 pragma Style_Checks (On); -- turn style checks back on
3278 NULL; -- this will generate an error message
3282 Finally the two argument form is allowed only if the first argument is
3283 @code{On} or @code{Off}. The effect is to turn of semantic style checks
3284 for the specified entity, as shown in the following example:
3286 @smallexample @c ada
3290 pragma Style_Checks ("r"); -- require consistency of identifier casing
3292 Rf1 : Integer := ARG; -- incorrect, wrong case
3293 pragma Style_Checks (Off, Arg);
3294 Rf2 : Integer := ARG; -- OK, no error
3297 @node Pragma Subtitle
3298 @unnumberedsec Pragma Subtitle
3303 @smallexample @c ada
3304 pragma Subtitle ([Subtitle =>] STRING_LITERAL);
3308 This pragma is recognized for compatibility with other Ada compilers
3309 but is ignored by GNAT@.
3311 @node Pragma Suppress_All
3312 @unnumberedsec Pragma Suppress_All
3313 @findex Suppress_All
3317 @smallexample @c ada
3318 pragma Suppress_All;
3322 This pragma can only appear immediately following a compilation
3323 unit. The effect is to apply @code{Suppress (All_Checks)} to the unit
3324 which it follows. This pragma is implemented for compatibility with DEC
3325 Ada 83 usage. The use of pragma @code{Suppress (All_Checks)} as a normal
3326 configuration pragma is the preferred usage in GNAT@.
3328 @node Pragma Suppress_Exception_Locations
3329 @unnumberedsec Pragma Suppress_Exception_Locations
3330 @findex Suppress_Exception_Locations
3334 @smallexample @c ada
3335 pragma Suppress_Exception_Locations;
3339 In normal mode, a raise statement for an exception by default generates
3340 an exception message giving the file name and line number for the location
3341 of the raise. This is useful for debugging and logging purposes, but this
3342 entails extra space for the strings for the messages. The configuration
3343 pragma @code{Suppress_Exception_Locations} can be used to suppress the
3344 generation of these strings, with the result that space is saved, but the
3345 exception message for such raises is null. This configuration pragma may
3346 appear in a global configuration pragma file, or in a specific unit as
3347 usual. It is not required that this pragma be used consistently within
3348 a partition, so it is fine to have some units within a partition compiled
3349 with this pragma and others compiled in normal mode without it.
3351 @node Pragma Suppress_Initialization
3352 @unnumberedsec Pragma Suppress_Initialization
3353 @findex Suppress_Initialization
3354 @cindex Suppressing initialization
3355 @cindex Initialization, suppression of
3359 @smallexample @c ada
3360 pragma Suppress_Initialization ([Entity =>] type_Name);
3364 This pragma suppresses any implicit or explicit initialization
3365 associated with the given type name for all variables of this type.
3367 @node Pragma Task_Info
3368 @unnumberedsec Pragma Task_Info
3373 @smallexample @c ada
3374 pragma Task_Info (EXPRESSION);
3378 This pragma appears within a task definition (like pragma
3379 @code{Priority}) and applies to the task in which it appears. The
3380 argument must be of type @code{System.Task_Info.Task_Info_Type}.
3381 The @code{Task_Info} pragma provides system dependent control over
3382 aspects of tasking implementation, for example, the ability to map
3383 tasks to specific processors. For details on the facilities available
3384 for the version of GNAT that you are using, see the documentation
3385 in the specification of package System.Task_Info in the runtime
3388 @node Pragma Task_Name
3389 @unnumberedsec Pragma Task_Name
3394 @smallexample @c ada
3395 pragma Task_Name (string_EXPRESSION);
3399 This pragma appears within a task definition (like pragma
3400 @code{Priority}) and applies to the task in which it appears. The
3401 argument must be of type String, and provides a name to be used for
3402 the task instance when the task is created. Note that this expression
3403 is not required to be static, and in particular, it can contain
3404 references to task discriminants. This facility can be used to
3405 provide different names for different tasks as they are created,
3406 as illustrated in the example below.
3408 The task name is recorded internally in the run-time structures
3409 and is accessible to tools like the debugger. In addition the
3410 routine @code{Ada.Task_Identification.Image} will return this
3411 string, with a unique task address appended.
3413 @smallexample @c ada
3414 -- Example of the use of pragma Task_Name
3416 with Ada.Task_Identification;
3417 use Ada.Task_Identification;
3418 with Text_IO; use Text_IO;
3421 type Astring is access String;
3423 task type Task_Typ (Name : access String) is
3424 pragma Task_Name (Name.all);
3427 task body Task_Typ is
3428 Nam : constant String := Image (Current_Task);
3430 Put_Line ("-->" & Nam (1 .. 14) & "<--");
3433 type Ptr_Task is access Task_Typ;
3434 Task_Var : Ptr_Task;
3438 new Task_Typ (new String'("This is task 1"));
3440 new Task_Typ (new String'("This is task 2"));
3444 @node Pragma Task_Storage
3445 @unnumberedsec Pragma Task_Storage
3446 @findex Task_Storage
3449 @smallexample @c ada
3450 pragma Task_Storage (
3451 [Task_Type =>] LOCAL_NAME,
3452 [Top_Guard =>] static_integer_EXPRESSION);
3456 This pragma specifies the length of the guard area for tasks. The guard
3457 area is an additional storage area allocated to a task. A value of zero
3458 means that either no guard area is created or a minimal guard area is
3459 created, depending on the target. This pragma can appear anywhere a
3460 @code{Storage_Size} attribute definition clause is allowed for a task
3463 @node Pragma Thread_Body
3464 @unnumberedsec Pragma Thread_Body
3468 @smallexample @c ada
3469 pragma Thread_Body (
3470 [Entity =>] LOCAL_NAME,
3471 [[Secondary_Stack_Size =>] static_integer_EXPRESSION)];
3475 This pragma specifies that the subprogram whose name is given as the
3476 @code{Entity} argument is a thread body, which will be activated
3477 by being called via its Address from foreign code. The purpose is
3478 to allow execution and registration of the foreign thread within the
3479 Ada run-time system.
3481 See the library unit @code{System.Threads} for details on the expansion of
3482 a thread body subprogram, including the calls made to subprograms
3483 within System.Threads to register the task. This unit also lists the
3484 targets and runtime systems for which this pragma is supported.
3486 A thread body subprogram may not be called directly from Ada code, and
3487 it is not permitted to apply the Access (or Unrestricted_Access) attributes
3488 to such a subprogram. The only legitimate way of calling such a subprogram
3489 is to pass its Address to foreign code and then make the call from the
3492 A thread body subprogram may have any parameters, and it may be a function
3493 returning a result. The convention of the thread body subprogram may be
3494 set in the usual manner using @code{pragma Convention}.
3496 The secondary stack size parameter, if given, is used to set the size
3497 of secondary stack for the thread. The secondary stack is allocated as
3498 a local variable of the expanded thread body subprogram, and thus is
3499 allocated out of the main thread stack size. If no secondary stack
3500 size parameter is present, the default size (from the declaration in
3501 @code{System.Secondary_Stack} is used.
3503 @node Pragma Time_Slice
3504 @unnumberedsec Pragma Time_Slice
3509 @smallexample @c ada
3510 pragma Time_Slice (static_duration_EXPRESSION);
3514 For implementations of GNAT on operating systems where it is possible
3515 to supply a time slice value, this pragma may be used for this purpose.
3516 It is ignored if it is used in a system that does not allow this control,
3517 or if it appears in other than the main program unit.
3519 Note that the effect of this pragma is identical to the effect of the
3520 DEC Ada 83 pragma of the same name when operating under OpenVMS systems.
3523 @unnumberedsec Pragma Title
3528 @smallexample @c ada
3529 pragma Title (TITLING_OPTION [, TITLING OPTION]);
3532 [Title =>] STRING_LITERAL,
3533 | [Subtitle =>] STRING_LITERAL
3537 Syntax checked but otherwise ignored by GNAT@. This is a listing control
3538 pragma used in DEC Ada 83 implementations to provide a title and/or
3539 subtitle for the program listing. The program listing generated by GNAT
3540 does not have titles or subtitles.
3542 Unlike other pragmas, the full flexibility of named notation is allowed
3543 for this pragma, i.e.@: the parameters may be given in any order if named
3544 notation is used, and named and positional notation can be mixed
3545 following the normal rules for procedure calls in Ada.
3547 @node Pragma Unchecked_Union
3548 @unnumberedsec Pragma Unchecked_Union
3550 @findex Unchecked_Union
3554 @smallexample @c ada
3555 pragma Unchecked_Union (first_subtype_LOCAL_NAME);
3559 This pragma is used to declare that the specified type should be represented
3561 equivalent to a C union type, and is intended only for use in
3562 interfacing with C code that uses union types. In Ada terms, the named
3563 type must obey the following rules:
3567 It is a non-tagged non-limited record type.
3569 It has a single discrete discriminant with a default value.
3571 The component list consists of a single variant part.
3573 Each variant has a component list with a single component.
3575 No nested variants are allowed.
3577 No component has an explicit default value.
3579 No component has a non-static constraint.
3583 In addition, given a type that meets the above requirements, the
3584 following restrictions apply to its use throughout the program:
3588 The discriminant name can be mentioned only in an aggregate.
3590 No subtypes may be created of this type.
3592 The type may not be constrained by giving a discriminant value.
3594 The type cannot be passed as the actual for a generic formal with a
3599 Equality and inequality operations on @code{unchecked_unions} are not
3600 available, since there is no discriminant to compare and the compiler
3601 does not even know how many bits to compare. It is implementation
3602 dependent whether this is detected at compile time as an illegality or
3603 whether it is undetected and considered to be an erroneous construct. In
3604 GNAT, a direct comparison is illegal, but GNAT does not attempt to catch
3605 the composite case (where two composites are compared that contain an
3606 unchecked union component), so such comparisons are simply considered
3609 The layout of the resulting type corresponds exactly to a C union, where
3610 each branch of the union corresponds to a single variant in the Ada
3611 record. The semantics of the Ada program is not changed in any way by
3612 the pragma, i.e.@: provided the above restrictions are followed, and no
3613 erroneous incorrect references to fields or erroneous comparisons occur,
3614 the semantics is exactly as described by the Ada reference manual.
3615 Pragma @code{Suppress (Discriminant_Check)} applies implicitly to the
3616 type and the default convention is C.
3618 @node Pragma Unimplemented_Unit
3619 @unnumberedsec Pragma Unimplemented_Unit
3620 @findex Unimplemented_Unit
3624 @smallexample @c ada
3625 pragma Unimplemented_Unit;
3629 If this pragma occurs in a unit that is processed by the compiler, GNAT
3630 aborts with the message @samp{@var{xxx} not implemented}, where
3631 @var{xxx} is the name of the current compilation unit. This pragma is
3632 intended to allow the compiler to handle unimplemented library units in
3635 The abort only happens if code is being generated. Thus you can use
3636 specs of unimplemented packages in syntax or semantic checking mode.
3638 @node Pragma Universal_Data
3639 @unnumberedsec Pragma Universal_Data
3640 @findex Universal_Data
3644 @smallexample @c ada
3645 pragma Universal_Data [(library_unit_Name)];
3649 This pragma is supported only for the AAMP target and is ignored for
3650 other targets. The pragma specifies that all library-level objects
3651 (Counter 0 data) associated with the library unit are to be accessed
3652 and updated using universal addressing (24-bit addresses for AAMP5)
3653 rather than the default of 16-bit Data Environment (DENV) addressing.
3654 Use of this pragma will generally result in less efficient code for
3655 references to global data associated with the library unit, but
3656 allows such data to be located anywhere in memory. This pragma is
3657 a library unit pragma, but can also be used as a configuration pragma
3658 (including use in the @file{gnat.adc} file). The functionality
3659 of this pragma is also available by applying the -univ switch on the
3660 compilations of units where universal addressing of the data is desired.
3662 @node Pragma Unreferenced
3663 @unnumberedsec Pragma Unreferenced
3664 @findex Unreferenced
3665 @cindex Warnings, unreferenced
3669 @smallexample @c ada
3670 pragma Unreferenced (local_Name @{, local_Name@});
3674 This pragma signals that the entities whose names are listed are
3675 deliberately not referenced in the current source unit. This
3676 suppresses warnings about the
3677 entities being unreferenced, and in addition a warning will be
3678 generated if one of these entities is in fact referenced in the
3679 same unit as the pragma (or in the corresponding body, or one
3682 This is particularly useful for clearly signaling that a particular
3683 parameter is not referenced in some particular subprogram implementation
3684 and that this is deliberate. It can also be useful in the case of
3685 objects declared only for their initialization or finalization side
3688 If @code{local_Name} identifies more than one matching homonym in the
3689 current scope, then the entity most recently declared is the one to which
3692 The left hand side of an assignment does not count as a reference for the
3693 purpose of this pragma. Thus it is fine to assign to an entity for which
3694 pragma Unreferenced is given.
3696 @node Pragma Unreserve_All_Interrupts
3697 @unnumberedsec Pragma Unreserve_All_Interrupts
3698 @findex Unreserve_All_Interrupts
3702 @smallexample @c ada
3703 pragma Unreserve_All_Interrupts;
3707 Normally certain interrupts are reserved to the implementation. Any attempt
3708 to attach an interrupt causes Program_Error to be raised, as described in
3709 RM C.3.2(22). A typical example is the @code{SIGINT} interrupt used in
3710 many systems for a @kbd{Ctrl-C} interrupt. Normally this interrupt is
3711 reserved to the implementation, so that @kbd{Ctrl-C} can be used to
3712 interrupt execution.
3714 If the pragma @code{Unreserve_All_Interrupts} appears anywhere in any unit in
3715 a program, then all such interrupts are unreserved. This allows the
3716 program to handle these interrupts, but disables their standard
3717 functions. For example, if this pragma is used, then pressing
3718 @kbd{Ctrl-C} will not automatically interrupt execution. However,
3719 a program can then handle the @code{SIGINT} interrupt as it chooses.
3721 For a full list of the interrupts handled in a specific implementation,
3722 see the source code for the specification of @code{Ada.Interrupts.Names} in
3723 file @file{a-intnam.ads}. This is a target dependent file that contains the
3724 list of interrupts recognized for a given target. The documentation in
3725 this file also specifies what interrupts are affected by the use of
3726 the @code{Unreserve_All_Interrupts} pragma.
3728 For a more general facility for controlling what interrupts can be
3729 handled, see pragma @code{Interrupt_State}, which subsumes the functionality
3730 of the @code{Unreserve_All_Interrupts} pragma.
3732 @node Pragma Unsuppress
3733 @unnumberedsec Pragma Unsuppress
3738 @smallexample @c ada
3739 pragma Unsuppress (IDENTIFIER [, [On =>] NAME]);
3743 This pragma undoes the effect of a previous pragma @code{Suppress}. If
3744 there is no corresponding pragma @code{Suppress} in effect, it has no
3745 effect. The range of the effect is the same as for pragma
3746 @code{Suppress}. The meaning of the arguments is identical to that used
3747 in pragma @code{Suppress}.
3749 One important application is to ensure that checks are on in cases where
3750 code depends on the checks for its correct functioning, so that the code
3751 will compile correctly even if the compiler switches are set to suppress
3754 @node Pragma Use_VADS_Size
3755 @unnumberedsec Pragma Use_VADS_Size
3756 @cindex @code{Size}, VADS compatibility
3757 @findex Use_VADS_Size
3761 @smallexample @c ada
3762 pragma Use_VADS_Size;
3766 This is a configuration pragma. In a unit to which it applies, any use
3767 of the 'Size attribute is automatically interpreted as a use of the
3768 'VADS_Size attribute. Note that this may result in incorrect semantic
3769 processing of valid Ada 95 programs. This is intended to aid in the
3770 handling of legacy code which depends on the interpretation of Size
3771 as implemented in the VADS compiler. See description of the VADS_Size
3772 attribute for further details.
3774 @node Pragma Validity_Checks
3775 @unnumberedsec Pragma Validity_Checks
3776 @findex Validity_Checks
3780 @smallexample @c ada
3781 pragma Validity_Checks (string_LITERAL | ALL_CHECKS | On | Off);
3785 This pragma is used in conjunction with compiler switches to control the
3786 built-in validity checking provided by GNAT@. The compiler switches, if set
3787 provide an initial setting for the switches, and this pragma may be used
3788 to modify these settings, or the settings may be provided entirely by
3789 the use of the pragma. This pragma can be used anywhere that a pragma
3790 is legal, including use as a configuration pragma (including use in
3791 the @file{gnat.adc} file).
3793 The form with a string literal specifies which validity options are to be
3794 activated. The validity checks are first set to include only the default
3795 reference manual settings, and then a string of letters in the string
3796 specifies the exact set of options required. The form of this string
3797 is exactly as described for the @code{-gnatVx} compiler switch (see the
3798 GNAT users guide for details). For example the following two methods
3799 can be used to enable validity checking for mode @code{in} and
3800 @code{in out} subprogram parameters:
3804 @smallexample @c ada
3805 pragma Validity_Checks ("im");
3810 gcc -c -gnatVim @dots{}
3815 The form ALL_CHECKS activates all standard checks (its use is equivalent
3816 to the use of the @code{gnatva} switch.
3818 The forms with @code{Off} and @code{On}
3819 can be used to temporarily disable validity checks
3820 as shown in the following example:
3822 @smallexample @c ada
3826 pragma Validity_Checks ("c"); -- validity checks for copies
3827 pragma Validity_Checks (Off); -- turn off validity checks
3828 A := B; -- B will not be validity checked
3829 pragma Validity_Checks (On); -- turn validity checks back on
3830 A := C; -- C will be validity checked
3833 @node Pragma Volatile
3834 @unnumberedsec Pragma Volatile
3839 @smallexample @c ada
3840 pragma Volatile (local_NAME);
3844 This pragma is defined by the Ada 95 Reference Manual, and the GNAT
3845 implementation is fully conformant with this definition. The reason it
3846 is mentioned in this section is that a pragma of the same name was supplied
3847 in some Ada 83 compilers, including DEC Ada 83. The Ada 95 implementation
3848 of pragma Volatile is upwards compatible with the implementation in
3851 @node Pragma Warnings
3852 @unnumberedsec Pragma Warnings
3857 @smallexample @c ada
3858 pragma Warnings (On | Off [, LOCAL_NAME]);
3862 Normally warnings are enabled, with the output being controlled by
3863 the command line switch. Warnings (@code{Off}) turns off generation of
3864 warnings until a Warnings (@code{On}) is encountered or the end of the
3865 current unit. If generation of warnings is turned off using this
3866 pragma, then no warning messages are output, regardless of the
3867 setting of the command line switches.
3869 The form with a single argument is a configuration pragma.
3871 If the @var{local_name} parameter is present, warnings are suppressed for
3872 the specified entity. This suppression is effective from the point where
3873 it occurs till the end of the extended scope of the variable (similar to
3874 the scope of @code{Suppress}).
3876 @node Pragma Weak_External
3877 @unnumberedsec Pragma Weak_External
3878 @findex Weak_External
3882 @smallexample @c ada
3883 pragma Weak_External ([Entity =>] LOCAL_NAME);
3887 This pragma specifies that the given entity should be marked as a weak
3888 external (one that does not have to be resolved) for the linker. For
3889 further details, consult the GCC manual.
3891 @node Implementation Defined Attributes
3892 @chapter Implementation Defined Attributes
3893 Ada 95 defines (throughout the Ada 95 reference manual,
3894 summarized in annex K),
3895 a set of attributes that provide useful additional functionality in all
3896 areas of the language. These language defined attributes are implemented
3897 in GNAT and work as described in the Ada 95 Reference Manual.
3899 In addition, Ada 95 allows implementations to define additional
3900 attributes whose meaning is defined by the implementation. GNAT provides
3901 a number of these implementation-dependent attributes which can be used
3902 to extend and enhance the functionality of the compiler. This section of
3903 the GNAT reference manual describes these additional attributes.
3905 Note that any program using these attributes may not be portable to
3906 other compilers (although GNAT implements this set of attributes on all
3907 platforms). Therefore if portability to other compilers is an important
3908 consideration, you should minimize the use of these attributes.
3919 * Default_Bit_Order::
3927 * Has_Discriminants::
3933 * Max_Interrupt_Priority::
3935 * Maximum_Alignment::
3939 * Passed_By_Reference::
3950 * Unconstrained_Array::
3951 * Universal_Literal_String::
3952 * Unrestricted_Access::
3960 @unnumberedsec Abort_Signal
3961 @findex Abort_Signal
3963 @code{Standard'Abort_Signal} (@code{Standard} is the only allowed
3964 prefix) provides the entity for the special exception used to signal
3965 task abort or asynchronous transfer of control. Normally this attribute
3966 should only be used in the tasking runtime (it is highly peculiar, and
3967 completely outside the normal semantics of Ada, for a user program to
3968 intercept the abort exception).
3971 @unnumberedsec Address_Size
3972 @cindex Size of @code{Address}
3973 @findex Address_Size
3975 @code{Standard'Address_Size} (@code{Standard} is the only allowed
3976 prefix) is a static constant giving the number of bits in an
3977 @code{Address}. It is the same value as System.Address'Size,
3978 but has the advantage of being static, while a direct
3979 reference to System.Address'Size is non-static because Address
3983 @unnumberedsec Asm_Input
3986 The @code{Asm_Input} attribute denotes a function that takes two
3987 parameters. The first is a string, the second is an expression of the
3988 type designated by the prefix. The first (string) argument is required
3989 to be a static expression, and is the constraint for the parameter,
3990 (e.g.@: what kind of register is required). The second argument is the
3991 value to be used as the input argument. The possible values for the
3992 constant are the same as those used in the RTL, and are dependent on
3993 the configuration file used to built the GCC back end.
3994 @ref{Machine Code Insertions}
3997 @unnumberedsec Asm_Output
4000 The @code{Asm_Output} attribute denotes a function that takes two
4001 parameters. The first is a string, the second is the name of a variable
4002 of the type designated by the attribute prefix. The first (string)
4003 argument is required to be a static expression and designates the
4004 constraint for the parameter (e.g.@: what kind of register is
4005 required). The second argument is the variable to be updated with the
4006 result. The possible values for constraint are the same as those used in
4007 the RTL, and are dependent on the configuration file used to build the
4008 GCC back end. If there are no output operands, then this argument may
4009 either be omitted, or explicitly given as @code{No_Output_Operands}.
4010 @ref{Machine Code Insertions}
4013 @unnumberedsec AST_Entry
4017 This attribute is implemented only in OpenVMS versions of GNAT@. Applied to
4018 the name of an entry, it yields a value of the predefined type AST_Handler
4019 (declared in the predefined package System, as extended by the use of
4020 pragma @code{Extend_System (Aux_DEC)}). This value enables the given entry to
4021 be called when an AST occurs. For further details, refer to the @cite{DEC Ada
4022 Language Reference Manual}, section 9.12a.
4027 @code{@var{obj}'Bit}, where @var{obj} is any object, yields the bit
4028 offset within the storage unit (byte) that contains the first bit of
4029 storage allocated for the object. The value of this attribute is of the
4030 type @code{Universal_Integer}, and is always a non-negative number not
4031 exceeding the value of @code{System.Storage_Unit}.
4033 For an object that is a variable or a constant allocated in a register,
4034 the value is zero. (The use of this attribute does not force the
4035 allocation of a variable to memory).
4037 For an object that is a formal parameter, this attribute applies
4038 to either the matching actual parameter or to a copy of the
4039 matching actual parameter.
4041 For an access object the value is zero. Note that
4042 @code{@var{obj}.all'Bit} is subject to an @code{Access_Check} for the
4043 designated object. Similarly for a record component
4044 @code{@var{X}.@var{C}'Bit} is subject to a discriminant check and
4045 @code{@var{X}(@var{I}).Bit} and @code{@var{X}(@var{I1}..@var{I2})'Bit}
4046 are subject to index checks.
4048 This attribute is designed to be compatible with the DEC Ada 83 definition
4049 and implementation of the @code{Bit} attribute.
4052 @unnumberedsec Bit_Position
4053 @findex Bit_Position
4055 @code{@var{R.C}'Bit}, where @var{R} is a record object and C is one
4056 of the fields of the record type, yields the bit
4057 offset within the record contains the first bit of
4058 storage allocated for the object. The value of this attribute is of the
4059 type @code{Universal_Integer}. The value depends only on the field
4060 @var{C} and is independent of the alignment of
4061 the containing record @var{R}.
4064 @unnumberedsec Code_Address
4065 @findex Code_Address
4066 @cindex Subprogram address
4067 @cindex Address of subprogram code
4070 attribute may be applied to subprograms in Ada 95, but the
4071 intended effect from the Ada 95 reference manual seems to be to provide
4072 an address value which can be used to call the subprogram by means of
4073 an address clause as in the following example:
4075 @smallexample @c ada
4076 procedure K is @dots{}
4079 for L'Address use K'Address;
4080 pragma Import (Ada, L);
4084 A call to @code{L} is then expected to result in a call to @code{K}@.
4085 In Ada 83, where there were no access-to-subprogram values, this was
4086 a common work around for getting the effect of an indirect call.
4087 GNAT implements the above use of @code{Address} and the technique
4088 illustrated by the example code works correctly.
4090 However, for some purposes, it is useful to have the address of the start
4091 of the generated code for the subprogram. On some architectures, this is
4092 not necessarily the same as the @code{Address} value described above.
4093 For example, the @code{Address} value may reference a subprogram
4094 descriptor rather than the subprogram itself.
4096 The @code{'Code_Address} attribute, which can only be applied to
4097 subprogram entities, always returns the address of the start of the
4098 generated code of the specified subprogram, which may or may not be
4099 the same value as is returned by the corresponding @code{'Address}
4102 @node Default_Bit_Order
4103 @unnumberedsec Default_Bit_Order
4105 @cindex Little endian
4106 @findex Default_Bit_Order
4108 @code{Standard'Default_Bit_Order} (@code{Standard} is the only
4109 permissible prefix), provides the value @code{System.Default_Bit_Order}
4110 as a @code{Pos} value (0 for @code{High_Order_First}, 1 for
4111 @code{Low_Order_First}). This is used to construct the definition of
4112 @code{Default_Bit_Order} in package @code{System}.
4115 @unnumberedsec Elaborated
4118 The prefix of the @code{'Elaborated} attribute must be a unit name. The
4119 value is a Boolean which indicates whether or not the given unit has been
4120 elaborated. This attribute is primarily intended for internal use by the
4121 generated code for dynamic elaboration checking, but it can also be used
4122 in user programs. The value will always be True once elaboration of all
4123 units has been completed. An exception is for units which need no
4124 elaboration, the value is always False for such units.
4127 @unnumberedsec Elab_Body
4130 This attribute can only be applied to a program unit name. It returns
4131 the entity for the corresponding elaboration procedure for elaborating
4132 the body of the referenced unit. This is used in the main generated
4133 elaboration procedure by the binder and is not normally used in any
4134 other context. However, there may be specialized situations in which it
4135 is useful to be able to call this elaboration procedure from Ada code,
4136 e.g.@: if it is necessary to do selective re-elaboration to fix some
4140 @unnumberedsec Elab_Spec
4143 This attribute can only be applied to a program unit name. It returns
4144 the entity for the corresponding elaboration procedure for elaborating
4145 the specification of the referenced unit. This is used in the main
4146 generated elaboration procedure by the binder and is not normally used
4147 in any other context. However, there may be specialized situations in
4148 which it is useful to be able to call this elaboration procedure from
4149 Ada code, e.g.@: if it is necessary to do selective re-elaboration to fix
4154 @cindex Ada 83 attributes
4157 The @code{Emax} attribute is provided for compatibility with Ada 83. See
4158 the Ada 83 reference manual for an exact description of the semantics of
4162 @unnumberedsec Enum_Rep
4163 @cindex Representation of enums
4166 For every enumeration subtype @var{S}, @code{@var{S}'Enum_Rep} denotes a
4167 function with the following spec:
4169 @smallexample @c ada
4170 function @var{S}'Enum_Rep (Arg : @var{S}'Base)
4171 return @i{Universal_Integer};
4175 It is also allowable to apply @code{Enum_Rep} directly to an object of an
4176 enumeration type or to a non-overloaded enumeration
4177 literal. In this case @code{@var{S}'Enum_Rep} is equivalent to
4178 @code{@var{typ}'Enum_Rep(@var{S})} where @var{typ} is the type of the
4179 enumeration literal or object.
4181 The function returns the representation value for the given enumeration
4182 value. This will be equal to value of the @code{Pos} attribute in the
4183 absence of an enumeration representation clause. This is a static
4184 attribute (i.e.@: the result is static if the argument is static).
4186 @code{@var{S}'Enum_Rep} can also be used with integer types and objects,
4187 in which case it simply returns the integer value. The reason for this
4188 is to allow it to be used for @code{(<>)} discrete formal arguments in
4189 a generic unit that can be instantiated with either enumeration types
4190 or integer types. Note that if @code{Enum_Rep} is used on a modular
4191 type whose upper bound exceeds the upper bound of the largest signed
4192 integer type, and the argument is a variable, so that the universal
4193 integer calculation is done at run-time, then the call to @code{Enum_Rep}
4194 may raise @code{Constraint_Error}.
4197 @unnumberedsec Epsilon
4198 @cindex Ada 83 attributes
4201 The @code{Epsilon} attribute is provided for compatibility with Ada 83. See
4202 the Ada 83 reference manual for an exact description of the semantics of
4206 @unnumberedsec Fixed_Value
4209 For every fixed-point type @var{S}, @code{@var{S}'Fixed_Value} denotes a
4210 function with the following specification:
4212 @smallexample @c ada
4213 function @var{S}'Fixed_Value (Arg : @i{Universal_Integer})
4218 The value returned is the fixed-point value @var{V} such that
4220 @smallexample @c ada
4221 @var{V} = Arg * @var{S}'Small
4225 The effect is thus similar to first converting the argument to the
4226 integer type used to represent @var{S}, and then doing an unchecked
4227 conversion to the fixed-point type. The difference is
4228 that there are full range checks, to ensure that the result is in range.
4229 This attribute is primarily intended for use in implementation of the
4230 input-output functions for fixed-point values.
4232 @node Has_Discriminants
4233 @unnumberedsec Has_Discriminants
4234 @cindex Discriminants, testing for
4235 @findex Has_Discriminants
4237 The prefix of the @code{Has_Discriminants} attribute is a type. The result
4238 is a Boolean value which is True if the type has discriminants, and False
4239 otherwise. The intended use of this attribute is in conjunction with generic
4240 definitions. If the attribute is applied to a generic private type, it
4241 indicates whether or not the corresponding actual type has discriminants.
4247 The @code{Img} attribute differs from @code{Image} in that it may be
4248 applied to objects as well as types, in which case it gives the
4249 @code{Image} for the subtype of the object. This is convenient for
4252 @smallexample @c ada
4253 Put_Line ("X = " & X'Img);
4257 has the same meaning as the more verbose:
4259 @smallexample @c ada
4260 Put_Line ("X = " & @var{T}'Image (X));
4264 where @var{T} is the (sub)type of the object @code{X}.
4267 @unnumberedsec Integer_Value
4268 @findex Integer_Value
4270 For every integer type @var{S}, @code{@var{S}'Integer_Value} denotes a
4271 function with the following spec:
4273 @smallexample @c ada
4274 function @var{S}'Integer_Value (Arg : @i{Universal_Fixed})
4279 The value returned is the integer value @var{V}, such that
4281 @smallexample @c ada
4282 Arg = @var{V} * @var{T}'Small
4286 where @var{T} is the type of @code{Arg}.
4287 The effect is thus similar to first doing an unchecked conversion from
4288 the fixed-point type to its corresponding implementation type, and then
4289 converting the result to the target integer type. The difference is
4290 that there are full range checks, to ensure that the result is in range.
4291 This attribute is primarily intended for use in implementation of the
4292 standard input-output functions for fixed-point values.
4295 @unnumberedsec Large
4296 @cindex Ada 83 attributes
4299 The @code{Large} attribute is provided for compatibility with Ada 83. See
4300 the Ada 83 reference manual for an exact description of the semantics of
4304 @unnumberedsec Machine_Size
4305 @findex Machine_Size
4307 This attribute is identical to the @code{Object_Size} attribute. It is
4308 provided for compatibility with the DEC Ada 83 attribute of this name.
4311 @unnumberedsec Mantissa
4312 @cindex Ada 83 attributes
4315 The @code{Mantissa} attribute is provided for compatibility with Ada 83. See
4316 the Ada 83 reference manual for an exact description of the semantics of
4319 @node Max_Interrupt_Priority
4320 @unnumberedsec Max_Interrupt_Priority
4321 @cindex Interrupt priority, maximum
4322 @findex Max_Interrupt_Priority
4324 @code{Standard'Max_Interrupt_Priority} (@code{Standard} is the only
4325 permissible prefix), provides the same value as
4326 @code{System.Max_Interrupt_Priority}.
4329 @unnumberedsec Max_Priority
4330 @cindex Priority, maximum
4331 @findex Max_Priority
4333 @code{Standard'Max_Priority} (@code{Standard} is the only permissible
4334 prefix) provides the same value as @code{System.Max_Priority}.
4336 @node Maximum_Alignment
4337 @unnumberedsec Maximum_Alignment
4338 @cindex Alignment, maximum
4339 @findex Maximum_Alignment
4341 @code{Standard'Maximum_Alignment} (@code{Standard} is the only
4342 permissible prefix) provides the maximum useful alignment value for the
4343 target. This is a static value that can be used to specify the alignment
4344 for an object, guaranteeing that it is properly aligned in all
4347 @node Mechanism_Code
4348 @unnumberedsec Mechanism_Code
4349 @cindex Return values, passing mechanism
4350 @cindex Parameters, passing mechanism
4351 @findex Mechanism_Code
4353 @code{@var{function}'Mechanism_Code} yields an integer code for the
4354 mechanism used for the result of function, and
4355 @code{@var{subprogram}'Mechanism_Code (@var{n})} yields the mechanism
4356 used for formal parameter number @var{n} (a static integer value with 1
4357 meaning the first parameter) of @var{subprogram}. The code returned is:
4365 by descriptor (default descriptor class)
4367 by descriptor (UBS: unaligned bit string)
4369 by descriptor (UBSB: aligned bit string with arbitrary bounds)
4371 by descriptor (UBA: unaligned bit array)
4373 by descriptor (S: string, also scalar access type parameter)
4375 by descriptor (SB: string with arbitrary bounds)
4377 by descriptor (A: contiguous array)
4379 by descriptor (NCA: non-contiguous array)
4383 Values from 3 through 10 are only relevant to Digital OpenVMS implementations.
4386 @node Null_Parameter
4387 @unnumberedsec Null_Parameter
4388 @cindex Zero address, passing
4389 @findex Null_Parameter
4391 A reference @code{@var{T}'Null_Parameter} denotes an imaginary object of
4392 type or subtype @var{T} allocated at machine address zero. The attribute
4393 is allowed only as the default expression of a formal parameter, or as
4394 an actual expression of a subprogram call. In either case, the
4395 subprogram must be imported.
4397 The identity of the object is represented by the address zero in the
4398 argument list, independent of the passing mechanism (explicit or
4401 This capability is needed to specify that a zero address should be
4402 passed for a record or other composite object passed by reference.
4403 There is no way of indicating this without the @code{Null_Parameter}
4407 @unnumberedsec Object_Size
4408 @cindex Size, used for objects
4411 The size of an object is not necessarily the same as the size of the type
4412 of an object. This is because by default object sizes are increased to be
4413 a multiple of the alignment of the object. For example,
4414 @code{Natural'Size} is
4415 31, but by default objects of type @code{Natural} will have a size of 32 bits.
4416 Similarly, a record containing an integer and a character:
4418 @smallexample @c ada
4426 will have a size of 40 (that is @code{Rec'Size} will be 40. The
4427 alignment will be 4, because of the
4428 integer field, and so the default size of record objects for this type
4429 will be 64 (8 bytes).
4431 The @code{@var{type}'Object_Size} attribute
4432 has been added to GNAT to allow the
4433 default object size of a type to be easily determined. For example,
4434 @code{Natural'Object_Size} is 32, and
4435 @code{Rec'Object_Size} (for the record type in the above example) will be
4436 64. Note also that, unlike the situation with the
4437 @code{Size} attribute as defined in the Ada RM, the
4438 @code{Object_Size} attribute can be specified individually
4439 for different subtypes. For example:
4441 @smallexample @c ada
4442 type R is new Integer;
4443 subtype R1 is R range 1 .. 10;
4444 subtype R2 is R range 1 .. 10;
4445 for R2'Object_Size use 8;
4449 In this example, @code{R'Object_Size} and @code{R1'Object_Size} are both
4450 32 since the default object size for a subtype is the same as the object size
4451 for the parent subtype. This means that objects of type @code{R}
4453 by default be 32 bits (four bytes). But objects of type
4454 @code{R2} will be only
4455 8 bits (one byte), since @code{R2'Object_Size} has been set to 8.
4457 @node Passed_By_Reference
4458 @unnumberedsec Passed_By_Reference
4459 @cindex Parameters, when passed by reference
4460 @findex Passed_By_Reference
4462 @code{@var{type}'Passed_By_Reference} for any subtype @var{type} returns
4463 a value of type @code{Boolean} value that is @code{True} if the type is
4464 normally passed by reference and @code{False} if the type is normally
4465 passed by copy in calls. For scalar types, the result is always @code{False}
4466 and is static. For non-scalar types, the result is non-static.
4469 @unnumberedsec Range_Length
4470 @findex Range_Length
4472 @code{@var{type}'Range_Length} for any discrete type @var{type} yields
4473 the number of values represented by the subtype (zero for a null
4474 range). The result is static for static subtypes. @code{Range_Length}
4475 applied to the index subtype of a one dimensional array always gives the
4476 same result as @code{Range} applied to the array itself.
4479 @unnumberedsec Safe_Emax
4480 @cindex Ada 83 attributes
4483 The @code{Safe_Emax} attribute is provided for compatibility with Ada 83. See
4484 the Ada 83 reference manual for an exact description of the semantics of
4488 @unnumberedsec Safe_Large
4489 @cindex Ada 83 attributes
4492 The @code{Safe_Large} attribute is provided for compatibility with Ada 83. See
4493 the Ada 83 reference manual for an exact description of the semantics of
4497 @unnumberedsec Small
4498 @cindex Ada 83 attributes
4501 The @code{Small} attribute is defined in Ada 95 only for fixed-point types.
4502 GNAT also allows this attribute to be applied to floating-point types
4503 for compatibility with Ada 83. See
4504 the Ada 83 reference manual for an exact description of the semantics of
4505 this attribute when applied to floating-point types.
4508 @unnumberedsec Storage_Unit
4509 @findex Storage_Unit
4511 @code{Standard'Storage_Unit} (@code{Standard} is the only permissible
4512 prefix) provides the same value as @code{System.Storage_Unit}.
4515 @unnumberedsec Target_Name
4518 @code{Standard'Target_Name} (@code{Standard} is the only permissible
4519 prefix) provides a static string value that identifies the target
4520 for the current compilation. For GCC implementations, this is the
4521 standard gcc target name without the terminating slash (for
4522 example, GNAT 5.0 on windows yields "i586-pc-mingw32msv").
4528 @code{Standard'Tick} (@code{Standard} is the only permissible prefix)
4529 provides the same value as @code{System.Tick},
4532 @unnumberedsec To_Address
4535 The @code{System'To_Address}
4536 (@code{System} is the only permissible prefix)
4537 denotes a function identical to
4538 @code{System.Storage_Elements.To_Address} except that
4539 it is a static attribute. This means that if its argument is
4540 a static expression, then the result of the attribute is a
4541 static expression. The result is that such an expression can be
4542 used in contexts (e.g.@: preelaborable packages) which require a
4543 static expression and where the function call could not be used
4544 (since the function call is always non-static, even if its
4545 argument is static).
4548 @unnumberedsec Type_Class
4551 @code{@var{type}'Type_Class} for any type or subtype @var{type} yields
4552 the value of the type class for the full type of @var{type}. If
4553 @var{type} is a generic formal type, the value is the value for the
4554 corresponding actual subtype. The value of this attribute is of type
4555 @code{System.Aux_DEC.Type_Class}, which has the following definition:
4557 @smallexample @c ada
4559 (Type_Class_Enumeration,
4561 Type_Class_Fixed_Point,
4562 Type_Class_Floating_Point,
4567 Type_Class_Address);
4571 Protected types yield the value @code{Type_Class_Task}, which thus
4572 applies to all concurrent types. This attribute is designed to
4573 be compatible with the DEC Ada 83 attribute of the same name.
4576 @unnumberedsec UET_Address
4579 The @code{UET_Address} attribute can only be used for a prefix which
4580 denotes a library package. It yields the address of the unit exception
4581 table when zero cost exception handling is used. This attribute is
4582 intended only for use within the GNAT implementation. See the unit
4583 @code{Ada.Exceptions} in files @file{a-except.ads} and @file{a-except.adb}
4584 for details on how this attribute is used in the implementation.
4586 @node Unconstrained_Array
4587 @unnumberedsec Unconstrained_Array
4588 @findex Unconstrained_Array
4590 The @code{Unconstrained_Array} attribute can be used with a prefix that
4591 denotes any type or subtype. It is a static attribute that yields
4592 @code{True} if the prefix designates an unconstrained array,
4593 and @code{False} otherwise. In a generic instance, the result is
4594 still static, and yields the result of applying this test to the
4597 @node Universal_Literal_String
4598 @unnumberedsec Universal_Literal_String
4599 @cindex Named numbers, representation of
4600 @findex Universal_Literal_String
4602 The prefix of @code{Universal_Literal_String} must be a named
4603 number. The static result is the string consisting of the characters of
4604 the number as defined in the original source. This allows the user
4605 program to access the actual text of named numbers without intermediate
4606 conversions and without the need to enclose the strings in quotes (which
4607 would preclude their use as numbers). This is used internally for the
4608 construction of values of the floating-point attributes from the file
4609 @file{ttypef.ads}, but may also be used by user programs.
4611 @node Unrestricted_Access
4612 @unnumberedsec Unrestricted_Access
4613 @cindex @code{Access}, unrestricted
4614 @findex Unrestricted_Access
4616 The @code{Unrestricted_Access} attribute is similar to @code{Access}
4617 except that all accessibility and aliased view checks are omitted. This
4618 is a user-beware attribute. It is similar to
4619 @code{Address}, for which it is a desirable replacement where the value
4620 desired is an access type. In other words, its effect is identical to
4621 first applying the @code{Address} attribute and then doing an unchecked
4622 conversion to a desired access type. In GNAT, but not necessarily in
4623 other implementations, the use of static chains for inner level
4624 subprograms means that @code{Unrestricted_Access} applied to a
4625 subprogram yields a value that can be called as long as the subprogram
4626 is in scope (normal Ada 95 accessibility rules restrict this usage).
4628 It is possible to use @code{Unrestricted_Access} for any type, but care
4629 must be excercised if it is used to create pointers to unconstrained
4630 objects. In this case, the resulting pointer has the same scope as the
4631 context of the attribute, and may not be returned to some enclosing
4632 scope. For instance, a function cannot use @code{Unrestricted_Access}
4633 to create a unconstrained pointer and then return that value to the
4637 @unnumberedsec VADS_Size
4638 @cindex @code{Size}, VADS compatibility
4641 The @code{'VADS_Size} attribute is intended to make it easier to port
4642 legacy code which relies on the semantics of @code{'Size} as implemented
4643 by the VADS Ada 83 compiler. GNAT makes a best effort at duplicating the
4644 same semantic interpretation. In particular, @code{'VADS_Size} applied
4645 to a predefined or other primitive type with no Size clause yields the
4646 Object_Size (for example, @code{Natural'Size} is 32 rather than 31 on
4647 typical machines). In addition @code{'VADS_Size} applied to an object
4648 gives the result that would be obtained by applying the attribute to
4649 the corresponding type.
4652 @unnumberedsec Value_Size
4653 @cindex @code{Size}, setting for not-first subtype
4655 @code{@var{type}'Value_Size} is the number of bits required to represent
4656 a value of the given subtype. It is the same as @code{@var{type}'Size},
4657 but, unlike @code{Size}, may be set for non-first subtypes.
4660 @unnumberedsec Wchar_T_Size
4661 @findex Wchar_T_Size
4662 @code{Standard'Wchar_T_Size} (@code{Standard} is the only permissible
4663 prefix) provides the size in bits of the C @code{wchar_t} type
4664 primarily for constructing the definition of this type in
4665 package @code{Interfaces.C}.
4668 @unnumberedsec Word_Size
4670 @code{Standard'Word_Size} (@code{Standard} is the only permissible
4671 prefix) provides the value @code{System.Word_Size}.
4673 @c ------------------------
4674 @node Implementation Advice
4675 @chapter Implementation Advice
4677 The main text of the Ada 95 Reference Manual describes the required
4678 behavior of all Ada 95 compilers, and the GNAT compiler conforms to
4681 In addition, there are sections throughout the Ada 95
4682 reference manual headed
4683 by the phrase ``implementation advice''. These sections are not normative,
4684 i.e.@: they do not specify requirements that all compilers must
4685 follow. Rather they provide advice on generally desirable behavior. You
4686 may wonder why they are not requirements. The most typical answer is
4687 that they describe behavior that seems generally desirable, but cannot
4688 be provided on all systems, or which may be undesirable on some systems.
4690 As far as practical, GNAT follows the implementation advice sections in
4691 the Ada 95 Reference Manual. This chapter contains a table giving the
4692 reference manual section number, paragraph number and several keywords
4693 for each advice. Each entry consists of the text of the advice followed
4694 by the GNAT interpretation of this advice. Most often, this simply says
4695 ``followed'', which means that GNAT follows the advice. However, in a
4696 number of cases, GNAT deliberately deviates from this advice, in which
4697 case the text describes what GNAT does and why.
4699 @cindex Error detection
4700 @unnumberedsec 1.1.3(20): Error Detection
4703 If an implementation detects the use of an unsupported Specialized Needs
4704 Annex feature at run time, it should raise @code{Program_Error} if
4707 Not relevant. All specialized needs annex features are either supported,
4708 or diagnosed at compile time.
4711 @unnumberedsec 1.1.3(31): Child Units
4714 If an implementation wishes to provide implementation-defined
4715 extensions to the functionality of a language-defined library unit, it
4716 should normally do so by adding children to the library unit.
4720 @cindex Bounded errors
4721 @unnumberedsec 1.1.5(12): Bounded Errors
4724 If an implementation detects a bounded error or erroneous
4725 execution, it should raise @code{Program_Error}.
4727 Followed in all cases in which the implementation detects a bounded
4728 error or erroneous execution. Not all such situations are detected at
4732 @unnumberedsec 2.8(16): Pragmas
4735 Normally, implementation-defined pragmas should have no semantic effect
4736 for error-free programs; that is, if the implementation-defined pragmas
4737 are removed from a working program, the program should still be legal,
4738 and should still have the same semantics.
4740 The following implementation defined pragmas are exceptions to this
4752 @item CPP_Constructor
4760 @item Interface_Name
4762 @item Machine_Attribute
4764 @item Unimplemented_Unit
4766 @item Unchecked_Union
4771 In each of the above cases, it is essential to the purpose of the pragma
4772 that this advice not be followed. For details see the separate section
4773 on implementation defined pragmas.
4775 @unnumberedsec 2.8(17-19): Pragmas
4778 Normally, an implementation should not define pragmas that can
4779 make an illegal program legal, except as follows:
4783 A pragma used to complete a declaration, such as a pragma @code{Import};
4787 A pragma used to configure the environment by adding, removing, or
4788 replacing @code{library_items}.
4790 See response to paragraph 16 of this same section.
4792 @cindex Character Sets
4793 @cindex Alternative Character Sets
4794 @unnumberedsec 3.5.2(5): Alternative Character Sets
4797 If an implementation supports a mode with alternative interpretations
4798 for @code{Character} and @code{Wide_Character}, the set of graphic
4799 characters of @code{Character} should nevertheless remain a proper
4800 subset of the set of graphic characters of @code{Wide_Character}. Any
4801 character set ``localizations'' should be reflected in the results of
4802 the subprograms defined in the language-defined package
4803 @code{Characters.Handling} (see A.3) available in such a mode. In a mode with
4804 an alternative interpretation of @code{Character}, the implementation should
4805 also support a corresponding change in what is a legal
4806 @code{identifier_letter}.
4808 Not all wide character modes follow this advice, in particular the JIS
4809 and IEC modes reflect standard usage in Japan, and in these encoding,
4810 the upper half of the Latin-1 set is not part of the wide-character
4811 subset, since the most significant bit is used for wide character
4812 encoding. However, this only applies to the external forms. Internally
4813 there is no such restriction.
4815 @cindex Integer types
4816 @unnumberedsec 3.5.4(28): Integer Types
4820 An implementation should support @code{Long_Integer} in addition to
4821 @code{Integer} if the target machine supports 32-bit (or longer)
4822 arithmetic. No other named integer subtypes are recommended for package
4823 @code{Standard}. Instead, appropriate named integer subtypes should be
4824 provided in the library package @code{Interfaces} (see B.2).
4826 @code{Long_Integer} is supported. Other standard integer types are supported
4827 so this advice is not fully followed. These types
4828 are supported for convenient interface to C, and so that all hardware
4829 types of the machine are easily available.
4830 @unnumberedsec 3.5.4(29): Integer Types
4834 An implementation for a two's complement machine should support
4835 modular types with a binary modulus up to @code{System.Max_Int*2+2}. An
4836 implementation should support a non-binary modules up to @code{Integer'Last}.
4840 @cindex Enumeration values
4841 @unnumberedsec 3.5.5(8): Enumeration Values
4844 For the evaluation of a call on @code{@var{S}'Pos} for an enumeration
4845 subtype, if the value of the operand does not correspond to the internal
4846 code for any enumeration literal of its type (perhaps due to an
4847 un-initialized variable), then the implementation should raise
4848 @code{Program_Error}. This is particularly important for enumeration
4849 types with noncontiguous internal codes specified by an
4850 enumeration_representation_clause.
4855 @unnumberedsec 3.5.7(17): Float Types
4858 An implementation should support @code{Long_Float} in addition to
4859 @code{Float} if the target machine supports 11 or more digits of
4860 precision. No other named floating point subtypes are recommended for
4861 package @code{Standard}. Instead, appropriate named floating point subtypes
4862 should be provided in the library package @code{Interfaces} (see B.2).
4864 @code{Short_Float} and @code{Long_Long_Float} are also provided. The
4865 former provides improved compatibility with other implementations
4866 supporting this type. The latter corresponds to the highest precision
4867 floating-point type supported by the hardware. On most machines, this
4868 will be the same as @code{Long_Float}, but on some machines, it will
4869 correspond to the IEEE extended form. The notable case is all ia32
4870 (x86) implementations, where @code{Long_Long_Float} corresponds to
4871 the 80-bit extended precision format supported in hardware on this
4872 processor. Note that the 128-bit format on SPARC is not supported,
4873 since this is a software rather than a hardware format.
4875 @cindex Multidimensional arrays
4876 @cindex Arrays, multidimensional
4877 @unnumberedsec 3.6.2(11): Multidimensional Arrays
4880 An implementation should normally represent multidimensional arrays in
4881 row-major order, consistent with the notation used for multidimensional
4882 array aggregates (see 4.3.3). However, if a pragma @code{Convention}
4883 (@code{Fortran}, @dots{}) applies to a multidimensional array type, then
4884 column-major order should be used instead (see B.5, ``Interfacing with
4889 @findex Duration'Small
4890 @unnumberedsec 9.6(30-31): Duration'Small
4893 Whenever possible in an implementation, the value of @code{Duration'Small}
4894 should be no greater than 100 microseconds.
4896 Followed. (@code{Duration'Small} = 10**(@minus{}9)).
4900 The time base for @code{delay_relative_statements} should be monotonic;
4901 it need not be the same time base as used for @code{Calendar.Clock}.
4905 @unnumberedsec 10.2.1(12): Consistent Representation
4908 In an implementation, a type declared in a pre-elaborated package should
4909 have the same representation in every elaboration of a given version of
4910 the package, whether the elaborations occur in distinct executions of
4911 the same program, or in executions of distinct programs or partitions
4912 that include the given version.
4914 Followed, except in the case of tagged types. Tagged types involve
4915 implicit pointers to a local copy of a dispatch table, and these pointers
4916 have representations which thus depend on a particular elaboration of the
4917 package. It is not easy to see how it would be possible to follow this
4918 advice without severely impacting efficiency of execution.
4920 @cindex Exception information
4921 @unnumberedsec 11.4.1(19): Exception Information
4924 @code{Exception_Message} by default and @code{Exception_Information}
4925 should produce information useful for
4926 debugging. @code{Exception_Message} should be short, about one
4927 line. @code{Exception_Information} can be long. @code{Exception_Message}
4928 should not include the
4929 @code{Exception_Name}. @code{Exception_Information} should include both
4930 the @code{Exception_Name} and the @code{Exception_Message}.
4932 Followed. For each exception that doesn't have a specified
4933 @code{Exception_Message}, the compiler generates one containing the location
4934 of the raise statement. This location has the form ``file:line'', where
4935 file is the short file name (without path information) and line is the line
4936 number in the file. Note that in the case of the Zero Cost Exception
4937 mechanism, these messages become redundant with the Exception_Information that
4938 contains a full backtrace of the calling sequence, so they are disabled.
4939 To disable explicitly the generation of the source location message, use the
4940 Pragma @code{Discard_Names}.
4942 @cindex Suppression of checks
4943 @cindex Checks, suppression of
4944 @unnumberedsec 11.5(28): Suppression of Checks
4947 The implementation should minimize the code executed for checks that
4948 have been suppressed.
4952 @cindex Representation clauses
4953 @unnumberedsec 13.1 (21-24): Representation Clauses
4956 The recommended level of support for all representation items is
4957 qualified as follows:
4961 An implementation need not support representation items containing
4962 non-static expressions, except that an implementation should support a
4963 representation item for a given entity if each non-static expression in
4964 the representation item is a name that statically denotes a constant
4965 declared before the entity.
4967 Followed. GNAT does not support non-static expressions in representation
4968 clauses unless they are constants declared before the entity. For
4971 @smallexample @c ada
4973 for X'Address use To_address (16#2000#);
4977 will be rejected, since the To_Address expression is non-static. Instead
4980 @smallexample @c ada
4981 X_Address : constant Address : = To_Address (16#2000#);
4983 for X'Address use X_Address;
4988 An implementation need not support a specification for the @code{Size}
4989 for a given composite subtype, nor the size or storage place for an
4990 object (including a component) of a given composite subtype, unless the
4991 constraints on the subtype and its composite subcomponents (if any) are
4992 all static constraints.
4994 Followed. Size Clauses are not permitted on non-static components, as
4999 An aliased component, or a component whose type is by-reference, should
5000 always be allocated at an addressable location.
5004 @cindex Packed types
5005 @unnumberedsec 13.2(6-8): Packed Types
5008 If a type is packed, then the implementation should try to minimize
5009 storage allocated to objects of the type, possibly at the expense of
5010 speed of accessing components, subject to reasonable complexity in
5011 addressing calculations.
5015 The recommended level of support pragma @code{Pack} is:
5017 For a packed record type, the components should be packed as tightly as
5018 possible subject to the Sizes of the component subtypes, and subject to
5019 any @code{record_representation_clause} that applies to the type; the
5020 implementation may, but need not, reorder components or cross aligned
5021 word boundaries to improve the packing. A component whose @code{Size} is
5022 greater than the word size may be allocated an integral number of words.
5024 Followed. Tight packing of arrays is supported for all component sizes
5025 up to 64-bits. If the array component size is 1 (that is to say, if
5026 the component is a boolean type or an enumeration type with two values)
5027 then values of the type are implicitly initialized to zero. This
5028 happens both for objects of the packed type, and for objects that have a
5029 subcomponent of the packed type.
5033 An implementation should support Address clauses for imported
5037 @cindex @code{Address} clauses
5038 @unnumberedsec 13.3(14-19): Address Clauses
5042 For an array @var{X}, @code{@var{X}'Address} should point at the first
5043 component of the array, and not at the array bounds.
5049 The recommended level of support for the @code{Address} attribute is:
5051 @code{@var{X}'Address} should produce a useful result if @var{X} is an
5052 object that is aliased or of a by-reference type, or is an entity whose
5053 @code{Address} has been specified.
5055 Followed. A valid address will be produced even if none of those
5056 conditions have been met. If necessary, the object is forced into
5057 memory to ensure the address is valid.
5061 An implementation should support @code{Address} clauses for imported
5068 Objects (including subcomponents) that are aliased or of a by-reference
5069 type should be allocated on storage element boundaries.
5075 If the @code{Address} of an object is specified, or it is imported or exported,
5076 then the implementation should not perform optimizations based on
5077 assumptions of no aliases.
5081 @cindex @code{Alignment} clauses
5082 @unnumberedsec 13.3(29-35): Alignment Clauses
5085 The recommended level of support for the @code{Alignment} attribute for
5088 An implementation should support specified Alignments that are factors
5089 and multiples of the number of storage elements per word, subject to the
5096 An implementation need not support specified @code{Alignment}s for
5097 combinations of @code{Size}s and @code{Alignment}s that cannot be easily
5098 loaded and stored by available machine instructions.
5104 An implementation need not support specified @code{Alignment}s that are
5105 greater than the maximum @code{Alignment} the implementation ever returns by
5112 The recommended level of support for the @code{Alignment} attribute for
5115 Same as above, for subtypes, but in addition:
5121 For stand-alone library-level objects of statically constrained
5122 subtypes, the implementation should support all @code{Alignment}s
5123 supported by the target linker. For example, page alignment is likely to
5124 be supported for such objects, but not for subtypes.
5128 @cindex @code{Size} clauses
5129 @unnumberedsec 13.3(42-43): Size Clauses
5132 The recommended level of support for the @code{Size} attribute of
5135 A @code{Size} clause should be supported for an object if the specified
5136 @code{Size} is at least as large as its subtype's @code{Size}, and
5137 corresponds to a size in storage elements that is a multiple of the
5138 object's @code{Alignment} (if the @code{Alignment} is nonzero).
5142 @unnumberedsec 13.3(50-56): Size Clauses
5145 If the @code{Size} of a subtype is specified, and allows for efficient
5146 independent addressability (see 9.10) on the target architecture, then
5147 the @code{Size} of the following objects of the subtype should equal the
5148 @code{Size} of the subtype:
5150 Aliased objects (including components).
5156 @code{Size} clause on a composite subtype should not affect the
5157 internal layout of components.
5163 The recommended level of support for the @code{Size} attribute of subtypes is:
5167 The @code{Size} (if not specified) of a static discrete or fixed point
5168 subtype should be the number of bits needed to represent each value
5169 belonging to the subtype using an unbiased representation, leaving space
5170 for a sign bit only if the subtype contains negative values. If such a
5171 subtype is a first subtype, then an implementation should support a
5172 specified @code{Size} for it that reflects this representation.
5178 For a subtype implemented with levels of indirection, the @code{Size}
5179 should include the size of the pointers, but not the size of what they
5184 @cindex @code{Component_Size} clauses
5185 @unnumberedsec 13.3(71-73): Component Size Clauses
5188 The recommended level of support for the @code{Component_Size}
5193 An implementation need not support specified @code{Component_Sizes} that are
5194 less than the @code{Size} of the component subtype.
5200 An implementation should support specified @code{Component_Size}s that
5201 are factors and multiples of the word size. For such
5202 @code{Component_Size}s, the array should contain no gaps between
5203 components. For other @code{Component_Size}s (if supported), the array
5204 should contain no gaps between components when packing is also
5205 specified; the implementation should forbid this combination in cases
5206 where it cannot support a no-gaps representation.
5210 @cindex Enumeration representation clauses
5211 @cindex Representation clauses, enumeration
5212 @unnumberedsec 13.4(9-10): Enumeration Representation Clauses
5215 The recommended level of support for enumeration representation clauses
5218 An implementation need not support enumeration representation clauses
5219 for boolean types, but should at minimum support the internal codes in
5220 the range @code{System.Min_Int.System.Max_Int}.
5224 @cindex Record representation clauses
5225 @cindex Representation clauses, records
5226 @unnumberedsec 13.5.1(17-22): Record Representation Clauses
5229 The recommended level of support for
5230 @*@code{record_representation_clauses} is:
5232 An implementation should support storage places that can be extracted
5233 with a load, mask, shift sequence of machine code, and set with a load,
5234 shift, mask, store sequence, given the available machine instructions
5241 A storage place should be supported if its size is equal to the
5242 @code{Size} of the component subtype, and it starts and ends on a
5243 boundary that obeys the @code{Alignment} of the component subtype.
5249 If the default bit ordering applies to the declaration of a given type,
5250 then for a component whose subtype's @code{Size} is less than the word
5251 size, any storage place that does not cross an aligned word boundary
5252 should be supported.
5258 An implementation may reserve a storage place for the tag field of a
5259 tagged type, and disallow other components from overlapping that place.
5261 Followed. The storage place for the tag field is the beginning of the tagged
5262 record, and its size is Address'Size. GNAT will reject an explicit component
5263 clause for the tag field.
5267 An implementation need not support a @code{component_clause} for a
5268 component of an extension part if the storage place is not after the
5269 storage places of all components of the parent type, whether or not
5270 those storage places had been specified.
5272 Followed. The above advice on record representation clauses is followed,
5273 and all mentioned features are implemented.
5275 @cindex Storage place attributes
5276 @unnumberedsec 13.5.2(5): Storage Place Attributes
5279 If a component is represented using some form of pointer (such as an
5280 offset) to the actual data of the component, and this data is contiguous
5281 with the rest of the object, then the storage place attributes should
5282 reflect the place of the actual data, not the pointer. If a component is
5283 allocated discontinuously from the rest of the object, then a warning
5284 should be generated upon reference to one of its storage place
5287 Followed. There are no such components in GNAT@.
5289 @cindex Bit ordering
5290 @unnumberedsec 13.5.3(7-8): Bit Ordering
5293 The recommended level of support for the non-default bit ordering is:
5297 If @code{Word_Size} = @code{Storage_Unit}, then the implementation
5298 should support the non-default bit ordering in addition to the default
5301 Followed. Word size does not equal storage size in this implementation.
5302 Thus non-default bit ordering is not supported.
5304 @cindex @code{Address}, as private type
5305 @unnumberedsec 13.7(37): Address as Private
5308 @code{Address} should be of a private type.
5312 @cindex Operations, on @code{Address}
5313 @cindex @code{Address}, operations of
5314 @unnumberedsec 13.7.1(16): Address Operations
5317 Operations in @code{System} and its children should reflect the target
5318 environment semantics as closely as is reasonable. For example, on most
5319 machines, it makes sense for address arithmetic to ``wrap around''.
5320 Operations that do not make sense should raise @code{Program_Error}.
5322 Followed. Address arithmetic is modular arithmetic that wraps around. No
5323 operation raises @code{Program_Error}, since all operations make sense.
5325 @cindex Unchecked conversion
5326 @unnumberedsec 13.9(14-17): Unchecked Conversion
5329 The @code{Size} of an array object should not include its bounds; hence,
5330 the bounds should not be part of the converted data.
5336 The implementation should not generate unnecessary run-time checks to
5337 ensure that the representation of @var{S} is a representation of the
5338 target type. It should take advantage of the permission to return by
5339 reference when possible. Restrictions on unchecked conversions should be
5340 avoided unless required by the target environment.
5342 Followed. There are no restrictions on unchecked conversion. A warning is
5343 generated if the source and target types do not have the same size since
5344 the semantics in this case may be target dependent.
5348 The recommended level of support for unchecked conversions is:
5352 Unchecked conversions should be supported and should be reversible in
5353 the cases where this clause defines the result. To enable meaningful use
5354 of unchecked conversion, a contiguous representation should be used for
5355 elementary subtypes, for statically constrained array subtypes whose
5356 component subtype is one of the subtypes described in this paragraph,
5357 and for record subtypes without discriminants whose component subtypes
5358 are described in this paragraph.
5362 @cindex Heap usage, implicit
5363 @unnumberedsec 13.11(23-25): Implicit Heap Usage
5366 An implementation should document any cases in which it dynamically
5367 allocates heap storage for a purpose other than the evaluation of an
5370 Followed, the only other points at which heap storage is dynamically
5371 allocated are as follows:
5375 At initial elaboration time, to allocate dynamically sized global
5379 To allocate space for a task when a task is created.
5382 To extend the secondary stack dynamically when needed. The secondary
5383 stack is used for returning variable length results.
5388 A default (implementation-provided) storage pool for an
5389 access-to-constant type should not have overhead to support deallocation of
5396 A storage pool for an anonymous access type should be created at the
5397 point of an allocator for the type, and be reclaimed when the designated
5398 object becomes inaccessible.
5402 @cindex Unchecked deallocation
5403 @unnumberedsec 13.11.2(17): Unchecked De-allocation
5406 For a standard storage pool, @code{Free} should actually reclaim the
5411 @cindex Stream oriented attributes
5412 @unnumberedsec 13.13.2(17): Stream Oriented Attributes
5415 If a stream element is the same size as a storage element, then the
5416 normal in-memory representation should be used by @code{Read} and
5417 @code{Write} for scalar objects. Otherwise, @code{Read} and @code{Write}
5418 should use the smallest number of stream elements needed to represent
5419 all values in the base range of the scalar type.
5422 Followed. By default, GNAT uses the interpretation suggested by AI-195,
5423 which specifies using the size of the first subtype.
5424 However, such an implementation is based on direct binary
5425 representations and is therefore target- and endianness-dependent.
5426 To address this issue, GNAT also supplies an alternate implementation
5427 of the stream attributes @code{Read} and @code{Write},
5428 which uses the target-independent XDR standard representation
5430 @cindex XDR representation
5431 @cindex @code{Read} attribute
5432 @cindex @code{Write} attribute
5433 @cindex Stream oriented attributes
5434 The XDR implementation is provided as an alternative body of the
5435 @code{System.Stream_Attributes} package, in the file
5436 @file{s-strxdr.adb} in the GNAT library.
5437 There is no @file{s-strxdr.ads} file.
5438 In order to install the XDR implementation, do the following:
5440 @item Replace the default implementation of the
5441 @code{System.Stream_Attributes} package with the XDR implementation.
5442 For example on a Unix platform issue the commands:
5444 $ mv s-stratt.adb s-strold.adb
5445 $ mv s-strxdr.adb s-stratt.adb
5449 Rebuild the GNAT run-time library as documented in the
5450 @cite{GNAT User's Guide}
5453 @unnumberedsec A.1(52): Names of Predefined Numeric Types
5456 If an implementation provides additional named predefined integer types,
5457 then the names should end with @samp{Integer} as in
5458 @samp{Long_Integer}. If an implementation provides additional named
5459 predefined floating point types, then the names should end with
5460 @samp{Float} as in @samp{Long_Float}.
5464 @findex Ada.Characters.Handling
5465 @unnumberedsec A.3.2(49): @code{Ada.Characters.Handling}
5468 If an implementation provides a localized definition of @code{Character}
5469 or @code{Wide_Character}, then the effects of the subprograms in
5470 @code{Characters.Handling} should reflect the localizations. See also
5473 Followed. GNAT provides no such localized definitions.
5475 @cindex Bounded-length strings
5476 @unnumberedsec A.4.4(106): Bounded-Length String Handling
5479 Bounded string objects should not be implemented by implicit pointers
5480 and dynamic allocation.
5482 Followed. No implicit pointers or dynamic allocation are used.
5484 @cindex Random number generation
5485 @unnumberedsec A.5.2(46-47): Random Number Generation
5488 Any storage associated with an object of type @code{Generator} should be
5489 reclaimed on exit from the scope of the object.
5495 If the generator period is sufficiently long in relation to the number
5496 of distinct initiator values, then each possible value of
5497 @code{Initiator} passed to @code{Reset} should initiate a sequence of
5498 random numbers that does not, in a practical sense, overlap the sequence
5499 initiated by any other value. If this is not possible, then the mapping
5500 between initiator values and generator states should be a rapidly
5501 varying function of the initiator value.
5503 Followed. The generator period is sufficiently long for the first
5504 condition here to hold true.
5506 @findex Get_Immediate
5507 @unnumberedsec A.10.7(23): @code{Get_Immediate}
5510 The @code{Get_Immediate} procedures should be implemented with
5511 unbuffered input. For a device such as a keyboard, input should be
5512 @dfn{available} if a key has already been typed, whereas for a disk
5513 file, input should always be available except at end of file. For a file
5514 associated with a keyboard-like device, any line-editing features of the
5515 underlying operating system should be disabled during the execution of
5516 @code{Get_Immediate}.
5518 Followed on all targets except VxWorks. For VxWorks, there is no way to
5519 provide this functionality that does not result in the input buffer being
5520 flushed before the @code{Get_Immediate} call. A special unit
5521 @code{Interfaces.Vxworks.IO} is provided that contains routines to enable
5525 @unnumberedsec B.1(39-41): Pragma @code{Export}
5528 If an implementation supports pragma @code{Export} to a given language,
5529 then it should also allow the main subprogram to be written in that
5530 language. It should support some mechanism for invoking the elaboration
5531 of the Ada library units included in the system, and for invoking the
5532 finalization of the environment task. On typical systems, the
5533 recommended mechanism is to provide two subprograms whose link names are
5534 @code{adainit} and @code{adafinal}. @code{adainit} should contain the
5535 elaboration code for library units. @code{adafinal} should contain the
5536 finalization code. These subprograms should have no effect the second
5537 and subsequent time they are called.
5543 Automatic elaboration of pre-elaborated packages should be
5544 provided when pragma @code{Export} is supported.
5546 Followed when the main program is in Ada. If the main program is in a
5547 foreign language, then
5548 @code{adainit} must be called to elaborate pre-elaborated
5553 For each supported convention @var{L} other than @code{Intrinsic}, an
5554 implementation should support @code{Import} and @code{Export} pragmas
5555 for objects of @var{L}-compatible types and for subprograms, and pragma
5556 @code{Convention} for @var{L}-eligible types and for subprograms,
5557 presuming the other language has corresponding features. Pragma
5558 @code{Convention} need not be supported for scalar types.
5562 @cindex Package @code{Interfaces}
5564 @unnumberedsec B.2(12-13): Package @code{Interfaces}
5567 For each implementation-defined convention identifier, there should be a
5568 child package of package Interfaces with the corresponding name. This
5569 package should contain any declarations that would be useful for
5570 interfacing to the language (implementation) represented by the
5571 convention. Any declarations useful for interfacing to any language on
5572 the given hardware architecture should be provided directly in
5575 Followed. An additional package not defined
5576 in the Ada 95 Reference Manual is @code{Interfaces.CPP}, used
5577 for interfacing to C++.
5581 An implementation supporting an interface to C, COBOL, or Fortran should
5582 provide the corresponding package or packages described in the following
5585 Followed. GNAT provides all the packages described in this section.
5587 @cindex C, interfacing with
5588 @unnumberedsec B.3(63-71): Interfacing with C
5591 An implementation should support the following interface correspondences
5598 An Ada procedure corresponds to a void-returning C function.
5604 An Ada function corresponds to a non-void C function.
5610 An Ada @code{in} scalar parameter is passed as a scalar argument to a C
5617 An Ada @code{in} parameter of an access-to-object type with designated
5618 type @var{T} is passed as a @code{@var{t}*} argument to a C function,
5619 where @var{t} is the C type corresponding to the Ada type @var{T}.
5625 An Ada access @var{T} parameter, or an Ada @code{out} or @code{in out}
5626 parameter of an elementary type @var{T}, is passed as a @code{@var{t}*}
5627 argument to a C function, where @var{t} is the C type corresponding to
5628 the Ada type @var{T}. In the case of an elementary @code{out} or
5629 @code{in out} parameter, a pointer to a temporary copy is used to
5630 preserve by-copy semantics.
5636 An Ada parameter of a record type @var{T}, of any mode, is passed as a
5637 @code{@var{t}*} argument to a C function, where @var{t} is the C
5638 structure corresponding to the Ada type @var{T}.
5640 Followed. This convention may be overridden by the use of the C_Pass_By_Copy
5641 pragma, or Convention, or by explicitly specifying the mechanism for a given
5642 call using an extended import or export pragma.
5646 An Ada parameter of an array type with component type @var{T}, of any
5647 mode, is passed as a @code{@var{t}*} argument to a C function, where
5648 @var{t} is the C type corresponding to the Ada type @var{T}.
5654 An Ada parameter of an access-to-subprogram type is passed as a pointer
5655 to a C function whose prototype corresponds to the designated
5656 subprogram's specification.
5660 @cindex COBOL, interfacing with
5661 @unnumberedsec B.4(95-98): Interfacing with COBOL
5664 An Ada implementation should support the following interface
5665 correspondences between Ada and COBOL@.
5671 An Ada access @var{T} parameter is passed as a @samp{BY REFERENCE} data item of
5672 the COBOL type corresponding to @var{T}.
5678 An Ada in scalar parameter is passed as a @samp{BY CONTENT} data item of
5679 the corresponding COBOL type.
5685 Any other Ada parameter is passed as a @samp{BY REFERENCE} data item of the
5686 COBOL type corresponding to the Ada parameter type; for scalars, a local
5687 copy is used if necessary to ensure by-copy semantics.
5691 @cindex Fortran, interfacing with
5692 @unnumberedsec B.5(22-26): Interfacing with Fortran
5695 An Ada implementation should support the following interface
5696 correspondences between Ada and Fortran:
5702 An Ada procedure corresponds to a Fortran subroutine.
5708 An Ada function corresponds to a Fortran function.
5714 An Ada parameter of an elementary, array, or record type @var{T} is
5715 passed as a @var{T} argument to a Fortran procedure, where @var{T} is
5716 the Fortran type corresponding to the Ada type @var{T}, and where the
5717 INTENT attribute of the corresponding dummy argument matches the Ada
5718 formal parameter mode; the Fortran implementation's parameter passing
5719 conventions are used. For elementary types, a local copy is used if
5720 necessary to ensure by-copy semantics.
5726 An Ada parameter of an access-to-subprogram type is passed as a
5727 reference to a Fortran procedure whose interface corresponds to the
5728 designated subprogram's specification.
5732 @cindex Machine operations
5733 @unnumberedsec C.1(3-5): Access to Machine Operations
5736 The machine code or intrinsic support should allow access to all
5737 operations normally available to assembly language programmers for the
5738 target environment, including privileged instructions, if any.
5744 The interfacing pragmas (see Annex B) should support interface to
5745 assembler; the default assembler should be associated with the
5746 convention identifier @code{Assembler}.
5752 If an entity is exported to assembly language, then the implementation
5753 should allocate it at an addressable location, and should ensure that it
5754 is retained by the linking process, even if not otherwise referenced
5755 from the Ada code. The implementation should assume that any call to a
5756 machine code or assembler subprogram is allowed to read or update every
5757 object that is specified as exported.
5761 @unnumberedsec C.1(10-16): Access to Machine Operations
5764 The implementation should ensure that little or no overhead is
5765 associated with calling intrinsic and machine-code subprograms.
5767 Followed for both intrinsics and machine-code subprograms.
5771 It is recommended that intrinsic subprograms be provided for convenient
5772 access to any machine operations that provide special capabilities or
5773 efficiency and that are not otherwise available through the language
5776 Followed. A full set of machine operation intrinsic subprograms is provided.
5780 Atomic read-modify-write operations---e.g.@:, test and set, compare and
5781 swap, decrement and test, enqueue/dequeue.
5783 Followed on any target supporting such operations.
5787 Standard numeric functions---e.g.@:, sin, log.
5789 Followed on any target supporting such operations.
5793 String manipulation operations---e.g.@:, translate and test.
5795 Followed on any target supporting such operations.
5799 Vector operations---e.g.@:, compare vector against thresholds.
5801 Followed on any target supporting such operations.
5805 Direct operations on I/O ports.
5807 Followed on any target supporting such operations.
5809 @cindex Interrupt support
5810 @unnumberedsec C.3(28): Interrupt Support
5813 If the @code{Ceiling_Locking} policy is not in effect, the
5814 implementation should provide means for the application to specify which
5815 interrupts are to be blocked during protected actions, if the underlying
5816 system allows for a finer-grain control of interrupt blocking.
5818 Followed. The underlying system does not allow for finer-grain control
5819 of interrupt blocking.
5821 @cindex Protected procedure handlers
5822 @unnumberedsec C.3.1(20-21): Protected Procedure Handlers
5825 Whenever possible, the implementation should allow interrupt handlers to
5826 be called directly by the hardware.
5830 This is never possible under IRIX, so this is followed by default.
5832 Followed on any target where the underlying operating system permits
5837 Whenever practical, violations of any
5838 implementation-defined restrictions should be detected before run time.
5840 Followed. Compile time warnings are given when possible.
5842 @cindex Package @code{Interrupts}
5844 @unnumberedsec C.3.2(25): Package @code{Interrupts}
5848 If implementation-defined forms of interrupt handler procedures are
5849 supported, such as protected procedures with parameters, then for each
5850 such form of a handler, a type analogous to @code{Parameterless_Handler}
5851 should be specified in a child package of @code{Interrupts}, with the
5852 same operations as in the predefined package Interrupts.
5856 @cindex Pre-elaboration requirements
5857 @unnumberedsec C.4(14): Pre-elaboration Requirements
5860 It is recommended that pre-elaborated packages be implemented in such a
5861 way that there should be little or no code executed at run time for the
5862 elaboration of entities not already covered by the Implementation
5865 Followed. Executable code is generated in some cases, e.g.@: loops
5866 to initialize large arrays.
5868 @unnumberedsec C.5(8): Pragma @code{Discard_Names}
5872 If the pragma applies to an entity, then the implementation should
5873 reduce the amount of storage used for storing names associated with that
5878 @cindex Package @code{Task_Attributes}
5879 @findex Task_Attributes
5880 @unnumberedsec C.7.2(30): The Package Task_Attributes
5883 Some implementations are targeted to domains in which memory use at run
5884 time must be completely deterministic. For such implementations, it is
5885 recommended that the storage for task attributes will be pre-allocated
5886 statically and not from the heap. This can be accomplished by either
5887 placing restrictions on the number and the size of the task's
5888 attributes, or by using the pre-allocated storage for the first @var{N}
5889 attribute objects, and the heap for the others. In the latter case,
5890 @var{N} should be documented.
5892 Not followed. This implementation is not targeted to such a domain.
5894 @cindex Locking Policies
5895 @unnumberedsec D.3(17): Locking Policies
5899 The implementation should use names that end with @samp{_Locking} for
5900 locking policies defined by the implementation.
5902 Followed. A single implementation-defined locking policy is defined,
5903 whose name (@code{Inheritance_Locking}) follows this suggestion.
5905 @cindex Entry queuing policies
5906 @unnumberedsec D.4(16): Entry Queuing Policies
5909 Names that end with @samp{_Queuing} should be used
5910 for all implementation-defined queuing policies.
5912 Followed. No such implementation-defined queuing policies exist.
5914 @cindex Preemptive abort
5915 @unnumberedsec D.6(9-10): Preemptive Abort
5918 Even though the @code{abort_statement} is included in the list of
5919 potentially blocking operations (see 9.5.1), it is recommended that this
5920 statement be implemented in a way that never requires the task executing
5921 the @code{abort_statement} to block.
5927 On a multi-processor, the delay associated with aborting a task on
5928 another processor should be bounded; the implementation should use
5929 periodic polling, if necessary, to achieve this.
5933 @cindex Tasking restrictions
5934 @unnumberedsec D.7(21): Tasking Restrictions
5937 When feasible, the implementation should take advantage of the specified
5938 restrictions to produce a more efficient implementation.
5940 GNAT currently takes advantage of these restrictions by providing an optimized
5941 run time when the Ravenscar profile and the GNAT restricted run time set
5942 of restrictions are specified. See pragma @code{Ravenscar} and pragma
5943 @code{Restricted_Run_Time} for more details.
5945 @cindex Time, monotonic
5946 @unnumberedsec D.8(47-49): Monotonic Time
5949 When appropriate, implementations should provide configuration
5950 mechanisms to change the value of @code{Tick}.
5952 Such configuration mechanisms are not appropriate to this implementation
5953 and are thus not supported.
5957 It is recommended that @code{Calendar.Clock} and @code{Real_Time.Clock}
5958 be implemented as transformations of the same time base.
5964 It is recommended that the @dfn{best} time base which exists in
5965 the underlying system be available to the application through
5966 @code{Clock}. @dfn{Best} may mean highest accuracy or largest range.
5970 @cindex Partition communication subsystem
5972 @unnumberedsec E.5(28-29): Partition Communication Subsystem
5975 Whenever possible, the PCS on the called partition should allow for
5976 multiple tasks to call the RPC-receiver with different messages and
5977 should allow them to block until the corresponding subprogram body
5980 Followed by GLADE, a separately supplied PCS that can be used with
5985 The @code{Write} operation on a stream of type @code{Params_Stream_Type}
5986 should raise @code{Storage_Error} if it runs out of space trying to
5987 write the @code{Item} into the stream.
5989 Followed by GLADE, a separately supplied PCS that can be used with
5992 @cindex COBOL support
5993 @unnumberedsec F(7): COBOL Support
5996 If COBOL (respectively, C) is widely supported in the target
5997 environment, implementations supporting the Information Systems Annex
5998 should provide the child package @code{Interfaces.COBOL} (respectively,
5999 @code{Interfaces.C}) specified in Annex B and should support a
6000 @code{convention_identifier} of COBOL (respectively, C) in the interfacing
6001 pragmas (see Annex B), thus allowing Ada programs to interface with
6002 programs written in that language.
6006 @cindex Decimal radix support
6007 @unnumberedsec F.1(2): Decimal Radix Support
6010 Packed decimal should be used as the internal representation for objects
6011 of subtype @var{S} when @var{S}'Machine_Radix = 10.
6013 Not followed. GNAT ignores @var{S}'Machine_Radix and always uses binary
6017 @unnumberedsec G: Numerics
6020 If Fortran (respectively, C) is widely supported in the target
6021 environment, implementations supporting the Numerics Annex
6022 should provide the child package @code{Interfaces.Fortran} (respectively,
6023 @code{Interfaces.C}) specified in Annex B and should support a
6024 @code{convention_identifier} of Fortran (respectively, C) in the interfacing
6025 pragmas (see Annex B), thus allowing Ada programs to interface with
6026 programs written in that language.
6030 @cindex Complex types
6031 @unnumberedsec G.1.1(56-58): Complex Types
6034 Because the usual mathematical meaning of multiplication of a complex
6035 operand and a real operand is that of the scaling of both components of
6036 the former by the latter, an implementation should not perform this
6037 operation by first promoting the real operand to complex type and then
6038 performing a full complex multiplication. In systems that, in the
6039 future, support an Ada binding to IEC 559:1989, the latter technique
6040 will not generate the required result when one of the components of the
6041 complex operand is infinite. (Explicit multiplication of the infinite
6042 component by the zero component obtained during promotion yields a NaN
6043 that propagates into the final result.) Analogous advice applies in the
6044 case of multiplication of a complex operand and a pure-imaginary
6045 operand, and in the case of division of a complex operand by a real or
6046 pure-imaginary operand.
6052 Similarly, because the usual mathematical meaning of addition of a
6053 complex operand and a real operand is that the imaginary operand remains
6054 unchanged, an implementation should not perform this operation by first
6055 promoting the real operand to complex type and then performing a full
6056 complex addition. In implementations in which the @code{Signed_Zeros}
6057 attribute of the component type is @code{True} (and which therefore
6058 conform to IEC 559:1989 in regard to the handling of the sign of zero in
6059 predefined arithmetic operations), the latter technique will not
6060 generate the required result when the imaginary component of the complex
6061 operand is a negatively signed zero. (Explicit addition of the negative
6062 zero to the zero obtained during promotion yields a positive zero.)
6063 Analogous advice applies in the case of addition of a complex operand
6064 and a pure-imaginary operand, and in the case of subtraction of a
6065 complex operand and a real or pure-imaginary operand.
6071 Implementations in which @code{Real'Signed_Zeros} is @code{True} should
6072 attempt to provide a rational treatment of the signs of zero results and
6073 result components. As one example, the result of the @code{Argument}
6074 function should have the sign of the imaginary component of the
6075 parameter @code{X} when the point represented by that parameter lies on
6076 the positive real axis; as another, the sign of the imaginary component
6077 of the @code{Compose_From_Polar} function should be the same as
6078 (respectively, the opposite of) that of the @code{Argument} parameter when that
6079 parameter has a value of zero and the @code{Modulus} parameter has a
6080 nonnegative (respectively, negative) value.
6084 @cindex Complex elementary functions
6085 @unnumberedsec G.1.2(49): Complex Elementary Functions
6088 Implementations in which @code{Complex_Types.Real'Signed_Zeros} is
6089 @code{True} should attempt to provide a rational treatment of the signs
6090 of zero results and result components. For example, many of the complex
6091 elementary functions have components that are odd functions of one of
6092 the parameter components; in these cases, the result component should
6093 have the sign of the parameter component at the origin. Other complex
6094 elementary functions have zero components whose sign is opposite that of
6095 a parameter component at the origin, or is always positive or always
6100 @cindex Accuracy requirements
6101 @unnumberedsec G.2.4(19): Accuracy Requirements
6104 The versions of the forward trigonometric functions without a
6105 @code{Cycle} parameter should not be implemented by calling the
6106 corresponding version with a @code{Cycle} parameter of
6107 @code{2.0*Numerics.Pi}, since this will not provide the required
6108 accuracy in some portions of the domain. For the same reason, the
6109 version of @code{Log} without a @code{Base} parameter should not be
6110 implemented by calling the corresponding version with a @code{Base}
6111 parameter of @code{Numerics.e}.
6115 @cindex Complex arithmetic accuracy
6116 @cindex Accuracy, complex arithmetic
6117 @unnumberedsec G.2.6(15): Complex Arithmetic Accuracy
6121 The version of the @code{Compose_From_Polar} function without a
6122 @code{Cycle} parameter should not be implemented by calling the
6123 corresponding version with a @code{Cycle} parameter of
6124 @code{2.0*Numerics.Pi}, since this will not provide the required
6125 accuracy in some portions of the domain.
6129 @c -----------------------------------------
6130 @node Implementation Defined Characteristics
6131 @chapter Implementation Defined Characteristics
6134 In addition to the implementation dependent pragmas and attributes, and
6135 the implementation advice, there are a number of other features of Ada
6136 95 that are potentially implementation dependent. These are mentioned
6137 throughout the Ada 95 Reference Manual, and are summarized in annex M@.
6139 A requirement for conforming Ada compilers is that they provide
6140 documentation describing how the implementation deals with each of these
6141 issues. In this chapter, you will find each point in annex M listed
6142 followed by a description in italic font of how GNAT
6146 implementation on IRIX 5.3 operating system or greater
6148 handles the implementation dependence.
6150 You can use this chapter as a guide to minimizing implementation
6151 dependent features in your programs if portability to other compilers
6152 and other operating systems is an important consideration. The numbers
6153 in each section below correspond to the paragraph number in the Ada 95
6159 @strong{2}. Whether or not each recommendation given in Implementation
6160 Advice is followed. See 1.1.2(37).
6163 @xref{Implementation Advice}.
6168 @strong{3}. Capacity limitations of the implementation. See 1.1.3(3).
6171 The complexity of programs that can be processed is limited only by the
6172 total amount of available virtual memory, and disk space for the
6173 generated object files.
6178 @strong{4}. Variations from the standard that are impractical to avoid
6179 given the implementation's execution environment. See 1.1.3(6).
6182 There are no variations from the standard.
6187 @strong{5}. Which @code{code_statement}s cause external
6188 interactions. See 1.1.3(10).
6191 Any @code{code_statement} can potentially cause external interactions.
6196 @strong{6}. The coded representation for the text of an Ada
6197 program. See 2.1(4).
6200 See separate section on source representation.
6205 @strong{7}. The control functions allowed in comments. See 2.1(14).
6208 See separate section on source representation.
6213 @strong{8}. The representation for an end of line. See 2.2(2).
6216 See separate section on source representation.
6221 @strong{9}. Maximum supported line length and lexical element
6222 length. See 2.2(15).
6225 The maximum line length is 255 characters an the maximum length of a
6226 lexical element is also 255 characters.
6231 @strong{10}. Implementation defined pragmas. See 2.8(14).
6235 @xref{Implementation Defined Pragmas}.
6240 @strong{11}. Effect of pragma @code{Optimize}. See 2.8(27).
6243 Pragma @code{Optimize}, if given with a @code{Time} or @code{Space}
6244 parameter, checks that the optimization flag is set, and aborts if it is
6250 @strong{12}. The sequence of characters of the value returned by
6251 @code{@var{S}'Image} when some of the graphic characters of
6252 @code{@var{S}'Wide_Image} are not defined in @code{Character}. See
6256 The sequence of characters is as defined by the wide character encoding
6257 method used for the source. See section on source representation for
6263 @strong{13}. The predefined integer types declared in
6264 @code{Standard}. See 3.5.4(25).
6268 @item Short_Short_Integer
6271 (Short) 16 bit signed
6275 64 bit signed (Alpha OpenVMS only)
6276 32 bit signed (all other targets)
6277 @item Long_Long_Integer
6284 @strong{14}. Any nonstandard integer types and the operators defined
6285 for them. See 3.5.4(26).
6288 There are no nonstandard integer types.
6293 @strong{15}. Any nonstandard real types and the operators defined for
6297 There are no nonstandard real types.
6302 @strong{16}. What combinations of requested decimal precision and range
6303 are supported for floating point types. See 3.5.7(7).
6306 The precision and range is as defined by the IEEE standard.
6311 @strong{17}. The predefined floating point types declared in
6312 @code{Standard}. See 3.5.7(16).
6319 (Short) 32 bit IEEE short
6322 @item Long_Long_Float
6323 64 bit IEEE long (80 bit IEEE long on x86 processors)
6329 @strong{18}. The small of an ordinary fixed point type. See 3.5.9(8).
6332 @code{Fine_Delta} is 2**(@minus{}63)
6337 @strong{19}. What combinations of small, range, and digits are
6338 supported for fixed point types. See 3.5.9(10).
6341 Any combinations are permitted that do not result in a small less than
6342 @code{Fine_Delta} and do not result in a mantissa larger than 63 bits.
6343 If the mantissa is larger than 53 bits on machines where Long_Long_Float
6344 is 64 bits (true of all architectures except ia32), then the output from
6345 Text_IO is accurate to only 53 bits, rather than the full mantissa. This
6346 is because floating-point conversions are used to convert fixed point.
6351 @strong{20}. The result of @code{Tags.Expanded_Name} for types declared
6352 within an unnamed @code{block_statement}. See 3.9(10).
6355 Block numbers of the form @code{B@var{nnn}}, where @var{nnn} is a
6356 decimal integer are allocated.
6361 @strong{21}. Implementation-defined attributes. See 4.1.4(12).
6364 @xref{Implementation Defined Attributes}.
6369 @strong{22}. Any implementation-defined time types. See 9.6(6).
6372 There are no implementation-defined time types.
6377 @strong{23}. The time base associated with relative delays.
6380 See 9.6(20). The time base used is that provided by the C library
6381 function @code{gettimeofday}.
6386 @strong{24}. The time base of the type @code{Calendar.Time}. See
6390 The time base used is that provided by the C library function
6391 @code{gettimeofday}.
6396 @strong{25}. The time zone used for package @code{Calendar}
6397 operations. See 9.6(24).
6400 The time zone used by package @code{Calendar} is the current system time zone
6401 setting for local time, as accessed by the C library function
6407 @strong{26}. Any limit on @code{delay_until_statements} of
6408 @code{select_statements}. See 9.6(29).
6411 There are no such limits.
6416 @strong{27}. Whether or not two non overlapping parts of a composite
6417 object are independently addressable, in the case where packing, record
6418 layout, or @code{Component_Size} is specified for the object. See
6422 Separate components are independently addressable if they do not share
6423 overlapping storage units.
6428 @strong{28}. The representation for a compilation. See 10.1(2).
6431 A compilation is represented by a sequence of files presented to the
6432 compiler in a single invocation of the @code{gcc} command.
6437 @strong{29}. Any restrictions on compilations that contain multiple
6438 compilation_units. See 10.1(4).
6441 No single file can contain more than one compilation unit, but any
6442 sequence of files can be presented to the compiler as a single
6448 @strong{30}. The mechanisms for creating an environment and for adding
6449 and replacing compilation units. See 10.1.4(3).
6452 See separate section on compilation model.
6457 @strong{31}. The manner of explicitly assigning library units to a
6458 partition. See 10.2(2).
6461 If a unit contains an Ada main program, then the Ada units for the partition
6462 are determined by recursive application of the rules in the Ada Reference
6463 Manual section 10.2(2-6). In other words, the Ada units will be those that
6464 are needed by the main program, and then this definition of need is applied
6465 recursively to those units, and the partition contains the transitive
6466 closure determined by this relationship. In short, all the necessary units
6467 are included, with no need to explicitly specify the list. If additional
6468 units are required, e.g.@: by foreign language units, then all units must be
6469 mentioned in the context clause of one of the needed Ada units.
6471 If the partition contains no main program, or if the main program is in
6472 a language other than Ada, then GNAT
6473 provides the binder options @code{-z} and @code{-n} respectively, and in
6474 this case a list of units can be explicitly supplied to the binder for
6475 inclusion in the partition (all units needed by these units will also
6476 be included automatically). For full details on the use of these
6477 options, refer to the @cite{GNAT User's Guide} sections on Binding
6483 @strong{32}. The implementation-defined means, if any, of specifying
6484 which compilation units are needed by a given compilation unit. See
6488 The units needed by a given compilation unit are as defined in
6489 the Ada Reference Manual section 10.2(2-6). There are no
6490 implementation-defined pragmas or other implementation-defined
6491 means for specifying needed units.
6496 @strong{33}. The manner of designating the main subprogram of a
6497 partition. See 10.2(7).
6500 The main program is designated by providing the name of the
6501 corresponding @file{ALI} file as the input parameter to the binder.
6506 @strong{34}. The order of elaboration of @code{library_items}. See
6510 The first constraint on ordering is that it meets the requirements of
6511 chapter 10 of the Ada 95 Reference Manual. This still leaves some
6512 implementation dependent choices, which are resolved by first
6513 elaborating bodies as early as possible (i.e.@: in preference to specs
6514 where there is a choice), and second by evaluating the immediate with
6515 clauses of a unit to determine the probably best choice, and
6516 third by elaborating in alphabetical order of unit names
6517 where a choice still remains.
6522 @strong{35}. Parameter passing and function return for the main
6523 subprogram. See 10.2(21).
6526 The main program has no parameters. It may be a procedure, or a function
6527 returning an integer type. In the latter case, the returned integer
6528 value is the return code of the program (overriding any value that
6529 may have been set by a call to @code{Ada.Command_Line.Set_Exit_Status}).
6534 @strong{36}. The mechanisms for building and running partitions. See
6538 GNAT itself supports programs with only a single partition. The GNATDIST
6539 tool provided with the GLADE package (which also includes an implementation
6540 of the PCS) provides a completely flexible method for building and running
6541 programs consisting of multiple partitions. See the separate GLADE manual
6547 @strong{37}. The details of program execution, including program
6548 termination. See 10.2(25).
6551 See separate section on compilation model.
6556 @strong{38}. The semantics of any non-active partitions supported by the
6557 implementation. See 10.2(28).
6560 Passive partitions are supported on targets where shared memory is
6561 provided by the operating system. See the GLADE reference manual for
6567 @strong{39}. The information returned by @code{Exception_Message}. See
6571 Exception message returns the null string unless a specific message has
6572 been passed by the program.
6577 @strong{40}. The result of @code{Exceptions.Exception_Name} for types
6578 declared within an unnamed @code{block_statement}. See 11.4.1(12).
6581 Blocks have implementation defined names of the form @code{B@var{nnn}}
6582 where @var{nnn} is an integer.
6587 @strong{41}. The information returned by
6588 @code{Exception_Information}. See 11.4.1(13).
6591 @code{Exception_Information} returns a string in the following format:
6594 @emph{Exception_Name:} nnnnn
6595 @emph{Message:} mmmmm
6597 @emph{Call stack traceback locations:}
6598 0xhhhh 0xhhhh 0xhhhh ... 0xhhh
6606 @code{nnnn} is the fully qualified name of the exception in all upper
6607 case letters. This line is always present.
6610 @code{mmmm} is the message (this line present only if message is non-null)
6613 @code{ppp} is the Process Id value as a decimal integer (this line is
6614 present only if the Process Id is non-zero). Currently we are
6615 not making use of this field.
6618 The Call stack traceback locations line and the following values
6619 are present only if at least one traceback location was recorded.
6620 The values are given in C style format, with lower case letters
6621 for a-f, and only as many digits present as are necessary.
6625 The line terminator sequence at the end of each line, including
6626 the last line is a single @code{LF} character (@code{16#0A#}).
6631 @strong{42}. Implementation-defined check names. See 11.5(27).
6634 No implementation-defined check names are supported.
6639 @strong{43}. The interpretation of each aspect of representation. See
6643 See separate section on data representations.
6648 @strong{44}. Any restrictions placed upon representation items. See
6652 See separate section on data representations.
6657 @strong{45}. The meaning of @code{Size} for indefinite subtypes. See
6661 Size for an indefinite subtype is the maximum possible size, except that
6662 for the case of a subprogram parameter, the size of the parameter object
6668 @strong{46}. The default external representation for a type tag. See
6672 The default external representation for a type tag is the fully expanded
6673 name of the type in upper case letters.
6678 @strong{47}. What determines whether a compilation unit is the same in
6679 two different partitions. See 13.3(76).
6682 A compilation unit is the same in two different partitions if and only
6683 if it derives from the same source file.
6688 @strong{48}. Implementation-defined components. See 13.5.1(15).
6691 The only implementation defined component is the tag for a tagged type,
6692 which contains a pointer to the dispatching table.
6697 @strong{49}. If @code{Word_Size} = @code{Storage_Unit}, the default bit
6698 ordering. See 13.5.3(5).
6701 @code{Word_Size} (32) is not the same as @code{Storage_Unit} (8) for this
6702 implementation, so no non-default bit ordering is supported. The default
6703 bit ordering corresponds to the natural endianness of the target architecture.
6708 @strong{50}. The contents of the visible part of package @code{System}
6709 and its language-defined children. See 13.7(2).
6712 See the definition of these packages in files @file{system.ads} and
6713 @file{s-stoele.ads}.
6718 @strong{51}. The contents of the visible part of package
6719 @code{System.Machine_Code}, and the meaning of
6720 @code{code_statements}. See 13.8(7).
6723 See the definition and documentation in file @file{s-maccod.ads}.
6728 @strong{52}. The effect of unchecked conversion. See 13.9(11).
6731 Unchecked conversion between types of the same size
6732 and results in an uninterpreted transmission of the bits from one type
6733 to the other. If the types are of unequal sizes, then in the case of
6734 discrete types, a shorter source is first zero or sign extended as
6735 necessary, and a shorter target is simply truncated on the left.
6736 For all non-discrete types, the source is first copied if necessary
6737 to ensure that the alignment requirements of the target are met, then
6738 a pointer is constructed to the source value, and the result is obtained
6739 by dereferencing this pointer after converting it to be a pointer to the
6745 @strong{53}. The manner of choosing a storage pool for an access type
6746 when @code{Storage_Pool} is not specified for the type. See 13.11(17).
6749 There are 3 different standard pools used by the compiler when
6750 @code{Storage_Pool} is not specified depending whether the type is local
6751 to a subprogram or defined at the library level and whether
6752 @code{Storage_Size}is specified or not. See documentation in the runtime
6753 library units @code{System.Pool_Global}, @code{System.Pool_Size} and
6754 @code{System.Pool_Local} in files @file{s-poosiz.ads},
6755 @file{s-pooglo.ads} and @file{s-pooloc.ads} for full details on the
6761 @strong{54}. Whether or not the implementation provides user-accessible
6762 names for the standard pool type(s). See 13.11(17).
6766 See documentation in the sources of the run time mentioned in paragraph
6767 @strong{53} . All these pools are accessible by means of @code{with}'ing
6773 @strong{55}. The meaning of @code{Storage_Size}. See 13.11(18).
6776 @code{Storage_Size} is measured in storage units, and refers to the
6777 total space available for an access type collection, or to the primary
6778 stack space for a task.
6783 @strong{56}. Implementation-defined aspects of storage pools. See
6787 See documentation in the sources of the run time mentioned in paragraph
6788 @strong{53} for details on GNAT-defined aspects of storage pools.
6793 @strong{57}. The set of restrictions allowed in a pragma
6794 @code{Restrictions}. See 13.12(7).
6797 All RM defined Restriction identifiers are implemented. The following
6798 additional restriction identifiers are provided. There are two separate
6799 lists of implementation dependent restriction identifiers. The first
6800 set requires consistency throughout a partition (in other words, if the
6801 restriction identifier is used for any compilation unit in the partition,
6802 then all compilation units in the partition must obey the restriction.
6806 @item Boolean_Entry_Barriers
6807 @findex Boolean_Entry_Barriers
6808 This restriction ensures at compile time that barriers in entry declarations
6809 for protected types are restricted to references to simple boolean variables
6810 defined in the private part of the protected type. No other form of entry
6811 barriers is permitted. This is one of the restrictions of the Ravenscar
6812 profile for limited tasking (see also pragma @code{Ravenscar}).
6814 @item Max_Entry_Queue_Depth => Expr
6815 @findex Max_Entry_Queue_Depth
6816 This restriction is a declaration that any protected entry compiled in
6817 the scope of the restriction has at most the specified number of
6818 tasks waiting on the entry
6819 at any one time, and so no queue is required. This restriction is not
6820 checked at compile time. A program execution is erroneous if an attempt
6821 is made to queue more than the specified number of tasks on such an entry.
6825 This restriction ensures at compile time that there is no implicit or
6826 explicit dependence on the package @code{Ada.Calendar}.
6828 @item No_Direct_Boolean_Operators
6829 @findex No_Direct_Boolean_Operators
6830 This restriction ensures that no logical (and/or/xor) or comparison
6831 operators are used on operands of type Boolean (or any type derived
6832 from Boolean). This is intended for use in safety critical programs
6833 where the certification protocol requires the use of short-circuit
6834 (and then, or else) forms for all composite boolean operations.
6836 @item No_Dynamic_Interrupts
6837 @findex No_Dynamic_Interrupts
6838 This restriction ensures at compile time that there is no attempt to
6839 dynamically associate interrupts. Only static association is allowed.
6841 @item No_Enumeration_Maps
6842 @findex No_Enumeration_Maps
6843 This restriction ensures at compile time that no operations requiring
6844 enumeration maps are used (that is Image and Value attributes applied
6845 to enumeration types).
6847 @item No_Entry_Calls_In_Elaboration_Code
6848 @findex No_Entry_Calls_In_Elaboration_Code
6849 This restriction ensures at compile time that no task or protected entry
6850 calls are made during elaboration code. As a result of the use of this
6851 restriction, the compiler can assume that no code past an accept statement
6852 in a task can be executed at elaboration time.
6854 @item No_Exception_Handlers
6855 @findex No_Exception_Handlers
6856 This restriction ensures at compile time that there are no explicit
6857 exception handlers. It also indicates that no exception propagation will
6858 be provided. In this mode, exceptions may be raised but will result in
6859 an immediate call to the last chance handler, a routine that the user
6860 must define with the following profile:
6862 procedure Last_Chance_Handler
6863 (Source_Location : System.Address; Line : Integer);
6864 pragma Export (C, Last_Chance_Handler,
6865 "__gnat_last_chance_handler");
6867 The parameter is a C null-terminated string representing a message to be
6868 associated with the exception (typically the source location of the raise
6869 statement generated by the compiler). The Line parameter when non-zero
6870 represents the line number in the source program where the raise occurs.
6872 @item No_Exception_Streams
6873 @findex No_Exception_Streams
6874 This restriction ensures at compile time that no stream operations for
6875 types Exception_Id or Exception_Occurrence are used. This also makes it
6876 impossible to pass exceptions to or from a partition with this restriction
6877 in a distributed environment. If this exception is active, then the generated
6878 code is simplified by omitting the otherwise-required global registration
6879 of exceptions when they are declared.
6881 @item No_Implicit_Conditionals
6882 @findex No_Implicit_Conditionals
6883 This restriction ensures that the generated code does not contain any
6884 implicit conditionals, either by modifying the generated code where possible,
6885 or by rejecting any construct that would otherwise generate an implicit
6888 @item No_Implicit_Dynamic_Code
6889 @findex No_Implicit_Dynamic_Code
6890 This restriction prevents the compiler from building ``trampolines''.
6891 This is a structure that is built on the stack and contains dynamic
6892 code to be executed at run time. A trampoline is needed to indirectly
6893 address a nested subprogram (that is a subprogram that is not at the
6894 library level). The restriction prevents the use of any of the
6895 attributes @code{Address}, @code{Access} or @code{Unrestricted_Access}
6896 being applied to a subprogram that is not at the library level.
6898 @item No_Implicit_Loops
6899 @findex No_Implicit_Loops
6900 This restriction ensures that the generated code does not contain any
6901 implicit @code{for} loops, either by modifying
6902 the generated code where possible,
6903 or by rejecting any construct that would otherwise generate an implicit
6906 @item No_Initialize_Scalars
6907 @findex No_Initialize_Scalars
6908 This restriction ensures that no unit in the partition is compiled with
6909 pragma Initialize_Scalars. This allows the generation of more efficient
6910 code, and in particular eliminates dummy null initialization routines that
6911 are otherwise generated for some record and array types.
6913 @item No_Local_Protected_Objects
6914 @findex No_Local_Protected_Objects
6915 This restriction ensures at compile time that protected objects are
6916 only declared at the library level.
6918 @item No_Protected_Type_Allocators
6919 @findex No_Protected_Type_Allocators
6920 This restriction ensures at compile time that there are no allocator
6921 expressions that attempt to allocate protected objects.
6923 @item No_Secondary_Stack
6924 @findex No_Secondary_Stack
6925 This restriction ensures at compile time that the generated code does not
6926 contain any reference to the secondary stack. The secondary stack is used
6927 to implement functions returning unconstrained objects (arrays or records)
6930 @item No_Select_Statements
6931 @findex No_Select_Statements
6932 This restriction ensures at compile time no select statements of any kind
6933 are permitted, that is the keyword @code{select} may not appear.
6934 This is one of the restrictions of the Ravenscar
6935 profile for limited tasking (see also pragma @code{Ravenscar}).
6937 @item No_Standard_Storage_Pools
6938 @findex No_Standard_Storage_Pools
6939 This restriction ensures at compile time that no access types
6940 use the standard default storage pool. Any access type declared must
6941 have an explicit Storage_Pool attribute defined specifying a
6942 user-defined storage pool.
6946 This restriction ensures at compile time that there are no implicit or
6947 explicit dependencies on the package @code{Ada.Streams}.
6949 @item No_Task_Attributes
6950 @findex No_Task_Attributes
6951 This restriction ensures at compile time that there are no implicit or
6952 explicit dependencies on the package @code{Ada.Task_Attributes}.
6954 @item No_Task_Termination
6955 @findex No_Task_Termination
6956 This restriction ensures at compile time that no terminate alternatives
6957 appear in any task body.
6961 This restriction prevents the declaration of tasks or task types throughout
6962 the partition. It is similar in effect to the use of @code{Max_Tasks => 0}
6963 except that violations are caught at compile time and cause an error message
6964 to be output either by the compiler or binder.
6966 @item No_Wide_Characters
6967 @findex No_Wide_Characters
6968 This restriction ensures at compile time that no uses of the types
6969 @code{Wide_Character} or @code{Wide_String}
6970 appear, and that no wide character literals
6971 appear in the program (that is literals representing characters not in
6972 type @code{Character}.
6974 @item Static_Priorities
6975 @findex Static_Priorities
6976 This restriction ensures at compile time that all priority expressions
6977 are static, and that there are no dependencies on the package
6978 @code{Ada.Dynamic_Priorities}.
6980 @item Static_Storage_Size
6981 @findex Static_Storage_Size
6982 This restriction ensures at compile time that any expression appearing
6983 in a Storage_Size pragma or attribute definition clause is static.
6988 The second set of implementation dependent restriction identifiers
6989 does not require partition-wide consistency.
6990 The restriction may be enforced for a single
6991 compilation unit without any effect on any of the
6992 other compilation units in the partition.
6996 @item No_Elaboration_Code
6997 @findex No_Elaboration_Code
6998 This restriction ensures at compile time that no elaboration code is
6999 generated. Note that this is not the same condition as is enforced
7000 by pragma @code{Preelaborate}. There are cases in which pragma
7001 @code{Preelaborate} still permits code to be generated (e.g.@: code
7002 to initialize a large array to all zeroes), and there are cases of units
7003 which do not meet the requirements for pragma @code{Preelaborate},
7004 but for which no elaboration code is generated. Generally, it is
7005 the case that preelaborable units will meet the restrictions, with
7006 the exception of large aggregates initialized with an others_clause,
7007 and exception declarations (which generate calls to a run-time
7008 registry procedure). Note that this restriction is enforced on
7009 a unit by unit basis, it need not be obeyed consistently
7010 throughout a partition.
7012 @item No_Entry_Queue
7013 @findex No_Entry_Queue
7014 This restriction is a declaration that any protected entry compiled in
7015 the scope of the restriction has at most one task waiting on the entry
7016 at any one time, and so no queue is required. This restriction is not
7017 checked at compile time. A program execution is erroneous if an attempt
7018 is made to queue a second task on such an entry.
7020 @item No_Implementation_Attributes
7021 @findex No_Implementation_Attributes
7022 This restriction checks at compile time that no GNAT-defined attributes
7023 are present. With this restriction, the only attributes that can be used
7024 are those defined in the Ada 95 Reference Manual.
7026 @item No_Implementation_Pragmas
7027 @findex No_Implementation_Pragmas
7028 This restriction checks at compile time that no GNAT-defined pragmas
7029 are present. With this restriction, the only pragmas that can be used
7030 are those defined in the Ada 95 Reference Manual.
7032 @item No_Implementation_Restrictions
7033 @findex No_Implementation_Restrictions
7034 This restriction checks at compile time that no GNAT-defined restriction
7035 identifiers (other than @code{No_Implementation_Restrictions} itself)
7036 are present. With this restriction, the only other restriction identifiers
7037 that can be used are those defined in the Ada 95 Reference Manual.
7044 @strong{58}. The consequences of violating limitations on
7045 @code{Restrictions} pragmas. See 13.12(9).
7048 Restrictions that can be checked at compile time result in illegalities
7049 if violated. Currently there are no other consequences of violating
7055 @strong{59}. The representation used by the @code{Read} and
7056 @code{Write} attributes of elementary types in terms of stream
7057 elements. See 13.13.2(9).
7060 The representation is the in-memory representation of the base type of
7061 the type, using the number of bits corresponding to the
7062 @code{@var{type}'Size} value, and the natural ordering of the machine.
7067 @strong{60}. The names and characteristics of the numeric subtypes
7068 declared in the visible part of package @code{Standard}. See A.1(3).
7071 See items describing the integer and floating-point types supported.
7076 @strong{61}. The accuracy actually achieved by the elementary
7077 functions. See A.5.1(1).
7080 The elementary functions correspond to the functions available in the C
7081 library. Only fast math mode is implemented.
7086 @strong{62}. The sign of a zero result from some of the operators or
7087 functions in @code{Numerics.Generic_Elementary_Functions}, when
7088 @code{Float_Type'Signed_Zeros} is @code{True}. See A.5.1(46).
7091 The sign of zeroes follows the requirements of the IEEE 754 standard on
7097 @strong{63}. The value of
7098 @code{Numerics.Float_Random.Max_Image_Width}. See A.5.2(27).
7101 Maximum image width is 649, see library file @file{a-numran.ads}.
7106 @strong{64}. The value of
7107 @code{Numerics.Discrete_Random.Max_Image_Width}. See A.5.2(27).
7110 Maximum image width is 80, see library file @file{a-nudira.ads}.
7115 @strong{65}. The algorithms for random number generation. See
7119 The algorithm is documented in the source files @file{a-numran.ads} and
7120 @file{a-numran.adb}.
7125 @strong{66}. The string representation of a random number generator's
7126 state. See A.5.2(38).
7129 See the documentation contained in the file @file{a-numran.adb}.
7134 @strong{67}. The minimum time interval between calls to the
7135 time-dependent Reset procedure that are guaranteed to initiate different
7136 random number sequences. See A.5.2(45).
7139 The minimum period between reset calls to guarantee distinct series of
7140 random numbers is one microsecond.
7145 @strong{68}. The values of the @code{Model_Mantissa},
7146 @code{Model_Emin}, @code{Model_Epsilon}, @code{Model},
7147 @code{Safe_First}, and @code{Safe_Last} attributes, if the Numerics
7148 Annex is not supported. See A.5.3(72).
7151 See the source file @file{ttypef.ads} for the values of all numeric
7157 @strong{69}. Any implementation-defined characteristics of the
7158 input-output packages. See A.7(14).
7161 There are no special implementation defined characteristics for these
7167 @strong{70}. The value of @code{Buffer_Size} in @code{Storage_IO}. See
7171 All type representations are contiguous, and the @code{Buffer_Size} is
7172 the value of @code{@var{type}'Size} rounded up to the next storage unit
7178 @strong{71}. External files for standard input, standard output, and
7179 standard error See A.10(5).
7182 These files are mapped onto the files provided by the C streams
7183 libraries. See source file @file{i-cstrea.ads} for further details.
7188 @strong{72}. The accuracy of the value produced by @code{Put}. See
7192 If more digits are requested in the output than are represented by the
7193 precision of the value, zeroes are output in the corresponding least
7194 significant digit positions.
7199 @strong{73}. The meaning of @code{Argument_Count}, @code{Argument}, and
7200 @code{Command_Name}. See A.15(1).
7203 These are mapped onto the @code{argv} and @code{argc} parameters of the
7204 main program in the natural manner.
7209 @strong{74}. Implementation-defined convention names. See B.1(11).
7212 The following convention names are supported
7220 Synonym for Assembler
7222 Synonym for Assembler
7225 @item C_Pass_By_Copy
7226 Allowed only for record types, like C, but also notes that record
7227 is to be passed by copy rather than reference.
7233 Treated the same as C
7235 Treated the same as C
7239 For support of pragma @code{Import} with convention Intrinsic, see
7240 separate section on Intrinsic Subprograms.
7242 Stdcall (used for Windows implementations only). This convention correspond
7243 to the WINAPI (previously called Pascal convention) C/C++ convention under
7244 Windows. A function with this convention cleans the stack before exit.
7250 Stubbed is a special convention used to indicate that the body of the
7251 subprogram will be entirely ignored. Any call to the subprogram
7252 is converted into a raise of the @code{Program_Error} exception. If a
7253 pragma @code{Import} specifies convention @code{stubbed} then no body need
7254 be present at all. This convention is useful during development for the
7255 inclusion of subprograms whose body has not yet been written.
7259 In addition, all otherwise unrecognized convention names are also
7260 treated as being synonymous with convention C@. In all implementations
7261 except for VMS, use of such other names results in a warning. In VMS
7262 implementations, these names are accepted silently.
7267 @strong{75}. The meaning of link names. See B.1(36).
7270 Link names are the actual names used by the linker.
7275 @strong{76}. The manner of choosing link names when neither the link
7276 name nor the address of an imported or exported entity is specified. See
7280 The default linker name is that which would be assigned by the relevant
7281 external language, interpreting the Ada name as being in all lower case
7287 @strong{77}. The effect of pragma @code{Linker_Options}. See B.1(37).
7290 The string passed to @code{Linker_Options} is presented uninterpreted as
7291 an argument to the link command, unless it contains Ascii.NUL characters.
7292 NUL characters if they appear act as argument separators, so for example
7294 @smallexample @c ada
7295 pragma Linker_Options ("-labc" & ASCII.Nul & "-ldef");
7299 causes two separate arguments @code{-labc} and @code{-ldef} to be passed to the
7300 linker. The order of linker options is preserved for a given unit. The final
7301 list of options passed to the linker is in reverse order of the elaboration
7302 order. For example, linker options fo a body always appear before the options
7303 from the corresponding package spec.
7308 @strong{78}. The contents of the visible part of package
7309 @code{Interfaces} and its language-defined descendants. See B.2(1).
7312 See files with prefix @file{i-} in the distributed library.
7317 @strong{79}. Implementation-defined children of package
7318 @code{Interfaces}. The contents of the visible part of package
7319 @code{Interfaces}. See B.2(11).
7322 See files with prefix @file{i-} in the distributed library.
7327 @strong{80}. The types @code{Floating}, @code{Long_Floating},
7328 @code{Binary}, @code{Long_Binary}, @code{Decimal_ Element}, and
7329 @code{COBOL_Character}; and the initialization of the variables
7330 @code{Ada_To_COBOL} and @code{COBOL_To_Ada}, in
7331 @code{Interfaces.COBOL}. See B.4(50).
7338 (Floating) Long_Float
7343 @item Decimal_Element
7345 @item COBOL_Character
7350 For initialization, see the file @file{i-cobol.ads} in the distributed library.
7355 @strong{81}. Support for access to machine instructions. See C.1(1).
7358 See documentation in file @file{s-maccod.ads} in the distributed library.
7363 @strong{82}. Implementation-defined aspects of access to machine
7364 operations. See C.1(9).
7367 See documentation in file @file{s-maccod.ads} in the distributed library.
7372 @strong{83}. Implementation-defined aspects of interrupts. See C.3(2).
7375 Interrupts are mapped to signals or conditions as appropriate. See
7377 @code{Ada.Interrupt_Names} in source file @file{a-intnam.ads} for details
7378 on the interrupts supported on a particular target.
7383 @strong{84}. Implementation-defined aspects of pre-elaboration. See
7387 GNAT does not permit a partition to be restarted without reloading,
7388 except under control of the debugger.
7393 @strong{85}. The semantics of pragma @code{Discard_Names}. See C.5(7).
7396 Pragma @code{Discard_Names} causes names of enumeration literals to
7397 be suppressed. In the presence of this pragma, the Image attribute
7398 provides the image of the Pos of the literal, and Value accepts
7404 @strong{86}. The result of the @code{Task_Identification.Image}
7405 attribute. See C.7.1(7).
7408 The result of this attribute is an 8-digit hexadecimal string
7409 representing the virtual address of the task control block.
7414 @strong{87}. The value of @code{Current_Task} when in a protected entry
7415 or interrupt handler. See C.7.1(17).
7418 Protected entries or interrupt handlers can be executed by any
7419 convenient thread, so the value of @code{Current_Task} is undefined.
7424 @strong{88}. The effect of calling @code{Current_Task} from an entry
7425 body or interrupt handler. See C.7.1(19).
7428 The effect of calling @code{Current_Task} from an entry body or
7429 interrupt handler is to return the identification of the task currently
7435 @strong{89}. Implementation-defined aspects of
7436 @code{Task_Attributes}. See C.7.2(19).
7439 There are no implementation-defined aspects of @code{Task_Attributes}.
7444 @strong{90}. Values of all @code{Metrics}. See D(2).
7447 The metrics information for GNAT depends on the performance of the
7448 underlying operating system. The sources of the run-time for tasking
7449 implementation, together with the output from @code{-gnatG} can be
7450 used to determine the exact sequence of operating systems calls made
7451 to implement various tasking constructs. Together with appropriate
7452 information on the performance of the underlying operating system,
7453 on the exact target in use, this information can be used to determine
7454 the required metrics.
7459 @strong{91}. The declarations of @code{Any_Priority} and
7460 @code{Priority}. See D.1(11).
7463 See declarations in file @file{system.ads}.
7468 @strong{92}. Implementation-defined execution resources. See D.1(15).
7471 There are no implementation-defined execution resources.
7476 @strong{93}. Whether, on a multiprocessor, a task that is waiting for
7477 access to a protected object keeps its processor busy. See D.2.1(3).
7480 On a multi-processor, a task that is waiting for access to a protected
7481 object does not keep its processor busy.
7486 @strong{94}. The affect of implementation defined execution resources
7487 on task dispatching. See D.2.1(9).
7492 Tasks map to IRIX threads, and the dispatching policy is as defined by
7493 the IRIX implementation of threads.
7495 Tasks map to threads in the threads package used by GNAT@. Where possible
7496 and appropriate, these threads correspond to native threads of the
7497 underlying operating system.
7502 @strong{95}. Implementation-defined @code{policy_identifiers} allowed
7503 in a pragma @code{Task_Dispatching_Policy}. See D.2.2(3).
7506 There are no implementation-defined policy-identifiers allowed in this
7512 @strong{96}. Implementation-defined aspects of priority inversion. See
7516 Execution of a task cannot be preempted by the implementation processing
7517 of delay expirations for lower priority tasks.
7522 @strong{97}. Implementation defined task dispatching. See D.2.2(18).
7527 Tasks map to IRIX threads, and the dispatching policy is as defied by
7528 the IRIX implementation of threads.
7530 The policy is the same as that of the underlying threads implementation.
7535 @strong{98}. Implementation-defined @code{policy_identifiers} allowed
7536 in a pragma @code{Locking_Policy}. See D.3(4).
7539 The only implementation defined policy permitted in GNAT is
7540 @code{Inheritance_Locking}. On targets that support this policy, locking
7541 is implemented by inheritance, i.e.@: the task owning the lock operates
7542 at a priority equal to the highest priority of any task currently
7543 requesting the lock.
7548 @strong{99}. Default ceiling priorities. See D.3(10).
7551 The ceiling priority of protected objects of the type
7552 @code{System.Interrupt_Priority'Last} as described in the Ada 95
7553 Reference Manual D.3(10),
7558 @strong{100}. The ceiling of any protected object used internally by
7559 the implementation. See D.3(16).
7562 The ceiling priority of internal protected objects is
7563 @code{System.Priority'Last}.
7568 @strong{101}. Implementation-defined queuing policies. See D.4(1).
7571 There are no implementation-defined queueing policies.
7576 @strong{102}. On a multiprocessor, any conditions that cause the
7577 completion of an aborted construct to be delayed later than what is
7578 specified for a single processor. See D.6(3).
7581 The semantics for abort on a multi-processor is the same as on a single
7582 processor, there are no further delays.
7587 @strong{103}. Any operations that implicitly require heap storage
7588 allocation. See D.7(8).
7591 The only operation that implicitly requires heap storage allocation is
7597 @strong{104}. Implementation-defined aspects of pragma
7598 @code{Restrictions}. See D.7(20).
7601 There are no such implementation-defined aspects.
7606 @strong{105}. Implementation-defined aspects of package
7607 @code{Real_Time}. See D.8(17).
7610 There are no implementation defined aspects of package @code{Real_Time}.
7615 @strong{106}. Implementation-defined aspects of
7616 @code{delay_statements}. See D.9(8).
7619 Any difference greater than one microsecond will cause the task to be
7620 delayed (see D.9(7)).
7625 @strong{107}. The upper bound on the duration of interrupt blocking
7626 caused by the implementation. See D.12(5).
7629 The upper bound is determined by the underlying operating system. In
7630 no cases is it more than 10 milliseconds.
7635 @strong{108}. The means for creating and executing distributed
7639 The GLADE package provides a utility GNATDIST for creating and executing
7640 distributed programs. See the GLADE reference manual for further details.
7645 @strong{109}. Any events that can result in a partition becoming
7646 inaccessible. See E.1(7).
7649 See the GLADE reference manual for full details on such events.
7654 @strong{110}. The scheduling policies, treatment of priorities, and
7655 management of shared resources between partitions in certain cases. See
7659 See the GLADE reference manual for full details on these aspects of
7660 multi-partition execution.
7665 @strong{111}. Events that cause the version of a compilation unit to
7669 Editing the source file of a compilation unit, or the source files of
7670 any units on which it is dependent in a significant way cause the version
7671 to change. No other actions cause the version number to change. All changes
7672 are significant except those which affect only layout, capitalization or
7678 @strong{112}. Whether the execution of the remote subprogram is
7679 immediately aborted as a result of cancellation. See E.4(13).
7682 See the GLADE reference manual for details on the effect of abort in
7683 a distributed application.
7688 @strong{113}. Implementation-defined aspects of the PCS@. See E.5(25).
7691 See the GLADE reference manual for a full description of all implementation
7692 defined aspects of the PCS@.
7697 @strong{114}. Implementation-defined interfaces in the PCS@. See
7701 See the GLADE reference manual for a full description of all
7702 implementation defined interfaces.
7707 @strong{115}. The values of named numbers in the package
7708 @code{Decimal}. See F.2(7).
7720 @item Max_Decimal_Digits
7727 @strong{116}. The value of @code{Max_Picture_Length} in the package
7728 @code{Text_IO.Editing}. See F.3.3(16).
7736 @strong{117}. The value of @code{Max_Picture_Length} in the package
7737 @code{Wide_Text_IO.Editing}. See F.3.4(5).
7745 @strong{118}. The accuracy actually achieved by the complex elementary
7746 functions and by other complex arithmetic operations. See G.1(1).
7749 Standard library functions are used for the complex arithmetic
7750 operations. Only fast math mode is currently supported.
7755 @strong{119}. The sign of a zero result (or a component thereof) from
7756 any operator or function in @code{Numerics.Generic_Complex_Types}, when
7757 @code{Real'Signed_Zeros} is True. See G.1.1(53).
7760 The signs of zero values are as recommended by the relevant
7761 implementation advice.
7766 @strong{120}. The sign of a zero result (or a component thereof) from
7767 any operator or function in
7768 @code{Numerics.Generic_Complex_Elementary_Functions}, when
7769 @code{Real'Signed_Zeros} is @code{True}. See G.1.2(45).
7772 The signs of zero values are as recommended by the relevant
7773 implementation advice.
7778 @strong{121}. Whether the strict mode or the relaxed mode is the
7779 default. See G.2(2).
7782 The strict mode is the default. There is no separate relaxed mode. GNAT
7783 provides a highly efficient implementation of strict mode.
7788 @strong{122}. The result interval in certain cases of fixed-to-float
7789 conversion. See G.2.1(10).
7792 For cases where the result interval is implementation dependent, the
7793 accuracy is that provided by performing all operations in 64-bit IEEE
7794 floating-point format.
7799 @strong{123}. The result of a floating point arithmetic operation in
7800 overflow situations, when the @code{Machine_Overflows} attribute of the
7801 result type is @code{False}. See G.2.1(13).
7804 Infinite and Nan values are produced as dictated by the IEEE
7805 floating-point standard.
7810 @strong{124}. The result interval for division (or exponentiation by a
7811 negative exponent), when the floating point hardware implements division
7812 as multiplication by a reciprocal. See G.2.1(16).
7815 Not relevant, division is IEEE exact.
7820 @strong{125}. The definition of close result set, which determines the
7821 accuracy of certain fixed point multiplications and divisions. See
7825 Operations in the close result set are performed using IEEE long format
7826 floating-point arithmetic. The input operands are converted to
7827 floating-point, the operation is done in floating-point, and the result
7828 is converted to the target type.
7833 @strong{126}. Conditions on a @code{universal_real} operand of a fixed
7834 point multiplication or division for which the result shall be in the
7835 perfect result set. See G.2.3(22).
7838 The result is only defined to be in the perfect result set if the result
7839 can be computed by a single scaling operation involving a scale factor
7840 representable in 64-bits.
7845 @strong{127}. The result of a fixed point arithmetic operation in
7846 overflow situations, when the @code{Machine_Overflows} attribute of the
7847 result type is @code{False}. See G.2.3(27).
7850 Not relevant, @code{Machine_Overflows} is @code{True} for fixed-point
7856 @strong{128}. The result of an elementary function reference in
7857 overflow situations, when the @code{Machine_Overflows} attribute of the
7858 result type is @code{False}. See G.2.4(4).
7861 IEEE infinite and Nan values are produced as appropriate.
7866 @strong{129}. The value of the angle threshold, within which certain
7867 elementary functions, complex arithmetic operations, and complex
7868 elementary functions yield results conforming to a maximum relative
7869 error bound. See G.2.4(10).
7872 Information on this subject is not yet available.
7877 @strong{130}. The accuracy of certain elementary functions for
7878 parameters beyond the angle threshold. See G.2.4(10).
7881 Information on this subject is not yet available.
7886 @strong{131}. The result of a complex arithmetic operation or complex
7887 elementary function reference in overflow situations, when the
7888 @code{Machine_Overflows} attribute of the corresponding real type is
7889 @code{False}. See G.2.6(5).
7892 IEEE infinite and Nan values are produced as appropriate.
7897 @strong{132}. The accuracy of certain complex arithmetic operations and
7898 certain complex elementary functions for parameters (or components
7899 thereof) beyond the angle threshold. See G.2.6(8).
7902 Information on those subjects is not yet available.
7907 @strong{133}. Information regarding bounded errors and erroneous
7908 execution. See H.2(1).
7911 Information on this subject is not yet available.
7916 @strong{134}. Implementation-defined aspects of pragma
7917 @code{Inspection_Point}. See H.3.2(8).
7920 Pragma @code{Inspection_Point} ensures that the variable is live and can
7921 be examined by the debugger at the inspection point.
7926 @strong{135}. Implementation-defined aspects of pragma
7927 @code{Restrictions}. See H.4(25).
7930 There are no implementation-defined aspects of pragma @code{Restrictions}. The
7931 use of pragma @code{Restrictions [No_Exceptions]} has no effect on the
7932 generated code. Checks must suppressed by use of pragma @code{Suppress}.
7937 @strong{136}. Any restrictions on pragma @code{Restrictions}. See
7941 There are no restrictions on pragma @code{Restrictions}.
7943 @node Intrinsic Subprograms
7944 @chapter Intrinsic Subprograms
7945 @cindex Intrinsic Subprograms
7948 * Intrinsic Operators::
7949 * Enclosing_Entity::
7950 * Exception_Information::
7951 * Exception_Message::
7959 * Shift_Right_Arithmetic::
7964 GNAT allows a user application program to write the declaration:
7966 @smallexample @c ada
7967 pragma Import (Intrinsic, name);
7971 providing that the name corresponds to one of the implemented intrinsic
7972 subprograms in GNAT, and that the parameter profile of the referenced
7973 subprogram meets the requirements. This chapter describes the set of
7974 implemented intrinsic subprograms, and the requirements on parameter profiles.
7975 Note that no body is supplied; as with other uses of pragma Import, the
7976 body is supplied elsewhere (in this case by the compiler itself). Note
7977 that any use of this feature is potentially non-portable, since the
7978 Ada standard does not require Ada compilers to implement this feature.
7980 @node Intrinsic Operators
7981 @section Intrinsic Operators
7982 @cindex Intrinsic operator
7985 All the predefined numeric operators in package Standard
7986 in @code{pragma Import (Intrinsic,..)}
7987 declarations. In the binary operator case, the operands must have the same
7988 size. The operand or operands must also be appropriate for
7989 the operator. For example, for addition, the operands must
7990 both be floating-point or both be fixed-point, and the
7991 right operand for @code{"**"} must have a root type of
7992 @code{Standard.Integer'Base}.
7993 You can use an intrinsic operator declaration as in the following example:
7995 @smallexample @c ada
7996 type Int1 is new Integer;
7997 type Int2 is new Integer;
7999 function "+" (X1 : Int1; X2 : Int2) return Int1;
8000 function "+" (X1 : Int1; X2 : Int2) return Int2;
8001 pragma Import (Intrinsic, "+");
8005 This declaration would permit ``mixed mode'' arithmetic on items
8006 of the differing types @code{Int1} and @code{Int2}.
8007 It is also possible to specify such operators for private types, if the
8008 full views are appropriate arithmetic types.
8010 @node Enclosing_Entity
8011 @section Enclosing_Entity
8012 @cindex Enclosing_Entity
8014 This intrinsic subprogram is used in the implementation of the
8015 library routine @code{GNAT.Source_Info}. The only useful use of the
8016 intrinsic import in this case is the one in this unit, so an
8017 application program should simply call the function
8018 @code{GNAT.Source_Info.Enclosing_Entity} to obtain the name of
8019 the current subprogram, package, task, entry, or protected subprogram.
8021 @node Exception_Information
8022 @section Exception_Information
8023 @cindex Exception_Information'
8025 This intrinsic subprogram is used in the implementation of the
8026 library routine @code{GNAT.Current_Exception}. The only useful
8027 use of the intrinsic import in this case is the one in this unit,
8028 so an application program should simply call the function
8029 @code{GNAT.Current_Exception.Exception_Information} to obtain
8030 the exception information associated with the current exception.
8032 @node Exception_Message
8033 @section Exception_Message
8034 @cindex Exception_Message
8036 This intrinsic subprogram is used in the implementation of the
8037 library routine @code{GNAT.Current_Exception}. The only useful
8038 use of the intrinsic import in this case is the one in this unit,
8039 so an application program should simply call the function
8040 @code{GNAT.Current_Exception.Exception_Message} to obtain
8041 the message associated with the current exception.
8043 @node Exception_Name
8044 @section Exception_Name
8045 @cindex Exception_Name
8047 This intrinsic subprogram is used in the implementation of the
8048 library routine @code{GNAT.Current_Exception}. The only useful
8049 use of the intrinsic import in this case is the one in this unit,
8050 so an application program should simply call the function
8051 @code{GNAT.Current_Exception.Exception_Name} to obtain
8052 the name of the current exception.
8058 This intrinsic subprogram is used in the implementation of the
8059 library routine @code{GNAT.Source_Info}. The only useful use of the
8060 intrinsic import in this case is the one in this unit, so an
8061 application program should simply call the function
8062 @code{GNAT.Source_Info.File} to obtain the name of the current
8069 This intrinsic subprogram is used in the implementation of the
8070 library routine @code{GNAT.Source_Info}. The only useful use of the
8071 intrinsic import in this case is the one in this unit, so an
8072 application program should simply call the function
8073 @code{GNAT.Source_Info.Line} to obtain the number of the current
8077 @section Rotate_Left
8080 In standard Ada 95, the @code{Rotate_Left} function is available only
8081 for the predefined modular types in package @code{Interfaces}. However, in
8082 GNAT it is possible to define a Rotate_Left function for a user
8083 defined modular type or any signed integer type as in this example:
8085 @smallexample @c ada
8087 (Value : My_Modular_Type;
8089 return My_Modular_Type;
8093 The requirements are that the profile be exactly as in the example
8094 above. The only modifications allowed are in the formal parameter
8095 names, and in the type of @code{Value} and the return type, which
8096 must be the same, and must be either a signed integer type, or
8097 a modular integer type with a binary modulus, and the size must
8098 be 8. 16, 32 or 64 bits.
8101 @section Rotate_Right
8102 @cindex Rotate_Right
8104 A @code{Rotate_Right} function can be defined for any user defined
8105 binary modular integer type, or signed integer type, as described
8106 above for @code{Rotate_Left}.
8112 A @code{Shift_Left} function can be defined for any user defined
8113 binary modular integer type, or signed integer type, as described
8114 above for @code{Rotate_Left}.
8117 @section Shift_Right
8120 A @code{Shift_Right} function can be defined for any user defined
8121 binary modular integer type, or signed integer type, as described
8122 above for @code{Rotate_Left}.
8124 @node Shift_Right_Arithmetic
8125 @section Shift_Right_Arithmetic
8126 @cindex Shift_Right_Arithmetic
8128 A @code{Shift_Right_Arithmetic} function can be defined for any user
8129 defined binary modular integer type, or signed integer type, as described
8130 above for @code{Rotate_Left}.
8132 @node Source_Location
8133 @section Source_Location
8134 @cindex Source_Location
8136 This intrinsic subprogram is used in the implementation of the
8137 library routine @code{GNAT.Source_Info}. The only useful use of the
8138 intrinsic import in this case is the one in this unit, so an
8139 application program should simply call the function
8140 @code{GNAT.Source_Info.Source_Location} to obtain the current
8141 source file location.
8143 @node Representation Clauses and Pragmas
8144 @chapter Representation Clauses and Pragmas
8145 @cindex Representation Clauses
8148 * Alignment Clauses::
8150 * Storage_Size Clauses::
8151 * Size of Variant Record Objects::
8152 * Biased Representation ::
8153 * Value_Size and Object_Size Clauses::
8154 * Component_Size Clauses::
8155 * Bit_Order Clauses::
8156 * Effect of Bit_Order on Byte Ordering::
8157 * Pragma Pack for Arrays::
8158 * Pragma Pack for Records::
8159 * Record Representation Clauses::
8160 * Enumeration Clauses::
8162 * Effect of Convention on Representation::
8163 * Determining the Representations chosen by GNAT::
8167 @cindex Representation Clause
8168 @cindex Representation Pragma
8169 @cindex Pragma, representation
8170 This section describes the representation clauses accepted by GNAT, and
8171 their effect on the representation of corresponding data objects.
8173 GNAT fully implements Annex C (Systems Programming). This means that all
8174 the implementation advice sections in chapter 13 are fully implemented.
8175 However, these sections only require a minimal level of support for
8176 representation clauses. GNAT provides much more extensive capabilities,
8177 and this section describes the additional capabilities provided.
8179 @node Alignment Clauses
8180 @section Alignment Clauses
8181 @cindex Alignment Clause
8184 GNAT requires that all alignment clauses specify a power of 2, and all
8185 default alignments are always a power of 2. The default alignment
8186 values are as follows:
8189 @item @emph{Primitive Types}.
8190 For primitive types, the alignment is the minimum of the actual size of
8191 objects of the type divided by @code{Storage_Unit},
8192 and the maximum alignment supported by the target.
8193 (This maximum alignment is given by the GNAT-specific attribute
8194 @code{Standard'Maximum_Alignment}; see @ref{Maximum_Alignment}.)
8195 @cindex @code{Maximum_Alignment} attribute
8196 For example, for type @code{Long_Float}, the object size is 8 bytes, and the
8197 default alignment will be 8 on any target that supports alignments
8198 this large, but on some targets, the maximum alignment may be smaller
8199 than 8, in which case objects of type @code{Long_Float} will be maximally
8202 @item @emph{Arrays}.
8203 For arrays, the alignment is equal to the alignment of the component type
8204 for the normal case where no packing or component size is given. If the
8205 array is packed, and the packing is effective (see separate section on
8206 packed arrays), then the alignment will be one for long packed arrays,
8207 or arrays whose length is not known at compile time. For short packed
8208 arrays, which are handled internally as modular types, the alignment
8209 will be as described for primitive types, e.g.@: a packed array of length
8210 31 bits will have an object size of four bytes, and an alignment of 4.
8212 @item @emph{Records}.
8213 For the normal non-packed case, the alignment of a record is equal to
8214 the maximum alignment of any of its components. For tagged records, this
8215 includes the implicit access type used for the tag. If a pragma @code{Pack} is
8216 used and all fields are packable (see separate section on pragma @code{Pack}),
8217 then the resulting alignment is 1.
8219 A special case is when:
8222 the size of the record is given explicitly, or a
8223 full record representation clause is given, and
8225 the size of the record is 2, 4, or 8 bytes.
8228 In this case, an alignment is chosen to match the
8229 size of the record. For example, if we have:
8231 @smallexample @c ada
8232 type Small is record
8235 for Small'Size use 16;
8239 then the default alignment of the record type @code{Small} is 2, not 1. This
8240 leads to more efficient code when the record is treated as a unit, and also
8241 allows the type to specified as @code{Atomic} on architectures requiring
8247 An alignment clause may
8248 always specify a larger alignment than the default value, up to some
8249 maximum value dependent on the target (obtainable by using the
8250 attribute reference @code{Standard'Maximum_Alignment}).
8252 it is permissible to specify a smaller alignment than the default value
8253 is for a record with a record representation clause.
8254 In this case, packable fields for which a component clause is
8255 given still result in a default alignment corresponding to the original
8256 type, but this may be overridden, since these components in fact only
8257 require an alignment of one byte. For example, given
8259 @smallexample @c ada
8265 A at 0 range 0 .. 31;
8268 for V'alignment use 1;
8272 @cindex Alignment, default
8273 The default alignment for the type @code{V} is 4, as a result of the
8274 Integer field in the record, but since this field is placed with a
8275 component clause, it is permissible, as shown, to override the default
8276 alignment of the record with a smaller value.
8279 @section Size Clauses
8283 The default size for a type @code{T} is obtainable through the
8284 language-defined attribute @code{T'Size} and also through the
8285 equivalent GNAT-defined attribute @code{T'Value_Size}.
8286 For objects of type @code{T}, GNAT will generally increase the type size
8287 so that the object size (obtainable through the GNAT-defined attribute
8288 @code{T'Object_Size})
8289 is a multiple of @code{T'Alignment * Storage_Unit}.
8292 @smallexample @c ada
8293 type Smallint is range 1 .. 6;
8302 In this example, @code{Smallint'Size} = @code{Smallint'Value_Size} = 3,
8303 as specified by the RM rules,
8304 but objects of this type will have a size of 8
8305 (@code{Smallint'Object_Size} = 8),
8306 since objects by default occupy an integral number
8307 of storage units. On some targets, notably older
8308 versions of the Digital Alpha, the size of stand
8309 alone objects of this type may be 32, reflecting
8310 the inability of the hardware to do byte load/stores.
8312 Similarly, the size of type @code{Rec} is 40 bits
8313 (@code{Rec'Size} = @code{Rec'Value_Size} = 40), but
8314 the alignment is 4, so objects of this type will have
8315 their size increased to 64 bits so that it is a multiple
8316 of the alignment (in bits). The reason for this decision, which is
8317 in accordance with the specific Implementation Advice in RM 13.3(43):
8320 A @code{Size} clause should be supported for an object if the specified
8321 @code{Size} is at least as large as its subtype's @code{Size}, and corresponds
8322 to a size in storage elements that is a multiple of the object's
8323 @code{Alignment} (if the @code{Alignment} is nonzero).
8327 An explicit size clause may be used to override the default size by
8328 increasing it. For example, if we have:
8330 @smallexample @c ada
8331 type My_Boolean is new Boolean;
8332 for My_Boolean'Size use 32;
8336 then values of this type will always be 32 bits long. In the case of
8337 discrete types, the size can be increased up to 64 bits, with the effect
8338 that the entire specified field is used to hold the value, sign- or
8339 zero-extended as appropriate. If more than 64 bits is specified, then
8340 padding space is allocated after the value, and a warning is issued that
8341 there are unused bits.
8343 Similarly the size of records and arrays may be increased, and the effect
8344 is to add padding bits after the value. This also causes a warning message
8347 The largest Size value permitted in GNAT is 2**31@minus{}1. Since this is a
8348 Size in bits, this corresponds to an object of size 256 megabytes (minus
8349 one). This limitation is true on all targets. The reason for this
8350 limitation is that it improves the quality of the code in many cases
8351 if it is known that a Size value can be accommodated in an object of
8354 @node Storage_Size Clauses
8355 @section Storage_Size Clauses
8356 @cindex Storage_Size Clause
8359 For tasks, the @code{Storage_Size} clause specifies the amount of space
8360 to be allocated for the task stack. This cannot be extended, and if the
8361 stack is exhausted, then @code{Storage_Error} will be raised (if stack
8362 checking is enabled). Use a @code{Storage_Size} attribute definition clause,
8363 or a @code{Storage_Size} pragma in the task definition to set the
8364 appropriate required size. A useful technique is to include in every
8365 task definition a pragma of the form:
8367 @smallexample @c ada
8368 pragma Storage_Size (Default_Stack_Size);
8372 Then @code{Default_Stack_Size} can be defined in a global package, and
8373 modified as required. Any tasks requiring stack sizes different from the
8374 default can have an appropriate alternative reference in the pragma.
8376 For access types, the @code{Storage_Size} clause specifies the maximum
8377 space available for allocation of objects of the type. If this space is
8378 exceeded then @code{Storage_Error} will be raised by an allocation attempt.
8379 In the case where the access type is declared local to a subprogram, the
8380 use of a @code{Storage_Size} clause triggers automatic use of a special
8381 predefined storage pool (@code{System.Pool_Size}) that ensures that all
8382 space for the pool is automatically reclaimed on exit from the scope in
8383 which the type is declared.
8385 A special case recognized by the compiler is the specification of a
8386 @code{Storage_Size} of zero for an access type. This means that no
8387 items can be allocated from the pool, and this is recognized at compile
8388 time, and all the overhead normally associated with maintaining a fixed
8389 size storage pool is eliminated. Consider the following example:
8391 @smallexample @c ada
8393 type R is array (Natural) of Character;
8394 type P is access all R;
8395 for P'Storage_Size use 0;
8396 -- Above access type intended only for interfacing purposes
8400 procedure g (m : P);
8401 pragma Import (C, g);
8412 As indicated in this example, these dummy storage pools are often useful in
8413 connection with interfacing where no object will ever be allocated. If you
8414 compile the above example, you get the warning:
8417 p.adb:16:09: warning: allocation from empty storage pool
8418 p.adb:16:09: warning: Storage_Error will be raised at run time
8422 Of course in practice, there will not be any explicit allocators in the
8423 case of such an access declaration.
8425 @node Size of Variant Record Objects
8426 @section Size of Variant Record Objects
8427 @cindex Size, variant record objects
8428 @cindex Variant record objects, size
8431 In the case of variant record objects, there is a question whether Size gives
8432 information about a particular variant, or the maximum size required
8433 for any variant. Consider the following program
8435 @smallexample @c ada
8436 with Text_IO; use Text_IO;
8438 type R1 (A : Boolean := False) is record
8440 when True => X : Character;
8449 Put_Line (Integer'Image (V1'Size));
8450 Put_Line (Integer'Image (V2'Size));
8455 Here we are dealing with a variant record, where the True variant
8456 requires 16 bits, and the False variant requires 8 bits.
8457 In the above example, both V1 and V2 contain the False variant,
8458 which is only 8 bits long. However, the result of running the
8467 The reason for the difference here is that the discriminant value of
8468 V1 is fixed, and will always be False. It is not possible to assign
8469 a True variant value to V1, therefore 8 bits is sufficient. On the
8470 other hand, in the case of V2, the initial discriminant value is
8471 False (from the default), but it is possible to assign a True
8472 variant value to V2, therefore 16 bits must be allocated for V2
8473 in the general case, even fewer bits may be needed at any particular
8474 point during the program execution.
8476 As can be seen from the output of this program, the @code{'Size}
8477 attribute applied to such an object in GNAT gives the actual allocated
8478 size of the variable, which is the largest size of any of the variants.
8479 The Ada Reference Manual is not completely clear on what choice should
8480 be made here, but the GNAT behavior seems most consistent with the
8481 language in the RM@.
8483 In some cases, it may be desirable to obtain the size of the current
8484 variant, rather than the size of the largest variant. This can be
8485 achieved in GNAT by making use of the fact that in the case of a
8486 subprogram parameter, GNAT does indeed return the size of the current
8487 variant (because a subprogram has no way of knowing how much space
8488 is actually allocated for the actual).
8490 Consider the following modified version of the above program:
8492 @smallexample @c ada
8493 with Text_IO; use Text_IO;
8495 type R1 (A : Boolean := False) is record
8497 when True => X : Character;
8504 function Size (V : R1) return Integer is
8510 Put_Line (Integer'Image (V2'Size));
8511 Put_Line (Integer'IMage (Size (V2)));
8513 Put_Line (Integer'Image (V2'Size));
8514 Put_Line (Integer'IMage (Size (V2)));
8519 The output from this program is
8529 Here we see that while the @code{'Size} attribute always returns
8530 the maximum size, regardless of the current variant value, the
8531 @code{Size} function does indeed return the size of the current
8534 @node Biased Representation
8535 @section Biased Representation
8536 @cindex Size for biased representation
8537 @cindex Biased representation
8540 In the case of scalars with a range starting at other than zero, it is
8541 possible in some cases to specify a size smaller than the default minimum
8542 value, and in such cases, GNAT uses an unsigned biased representation,
8543 in which zero is used to represent the lower bound, and successive values
8544 represent successive values of the type.
8546 For example, suppose we have the declaration:
8548 @smallexample @c ada
8549 type Small is range -7 .. -4;
8550 for Small'Size use 2;
8554 Although the default size of type @code{Small} is 4, the @code{Size}
8555 clause is accepted by GNAT and results in the following representation
8559 -7 is represented as 2#00#
8560 -6 is represented as 2#01#
8561 -5 is represented as 2#10#
8562 -4 is represented as 2#11#
8566 Biased representation is only used if the specified @code{Size} clause
8567 cannot be accepted in any other manner. These reduced sizes that force
8568 biased representation can be used for all discrete types except for
8569 enumeration types for which a representation clause is given.
8571 @node Value_Size and Object_Size Clauses
8572 @section Value_Size and Object_Size Clauses
8575 @cindex Size, of objects
8578 In Ada 95, @code{T'Size} for a type @code{T} is the minimum number of bits
8579 required to hold values of type @code{T}. Although this interpretation was
8580 allowed in Ada 83, it was not required, and this requirement in practice
8581 can cause some significant difficulties. For example, in most Ada 83
8582 compilers, @code{Natural'Size} was 32. However, in Ada 95,
8583 @code{Natural'Size} is
8584 typically 31. This means that code may change in behavior when moving
8585 from Ada 83 to Ada 95. For example, consider:
8587 @smallexample @c ada
8594 at 0 range 0 .. Natural'Size - 1;
8595 at 0 range Natural'Size .. 2 * Natural'Size - 1;
8600 In the above code, since the typical size of @code{Natural} objects
8601 is 32 bits and @code{Natural'Size} is 31, the above code can cause
8602 unexpected inefficient packing in Ada 95, and in general there are
8603 cases where the fact that the object size can exceed the
8604 size of the type causes surprises.
8606 To help get around this problem GNAT provides two implementation
8607 defined attributes, @code{Value_Size} and @code{Object_Size}. When
8608 applied to a type, these attributes yield the size of the type
8609 (corresponding to the RM defined size attribute), and the size of
8610 objects of the type respectively.
8612 The @code{Object_Size} is used for determining the default size of
8613 objects and components. This size value can be referred to using the
8614 @code{Object_Size} attribute. The phrase ``is used'' here means that it is
8615 the basis of the determination of the size. The backend is free to
8616 pad this up if necessary for efficiency, e.g.@: an 8-bit stand-alone
8617 character might be stored in 32 bits on a machine with no efficient
8618 byte access instructions such as the Alpha.
8620 The default rules for the value of @code{Object_Size} for
8621 discrete types are as follows:
8625 The @code{Object_Size} for base subtypes reflect the natural hardware
8626 size in bits (run the compiler with @option{-gnatS} to find those values
8627 for numeric types). Enumeration types and fixed-point base subtypes have
8628 8, 16, 32 or 64 bits for this size, depending on the range of values
8632 The @code{Object_Size} of a subtype is the same as the
8633 @code{Object_Size} of
8634 the type from which it is obtained.
8637 The @code{Object_Size} of a derived base type is copied from the parent
8638 base type, and the @code{Object_Size} of a derived first subtype is copied
8639 from the parent first subtype.
8643 The @code{Value_Size} attribute
8644 is the (minimum) number of bits required to store a value
8646 This value is used to determine how tightly to pack
8647 records or arrays with components of this type, and also affects
8648 the semantics of unchecked conversion (unchecked conversions where
8649 the @code{Value_Size} values differ generate a warning, and are potentially
8652 The default rules for the value of @code{Value_Size} are as follows:
8656 The @code{Value_Size} for a base subtype is the minimum number of bits
8657 required to store all values of the type (including the sign bit
8658 only if negative values are possible).
8661 If a subtype statically matches the first subtype of a given type, then it has
8662 by default the same @code{Value_Size} as the first subtype. This is a
8663 consequence of RM 13.1(14) (``if two subtypes statically match,
8664 then their subtype-specific aspects are the same''.)
8667 All other subtypes have a @code{Value_Size} corresponding to the minimum
8668 number of bits required to store all values of the subtype. For
8669 dynamic bounds, it is assumed that the value can range down or up
8670 to the corresponding bound of the ancestor
8674 The RM defined attribute @code{Size} corresponds to the
8675 @code{Value_Size} attribute.
8677 The @code{Size} attribute may be defined for a first-named subtype. This sets
8678 the @code{Value_Size} of
8679 the first-named subtype to the given value, and the
8680 @code{Object_Size} of this first-named subtype to the given value padded up
8681 to an appropriate boundary. It is a consequence of the default rules
8682 above that this @code{Object_Size} will apply to all further subtypes. On the
8683 other hand, @code{Value_Size} is affected only for the first subtype, any
8684 dynamic subtypes obtained from it directly, and any statically matching
8685 subtypes. The @code{Value_Size} of any other static subtypes is not affected.
8687 @code{Value_Size} and
8688 @code{Object_Size} may be explicitly set for any subtype using
8689 an attribute definition clause. Note that the use of these attributes
8690 can cause the RM 13.1(14) rule to be violated. If two access types
8691 reference aliased objects whose subtypes have differing @code{Object_Size}
8692 values as a result of explicit attribute definition clauses, then it
8693 is erroneous to convert from one access subtype to the other.
8695 At the implementation level, Esize stores the Object_Size and the
8696 RM_Size field stores the @code{Value_Size} (and hence the value of the
8697 @code{Size} attribute,
8698 which, as noted above, is equivalent to @code{Value_Size}).
8700 To get a feel for the difference, consider the following examples (note
8701 that in each case the base is @code{Short_Short_Integer} with a size of 8):
8704 Object_Size Value_Size
8706 type x1 is range 0 .. 5; 8 3
8708 type x2 is range 0 .. 5;
8709 for x2'size use 12; 16 12
8711 subtype x3 is x2 range 0 .. 3; 16 2
8713 subtype x4 is x2'base range 0 .. 10; 8 4
8715 subtype x5 is x2 range 0 .. dynamic; 16 3*
8717 subtype x6 is x2'base range 0 .. dynamic; 8 3*
8722 Note: the entries marked ``3*'' are not actually specified by the Ada 95 RM,
8723 but it seems in the spirit of the RM rules to allocate the minimum number
8724 of bits (here 3, given the range for @code{x2})
8725 known to be large enough to hold the given range of values.
8727 So far, so good, but GNAT has to obey the RM rules, so the question is
8728 under what conditions must the RM @code{Size} be used.
8729 The following is a list
8730 of the occasions on which the RM @code{Size} must be used:
8734 Component size for packed arrays or records
8737 Value of the attribute @code{Size} for a type
8740 Warning about sizes not matching for unchecked conversion
8744 For record types, the @code{Object_Size} is always a multiple of the
8745 alignment of the type (this is true for all types). In some cases the
8746 @code{Value_Size} can be smaller. Consider:
8756 On a typical 32-bit architecture, the X component will be four bytes, and
8757 require four-byte alignment, and the Y component will be one byte. In this
8758 case @code{R'Value_Size} will be 40 (bits) since this is the minimum size
8759 required to store a value of this type, and for example, it is permissible
8760 to have a component of type R in an outer record whose component size is
8761 specified to be 48 bits. However, @code{R'Object_Size} will be 64 (bits),
8762 since it must be rounded up so that this value is a multiple of the
8763 alignment (4 bytes = 32 bits).
8766 For all other types, the @code{Object_Size}
8767 and Value_Size are the same (and equivalent to the RM attribute @code{Size}).
8768 Only @code{Size} may be specified for such types.
8770 @node Component_Size Clauses
8771 @section Component_Size Clauses
8772 @cindex Component_Size Clause
8775 Normally, the value specified in a component clause must be consistent
8776 with the subtype of the array component with regard to size and alignment.
8777 In other words, the value specified must be at least equal to the size
8778 of this subtype, and must be a multiple of the alignment value.
8780 In addition, component size clauses are allowed which cause the array
8781 to be packed, by specifying a smaller value. The cases in which this
8782 is allowed are for component size values in the range 1 through 63. The value
8783 specified must not be smaller than the Size of the subtype. GNAT will
8784 accurately honor all packing requests in this range. For example, if
8787 @smallexample @c ada
8788 type r is array (1 .. 8) of Natural;
8789 for r'Component_Size use 31;
8793 then the resulting array has a length of 31 bytes (248 bits = 8 * 31).
8794 Of course access to the components of such an array is considerably
8795 less efficient than if the natural component size of 32 is used.
8797 @node Bit_Order Clauses
8798 @section Bit_Order Clauses
8799 @cindex Bit_Order Clause
8800 @cindex bit ordering
8801 @cindex ordering, of bits
8804 For record subtypes, GNAT permits the specification of the @code{Bit_Order}
8805 attribute. The specification may either correspond to the default bit
8806 order for the target, in which case the specification has no effect and
8807 places no additional restrictions, or it may be for the non-standard
8808 setting (that is the opposite of the default).
8810 In the case where the non-standard value is specified, the effect is
8811 to renumber bits within each byte, but the ordering of bytes is not
8812 affected. There are certain
8813 restrictions placed on component clauses as follows:
8817 @item Components fitting within a single storage unit.
8819 These are unrestricted, and the effect is merely to renumber bits. For
8820 example if we are on a little-endian machine with @code{Low_Order_First}
8821 being the default, then the following two declarations have exactly
8824 @smallexample @c ada
8827 B : Integer range 1 .. 120;
8831 A at 0 range 0 .. 0;
8832 B at 0 range 1 .. 7;
8837 B : Integer range 1 .. 120;
8840 for R2'Bit_Order use High_Order_First;
8843 A at 0 range 7 .. 7;
8844 B at 0 range 0 .. 6;
8849 The useful application here is to write the second declaration with the
8850 @code{Bit_Order} attribute definition clause, and know that it will be treated
8851 the same, regardless of whether the target is little-endian or big-endian.
8853 @item Components occupying an integral number of bytes.
8855 These are components that exactly fit in two or more bytes. Such component
8856 declarations are allowed, but have no effect, since it is important to realize
8857 that the @code{Bit_Order} specification does not affect the ordering of bytes.
8858 In particular, the following attempt at getting an endian-independent integer
8861 @smallexample @c ada
8866 for R2'Bit_Order use High_Order_First;
8869 A at 0 range 0 .. 31;
8874 This declaration will result in a little-endian integer on a
8875 little-endian machine, and a big-endian integer on a big-endian machine.
8876 If byte flipping is required for interoperability between big- and
8877 little-endian machines, this must be explicitly programmed. This capability
8878 is not provided by @code{Bit_Order}.
8880 @item Components that are positioned across byte boundaries
8882 but do not occupy an integral number of bytes. Given that bytes are not
8883 reordered, such fields would occupy a non-contiguous sequence of bits
8884 in memory, requiring non-trivial code to reassemble. They are for this
8885 reason not permitted, and any component clause specifying such a layout
8886 will be flagged as illegal by GNAT@.
8891 Since the misconception that Bit_Order automatically deals with all
8892 endian-related incompatibilities is a common one, the specification of
8893 a component field that is an integral number of bytes will always
8894 generate a warning. This warning may be suppressed using
8895 @code{pragma Suppress} if desired. The following section contains additional
8896 details regarding the issue of byte ordering.
8898 @node Effect of Bit_Order on Byte Ordering
8899 @section Effect of Bit_Order on Byte Ordering
8900 @cindex byte ordering
8901 @cindex ordering, of bytes
8904 In this section we will review the effect of the @code{Bit_Order} attribute
8905 definition clause on byte ordering. Briefly, it has no effect at all, but
8906 a detailed example will be helpful. Before giving this
8907 example, let us review the precise
8908 definition of the effect of defining @code{Bit_Order}. The effect of a
8909 non-standard bit order is described in section 15.5.3 of the Ada
8913 2 A bit ordering is a method of interpreting the meaning of
8914 the storage place attributes.
8918 To understand the precise definition of storage place attributes in
8919 this context, we visit section 13.5.1 of the manual:
8922 13 A record_representation_clause (without the mod_clause)
8923 specifies the layout. The storage place attributes (see 13.5.2)
8924 are taken from the values of the position, first_bit, and last_bit
8925 expressions after normalizing those values so that first_bit is
8926 less than Storage_Unit.
8930 The critical point here is that storage places are taken from
8931 the values after normalization, not before. So the @code{Bit_Order}
8932 interpretation applies to normalized values. The interpretation
8933 is described in the later part of the 15.5.3 paragraph:
8936 2 A bit ordering is a method of interpreting the meaning of
8937 the storage place attributes. High_Order_First (known in the
8938 vernacular as ``big endian'') means that the first bit of a
8939 storage element (bit 0) is the most significant bit (interpreting
8940 the sequence of bits that represent a component as an unsigned
8941 integer value). Low_Order_First (known in the vernacular as
8942 ``little endian'') means the opposite: the first bit is the
8947 Note that the numbering is with respect to the bits of a storage
8948 unit. In other words, the specification affects only the numbering
8949 of bits within a single storage unit.
8951 We can make the effect clearer by giving an example.
8953 Suppose that we have an external device which presents two bytes, the first
8954 byte presented, which is the first (low addressed byte) of the two byte
8955 record is called Master, and the second byte is called Slave.
8957 The left most (most significant bit is called Control for each byte, and
8958 the remaining 7 bits are called V1, V2, @dots{} V7, where V7 is the rightmost
8959 (least significant) bit.
8961 On a big-endian machine, we can write the following representation clause
8963 @smallexample @c ada
8965 Master_Control : Bit;
8973 Slave_Control : Bit;
8984 Master_Control at 0 range 0 .. 0;
8985 Master_V1 at 0 range 1 .. 1;
8986 Master_V2 at 0 range 2 .. 2;
8987 Master_V3 at 0 range 3 .. 3;
8988 Master_V4 at 0 range 4 .. 4;
8989 Master_V5 at 0 range 5 .. 5;
8990 Master_V6 at 0 range 6 .. 6;
8991 Master_V7 at 0 range 7 .. 7;
8992 Slave_Control at 1 range 0 .. 0;
8993 Slave_V1 at 1 range 1 .. 1;
8994 Slave_V2 at 1 range 2 .. 2;
8995 Slave_V3 at 1 range 3 .. 3;
8996 Slave_V4 at 1 range 4 .. 4;
8997 Slave_V5 at 1 range 5 .. 5;
8998 Slave_V6 at 1 range 6 .. 6;
8999 Slave_V7 at 1 range 7 .. 7;
9004 Now if we move this to a little endian machine, then the bit ordering within
9005 the byte is backwards, so we have to rewrite the record rep clause as:
9007 @smallexample @c ada
9009 Master_Control at 0 range 7 .. 7;
9010 Master_V1 at 0 range 6 .. 6;
9011 Master_V2 at 0 range 5 .. 5;
9012 Master_V3 at 0 range 4 .. 4;
9013 Master_V4 at 0 range 3 .. 3;
9014 Master_V5 at 0 range 2 .. 2;
9015 Master_V6 at 0 range 1 .. 1;
9016 Master_V7 at 0 range 0 .. 0;
9017 Slave_Control at 1 range 7 .. 7;
9018 Slave_V1 at 1 range 6 .. 6;
9019 Slave_V2 at 1 range 5 .. 5;
9020 Slave_V3 at 1 range 4 .. 4;
9021 Slave_V4 at 1 range 3 .. 3;
9022 Slave_V5 at 1 range 2 .. 2;
9023 Slave_V6 at 1 range 1 .. 1;
9024 Slave_V7 at 1 range 0 .. 0;
9029 It is a nuisance to have to rewrite the clause, especially if
9030 the code has to be maintained on both machines. However,
9031 this is a case that we can handle with the
9032 @code{Bit_Order} attribute if it is implemented.
9033 Note that the implementation is not required on byte addressed
9034 machines, but it is indeed implemented in GNAT.
9035 This means that we can simply use the
9036 first record clause, together with the declaration
9038 @smallexample @c ada
9039 for Data'Bit_Order use High_Order_First;
9043 and the effect is what is desired, namely the layout is exactly the same,
9044 independent of whether the code is compiled on a big-endian or little-endian
9047 The important point to understand is that byte ordering is not affected.
9048 A @code{Bit_Order} attribute definition never affects which byte a field
9049 ends up in, only where it ends up in that byte.
9050 To make this clear, let us rewrite the record rep clause of the previous
9053 @smallexample @c ada
9054 for Data'Bit_Order use High_Order_First;
9056 Master_Control at 0 range 0 .. 0;
9057 Master_V1 at 0 range 1 .. 1;
9058 Master_V2 at 0 range 2 .. 2;
9059 Master_V3 at 0 range 3 .. 3;
9060 Master_V4 at 0 range 4 .. 4;
9061 Master_V5 at 0 range 5 .. 5;
9062 Master_V6 at 0 range 6 .. 6;
9063 Master_V7 at 0 range 7 .. 7;
9064 Slave_Control at 0 range 8 .. 8;
9065 Slave_V1 at 0 range 9 .. 9;
9066 Slave_V2 at 0 range 10 .. 10;
9067 Slave_V3 at 0 range 11 .. 11;
9068 Slave_V4 at 0 range 12 .. 12;
9069 Slave_V5 at 0 range 13 .. 13;
9070 Slave_V6 at 0 range 14 .. 14;
9071 Slave_V7 at 0 range 15 .. 15;
9076 This is exactly equivalent to saying (a repeat of the first example):
9078 @smallexample @c ada
9079 for Data'Bit_Order use High_Order_First;
9081 Master_Control at 0 range 0 .. 0;
9082 Master_V1 at 0 range 1 .. 1;
9083 Master_V2 at 0 range 2 .. 2;
9084 Master_V3 at 0 range 3 .. 3;
9085 Master_V4 at 0 range 4 .. 4;
9086 Master_V5 at 0 range 5 .. 5;
9087 Master_V6 at 0 range 6 .. 6;
9088 Master_V7 at 0 range 7 .. 7;
9089 Slave_Control at 1 range 0 .. 0;
9090 Slave_V1 at 1 range 1 .. 1;
9091 Slave_V2 at 1 range 2 .. 2;
9092 Slave_V3 at 1 range 3 .. 3;
9093 Slave_V4 at 1 range 4 .. 4;
9094 Slave_V5 at 1 range 5 .. 5;
9095 Slave_V6 at 1 range 6 .. 6;
9096 Slave_V7 at 1 range 7 .. 7;
9101 Why are they equivalent? Well take a specific field, the @code{Slave_V2}
9102 field. The storage place attributes are obtained by normalizing the
9103 values given so that the @code{First_Bit} value is less than 8. After
9104 normalizing the values (0,10,10) we get (1,2,2) which is exactly what
9105 we specified in the other case.
9107 Now one might expect that the @code{Bit_Order} attribute might affect
9108 bit numbering within the entire record component (two bytes in this
9109 case, thus affecting which byte fields end up in), but that is not
9110 the way this feature is defined, it only affects numbering of bits,
9111 not which byte they end up in.
9113 Consequently it never makes sense to specify a starting bit number
9114 greater than 7 (for a byte addressable field) if an attribute
9115 definition for @code{Bit_Order} has been given, and indeed it
9116 may be actively confusing to specify such a value, so the compiler
9117 generates a warning for such usage.
9119 If you do need to control byte ordering then appropriate conditional
9120 values must be used. If in our example, the slave byte came first on
9121 some machines we might write:
9123 @smallexample @c ada
9124 Master_Byte_First constant Boolean := @dots{};
9126 Master_Byte : constant Natural :=
9127 1 - Boolean'Pos (Master_Byte_First);
9128 Slave_Byte : constant Natural :=
9129 Boolean'Pos (Master_Byte_First);
9131 for Data'Bit_Order use High_Order_First;
9133 Master_Control at Master_Byte range 0 .. 0;
9134 Master_V1 at Master_Byte range 1 .. 1;
9135 Master_V2 at Master_Byte range 2 .. 2;
9136 Master_V3 at Master_Byte range 3 .. 3;
9137 Master_V4 at Master_Byte range 4 .. 4;
9138 Master_V5 at Master_Byte range 5 .. 5;
9139 Master_V6 at Master_Byte range 6 .. 6;
9140 Master_V7 at Master_Byte range 7 .. 7;
9141 Slave_Control at Slave_Byte range 0 .. 0;
9142 Slave_V1 at Slave_Byte range 1 .. 1;
9143 Slave_V2 at Slave_Byte range 2 .. 2;
9144 Slave_V3 at Slave_Byte range 3 .. 3;
9145 Slave_V4 at Slave_Byte range 4 .. 4;
9146 Slave_V5 at Slave_Byte range 5 .. 5;
9147 Slave_V6 at Slave_Byte range 6 .. 6;
9148 Slave_V7 at Slave_Byte range 7 .. 7;
9153 Now to switch between machines, all that is necessary is
9154 to set the boolean constant @code{Master_Byte_First} in
9155 an appropriate manner.
9157 @node Pragma Pack for Arrays
9158 @section Pragma Pack for Arrays
9159 @cindex Pragma Pack (for arrays)
9162 Pragma @code{Pack} applied to an array has no effect unless the component type
9163 is packable. For a component type to be packable, it must be one of the
9170 Any type whose size is specified with a size clause
9172 Any packed array type with a static size
9176 For all these cases, if the component subtype size is in the range
9177 1 through 63, then the effect of the pragma @code{Pack} is exactly as though a
9178 component size were specified giving the component subtype size.
9179 For example if we have:
9181 @smallexample @c ada
9182 type r is range 0 .. 17;
9184 type ar is array (1 .. 8) of r;
9189 Then the component size of @code{ar} will be set to 5 (i.e.@: to @code{r'size},
9190 and the size of the array @code{ar} will be exactly 40 bits.
9192 Note that in some cases this rather fierce approach to packing can produce
9193 unexpected effects. For example, in Ada 95, type Natural typically has a
9194 size of 31, meaning that if you pack an array of Natural, you get 31-bit
9195 close packing, which saves a few bits, but results in far less efficient
9196 access. Since many other Ada compilers will ignore such a packing request,
9197 GNAT will generate a warning on some uses of pragma @code{Pack} that it guesses
9198 might not be what is intended. You can easily remove this warning by
9199 using an explicit @code{Component_Size} setting instead, which never generates
9200 a warning, since the intention of the programmer is clear in this case.
9202 GNAT treats packed arrays in one of two ways. If the size of the array is
9203 known at compile time and is less than 64 bits, then internally the array
9204 is represented as a single modular type, of exactly the appropriate number
9205 of bits. If the length is greater than 63 bits, or is not known at compile
9206 time, then the packed array is represented as an array of bytes, and the
9207 length is always a multiple of 8 bits.
9209 Note that to represent a packed array as a modular type, the alignment must
9210 be suitable for the modular type involved. For example, on typical machines
9211 a 32-bit packed array will be represented by a 32-bit modular integer with
9212 an alignment of four bytes. If you explicitly override the default alignment
9213 with an alignment clause that is too small, the modular representation
9214 cannot be used. For example, consider the following set of declarations:
9216 @smallexample @c ada
9217 type R is range 1 .. 3;
9218 type S is array (1 .. 31) of R;
9219 for S'Component_Size use 2;
9221 for S'Alignment use 1;
9225 If the alignment clause were not present, then a 62-bit modular
9226 representation would be chosen (typically with an alignment of 4 or 8
9227 bytes depending on the target). But the default alignment is overridden
9228 with the explicit alignment clause. This means that the modular
9229 representation cannot be used, and instead the array of bytes
9230 representation must be used, meaning that the length must be a multiple
9231 of 8. Thus the above set of declarations will result in a diagnostic
9232 rejecting the size clause and noting that the minimum size allowed is 64.
9234 @cindex Pragma Pack (for type Natural)
9235 @cindex Pragma Pack warning
9237 One special case that is worth noting occurs when the base type of the
9238 component size is 8/16/32 and the subtype is one bit less. Notably this
9239 occurs with subtype @code{Natural}. Consider:
9241 @smallexample @c ada
9242 type Arr is array (1 .. 32) of Natural;
9247 In all commonly used Ada 83 compilers, this pragma Pack would be ignored,
9248 since typically @code{Natural'Size} is 32 in Ada 83, and in any case most
9249 Ada 83 compilers did not attempt 31 bit packing.
9251 In Ada 95, @code{Natural'Size} is required to be 31. Furthermore, GNAT really
9252 does pack 31-bit subtype to 31 bits. This may result in a substantial
9253 unintended performance penalty when porting legacy Ada 83 code. To help
9254 prevent this, GNAT generates a warning in such cases. If you really want 31
9255 bit packing in a case like this, you can set the component size explicitly:
9257 @smallexample @c ada
9258 type Arr is array (1 .. 32) of Natural;
9259 for Arr'Component_Size use 31;
9263 Here 31-bit packing is achieved as required, and no warning is generated,
9264 since in this case the programmer intention is clear.
9266 @node Pragma Pack for Records
9267 @section Pragma Pack for Records
9268 @cindex Pragma Pack (for records)
9271 Pragma @code{Pack} applied to a record will pack the components to reduce
9272 wasted space from alignment gaps and by reducing the amount of space
9273 taken by components. We distinguish between @emph{packable} components and
9274 @emph{non-packable} components.
9275 Components of the following types are considered packable:
9278 All primitive types are packable.
9281 Small packed arrays, whose size does not exceed 64 bits, and where the
9282 size is statically known at compile time, are represented internally
9283 as modular integers, and so they are also packable.
9288 All packable components occupy the exact number of bits corresponding to
9289 their @code{Size} value, and are packed with no padding bits, i.e.@: they
9290 can start on an arbitrary bit boundary.
9292 All other types are non-packable, they occupy an integral number of
9294 are placed at a boundary corresponding to their alignment requirements.
9296 For example, consider the record
9298 @smallexample @c ada
9299 type Rb1 is array (1 .. 13) of Boolean;
9302 type Rb2 is array (1 .. 65) of Boolean;
9317 The representation for the record x2 is as follows:
9319 @smallexample @c ada
9320 for x2'Size use 224;
9322 l1 at 0 range 0 .. 0;
9323 l2 at 0 range 1 .. 64;
9324 l3 at 12 range 0 .. 31;
9325 l4 at 16 range 0 .. 0;
9326 l5 at 16 range 1 .. 13;
9327 l6 at 18 range 0 .. 71;
9332 Studying this example, we see that the packable fields @code{l1}
9334 of length equal to their sizes, and placed at specific bit boundaries (and
9335 not byte boundaries) to
9336 eliminate padding. But @code{l3} is of a non-packable float type, so
9337 it is on the next appropriate alignment boundary.
9339 The next two fields are fully packable, so @code{l4} and @code{l5} are
9340 minimally packed with no gaps. However, type @code{Rb2} is a packed
9341 array that is longer than 64 bits, so it is itself non-packable. Thus
9342 the @code{l6} field is aligned to the next byte boundary, and takes an
9343 integral number of bytes, i.e.@: 72 bits.
9345 @node Record Representation Clauses
9346 @section Record Representation Clauses
9347 @cindex Record Representation Clause
9350 Record representation clauses may be given for all record types, including
9351 types obtained by record extension. Component clauses are allowed for any
9352 static component. The restrictions on component clauses depend on the type
9355 @cindex Component Clause
9356 For all components of an elementary type, the only restriction on component
9357 clauses is that the size must be at least the 'Size value of the type
9358 (actually the Value_Size). There are no restrictions due to alignment,
9359 and such components may freely cross storage boundaries.
9361 Packed arrays with a size up to and including 64 bits are represented
9362 internally using a modular type with the appropriate number of bits, and
9363 thus the same lack of restriction applies. For example, if you declare:
9365 @smallexample @c ada
9366 type R is array (1 .. 49) of Boolean;
9372 then a component clause for a component of type R may start on any
9373 specified bit boundary, and may specify a value of 49 bits or greater.
9375 The rules for other types are different for GNAT 3 and GNAT 5 versions
9376 (based on GCC 2 and GCC 3 respectively). In GNAT 5, larger components
9377 may also be placed on arbitrary boundaries, so for example, the following
9380 @smallexample @c ada
9381 type R is array (1 .. 79) of Boolean;
9391 G at 0 range 0 .. 0;
9392 H at 0 range 1 .. 1;
9393 L at 0 range 2 .. 80;
9394 R at 0 range 81 .. 159;
9399 In GNAT 3, there are more severe restrictions on larger components.
9400 For non-primitive types, including packed arrays with a size greater than
9401 64 bits, component clauses must respect the alignment requirement of the
9402 type, in particular, always starting on a byte boundary, and the length
9403 must be a multiple of the storage unit.
9405 The following rules regarding tagged types are enforced in both GNAT 3 and
9408 The tag field of a tagged type always occupies an address sized field at
9409 the start of the record. No component clause may attempt to overlay this
9412 In the case of a record extension T1, of a type T, no component clause applied
9413 to the type T1 can specify a storage location that would overlap the first
9414 T'Size bytes of the record.
9416 @node Enumeration Clauses
9417 @section Enumeration Clauses
9419 The only restriction on enumeration clauses is that the range of values
9420 must be representable. For the signed case, if one or more of the
9421 representation values are negative, all values must be in the range:
9423 @smallexample @c ada
9424 System.Min_Int .. System.Max_Int
9428 For the unsigned case, where all values are non negative, the values must
9431 @smallexample @c ada
9432 0 .. System.Max_Binary_Modulus;
9436 A @emph{confirming} representation clause is one in which the values range
9437 from 0 in sequence, i.e.@: a clause that confirms the default representation
9438 for an enumeration type.
9439 Such a confirming representation
9440 is permitted by these rules, and is specially recognized by the compiler so
9441 that no extra overhead results from the use of such a clause.
9443 If an array has an index type which is an enumeration type to which an
9444 enumeration clause has been applied, then the array is stored in a compact
9445 manner. Consider the declarations:
9447 @smallexample @c ada
9448 type r is (A, B, C);
9449 for r use (A => 1, B => 5, C => 10);
9450 type t is array (r) of Character;
9454 The array type t corresponds to a vector with exactly three elements and
9455 has a default size equal to @code{3*Character'Size}. This ensures efficient
9456 use of space, but means that accesses to elements of the array will incur
9457 the overhead of converting representation values to the corresponding
9458 positional values, (i.e.@: the value delivered by the @code{Pos} attribute).
9460 @node Address Clauses
9461 @section Address Clauses
9462 @cindex Address Clause
9464 The reference manual allows a general restriction on representation clauses,
9465 as found in RM 13.1(22):
9468 An implementation need not support representation
9469 items containing nonstatic expressions, except that
9470 an implementation should support a representation item
9471 for a given entity if each nonstatic expression in the
9472 representation item is a name that statically denotes
9473 a constant declared before the entity.
9477 In practice this is applicable only to address clauses, since this is the
9478 only case in which a non-static expression is permitted by the syntax. As
9479 the AARM notes in sections 13.1 (22.a-22.h):
9482 22.a Reason: This is to avoid the following sort of thing:
9484 22.b X : Integer := F(@dots{});
9485 Y : Address := G(@dots{});
9486 for X'Address use Y;
9488 22.c In the above, we have to evaluate the
9489 initialization expression for X before we
9490 know where to put the result. This seems
9491 like an unreasonable implementation burden.
9493 22.d The above code should instead be written
9496 22.e Y : constant Address := G(@dots{});
9497 X : Integer := F(@dots{});
9498 for X'Address use Y;
9500 22.f This allows the expression ``Y'' to be safely
9501 evaluated before X is created.
9503 22.g The constant could be a formal parameter of mode in.
9505 22.h An implementation can support other nonstatic
9506 expressions if it wants to. Expressions of type
9507 Address are hardly ever static, but their value
9508 might be known at compile time anyway in many
9513 GNAT does indeed permit many additional cases of non-static expressions. In
9514 particular, if the type involved is elementary there are no restrictions
9515 (since in this case, holding a temporary copy of the initialization value,
9516 if one is present, is inexpensive). In addition, if there is no implicit or
9517 explicit initialization, then there are no restrictions. GNAT will reject
9518 only the case where all three of these conditions hold:
9523 The type of the item is non-elementary (e.g.@: a record or array).
9526 There is explicit or implicit initialization required for the object.
9527 Note that access values are always implicitly initialized, and also
9528 in GNAT, certain bit-packed arrays (those having a dynamic length or
9529 a length greater than 64) will also be implicitly initialized to zero.
9532 The address value is non-static. Here GNAT is more permissive than the
9533 RM, and allows the address value to be the address of a previously declared
9534 stand-alone variable, as long as it does not itself have an address clause.
9536 @smallexample @c ada
9537 Anchor : Some_Initialized_Type;
9538 Overlay : Some_Initialized_Type;
9539 for Overlay'Address use Anchor'Address;
9543 However, the prefix of the address clause cannot be an array component, or
9544 a component of a discriminated record.
9549 As noted above in section 22.h, address values are typically non-static. In
9550 particular the To_Address function, even if applied to a literal value, is
9551 a non-static function call. To avoid this minor annoyance, GNAT provides
9552 the implementation defined attribute 'To_Address. The following two
9553 expressions have identical values:
9557 @smallexample @c ada
9558 To_Address (16#1234_0000#)
9559 System'To_Address (16#1234_0000#);
9563 except that the second form is considered to be a static expression, and
9564 thus when used as an address clause value is always permitted.
9567 Additionally, GNAT treats as static an address clause that is an
9568 unchecked_conversion of a static integer value. This simplifies the porting
9569 of legacy code, and provides a portable equivalent to the GNAT attribute
9572 Another issue with address clauses is the interaction with alignment
9573 requirements. When an address clause is given for an object, the address
9574 value must be consistent with the alignment of the object (which is usually
9575 the same as the alignment of the type of the object). If an address clause
9576 is given that specifies an inappropriately aligned address value, then the
9577 program execution is erroneous.
9579 Since this source of erroneous behavior can have unfortunate effects, GNAT
9580 checks (at compile time if possible, generating a warning, or at execution
9581 time with a run-time check) that the alignment is appropriate. If the
9582 run-time check fails, then @code{Program_Error} is raised. This run-time
9583 check is suppressed if range checks are suppressed, or if
9584 @code{pragma Restrictions (No_Elaboration_Code)} is in effect.
9587 An address clause cannot be given for an exported object. More
9588 understandably the real restriction is that objects with an address
9589 clause cannot be exported. This is because such variables are not
9590 defined by the Ada program, so there is no external object to export.
9593 It is permissible to give an address clause and a pragma Import for the
9594 same object. In this case, the variable is not really defined by the
9595 Ada program, so there is no external symbol to be linked. The link name
9596 and the external name are ignored in this case. The reason that we allow this
9597 combination is that it provides a useful idiom to avoid unwanted
9598 initializations on objects with address clauses.
9600 When an address clause is given for an object that has implicit or
9601 explicit initialization, then by default initialization takes place. This
9602 means that the effect of the object declaration is to overwrite the
9603 memory at the specified address. This is almost always not what the
9604 programmer wants, so GNAT will output a warning:
9614 for Ext'Address use System'To_Address (16#1234_1234#);
9616 >>> warning: implicit initialization of "Ext" may
9617 modify overlaid storage
9618 >>> warning: use pragma Import for "Ext" to suppress
9619 initialization (RM B(24))
9625 As indicated by the warning message, the solution is to use a (dummy) pragma
9626 Import to suppress this initialization. The pragma tell the compiler that the
9627 object is declared and initialized elsewhere. The following package compiles
9628 without warnings (and the initialization is suppressed):
9630 @smallexample @c ada
9638 for Ext'Address use System'To_Address (16#1234_1234#);
9639 pragma Import (Ada, Ext);
9644 A final issue with address clauses involves their use for overlaying
9645 variables, as in the following example:
9646 @cindex Overlaying of objects
9648 @smallexample @c ada
9651 for B'Address use A'Address;
9655 or alternatively, using the form recommended by the RM:
9657 @smallexample @c ada
9659 Addr : constant Address := A'Address;
9661 for B'Address use Addr;
9665 In both of these cases, @code{A}
9666 and @code{B} become aliased to one another via the
9667 address clause. This use of address clauses to overlay
9668 variables, achieving an effect similar to unchecked
9669 conversion was erroneous in Ada 83, but in Ada 95
9670 the effect is implementation defined. Furthermore, the
9671 Ada 95 RM specifically recommends that in a situation
9672 like this, @code{B} should be subject to the following
9673 implementation advice (RM 13.3(19)):
9676 19 If the Address of an object is specified, or it is imported
9677 or exported, then the implementation should not perform
9678 optimizations based on assumptions of no aliases.
9682 GNAT follows this recommendation, and goes further by also applying
9683 this recommendation to the overlaid variable (@code{A}
9684 in the above example) in this case. This means that the overlay
9685 works "as expected", in that a modification to one of the variables
9686 will affect the value of the other.
9688 @node Effect of Convention on Representation
9689 @section Effect of Convention on Representation
9690 @cindex Convention, effect on representation
9693 Normally the specification of a foreign language convention for a type or
9694 an object has no effect on the chosen representation. In particular, the
9695 representation chosen for data in GNAT generally meets the standard system
9696 conventions, and for example records are laid out in a manner that is
9697 consistent with C@. This means that specifying convention C (for example)
9700 There are three exceptions to this general rule:
9704 @item Convention Fortran and array subtypes
9705 If pragma Convention Fortran is specified for an array subtype, then in
9706 accordance with the implementation advice in section 3.6.2(11) of the
9707 Ada Reference Manual, the array will be stored in a Fortran-compatible
9708 column-major manner, instead of the normal default row-major order.
9710 @item Convention C and enumeration types
9711 GNAT normally stores enumeration types in 8, 16, or 32 bits as required
9712 to accommodate all values of the type. For example, for the enumeration
9715 @smallexample @c ada
9716 type Color is (Red, Green, Blue);
9720 8 bits is sufficient to store all values of the type, so by default, objects
9721 of type @code{Color} will be represented using 8 bits. However, normal C
9722 convention is to use 32 bits for all enum values in C, since enum values
9723 are essentially of type int. If pragma @code{Convention C} is specified for an
9724 Ada enumeration type, then the size is modified as necessary (usually to
9725 32 bits) to be consistent with the C convention for enum values.
9727 @item Convention C/Fortran and Boolean types
9728 In C, the usual convention for boolean values, that is values used for
9729 conditions, is that zero represents false, and nonzero values represent
9730 true. In Ada, the normal convention is that two specific values, typically
9731 0/1, are used to represent false/true respectively.
9733 Fortran has a similar convention for @code{LOGICAL} values (any nonzero
9734 value represents true).
9736 To accommodate the Fortran and C conventions, if a pragma Convention specifies
9737 C or Fortran convention for a derived Boolean, as in the following example:
9739 @smallexample @c ada
9740 type C_Switch is new Boolean;
9741 pragma Convention (C, C_Switch);
9745 then the GNAT generated code will treat any nonzero value as true. For truth
9746 values generated by GNAT, the conventional value 1 will be used for True, but
9747 when one of these values is read, any nonzero value is treated as True.
9751 @node Determining the Representations chosen by GNAT
9752 @section Determining the Representations chosen by GNAT
9753 @cindex Representation, determination of
9754 @cindex @code{-gnatR} switch
9757 Although the descriptions in this section are intended to be complete, it is
9758 often easier to simply experiment to see what GNAT accepts and what the
9759 effect is on the layout of types and objects.
9761 As required by the Ada RM, if a representation clause is not accepted, then
9762 it must be rejected as illegal by the compiler. However, when a
9763 representation clause or pragma is accepted, there can still be questions
9764 of what the compiler actually does. For example, if a partial record
9765 representation clause specifies the location of some components and not
9766 others, then where are the non-specified components placed? Or if pragma
9767 @code{Pack} is used on a record, then exactly where are the resulting
9768 fields placed? The section on pragma @code{Pack} in this chapter can be
9769 used to answer the second question, but it is often easier to just see
9770 what the compiler does.
9772 For this purpose, GNAT provides the option @code{-gnatR}. If you compile
9773 with this option, then the compiler will output information on the actual
9774 representations chosen, in a format similar to source representation
9775 clauses. For example, if we compile the package:
9777 @smallexample @c ada
9779 type r (x : boolean) is tagged record
9781 when True => S : String (1 .. 100);
9786 type r2 is new r (false) with record
9791 y2 at 16 range 0 .. 31;
9798 type x1 is array (1 .. 10) of x;
9799 for x1'component_size use 11;
9801 type ia is access integer;
9803 type Rb1 is array (1 .. 13) of Boolean;
9806 type Rb2 is array (1 .. 65) of Boolean;
9822 using the switch @code{-gnatR} we obtain the following output:
9825 Representation information for unit q
9826 -------------------------------------
9829 for r'Alignment use 4;
9831 x at 4 range 0 .. 7;
9832 _tag at 0 range 0 .. 31;
9833 s at 5 range 0 .. 799;
9836 for r2'Size use 160;
9837 for r2'Alignment use 4;
9839 x at 4 range 0 .. 7;
9840 _tag at 0 range 0 .. 31;
9841 _parent at 0 range 0 .. 63;
9842 y2 at 16 range 0 .. 31;
9846 for x'Alignment use 1;
9848 y at 0 range 0 .. 7;
9851 for x1'Size use 112;
9852 for x1'Alignment use 1;
9853 for x1'Component_Size use 11;
9855 for rb1'Size use 13;
9856 for rb1'Alignment use 2;
9857 for rb1'Component_Size use 1;
9859 for rb2'Size use 72;
9860 for rb2'Alignment use 1;
9861 for rb2'Component_Size use 1;
9863 for x2'Size use 224;
9864 for x2'Alignment use 4;
9866 l1 at 0 range 0 .. 0;
9867 l2 at 0 range 1 .. 64;
9868 l3 at 12 range 0 .. 31;
9869 l4 at 16 range 0 .. 0;
9870 l5 at 16 range 1 .. 13;
9871 l6 at 18 range 0 .. 71;
9876 The Size values are actually the Object_Size, i.e.@: the default size that
9877 will be allocated for objects of the type.
9878 The ?? size for type r indicates that we have a variant record, and the
9879 actual size of objects will depend on the discriminant value.
9881 The Alignment values show the actual alignment chosen by the compiler
9882 for each record or array type.
9884 The record representation clause for type r shows where all fields
9885 are placed, including the compiler generated tag field (whose location
9886 cannot be controlled by the programmer).
9888 The record representation clause for the type extension r2 shows all the
9889 fields present, including the parent field, which is a copy of the fields
9890 of the parent type of r2, i.e.@: r1.
9892 The component size and size clauses for types rb1 and rb2 show
9893 the exact effect of pragma @code{Pack} on these arrays, and the record
9894 representation clause for type x2 shows how pragma @code{Pack} affects
9897 In some cases, it may be useful to cut and paste the representation clauses
9898 generated by the compiler into the original source to fix and guarantee
9899 the actual representation to be used.
9901 @node Standard Library Routines
9902 @chapter Standard Library Routines
9905 The Ada 95 Reference Manual contains in Annex A a full description of an
9906 extensive set of standard library routines that can be used in any Ada
9907 program, and which must be provided by all Ada compilers. They are
9908 analogous to the standard C library used by C programs.
9910 GNAT implements all of the facilities described in annex A, and for most
9911 purposes the description in the Ada 95
9912 reference manual, or appropriate Ada
9913 text book, will be sufficient for making use of these facilities.
9915 In the case of the input-output facilities, @xref{The Implementation of
9916 Standard I/O}, gives details on exactly how GNAT interfaces to the
9917 file system. For the remaining packages, the Ada 95 reference manual
9918 should be sufficient. The following is a list of the packages included,
9919 together with a brief description of the functionality that is provided.
9921 For completeness, references are included to other predefined library
9922 routines defined in other sections of the Ada 95 reference manual (these are
9923 cross-indexed from annex A).
9927 This is a parent package for all the standard library packages. It is
9928 usually included implicitly in your program, and itself contains no
9929 useful data or routines.
9931 @item Ada.Calendar (9.6)
9932 @code{Calendar} provides time of day access, and routines for
9933 manipulating times and durations.
9935 @item Ada.Characters (A.3.1)
9936 This is a dummy parent package that contains no useful entities
9938 @item Ada.Characters.Handling (A.3.2)
9939 This package provides some basic character handling capabilities,
9940 including classification functions for classes of characters (e.g.@: test
9941 for letters, or digits).
9943 @item Ada.Characters.Latin_1 (A.3.3)
9944 This package includes a complete set of definitions of the characters
9945 that appear in type CHARACTER@. It is useful for writing programs that
9946 will run in international environments. For example, if you want an
9947 upper case E with an acute accent in a string, it is often better to use
9948 the definition of @code{UC_E_Acute} in this package. Then your program
9949 will print in an understandable manner even if your environment does not
9950 support these extended characters.
9952 @item Ada.Command_Line (A.15)
9953 This package provides access to the command line parameters and the name
9954 of the current program (analogous to the use of @code{argc} and @code{argv}
9955 in C), and also allows the exit status for the program to be set in a
9956 system-independent manner.
9958 @item Ada.Decimal (F.2)
9959 This package provides constants describing the range of decimal numbers
9960 implemented, and also a decimal divide routine (analogous to the COBOL
9961 verb DIVIDE .. GIVING .. REMAINDER ..)
9963 @item Ada.Direct_IO (A.8.4)
9964 This package provides input-output using a model of a set of records of
9965 fixed-length, containing an arbitrary definite Ada type, indexed by an
9966 integer record number.
9968 @item Ada.Dynamic_Priorities (D.5)
9969 This package allows the priorities of a task to be adjusted dynamically
9970 as the task is running.
9972 @item Ada.Exceptions (11.4.1)
9973 This package provides additional information on exceptions, and also
9974 contains facilities for treating exceptions as data objects, and raising
9975 exceptions with associated messages.
9977 @item Ada.Finalization (7.6)
9978 This package contains the declarations and subprograms to support the
9979 use of controlled types, providing for automatic initialization and
9980 finalization (analogous to the constructors and destructors of C++)
9982 @item Ada.Interrupts (C.3.2)
9983 This package provides facilities for interfacing to interrupts, which
9984 includes the set of signals or conditions that can be raised and
9985 recognized as interrupts.
9987 @item Ada.Interrupts.Names (C.3.2)
9988 This package provides the set of interrupt names (actually signal
9989 or condition names) that can be handled by GNAT@.
9991 @item Ada.IO_Exceptions (A.13)
9992 This package defines the set of exceptions that can be raised by use of
9993 the standard IO packages.
9996 This package contains some standard constants and exceptions used
9997 throughout the numerics packages. Note that the constants pi and e are
9998 defined here, and it is better to use these definitions than rolling
10001 @item Ada.Numerics.Complex_Elementary_Functions
10002 Provides the implementation of standard elementary functions (such as
10003 log and trigonometric functions) operating on complex numbers using the
10004 standard @code{Float} and the @code{Complex} and @code{Imaginary} types
10005 created by the package @code{Numerics.Complex_Types}.
10007 @item Ada.Numerics.Complex_Types
10008 This is a predefined instantiation of
10009 @code{Numerics.Generic_Complex_Types} using @code{Standard.Float} to
10010 build the type @code{Complex} and @code{Imaginary}.
10012 @item Ada.Numerics.Discrete_Random
10013 This package provides a random number generator suitable for generating
10014 random integer values from a specified range.
10016 @item Ada.Numerics.Float_Random
10017 This package provides a random number generator suitable for generating
10018 uniformly distributed floating point values.
10020 @item Ada.Numerics.Generic_Complex_Elementary_Functions
10021 This is a generic version of the package that provides the
10022 implementation of standard elementary functions (such as log and
10023 trigonometric functions) for an arbitrary complex type.
10025 The following predefined instantiations of this package are provided:
10029 @code{Ada.Numerics.Short_Complex_Elementary_Functions}
10031 @code{Ada.Numerics.Complex_Elementary_Functions}
10033 @code{Ada.Numerics.
10034 Long_Complex_Elementary_Functions}
10037 @item Ada.Numerics.Generic_Complex_Types
10038 This is a generic package that allows the creation of complex types,
10039 with associated complex arithmetic operations.
10041 The following predefined instantiations of this package exist
10044 @code{Ada.Numerics.Short_Complex_Complex_Types}
10046 @code{Ada.Numerics.Complex_Complex_Types}
10048 @code{Ada.Numerics.Long_Complex_Complex_Types}
10051 @item Ada.Numerics.Generic_Elementary_Functions
10052 This is a generic package that provides the implementation of standard
10053 elementary functions (such as log an trigonometric functions) for an
10054 arbitrary float type.
10056 The following predefined instantiations of this package exist
10060 @code{Ada.Numerics.Short_Elementary_Functions}
10062 @code{Ada.Numerics.Elementary_Functions}
10064 @code{Ada.Numerics.Long_Elementary_Functions}
10067 @item Ada.Real_Time (D.8)
10068 This package provides facilities similar to those of @code{Calendar}, but
10069 operating with a finer clock suitable for real time control. Note that
10070 annex D requires that there be no backward clock jumps, and GNAT generally
10071 guarantees this behavior, but of course if the external clock on which
10072 the GNAT runtime depends is deliberately reset by some external event,
10073 then such a backward jump may occur.
10075 @item Ada.Sequential_IO (A.8.1)
10076 This package provides input-output facilities for sequential files,
10077 which can contain a sequence of values of a single type, which can be
10078 any Ada type, including indefinite (unconstrained) types.
10080 @item Ada.Storage_IO (A.9)
10081 This package provides a facility for mapping arbitrary Ada types to and
10082 from a storage buffer. It is primarily intended for the creation of new
10085 @item Ada.Streams (13.13.1)
10086 This is a generic package that provides the basic support for the
10087 concept of streams as used by the stream attributes (@code{Input},
10088 @code{Output}, @code{Read} and @code{Write}).
10090 @item Ada.Streams.Stream_IO (A.12.1)
10091 This package is a specialization of the type @code{Streams} defined in
10092 package @code{Streams} together with a set of operations providing
10093 Stream_IO capability. The Stream_IO model permits both random and
10094 sequential access to a file which can contain an arbitrary set of values
10095 of one or more Ada types.
10097 @item Ada.Strings (A.4.1)
10098 This package provides some basic constants used by the string handling
10101 @item Ada.Strings.Bounded (A.4.4)
10102 This package provides facilities for handling variable length
10103 strings. The bounded model requires a maximum length. It is thus
10104 somewhat more limited than the unbounded model, but avoids the use of
10105 dynamic allocation or finalization.
10107 @item Ada.Strings.Fixed (A.4.3)
10108 This package provides facilities for handling fixed length strings.
10110 @item Ada.Strings.Maps (A.4.2)
10111 This package provides facilities for handling character mappings and
10112 arbitrarily defined subsets of characters. For instance it is useful in
10113 defining specialized translation tables.
10115 @item Ada.Strings.Maps.Constants (A.4.6)
10116 This package provides a standard set of predefined mappings and
10117 predefined character sets. For example, the standard upper to lower case
10118 conversion table is found in this package. Note that upper to lower case
10119 conversion is non-trivial if you want to take the entire set of
10120 characters, including extended characters like E with an acute accent,
10121 into account. You should use the mappings in this package (rather than
10122 adding 32 yourself) to do case mappings.
10124 @item Ada.Strings.Unbounded (A.4.5)
10125 This package provides facilities for handling variable length
10126 strings. The unbounded model allows arbitrary length strings, but
10127 requires the use of dynamic allocation and finalization.
10129 @item Ada.Strings.Wide_Bounded (A.4.7)
10130 @itemx Ada.Strings.Wide_Fixed (A.4.7)
10131 @itemx Ada.Strings.Wide_Maps (A.4.7)
10132 @itemx Ada.Strings.Wide_Maps.Constants (A.4.7)
10133 @itemx Ada.Strings.Wide_Unbounded (A.4.7)
10134 These packages provide analogous capabilities to the corresponding
10135 packages without @samp{Wide_} in the name, but operate with the types
10136 @code{Wide_String} and @code{Wide_Character} instead of @code{String}
10137 and @code{Character}.
10139 @item Ada.Synchronous_Task_Control (D.10)
10140 This package provides some standard facilities for controlling task
10141 communication in a synchronous manner.
10144 This package contains definitions for manipulation of the tags of tagged
10147 @item Ada.Task_Attributes
10148 This package provides the capability of associating arbitrary
10149 task-specific data with separate tasks.
10152 This package provides basic text input-output capabilities for
10153 character, string and numeric data. The subpackages of this
10154 package are listed next.
10156 @item Ada.Text_IO.Decimal_IO
10157 Provides input-output facilities for decimal fixed-point types
10159 @item Ada.Text_IO.Enumeration_IO
10160 Provides input-output facilities for enumeration types.
10162 @item Ada.Text_IO.Fixed_IO
10163 Provides input-output facilities for ordinary fixed-point types.
10165 @item Ada.Text_IO.Float_IO
10166 Provides input-output facilities for float types. The following
10167 predefined instantiations of this generic package are available:
10171 @code{Short_Float_Text_IO}
10173 @code{Float_Text_IO}
10175 @code{Long_Float_Text_IO}
10178 @item Ada.Text_IO.Integer_IO
10179 Provides input-output facilities for integer types. The following
10180 predefined instantiations of this generic package are available:
10183 @item Short_Short_Integer
10184 @code{Ada.Short_Short_Integer_Text_IO}
10185 @item Short_Integer
10186 @code{Ada.Short_Integer_Text_IO}
10188 @code{Ada.Integer_Text_IO}
10190 @code{Ada.Long_Integer_Text_IO}
10191 @item Long_Long_Integer
10192 @code{Ada.Long_Long_Integer_Text_IO}
10195 @item Ada.Text_IO.Modular_IO
10196 Provides input-output facilities for modular (unsigned) types
10198 @item Ada.Text_IO.Complex_IO (G.1.3)
10199 This package provides basic text input-output capabilities for complex
10202 @item Ada.Text_IO.Editing (F.3.3)
10203 This package contains routines for edited output, analogous to the use
10204 of pictures in COBOL@. The picture formats used by this package are a
10205 close copy of the facility in COBOL@.
10207 @item Ada.Text_IO.Text_Streams (A.12.2)
10208 This package provides a facility that allows Text_IO files to be treated
10209 as streams, so that the stream attributes can be used for writing
10210 arbitrary data, including binary data, to Text_IO files.
10212 @item Ada.Unchecked_Conversion (13.9)
10213 This generic package allows arbitrary conversion from one type to
10214 another of the same size, providing for breaking the type safety in
10215 special circumstances.
10217 If the types have the same Size (more accurately the same Value_Size),
10218 then the effect is simply to transfer the bits from the source to the
10219 target type without any modification. This usage is well defined, and
10220 for simple types whose representation is typically the same across
10221 all implementations, gives a portable method of performing such
10224 If the types do not have the same size, then the result is implementation
10225 defined, and thus may be non-portable. The following describes how GNAT
10226 handles such unchecked conversion cases.
10228 If the types are of different sizes, and are both discrete types, then
10229 the effect is of a normal type conversion without any constraint checking.
10230 In particular if the result type has a larger size, the result will be
10231 zero or sign extended. If the result type has a smaller size, the result
10232 will be truncated by ignoring high order bits.
10234 If the types are of different sizes, and are not both discrete types,
10235 then the conversion works as though pointers were created to the source
10236 and target, and the pointer value is converted. The effect is that bits
10237 are copied from successive low order storage units and bits of the source
10238 up to the length of the target type.
10240 A warning is issued if the lengths differ, since the effect in this
10241 case is implementation dependent, and the above behavior may not match
10242 that of some other compiler.
10244 A pointer to one type may be converted to a pointer to another type using
10245 unchecked conversion. The only case in which the effect is undefined is
10246 when one or both pointers are pointers to unconstrained array types. In
10247 this case, the bounds information may get incorrectly transferred, and in
10248 particular, GNAT uses double size pointers for such types, and it is
10249 meaningless to convert between such pointer types. GNAT will issue a
10250 warning if the alignment of the target designated type is more strict
10251 than the alignment of the source designated type (since the result may
10252 be unaligned in this case).
10254 A pointer other than a pointer to an unconstrained array type may be
10255 converted to and from System.Address. Such usage is common in Ada 83
10256 programs, but note that Ada.Address_To_Access_Conversions is the
10257 preferred method of performing such conversions in Ada 95. Neither
10258 unchecked conversion nor Ada.Address_To_Access_Conversions should be
10259 used in conjunction with pointers to unconstrained objects, since
10260 the bounds information cannot be handled correctly in this case.
10262 @item Ada.Unchecked_Deallocation (13.11.2)
10263 This generic package allows explicit freeing of storage previously
10264 allocated by use of an allocator.
10266 @item Ada.Wide_Text_IO (A.11)
10267 This package is similar to @code{Ada.Text_IO}, except that the external
10268 file supports wide character representations, and the internal types are
10269 @code{Wide_Character} and @code{Wide_String} instead of @code{Character}
10270 and @code{String}. It contains generic subpackages listed next.
10272 @item Ada.Wide_Text_IO.Decimal_IO
10273 Provides input-output facilities for decimal fixed-point types
10275 @item Ada.Wide_Text_IO.Enumeration_IO
10276 Provides input-output facilities for enumeration types.
10278 @item Ada.Wide_Text_IO.Fixed_IO
10279 Provides input-output facilities for ordinary fixed-point types.
10281 @item Ada.Wide_Text_IO.Float_IO
10282 Provides input-output facilities for float types. The following
10283 predefined instantiations of this generic package are available:
10287 @code{Short_Float_Wide_Text_IO}
10289 @code{Float_Wide_Text_IO}
10291 @code{Long_Float_Wide_Text_IO}
10294 @item Ada.Wide_Text_IO.Integer_IO
10295 Provides input-output facilities for integer types. The following
10296 predefined instantiations of this generic package are available:
10299 @item Short_Short_Integer
10300 @code{Ada.Short_Short_Integer_Wide_Text_IO}
10301 @item Short_Integer
10302 @code{Ada.Short_Integer_Wide_Text_IO}
10304 @code{Ada.Integer_Wide_Text_IO}
10306 @code{Ada.Long_Integer_Wide_Text_IO}
10307 @item Long_Long_Integer
10308 @code{Ada.Long_Long_Integer_Wide_Text_IO}
10311 @item Ada.Wide_Text_IO.Modular_IO
10312 Provides input-output facilities for modular (unsigned) types
10314 @item Ada.Wide_Text_IO.Complex_IO (G.1.3)
10315 This package is similar to @code{Ada.Text_IO.Complex_IO}, except that the
10316 external file supports wide character representations.
10318 @item Ada.Wide_Text_IO.Editing (F.3.4)
10319 This package is similar to @code{Ada.Text_IO.Editing}, except that the
10320 types are @code{Wide_Character} and @code{Wide_String} instead of
10321 @code{Character} and @code{String}.
10323 @item Ada.Wide_Text_IO.Streams (A.12.3)
10324 This package is similar to @code{Ada.Text_IO.Streams}, except that the
10325 types are @code{Wide_Character} and @code{Wide_String} instead of
10326 @code{Character} and @code{String}.
10329 @node The Implementation of Standard I/O
10330 @chapter The Implementation of Standard I/O
10333 GNAT implements all the required input-output facilities described in
10334 A.6 through A.14. These sections of the Ada 95 reference manual describe the
10335 required behavior of these packages from the Ada point of view, and if
10336 you are writing a portable Ada program that does not need to know the
10337 exact manner in which Ada maps to the outside world when it comes to
10338 reading or writing external files, then you do not need to read this
10339 chapter. As long as your files are all regular files (not pipes or
10340 devices), and as long as you write and read the files only from Ada, the
10341 description in the Ada 95 reference manual is sufficient.
10343 However, if you want to do input-output to pipes or other devices, such
10344 as the keyboard or screen, or if the files you are dealing with are
10345 either generated by some other language, or to be read by some other
10346 language, then you need to know more about the details of how the GNAT
10347 implementation of these input-output facilities behaves.
10349 In this chapter we give a detailed description of exactly how GNAT
10350 interfaces to the file system. As always, the sources of the system are
10351 available to you for answering questions at an even more detailed level,
10352 but for most purposes the information in this chapter will suffice.
10354 Another reason that you may need to know more about how input-output is
10355 implemented arises when you have a program written in mixed languages
10356 where, for example, files are shared between the C and Ada sections of
10357 the same program. GNAT provides some additional facilities, in the form
10358 of additional child library packages, that facilitate this sharing, and
10359 these additional facilities are also described in this chapter.
10362 * Standard I/O Packages::
10371 * Operations on C Streams::
10372 * Interfacing to C Streams::
10375 @node Standard I/O Packages
10376 @section Standard I/O Packages
10379 The Standard I/O packages described in Annex A for
10385 Ada.Text_IO.Complex_IO
10387 Ada.Text_IO.Text_Streams,
10391 Ada.Wide_Text_IO.Complex_IO,
10393 Ada.Wide_Text_IO.Text_Streams
10403 are implemented using the C
10404 library streams facility; where
10408 All files are opened using @code{fopen}.
10410 All input/output operations use @code{fread}/@code{fwrite}.
10414 There is no internal buffering of any kind at the Ada library level. The
10415 only buffering is that provided at the system level in the
10416 implementation of the C library routines that support streams. This
10417 facilitates shared use of these streams by mixed language programs.
10420 @section FORM Strings
10423 The format of a FORM string in GNAT is:
10426 "keyword=value,keyword=value,@dots{},keyword=value"
10430 where letters may be in upper or lower case, and there are no spaces
10431 between values. The order of the entries is not important. Currently
10432 there are two keywords defined.
10440 The use of these parameters is described later in this section.
10446 Direct_IO can only be instantiated for definite types. This is a
10447 restriction of the Ada language, which means that the records are fixed
10448 length (the length being determined by @code{@var{type}'Size}, rounded
10449 up to the next storage unit boundary if necessary).
10451 The records of a Direct_IO file are simply written to the file in index
10452 sequence, with the first record starting at offset zero, and subsequent
10453 records following. There is no control information of any kind. For
10454 example, if 32-bit integers are being written, each record takes
10455 4-bytes, so the record at index @var{K} starts at offset
10456 (@var{K}@minus{}1)*4.
10458 There is no limit on the size of Direct_IO files, they are expanded as
10459 necessary to accommodate whatever records are written to the file.
10461 @node Sequential_IO
10462 @section Sequential_IO
10465 Sequential_IO may be instantiated with either a definite (constrained)
10466 or indefinite (unconstrained) type.
10468 For the definite type case, the elements written to the file are simply
10469 the memory images of the data values with no control information of any
10470 kind. The resulting file should be read using the same type, no validity
10471 checking is performed on input.
10473 For the indefinite type case, the elements written consist of two
10474 parts. First is the size of the data item, written as the memory image
10475 of a @code{Interfaces.C.size_t} value, followed by the memory image of
10476 the data value. The resulting file can only be read using the same
10477 (unconstrained) type. Normal assignment checks are performed on these
10478 read operations, and if these checks fail, @code{Data_Error} is
10479 raised. In particular, in the array case, the lengths must match, and in
10480 the variant record case, if the variable for a particular read operation
10481 is constrained, the discriminants must match.
10483 Note that it is not possible to use Sequential_IO to write variable
10484 length array items, and then read the data back into different length
10485 arrays. For example, the following will raise @code{Data_Error}:
10487 @smallexample @c ada
10488 package IO is new Sequential_IO (String);
10493 IO.Write (F, "hello!")
10494 IO.Reset (F, Mode=>In_File);
10501 On some Ada implementations, this will print @code{hell}, but the program is
10502 clearly incorrect, since there is only one element in the file, and that
10503 element is the string @code{hello!}.
10505 In Ada 95, this kind of behavior can be legitimately achieved using
10506 Stream_IO, and this is the preferred mechanism. In particular, the above
10507 program fragment rewritten to use Stream_IO will work correctly.
10513 Text_IO files consist of a stream of characters containing the following
10514 special control characters:
10517 LF (line feed, 16#0A#) Line Mark
10518 FF (form feed, 16#0C#) Page Mark
10522 A canonical Text_IO file is defined as one in which the following
10523 conditions are met:
10527 The character @code{LF} is used only as a line mark, i.e.@: to mark the end
10531 The character @code{FF} is used only as a page mark, i.e.@: to mark the
10532 end of a page and consequently can appear only immediately following a
10533 @code{LF} (line mark) character.
10536 The file ends with either @code{LF} (line mark) or @code{LF}-@code{FF}
10537 (line mark, page mark). In the former case, the page mark is implicitly
10538 assumed to be present.
10542 A file written using Text_IO will be in canonical form provided that no
10543 explicit @code{LF} or @code{FF} characters are written using @code{Put}
10544 or @code{Put_Line}. There will be no @code{FF} character at the end of
10545 the file unless an explicit @code{New_Page} operation was performed
10546 before closing the file.
10548 A canonical Text_IO file that is a regular file, i.e.@: not a device or a
10549 pipe, can be read using any of the routines in Text_IO@. The
10550 semantics in this case will be exactly as defined in the Ada 95 reference
10551 manual and all the routines in Text_IO are fully implemented.
10553 A text file that does not meet the requirements for a canonical Text_IO
10554 file has one of the following:
10558 The file contains @code{FF} characters not immediately following a
10559 @code{LF} character.
10562 The file contains @code{LF} or @code{FF} characters written by
10563 @code{Put} or @code{Put_Line}, which are not logically considered to be
10564 line marks or page marks.
10567 The file ends in a character other than @code{LF} or @code{FF},
10568 i.e.@: there is no explicit line mark or page mark at the end of the file.
10572 Text_IO can be used to read such non-standard text files but subprograms
10573 to do with line or page numbers do not have defined meanings. In
10574 particular, a @code{FF} character that does not follow a @code{LF}
10575 character may or may not be treated as a page mark from the point of
10576 view of page and line numbering. Every @code{LF} character is considered
10577 to end a line, and there is an implied @code{LF} character at the end of
10581 * Text_IO Stream Pointer Positioning::
10582 * Text_IO Reading and Writing Non-Regular Files::
10584 * Treating Text_IO Files as Streams::
10585 * Text_IO Extensions::
10586 * Text_IO Facilities for Unbounded Strings::
10589 @node Text_IO Stream Pointer Positioning
10590 @subsection Stream Pointer Positioning
10593 @code{Ada.Text_IO} has a definition of current position for a file that
10594 is being read. No internal buffering occurs in Text_IO, and usually the
10595 physical position in the stream used to implement the file corresponds
10596 to this logical position defined by Text_IO@. There are two exceptions:
10600 After a call to @code{End_Of_Page} that returns @code{True}, the stream
10601 is positioned past the @code{LF} (line mark) that precedes the page
10602 mark. Text_IO maintains an internal flag so that subsequent read
10603 operations properly handle the logical position which is unchanged by
10604 the @code{End_Of_Page} call.
10607 After a call to @code{End_Of_File} that returns @code{True}, if the
10608 Text_IO file was positioned before the line mark at the end of file
10609 before the call, then the logical position is unchanged, but the stream
10610 is physically positioned right at the end of file (past the line mark,
10611 and past a possible page mark following the line mark. Again Text_IO
10612 maintains internal flags so that subsequent read operations properly
10613 handle the logical position.
10617 These discrepancies have no effect on the observable behavior of
10618 Text_IO, but if a single Ada stream is shared between a C program and
10619 Ada program, or shared (using @samp{shared=yes} in the form string)
10620 between two Ada files, then the difference may be observable in some
10623 @node Text_IO Reading and Writing Non-Regular Files
10624 @subsection Reading and Writing Non-Regular Files
10627 A non-regular file is a device (such as a keyboard), or a pipe. Text_IO
10628 can be used for reading and writing. Writing is not affected and the
10629 sequence of characters output is identical to the normal file case, but
10630 for reading, the behavior of Text_IO is modified to avoid undesirable
10631 look-ahead as follows:
10633 An input file that is not a regular file is considered to have no page
10634 marks. Any @code{Ascii.FF} characters (the character normally used for a
10635 page mark) appearing in the file are considered to be data
10636 characters. In particular:
10640 @code{Get_Line} and @code{Skip_Line} do not test for a page mark
10641 following a line mark. If a page mark appears, it will be treated as a
10645 This avoids the need to wait for an extra character to be typed or
10646 entered from the pipe to complete one of these operations.
10649 @code{End_Of_Page} always returns @code{False}
10652 @code{End_Of_File} will return @code{False} if there is a page mark at
10653 the end of the file.
10657 Output to non-regular files is the same as for regular files. Page marks
10658 may be written to non-regular files using @code{New_Page}, but as noted
10659 above they will not be treated as page marks on input if the output is
10660 piped to another Ada program.
10662 Another important discrepancy when reading non-regular files is that the end
10663 of file indication is not ``sticky''. If an end of file is entered, e.g.@: by
10664 pressing the @key{EOT} key,
10666 is signaled once (i.e.@: the test @code{End_Of_File}
10667 will yield @code{True}, or a read will
10668 raise @code{End_Error}), but then reading can resume
10669 to read data past that end of
10670 file indication, until another end of file indication is entered.
10672 @node Get_Immediate
10673 @subsection Get_Immediate
10674 @cindex Get_Immediate
10677 Get_Immediate returns the next character (including control characters)
10678 from the input file. In particular, Get_Immediate will return LF or FF
10679 characters used as line marks or page marks. Such operations leave the
10680 file positioned past the control character, and it is thus not treated
10681 as having its normal function. This means that page, line and column
10682 counts after this kind of Get_Immediate call are set as though the mark
10683 did not occur. In the case where a Get_Immediate leaves the file
10684 positioned between the line mark and page mark (which is not normally
10685 possible), it is undefined whether the FF character will be treated as a
10688 @node Treating Text_IO Files as Streams
10689 @subsection Treating Text_IO Files as Streams
10690 @cindex Stream files
10693 The package @code{Text_IO.Streams} allows a Text_IO file to be treated
10694 as a stream. Data written to a Text_IO file in this stream mode is
10695 binary data. If this binary data contains bytes 16#0A# (@code{LF}) or
10696 16#0C# (@code{FF}), the resulting file may have non-standard
10697 format. Similarly if read operations are used to read from a Text_IO
10698 file treated as a stream, then @code{LF} and @code{FF} characters may be
10699 skipped and the effect is similar to that described above for
10700 @code{Get_Immediate}.
10702 @node Text_IO Extensions
10703 @subsection Text_IO Extensions
10704 @cindex Text_IO extensions
10707 A package GNAT.IO_Aux in the GNAT library provides some useful extensions
10708 to the standard @code{Text_IO} package:
10711 @item function File_Exists (Name : String) return Boolean;
10712 Determines if a file of the given name exists.
10714 @item function Get_Line return String;
10715 Reads a string from the standard input file. The value returned is exactly
10716 the length of the line that was read.
10718 @item function Get_Line (File : Ada.Text_IO.File_Type) return String;
10719 Similar, except that the parameter File specifies the file from which
10720 the string is to be read.
10724 @node Text_IO Facilities for Unbounded Strings
10725 @subsection Text_IO Facilities for Unbounded Strings
10726 @cindex Text_IO for unbounded strings
10727 @cindex Unbounded_String, Text_IO operations
10730 The package @code{Ada.Strings.Unbounded.Text_IO}
10731 in library files @code{a-suteio.ads/adb} contains some GNAT-specific
10732 subprograms useful for Text_IO operations on unbounded strings:
10736 @item function Get_Line (File : File_Type) return Unbounded_String;
10737 Reads a line from the specified file
10738 and returns the result as an unbounded string.
10740 @item procedure Put (File : File_Type; U : Unbounded_String);
10741 Writes the value of the given unbounded string to the specified file
10742 Similar to the effect of
10743 @code{Put (To_String (U))} except that an extra copy is avoided.
10745 @item procedure Put_Line (File : File_Type; U : Unbounded_String);
10746 Writes the value of the given unbounded string to the specified file,
10747 followed by a @code{New_Line}.
10748 Similar to the effect of @code{Put_Line (To_String (U))} except
10749 that an extra copy is avoided.
10753 In the above procedures, @code{File} is of type @code{Ada.Text_IO.File_Type}
10754 and is optional. If the parameter is omitted, then the standard input or
10755 output file is referenced as appropriate.
10757 The package @code{Ada.Strings.Wide_Unbounded.Wide_Text_IO} in library
10758 files @file{a-swuwti.ads} and @file{a-swuwti.adb} provides similar extended
10759 @code{Wide_Text_IO} functionality for unbounded wide strings.
10762 @section Wide_Text_IO
10765 @code{Wide_Text_IO} is similar in most respects to Text_IO, except that
10766 both input and output files may contain special sequences that represent
10767 wide character values. The encoding scheme for a given file may be
10768 specified using a FORM parameter:
10775 as part of the FORM string (WCEM = wide character encoding method),
10776 where @var{x} is one of the following characters
10782 Upper half encoding
10794 The encoding methods match those that
10795 can be used in a source
10796 program, but there is no requirement that the encoding method used for
10797 the source program be the same as the encoding method used for files,
10798 and different files may use different encoding methods.
10800 The default encoding method for the standard files, and for opened files
10801 for which no WCEM parameter is given in the FORM string matches the
10802 wide character encoding specified for the main program (the default
10803 being brackets encoding if no coding method was specified with -gnatW).
10807 In this encoding, a wide character is represented by a five character
10815 where @var{a}, @var{b}, @var{c}, @var{d} are the four hexadecimal
10816 characters (using upper case letters) of the wide character code. For
10817 example, ESC A345 is used to represent the wide character with code
10818 16#A345#. This scheme is compatible with use of the full
10819 @code{Wide_Character} set.
10821 @item Upper Half Coding
10822 The wide character with encoding 16#abcd#, where the upper bit is on
10823 (i.e.@: a is in the range 8-F) is represented as two bytes 16#ab# and
10824 16#cd#. The second byte may never be a format control character, but is
10825 not required to be in the upper half. This method can be also used for
10826 shift-JIS or EUC where the internal coding matches the external coding.
10828 @item Shift JIS Coding
10829 A wide character is represented by a two character sequence 16#ab# and
10830 16#cd#, with the restrictions described for upper half encoding as
10831 described above. The internal character code is the corresponding JIS
10832 character according to the standard algorithm for Shift-JIS
10833 conversion. Only characters defined in the JIS code set table can be
10834 used with this encoding method.
10837 A wide character is represented by a two character sequence 16#ab# and
10838 16#cd#, with both characters being in the upper half. The internal
10839 character code is the corresponding JIS character according to the EUC
10840 encoding algorithm. Only characters defined in the JIS code set table
10841 can be used with this encoding method.
10844 A wide character is represented using
10845 UCS Transformation Format 8 (UTF-8) as defined in Annex R of ISO
10846 10646-1/Am.2. Depending on the character value, the representation
10847 is a one, two, or three byte sequence:
10850 16#0000#-16#007f#: 2#0xxxxxxx#
10851 16#0080#-16#07ff#: 2#110xxxxx# 2#10xxxxxx#
10852 16#0800#-16#ffff#: 2#1110xxxx# 2#10xxxxxx# 2#10xxxxxx#
10856 where the xxx bits correspond to the left-padded bits of the
10857 16-bit character value. Note that all lower half ASCII characters
10858 are represented as ASCII bytes and all upper half characters and
10859 other wide characters are represented as sequences of upper-half
10860 (The full UTF-8 scheme allows for encoding 31-bit characters as
10861 6-byte sequences, but in this implementation, all UTF-8 sequences
10862 of four or more bytes length will raise a Constraint_Error, as
10863 will all invalid UTF-8 sequences.)
10865 @item Brackets Coding
10866 In this encoding, a wide character is represented by the following eight
10867 character sequence:
10874 where @code{a}, @code{b}, @code{c}, @code{d} are the four hexadecimal
10875 characters (using uppercase letters) of the wide character code. For
10876 example, @code{["A345"]} is used to represent the wide character with code
10878 This scheme is compatible with use of the full Wide_Character set.
10879 On input, brackets coding can also be used for upper half characters,
10880 e.g.@: @code{["C1"]} for lower case a. However, on output, brackets notation
10881 is only used for wide characters with a code greater than @code{16#FF#}.
10886 For the coding schemes other than Hex and Brackets encoding,
10887 not all wide character
10888 values can be represented. An attempt to output a character that cannot
10889 be represented using the encoding scheme for the file causes
10890 Constraint_Error to be raised. An invalid wide character sequence on
10891 input also causes Constraint_Error to be raised.
10894 * Wide_Text_IO Stream Pointer Positioning::
10895 * Wide_Text_IO Reading and Writing Non-Regular Files::
10898 @node Wide_Text_IO Stream Pointer Positioning
10899 @subsection Stream Pointer Positioning
10902 @code{Ada.Wide_Text_IO} is similar to @code{Ada.Text_IO} in its handling
10903 of stream pointer positioning (@pxref{Text_IO}). There is one additional
10906 If @code{Ada.Wide_Text_IO.Look_Ahead} reads a character outside the
10907 normal lower ASCII set (i.e.@: a character in the range:
10909 @smallexample @c ada
10910 Wide_Character'Val (16#0080#) .. Wide_Character'Val (16#FFFF#)
10914 then although the logical position of the file pointer is unchanged by
10915 the @code{Look_Ahead} call, the stream is physically positioned past the
10916 wide character sequence. Again this is to avoid the need for buffering
10917 or backup, and all @code{Wide_Text_IO} routines check the internal
10918 indication that this situation has occurred so that this is not visible
10919 to a normal program using @code{Wide_Text_IO}. However, this discrepancy
10920 can be observed if the wide text file shares a stream with another file.
10922 @node Wide_Text_IO Reading and Writing Non-Regular Files
10923 @subsection Reading and Writing Non-Regular Files
10926 As in the case of Text_IO, when a non-regular file is read, it is
10927 assumed that the file contains no page marks (any form characters are
10928 treated as data characters), and @code{End_Of_Page} always returns
10929 @code{False}. Similarly, the end of file indication is not sticky, so
10930 it is possible to read beyond an end of file.
10936 A stream file is a sequence of bytes, where individual elements are
10937 written to the file as described in the Ada 95 reference manual. The type
10938 @code{Stream_Element} is simply a byte. There are two ways to read or
10939 write a stream file.
10943 The operations @code{Read} and @code{Write} directly read or write a
10944 sequence of stream elements with no control information.
10947 The stream attributes applied to a stream file transfer data in the
10948 manner described for stream attributes.
10952 @section Shared Files
10955 Section A.14 of the Ada 95 Reference Manual allows implementations to
10956 provide a wide variety of behavior if an attempt is made to access the
10957 same external file with two or more internal files.
10959 To provide a full range of functionality, while at the same time
10960 minimizing the problems of portability caused by this implementation
10961 dependence, GNAT handles file sharing as follows:
10965 In the absence of a @samp{shared=@var{xxx}} form parameter, an attempt
10966 to open two or more files with the same full name is considered an error
10967 and is not supported. The exception @code{Use_Error} will be
10968 raised. Note that a file that is not explicitly closed by the program
10969 remains open until the program terminates.
10972 If the form parameter @samp{shared=no} appears in the form string, the
10973 file can be opened or created with its own separate stream identifier,
10974 regardless of whether other files sharing the same external file are
10975 opened. The exact effect depends on how the C stream routines handle
10976 multiple accesses to the same external files using separate streams.
10979 If the form parameter @samp{shared=yes} appears in the form string for
10980 each of two or more files opened using the same full name, the same
10981 stream is shared between these files, and the semantics are as described
10982 in Ada 95 Reference Manual, Section A.14.
10986 When a program that opens multiple files with the same name is ported
10987 from another Ada compiler to GNAT, the effect will be that
10988 @code{Use_Error} is raised.
10990 The documentation of the original compiler and the documentation of the
10991 program should then be examined to determine if file sharing was
10992 expected, and @samp{shared=@var{xxx}} parameters added to @code{Open}
10993 and @code{Create} calls as required.
10995 When a program is ported from GNAT to some other Ada compiler, no
10996 special attention is required unless the @samp{shared=@var{xxx}} form
10997 parameter is used in the program. In this case, you must examine the
10998 documentation of the new compiler to see if it supports the required
10999 file sharing semantics, and form strings modified appropriately. Of
11000 course it may be the case that the program cannot be ported if the
11001 target compiler does not support the required functionality. The best
11002 approach in writing portable code is to avoid file sharing (and hence
11003 the use of the @samp{shared=@var{xxx}} parameter in the form string)
11006 One common use of file sharing in Ada 83 is the use of instantiations of
11007 Sequential_IO on the same file with different types, to achieve
11008 heterogeneous input-output. Although this approach will work in GNAT if
11009 @samp{shared=yes} is specified, it is preferable in Ada 95 to use Stream_IO
11010 for this purpose (using the stream attributes)
11013 @section Open Modes
11016 @code{Open} and @code{Create} calls result in a call to @code{fopen}
11017 using the mode shown in the following table:
11020 @center @code{Open} and @code{Create} Call Modes
11022 @b{OPEN } @b{CREATE}
11023 Append_File "r+" "w+"
11025 Out_File (Direct_IO) "r+" "w"
11026 Out_File (all other cases) "w" "w"
11027 Inout_File "r+" "w+"
11031 If text file translation is required, then either @samp{b} or @samp{t}
11032 is added to the mode, depending on the setting of Text. Text file
11033 translation refers to the mapping of CR/LF sequences in an external file
11034 to LF characters internally. This mapping only occurs in DOS and
11035 DOS-like systems, and is not relevant to other systems.
11037 A special case occurs with Stream_IO@. As shown in the above table, the
11038 file is initially opened in @samp{r} or @samp{w} mode for the
11039 @code{In_File} and @code{Out_File} cases. If a @code{Set_Mode} operation
11040 subsequently requires switching from reading to writing or vice-versa,
11041 then the file is reopened in @samp{r+} mode to permit the required operation.
11043 @node Operations on C Streams
11044 @section Operations on C Streams
11045 The package @code{Interfaces.C_Streams} provides an Ada program with direct
11046 access to the C library functions for operations on C streams:
11048 @smallexample @c adanocomment
11049 package Interfaces.C_Streams is
11050 -- Note: the reason we do not use the types that are in
11051 -- Interfaces.C is that we want to avoid dragging in the
11052 -- code in this unit if possible.
11053 subtype chars is System.Address;
11054 -- Pointer to null-terminated array of characters
11055 subtype FILEs is System.Address;
11056 -- Corresponds to the C type FILE*
11057 subtype voids is System.Address;
11058 -- Corresponds to the C type void*
11059 subtype int is Integer;
11060 subtype long is Long_Integer;
11061 -- Note: the above types are subtypes deliberately, and it
11062 -- is part of this spec that the above correspondences are
11063 -- guaranteed. This means that it is legitimate to, for
11064 -- example, use Integer instead of int. We provide these
11065 -- synonyms for clarity, but in some cases it may be
11066 -- convenient to use the underlying types (for example to
11067 -- avoid an unnecessary dependency of a spec on the spec
11069 type size_t is mod 2 ** Standard'Address_Size;
11070 NULL_Stream : constant FILEs;
11071 -- Value returned (NULL in C) to indicate an
11072 -- fdopen/fopen/tmpfile error
11073 ----------------------------------
11074 -- Constants Defined in stdio.h --
11075 ----------------------------------
11076 EOF : constant int;
11077 -- Used by a number of routines to indicate error or
11079 IOFBF : constant int;
11080 IOLBF : constant int;
11081 IONBF : constant int;
11082 -- Used to indicate buffering mode for setvbuf call
11083 SEEK_CUR : constant int;
11084 SEEK_END : constant int;
11085 SEEK_SET : constant int;
11086 -- Used to indicate origin for fseek call
11087 function stdin return FILEs;
11088 function stdout return FILEs;
11089 function stderr return FILEs;
11090 -- Streams associated with standard files
11091 --------------------------
11092 -- Standard C functions --
11093 --------------------------
11094 -- The functions selected below are ones that are
11095 -- available in DOS, OS/2, UNIX and Xenix (but not
11096 -- necessarily in ANSI C). These are very thin interfaces
11097 -- which copy exactly the C headers. For more
11098 -- documentation on these functions, see the Microsoft C
11099 -- "Run-Time Library Reference" (Microsoft Press, 1990,
11100 -- ISBN 1-55615-225-6), which includes useful information
11101 -- on system compatibility.
11102 procedure clearerr (stream : FILEs);
11103 function fclose (stream : FILEs) return int;
11104 function fdopen (handle : int; mode : chars) return FILEs;
11105 function feof (stream : FILEs) return int;
11106 function ferror (stream : FILEs) return int;
11107 function fflush (stream : FILEs) return int;
11108 function fgetc (stream : FILEs) return int;
11109 function fgets (strng : chars; n : int; stream : FILEs)
11111 function fileno (stream : FILEs) return int;
11112 function fopen (filename : chars; Mode : chars)
11114 -- Note: to maintain target independence, use
11115 -- text_translation_required, a boolean variable defined in
11116 -- a-sysdep.c to deal with the target dependent text
11117 -- translation requirement. If this variable is set,
11118 -- then b/t should be appended to the standard mode
11119 -- argument to set the text translation mode off or on
11121 function fputc (C : int; stream : FILEs) return int;
11122 function fputs (Strng : chars; Stream : FILEs) return int;
11139 function ftell (stream : FILEs) return long;
11146 function isatty (handle : int) return int;
11147 procedure mktemp (template : chars);
11148 -- The return value (which is just a pointer to template)
11150 procedure rewind (stream : FILEs);
11151 function rmtmp return int;
11159 function tmpfile return FILEs;
11160 function ungetc (c : int; stream : FILEs) return int;
11161 function unlink (filename : chars) return int;
11162 ---------------------
11163 -- Extra functions --
11164 ---------------------
11165 -- These functions supply slightly thicker bindings than
11166 -- those above. They are derived from functions in the
11167 -- C Run-Time Library, but may do a bit more work than
11168 -- just directly calling one of the Library functions.
11169 function is_regular_file (handle : int) return int;
11170 -- Tests if given handle is for a regular file (result 1)
11171 -- or for a non-regular file (pipe or device, result 0).
11172 ---------------------------------
11173 -- Control of Text/Binary Mode --
11174 ---------------------------------
11175 -- If text_translation_required is true, then the following
11176 -- functions may be used to dynamically switch a file from
11177 -- binary to text mode or vice versa. These functions have
11178 -- no effect if text_translation_required is false (i.e. in
11179 -- normal UNIX mode). Use fileno to get a stream handle.
11180 procedure set_binary_mode (handle : int);
11181 procedure set_text_mode (handle : int);
11182 ----------------------------
11183 -- Full Path Name support --
11184 ----------------------------
11185 procedure full_name (nam : chars; buffer : chars);
11186 -- Given a NUL terminated string representing a file
11187 -- name, returns in buffer a NUL terminated string
11188 -- representing the full path name for the file name.
11189 -- On systems where it is relevant the drive is also
11190 -- part of the full path name. It is the responsibility
11191 -- of the caller to pass an actual parameter for buffer
11192 -- that is big enough for any full path name. Use
11193 -- max_path_len given below as the size of buffer.
11194 max_path_len : integer;
11195 -- Maximum length of an allowable full path name on the
11196 -- system, including a terminating NUL character.
11197 end Interfaces.C_Streams;
11200 @node Interfacing to C Streams
11201 @section Interfacing to C Streams
11204 The packages in this section permit interfacing Ada files to C Stream
11207 @smallexample @c ada
11208 with Interfaces.C_Streams;
11209 package Ada.Sequential_IO.C_Streams is
11210 function C_Stream (F : File_Type)
11211 return Interfaces.C_Streams.FILEs;
11213 (File : in out File_Type;
11214 Mode : in File_Mode;
11215 C_Stream : in Interfaces.C_Streams.FILEs;
11216 Form : in String := "");
11217 end Ada.Sequential_IO.C_Streams;
11219 with Interfaces.C_Streams;
11220 package Ada.Direct_IO.C_Streams is
11221 function C_Stream (F : File_Type)
11222 return Interfaces.C_Streams.FILEs;
11224 (File : in out File_Type;
11225 Mode : in File_Mode;
11226 C_Stream : in Interfaces.C_Streams.FILEs;
11227 Form : in String := "");
11228 end Ada.Direct_IO.C_Streams;
11230 with Interfaces.C_Streams;
11231 package Ada.Text_IO.C_Streams is
11232 function C_Stream (F : File_Type)
11233 return Interfaces.C_Streams.FILEs;
11235 (File : in out File_Type;
11236 Mode : in File_Mode;
11237 C_Stream : in Interfaces.C_Streams.FILEs;
11238 Form : in String := "");
11239 end Ada.Text_IO.C_Streams;
11241 with Interfaces.C_Streams;
11242 package Ada.Wide_Text_IO.C_Streams is
11243 function C_Stream (F : File_Type)
11244 return Interfaces.C_Streams.FILEs;
11246 (File : in out File_Type;
11247 Mode : in File_Mode;
11248 C_Stream : in Interfaces.C_Streams.FILEs;
11249 Form : in String := "");
11250 end Ada.Wide_Text_IO.C_Streams;
11252 with Interfaces.C_Streams;
11253 package Ada.Stream_IO.C_Streams is
11254 function C_Stream (F : File_Type)
11255 return Interfaces.C_Streams.FILEs;
11257 (File : in out File_Type;
11258 Mode : in File_Mode;
11259 C_Stream : in Interfaces.C_Streams.FILEs;
11260 Form : in String := "");
11261 end Ada.Stream_IO.C_Streams;
11265 In each of these five packages, the @code{C_Stream} function obtains the
11266 @code{FILE} pointer from a currently opened Ada file. It is then
11267 possible to use the @code{Interfaces.C_Streams} package to operate on
11268 this stream, or the stream can be passed to a C program which can
11269 operate on it directly. Of course the program is responsible for
11270 ensuring that only appropriate sequences of operations are executed.
11272 One particular use of relevance to an Ada program is that the
11273 @code{setvbuf} function can be used to control the buffering of the
11274 stream used by an Ada file. In the absence of such a call the standard
11275 default buffering is used.
11277 The @code{Open} procedures in these packages open a file giving an
11278 existing C Stream instead of a file name. Typically this stream is
11279 imported from a C program, allowing an Ada file to operate on an
11282 @node The GNAT Library
11283 @chapter The GNAT Library
11286 The GNAT library contains a number of general and special purpose packages.
11287 It represents functionality that the GNAT developers have found useful, and
11288 which is made available to GNAT users. The packages described here are fully
11289 supported, and upwards compatibility will be maintained in future releases,
11290 so you can use these facilities with the confidence that the same functionality
11291 will be available in future releases.
11293 The chapter here simply gives a brief summary of the facilities available.
11294 The full documentation is found in the spec file for the package. The full
11295 sources of these library packages, including both spec and body, are provided
11296 with all GNAT releases. For example, to find out the full specifications of
11297 the SPITBOL pattern matching capability, including a full tutorial and
11298 extensive examples, look in the @file{g-spipat.ads} file in the library.
11300 For each entry here, the package name (as it would appear in a @code{with}
11301 clause) is given, followed by the name of the corresponding spec file in
11302 parentheses. The packages are children in four hierarchies, @code{Ada},
11303 @code{Interfaces}, @code{System}, and @code{GNAT}, the latter being a
11304 GNAT-specific hierarchy.
11306 Note that an application program should only use packages in one of these
11307 four hierarchies if the package is defined in the Ada Reference Manual,
11308 or is listed in this section of the GNAT Programmers Reference Manual.
11309 All other units should be considered internal implementation units and
11310 should not be directly @code{with}'ed by application code. The use of
11311 a @code{with} statement that references one of these internal implementation
11312 units makes an application potentially dependent on changes in versions
11313 of GNAT, and will generate a warning message.
11316 * Ada.Characters.Latin_9 (a-chlat9.ads)::
11317 * Ada.Characters.Wide_Latin_1 (a-cwila1.ads)::
11318 * Ada.Characters.Wide_Latin_9 (a-cwila9.ads)::
11319 * Ada.Command_Line.Remove (a-colire.ads)::
11320 * Ada.Command_Line.Environment (a-colien.ads)::
11321 * Ada.Direct_IO.C_Streams (a-diocst.ads)::
11322 * Ada.Exceptions.Is_Null_Occurrence (a-einuoc.ads)::
11323 * Ada.Exceptions.Traceback (a-exctra.ads)::
11324 * Ada.Sequential_IO.C_Streams (a-siocst.ads)::
11325 * Ada.Streams.Stream_IO.C_Streams (a-ssicst.ads)::
11326 * Ada.Strings.Unbounded.Text_IO (a-suteio.ads)::
11327 * Ada.Strings.Wide_Unbounded.Wide_Text_IO (a-swuwti.ads)::
11328 * Ada.Text_IO.C_Streams (a-tiocst.ads)::
11329 * Ada.Wide_Text_IO.C_Streams (a-wtcstr.ads)::
11330 * GNAT.Array_Split (g-arrspl.ads)::
11331 * GNAT.AWK (g-awk.ads)::
11332 * GNAT.Bounded_Buffers (g-boubuf.ads)::
11333 * GNAT.Bounded_Mailboxes (g-boumai.ads)::
11334 * GNAT.Bubble_Sort (g-bubsor.ads)::
11335 * GNAT.Bubble_Sort_A (g-busora.ads)::
11336 * GNAT.Bubble_Sort_G (g-busorg.ads)::
11337 * GNAT.Calendar (g-calend.ads)::
11338 * GNAT.Calendar.Time_IO (g-catiio.ads)::
11339 * GNAT.CRC32 (g-crc32.ads)::
11340 * GNAT.Case_Util (g-casuti.ads)::
11341 * GNAT.CGI (g-cgi.ads)::
11342 * GNAT.CGI.Cookie (g-cgicoo.ads)::
11343 * GNAT.CGI.Debug (g-cgideb.ads)::
11344 * GNAT.Command_Line (g-comlin.ads)::
11345 * GNAT.Compiler_Version (g-comver.ads)::
11346 * GNAT.Ctrl_C (g-ctrl_c.ads)::
11347 * GNAT.Current_Exception (g-curexc.ads)::
11348 * GNAT.Debug_Pools (g-debpoo.ads)::
11349 * GNAT.Debug_Utilities (g-debuti.ads)::
11350 * GNAT.Directory_Operations (g-dirope.ads)::
11351 * GNAT.Dynamic_HTables (g-dynhta.ads)::
11352 * GNAT.Dynamic_Tables (g-dyntab.ads)::
11353 * GNAT.Exception_Actions (g-excact.ads)::
11354 * GNAT.Exception_Traces (g-exctra.ads)::
11355 * GNAT.Exceptions (g-except.ads)::
11356 * GNAT.Expect (g-expect.ads)::
11357 * GNAT.Float_Control (g-flocon.ads)::
11358 * GNAT.Heap_Sort (g-heasor.ads)::
11359 * GNAT.Heap_Sort_A (g-hesora.ads)::
11360 * GNAT.Heap_Sort_G (g-hesorg.ads)::
11361 * GNAT.HTable (g-htable.ads)::
11362 * GNAT.IO (g-io.ads)::
11363 * GNAT.IO_Aux (g-io_aux.ads)::
11364 * GNAT.Lock_Files (g-locfil.ads)::
11365 * GNAT.MD5 (g-md5.ads)::
11366 * GNAT.Memory_Dump (g-memdum.ads)::
11367 * GNAT.Most_Recent_Exception (g-moreex.ads)::
11368 * GNAT.OS_Lib (g-os_lib.ads)::
11369 * GNAT.Perfect_Hash.Generators (g-pehage.ads)::
11370 * GNAT.Regexp (g-regexp.ads)::
11371 * GNAT.Registry (g-regist.ads)::
11372 * GNAT.Regpat (g-regpat.ads)::
11373 * GNAT.Secondary_Stack_Info (g-sestin.ads)::
11374 * GNAT.Semaphores (g-semaph.ads)::
11375 * GNAT.Signals (g-signal.ads)::
11376 * GNAT.Sockets (g-socket.ads)::
11377 * GNAT.Source_Info (g-souinf.ads)::
11378 * GNAT.Spell_Checker (g-speche.ads)::
11379 * GNAT.Spitbol.Patterns (g-spipat.ads)::
11380 * GNAT.Spitbol (g-spitbo.ads)::
11381 * GNAT.Spitbol.Table_Boolean (g-sptabo.ads)::
11382 * GNAT.Spitbol.Table_Integer (g-sptain.ads)::
11383 * GNAT.Spitbol.Table_VString (g-sptavs.ads)::
11384 * GNAT.Strings (g-string.ads)::
11385 * GNAT.String_Split (g-strspl.ads)::
11386 * GNAT.Table (g-table.ads)::
11387 * GNAT.Task_Lock (g-tasloc.ads)::
11388 * GNAT.Threads (g-thread.ads)::
11389 * GNAT.Traceback (g-traceb.ads)::
11390 * GNAT.Traceback.Symbolic (g-trasym.ads)::
11391 * GNAT.Wide_String_Split (g-wistsp.ads)::
11392 * Interfaces.C.Extensions (i-cexten.ads)::
11393 * Interfaces.C.Streams (i-cstrea.ads)::
11394 * Interfaces.CPP (i-cpp.ads)::
11395 * Interfaces.Os2lib (i-os2lib.ads)::
11396 * Interfaces.Os2lib.Errors (i-os2err.ads)::
11397 * Interfaces.Os2lib.Synchronization (i-os2syn.ads)::
11398 * Interfaces.Os2lib.Threads (i-os2thr.ads)::
11399 * Interfaces.Packed_Decimal (i-pacdec.ads)::
11400 * Interfaces.VxWorks (i-vxwork.ads)::
11401 * Interfaces.VxWorks.IO (i-vxwoio.ads)::
11402 * System.Address_Image (s-addima.ads)::
11403 * System.Assertions (s-assert.ads)::
11404 * System.Memory (s-memory.ads)::
11405 * System.Partition_Interface (s-parint.ads)::
11406 * System.Restrictions (s-restri.ads)::
11407 * System.Rident (s-rident.ads)::
11408 * System.Task_Info (s-tasinf.ads)::
11409 * System.Wch_Cnv (s-wchcnv.ads)::
11410 * System.Wch_Con (s-wchcon.ads)::
11413 @node Ada.Characters.Latin_9 (a-chlat9.ads)
11414 @section @code{Ada.Characters.Latin_9} (@file{a-chlat9.ads})
11415 @cindex @code{Ada.Characters.Latin_9} (@file{a-chlat9.ads})
11416 @cindex Latin_9 constants for Character
11419 This child of @code{Ada.Characters}
11420 provides a set of definitions corresponding to those in the
11421 RM-defined package @code{Ada.Characters.Latin_1} but with the
11422 few modifications required for @code{Latin-9}
11423 The provision of such a package
11424 is specifically authorized by the Ada Reference Manual
11427 @node Ada.Characters.Wide_Latin_1 (a-cwila1.ads)
11428 @section @code{Ada.Characters.Wide_Latin_1} (@file{a-cwila1.ads})
11429 @cindex @code{Ada.Characters.Wide_Latin_1} (@file{a-cwila1.ads})
11430 @cindex Latin_1 constants for Wide_Character
11433 This child of @code{Ada.Characters}
11434 provides a set of definitions corresponding to those in the
11435 RM-defined package @code{Ada.Characters.Latin_1} but with the
11436 types of the constants being @code{Wide_Character}
11437 instead of @code{Character}. The provision of such a package
11438 is specifically authorized by the Ada Reference Manual
11441 @node Ada.Characters.Wide_Latin_9 (a-cwila9.ads)
11442 @section @code{Ada.Characters.Wide_Latin_9} (@file{a-cwila1.ads})
11443 @cindex @code{Ada.Characters.Wide_Latin_9} (@file{a-cwila1.ads})
11444 @cindex Latin_9 constants for Wide_Character
11447 This child of @code{Ada.Characters}
11448 provides a set of definitions corresponding to those in the
11449 GNAT defined package @code{Ada.Characters.Latin_9} but with the
11450 types of the constants being @code{Wide_Character}
11451 instead of @code{Character}. The provision of such a package
11452 is specifically authorized by the Ada Reference Manual
11455 @node Ada.Command_Line.Remove (a-colire.ads)
11456 @section @code{Ada.Command_Line.Remove} (@file{a-colire.ads})
11457 @cindex @code{Ada.Command_Line.Remove} (@file{a-colire.ads})
11458 @cindex Removing command line arguments
11459 @cindex Command line, argument removal
11462 This child of @code{Ada.Command_Line}
11463 provides a mechanism for logically removing
11464 arguments from the argument list. Once removed, an argument is not visible
11465 to further calls on the subprograms in @code{Ada.Command_Line} will not
11466 see the removed argument.
11468 @node Ada.Command_Line.Environment (a-colien.ads)
11469 @section @code{Ada.Command_Line.Environment} (@file{a-colien.ads})
11470 @cindex @code{Ada.Command_Line.Environment} (@file{a-colien.ads})
11471 @cindex Environment entries
11474 This child of @code{Ada.Command_Line}
11475 provides a mechanism for obtaining environment values on systems
11476 where this concept makes sense.
11478 @node Ada.Direct_IO.C_Streams (a-diocst.ads)
11479 @section @code{Ada.Direct_IO.C_Streams} (@file{a-diocst.ads})
11480 @cindex @code{Ada.Direct_IO.C_Streams} (@file{a-diocst.ads})
11481 @cindex C Streams, Interfacing with Direct_IO
11484 This package provides subprograms that allow interfacing between
11485 C streams and @code{Direct_IO}. The stream identifier can be
11486 extracted from a file opened on the Ada side, and an Ada file
11487 can be constructed from a stream opened on the C side.
11489 @node Ada.Exceptions.Is_Null_Occurrence (a-einuoc.ads)
11490 @section @code{Ada.Exceptions.Is_Null_Occurrence} (@file{a-einuoc.ads})
11491 @cindex @code{Ada.Exceptions.Is_Null_Occurrence} (@file{a-einuoc.ads})
11492 @cindex Null_Occurrence, testing for
11495 This child subprogram provides a way of testing for the null
11496 exception occurrence (@code{Null_Occurrence}) without raising
11499 @node Ada.Exceptions.Traceback (a-exctra.ads)
11500 @section @code{Ada.Exceptions.Traceback} (@file{a-exctra.ads})
11501 @cindex @code{Ada.Exceptions.Traceback} (@file{a-exctra.ads})
11502 @cindex Traceback for Exception Occurrence
11505 This child package provides the subprogram (@code{Tracebacks}) to
11506 give a traceback array of addresses based on an exception
11509 @node Ada.Sequential_IO.C_Streams (a-siocst.ads)
11510 @section @code{Ada.Sequential_IO.C_Streams} (@file{a-siocst.ads})
11511 @cindex @code{Ada.Sequential_IO.C_Streams} (@file{a-siocst.ads})
11512 @cindex C Streams, Interfacing with Sequential_IO
11515 This package provides subprograms that allow interfacing between
11516 C streams and @code{Sequential_IO}. The stream identifier can be
11517 extracted from a file opened on the Ada side, and an Ada file
11518 can be constructed from a stream opened on the C side.
11520 @node Ada.Streams.Stream_IO.C_Streams (a-ssicst.ads)
11521 @section @code{Ada.Streams.Stream_IO.C_Streams} (@file{a-ssicst.ads})
11522 @cindex @code{Ada.Streams.Stream_IO.C_Streams} (@file{a-ssicst.ads})
11523 @cindex C Streams, Interfacing with Stream_IO
11526 This package provides subprograms that allow interfacing between
11527 C streams and @code{Stream_IO}. The stream identifier can be
11528 extracted from a file opened on the Ada side, and an Ada file
11529 can be constructed from a stream opened on the C side.
11531 @node Ada.Strings.Unbounded.Text_IO (a-suteio.ads)
11532 @section @code{Ada.Strings.Unbounded.Text_IO} (@file{a-suteio.ads})
11533 @cindex @code{Ada.Strings.Unbounded.Text_IO} (@file{a-suteio.ads})
11534 @cindex @code{Unbounded_String}, IO support
11535 @cindex @code{Text_IO}, extensions for unbounded strings
11538 This package provides subprograms for Text_IO for unbounded
11539 strings, avoiding the necessity for an intermediate operation
11540 with ordinary strings.
11542 @node Ada.Strings.Wide_Unbounded.Wide_Text_IO (a-swuwti.ads)
11543 @section @code{Ada.Strings.Wide_Unbounded.Wide_Text_IO} (@file{a-swuwti.ads})
11544 @cindex @code{Ada.Strings.Wide_Unbounded.Wide_Text_IO} (@file{a-swuwti.ads})
11545 @cindex @code{Unbounded_Wide_String}, IO support
11546 @cindex @code{Text_IO}, extensions for unbounded wide strings
11549 This package provides subprograms for Text_IO for unbounded
11550 wide strings, avoiding the necessity for an intermediate operation
11551 with ordinary wide strings.
11553 @node Ada.Text_IO.C_Streams (a-tiocst.ads)
11554 @section @code{Ada.Text_IO.C_Streams} (@file{a-tiocst.ads})
11555 @cindex @code{Ada.Text_IO.C_Streams} (@file{a-tiocst.ads})
11556 @cindex C Streams, Interfacing with @code{Text_IO}
11559 This package provides subprograms that allow interfacing between
11560 C streams and @code{Text_IO}. The stream identifier can be
11561 extracted from a file opened on the Ada side, and an Ada file
11562 can be constructed from a stream opened on the C side.
11564 @node Ada.Wide_Text_IO.C_Streams (a-wtcstr.ads)
11565 @section @code{Ada.Wide_Text_IO.C_Streams} (@file{a-wtcstr.ads})
11566 @cindex @code{Ada.Wide_Text_IO.C_Streams} (@file{a-wtcstr.ads})
11567 @cindex C Streams, Interfacing with @code{Wide_Text_IO}
11570 This package provides subprograms that allow interfacing between
11571 C streams and @code{Wide_Text_IO}. The stream identifier can be
11572 extracted from a file opened on the Ada side, and an Ada file
11573 can be constructed from a stream opened on the C side.
11575 @node GNAT.Array_Split (g-arrspl.ads)
11576 @section @code{GNAT.Array_Split} (@file{g-arrspl.ads})
11577 @cindex @code{GNAT.Array_Split} (@file{g-arrspl.ads})
11578 @cindex Array splitter
11581 Useful array-manipulation routines: given a set of separators, split
11582 an array wherever the separators appear, and provide direct access
11583 to the resulting slices.
11585 @node GNAT.AWK (g-awk.ads)
11586 @section @code{GNAT.AWK} (@file{g-awk.ads})
11587 @cindex @code{GNAT.AWK} (@file{g-awk.ads})
11592 Provides AWK-like parsing functions, with an easy interface for parsing one
11593 or more files containing formatted data. The file is viewed as a database
11594 where each record is a line and a field is a data element in this line.
11596 @node GNAT.Bounded_Buffers (g-boubuf.ads)
11597 @section @code{GNAT.Bounded_Buffers} (@file{g-boubuf.ads})
11598 @cindex @code{GNAT.Bounded_Buffers} (@file{g-boubuf.ads})
11600 @cindex Bounded Buffers
11603 Provides a concurrent generic bounded buffer abstraction. Instances are
11604 useful directly or as parts of the implementations of other abstractions,
11607 @node GNAT.Bounded_Mailboxes (g-boumai.ads)
11608 @section @code{GNAT.Bounded_Mailboxes} (@file{g-boumai.ads})
11609 @cindex @code{GNAT.Bounded_Mailboxes} (@file{g-boumai.ads})
11614 Provides a thread-safe asynchronous intertask mailbox communication facility.
11616 @node GNAT.Bubble_Sort (g-bubsor.ads)
11617 @section @code{GNAT.Bubble_Sort} (@file{g-bubsor.ads})
11618 @cindex @code{GNAT.Bubble_Sort} (@file{g-bubsor.ads})
11620 @cindex Bubble sort
11623 Provides a general implementation of bubble sort usable for sorting arbitrary
11624 data items. Exchange and comparison procedures are provided by passing
11625 access-to-procedure values.
11627 @node GNAT.Bubble_Sort_A (g-busora.ads)
11628 @section @code{GNAT.Bubble_Sort_A} (@file{g-busora.ads})
11629 @cindex @code{GNAT.Bubble_Sort_A} (@file{g-busora.ads})
11631 @cindex Bubble sort
11634 Provides a general implementation of bubble sort usable for sorting arbitrary
11635 data items. Move and comparison procedures are provided by passing
11636 access-to-procedure values. This is an older version, retained for
11637 compatibility. Usually @code{GNAT.Bubble_Sort} will be preferable.
11639 @node GNAT.Bubble_Sort_G (g-busorg.ads)
11640 @section @code{GNAT.Bubble_Sort_G} (@file{g-busorg.ads})
11641 @cindex @code{GNAT.Bubble_Sort_G} (@file{g-busorg.ads})
11643 @cindex Bubble sort
11646 Similar to @code{Bubble_Sort_A} except that the move and sorting procedures
11647 are provided as generic parameters, this improves efficiency, especially
11648 if the procedures can be inlined, at the expense of duplicating code for
11649 multiple instantiations.
11651 @node GNAT.Calendar (g-calend.ads)
11652 @section @code{GNAT.Calendar} (@file{g-calend.ads})
11653 @cindex @code{GNAT.Calendar} (@file{g-calend.ads})
11654 @cindex @code{Calendar}
11657 Extends the facilities provided by @code{Ada.Calendar} to include handling
11658 of days of the week, an extended @code{Split} and @code{Time_Of} capability.
11659 Also provides conversion of @code{Ada.Calendar.Time} values to and from the
11660 C @code{timeval} format.
11662 @node GNAT.Calendar.Time_IO (g-catiio.ads)
11663 @section @code{GNAT.Calendar.Time_IO} (@file{g-catiio.ads})
11664 @cindex @code{Calendar}
11666 @cindex @code{GNAT.Calendar.Time_IO} (@file{g-catiio.ads})
11668 @node GNAT.CRC32 (g-crc32.ads)
11669 @section @code{GNAT.CRC32} (@file{g-crc32.ads})
11670 @cindex @code{GNAT.CRC32} (@file{g-crc32.ads})
11672 @cindex Cyclic Redundancy Check
11675 This package implements the CRC-32 algorithm. For a full description
11676 of this algorithm see
11677 ``Computation of Cyclic Redundancy Checks via Table Look-Up'',
11678 @cite{Communications of the ACM}, Vol.@: 31 No.@: 8, pp.@: 1008-1013,
11679 Aug.@: 1988. Sarwate, D.V@.
11682 Provides an extended capability for formatted output of time values with
11683 full user control over the format. Modeled on the GNU Date specification.
11685 @node GNAT.Case_Util (g-casuti.ads)
11686 @section @code{GNAT.Case_Util} (@file{g-casuti.ads})
11687 @cindex @code{GNAT.Case_Util} (@file{g-casuti.ads})
11688 @cindex Casing utilities
11689 @cindex Character handling (@code{GNAT.Case_Util})
11692 A set of simple routines for handling upper and lower casing of strings
11693 without the overhead of the full casing tables
11694 in @code{Ada.Characters.Handling}.
11696 @node GNAT.CGI (g-cgi.ads)
11697 @section @code{GNAT.CGI} (@file{g-cgi.ads})
11698 @cindex @code{GNAT.CGI} (@file{g-cgi.ads})
11699 @cindex CGI (Common Gateway Interface)
11702 This is a package for interfacing a GNAT program with a Web server via the
11703 Common Gateway Interface (CGI)@. Basically this package parses the CGI
11704 parameters, which are a set of key/value pairs sent by the Web server. It
11705 builds a table whose index is the key and provides some services to deal
11708 @node GNAT.CGI.Cookie (g-cgicoo.ads)
11709 @section @code{GNAT.CGI.Cookie} (@file{g-cgicoo.ads})
11710 @cindex @code{GNAT.CGI.Cookie} (@file{g-cgicoo.ads})
11711 @cindex CGI (Common Gateway Interface) cookie support
11712 @cindex Cookie support in CGI
11715 This is a package to interface a GNAT program with a Web server via the
11716 Common Gateway Interface (CGI). It exports services to deal with Web
11717 cookies (piece of information kept in the Web client software).
11719 @node GNAT.CGI.Debug (g-cgideb.ads)
11720 @section @code{GNAT.CGI.Debug} (@file{g-cgideb.ads})
11721 @cindex @code{GNAT.CGI.Debug} (@file{g-cgideb.ads})
11722 @cindex CGI (Common Gateway Interface) debugging
11725 This is a package to help debugging CGI (Common Gateway Interface)
11726 programs written in Ada.
11728 @node GNAT.Command_Line (g-comlin.ads)
11729 @section @code{GNAT.Command_Line} (@file{g-comlin.ads})
11730 @cindex @code{GNAT.Command_Line} (@file{g-comlin.ads})
11731 @cindex Command line
11734 Provides a high level interface to @code{Ada.Command_Line} facilities,
11735 including the ability to scan for named switches with optional parameters
11736 and expand file names using wild card notations.
11738 @node GNAT.Compiler_Version (g-comver.ads)
11739 @section @code{GNAT.Compiler_Version} (@file{g-comver.ads})
11740 @cindex @code{GNAT.Compiler_Version} (@file{g-comver.ads})
11741 @cindex Compiler Version
11742 @cindex Version, of compiler
11745 Provides a routine for obtaining the version of the compiler used to
11746 compile the program. More accurately this is the version of the binder
11747 used to bind the program (this will normally be the same as the version
11748 of the compiler if a consistent tool set is used to compile all units
11751 @node GNAT.Ctrl_C (g-ctrl_c.ads)
11752 @section @code{GNAT.Ctrl_C} (@file{g-ctrl_c.ads})
11753 @cindex @code{GNAT.Ctrl_C} (@file{g-ctrl_c.ads})
11757 Provides a simple interface to handle Ctrl-C keyboard events.
11759 @node GNAT.Current_Exception (g-curexc.ads)
11760 @section @code{GNAT.Current_Exception} (@file{g-curexc.ads})
11761 @cindex @code{GNAT.Current_Exception} (@file{g-curexc.ads})
11762 @cindex Current exception
11763 @cindex Exception retrieval
11766 Provides access to information on the current exception that has been raised
11767 without the need for using the Ada-95 exception choice parameter specification
11768 syntax. This is particularly useful in simulating typical facilities for
11769 obtaining information about exceptions provided by Ada 83 compilers.
11771 @node GNAT.Debug_Pools (g-debpoo.ads)
11772 @section @code{GNAT.Debug_Pools} (@file{g-debpoo.ads})
11773 @cindex @code{GNAT.Debug_Pools} (@file{g-debpoo.ads})
11775 @cindex Debug pools
11776 @cindex Memory corruption debugging
11779 Provide a debugging storage pools that helps tracking memory corruption
11780 problems. See section ``Finding memory problems with GNAT Debug Pool'' in
11781 the @cite{GNAT User's Guide}.
11783 @node GNAT.Debug_Utilities (g-debuti.ads)
11784 @section @code{GNAT.Debug_Utilities} (@file{g-debuti.ads})
11785 @cindex @code{GNAT.Debug_Utilities} (@file{g-debuti.ads})
11789 Provides a few useful utilities for debugging purposes, including conversion
11790 to and from string images of address values. Supports both C and Ada formats
11791 for hexadecimal literals.
11793 @node GNAT.Directory_Operations (g-dirope.ads)
11794 @section @code{GNAT.Directory_Operations} (g-dirope.ads)
11795 @cindex @code{GNAT.Directory_Operations} (g-dirope.ads)
11796 @cindex Directory operations
11799 Provides a set of routines for manipulating directories, including changing
11800 the current directory, making new directories, and scanning the files in a
11803 @node GNAT.Dynamic_HTables (g-dynhta.ads)
11804 @section @code{GNAT.Dynamic_HTables} (@file{g-dynhta.ads})
11805 @cindex @code{GNAT.Dynamic_HTables} (@file{g-dynhta.ads})
11806 @cindex Hash tables
11809 A generic implementation of hash tables that can be used to hash arbitrary
11810 data. Provided in two forms, a simple form with built in hash functions,
11811 and a more complex form in which the hash function is supplied.
11814 This package provides a facility similar to that of @code{GNAT.HTable},
11815 except that this package declares a type that can be used to define
11816 dynamic instances of the hash table, while an instantiation of
11817 @code{GNAT.HTable} creates a single instance of the hash table.
11819 @node GNAT.Dynamic_Tables (g-dyntab.ads)
11820 @section @code{GNAT.Dynamic_Tables} (@file{g-dyntab.ads})
11821 @cindex @code{GNAT.Dynamic_Tables} (@file{g-dyntab.ads})
11822 @cindex Table implementation
11823 @cindex Arrays, extendable
11826 A generic package providing a single dimension array abstraction where the
11827 length of the array can be dynamically modified.
11830 This package provides a facility similar to that of @code{GNAT.Table},
11831 except that this package declares a type that can be used to define
11832 dynamic instances of the table, while an instantiation of
11833 @code{GNAT.Table} creates a single instance of the table type.
11835 @node GNAT.Exception_Actions (g-excact.ads)
11836 @section @code{GNAT.Exception_Actions} (@file{g-excact.ads})
11837 @cindex @code{GNAT.Exception_Actions} (@file{g-excact.ads})
11838 @cindex Exception actions
11841 Provides callbacks when an exception is raised. Callbacks can be registered
11842 for specific exceptions, or when any exception is raised. This
11843 can be used for instance to force a core dump to ease debugging.
11845 @node GNAT.Exception_Traces (g-exctra.ads)
11846 @section @code{GNAT.Exception_Traces} (@file{g-exctra.ads})
11847 @cindex @code{GNAT.Exception_Traces} (@file{g-exctra.ads})
11848 @cindex Exception traces
11852 Provides an interface allowing to control automatic output upon exception
11855 @node GNAT.Exceptions (g-except.ads)
11856 @section @code{GNAT.Exceptions} (@file{g-expect.ads})
11857 @cindex @code{GNAT.Exceptions} (@file{g-expect.ads})
11858 @cindex Exceptions, Pure
11859 @cindex Pure packages, exceptions
11862 Normally it is not possible to raise an exception with
11863 a message from a subprogram in a pure package, since the
11864 necessary types and subprograms are in @code{Ada.Exceptions}
11865 which is not a pure unit. @code{GNAT.Exceptions} provides a
11866 facility for getting around this limitation for a few
11867 predefined exceptions, and for example allow raising
11868 @code{Constraint_Error} with a message from a pure subprogram.
11870 @node GNAT.Expect (g-expect.ads)
11871 @section @code{GNAT.Expect} (@file{g-expect.ads})
11872 @cindex @code{GNAT.Expect} (@file{g-expect.ads})
11875 Provides a set of subprograms similar to what is available
11876 with the standard Tcl Expect tool.
11877 It allows you to easily spawn and communicate with an external process.
11878 You can send commands or inputs to the process, and compare the output
11879 with some expected regular expression. Currently @code{GNAT.Expect}
11880 is implemented on all native GNAT ports except for OpenVMS@.
11881 It is not implemented for cross ports, and in particular is not
11882 implemented for VxWorks or LynxOS@.
11884 @node GNAT.Float_Control (g-flocon.ads)
11885 @section @code{GNAT.Float_Control} (@file{g-flocon.ads})
11886 @cindex @code{GNAT.Float_Control} (@file{g-flocon.ads})
11887 @cindex Floating-Point Processor
11890 Provides an interface for resetting the floating-point processor into the
11891 mode required for correct semantic operation in Ada. Some third party
11892 library calls may cause this mode to be modified, and the Reset procedure
11893 in this package can be used to reestablish the required mode.
11895 @node GNAT.Heap_Sort (g-heasor.ads)
11896 @section @code{GNAT.Heap_Sort} (@file{g-heasor.ads})
11897 @cindex @code{GNAT.Heap_Sort} (@file{g-heasor.ads})
11901 Provides a general implementation of heap sort usable for sorting arbitrary
11902 data items. Exchange and comparison procedures are provided by passing
11903 access-to-procedure values. The algorithm used is a modified heap sort
11904 that performs approximately N*log(N) comparisons in the worst case.
11906 @node GNAT.Heap_Sort_A (g-hesora.ads)
11907 @section @code{GNAT.Heap_Sort_A} (@file{g-hesora.ads})
11908 @cindex @code{GNAT.Heap_Sort_A} (@file{g-hesora.ads})
11912 Provides a general implementation of heap sort usable for sorting arbitrary
11913 data items. Move and comparison procedures are provided by passing
11914 access-to-procedure values. The algorithm used is a modified heap sort
11915 that performs approximately N*log(N) comparisons in the worst case.
11916 This differs from @code{GNAT.Heap_Sort} in having a less convenient
11917 interface, but may be slightly more efficient.
11919 @node GNAT.Heap_Sort_G (g-hesorg.ads)
11920 @section @code{GNAT.Heap_Sort_G} (@file{g-hesorg.ads})
11921 @cindex @code{GNAT.Heap_Sort_G} (@file{g-hesorg.ads})
11925 Similar to @code{Heap_Sort_A} except that the move and sorting procedures
11926 are provided as generic parameters, this improves efficiency, especially
11927 if the procedures can be inlined, at the expense of duplicating code for
11928 multiple instantiations.
11930 @node GNAT.HTable (g-htable.ads)
11931 @section @code{GNAT.HTable} (@file{g-htable.ads})
11932 @cindex @code{GNAT.HTable} (@file{g-htable.ads})
11933 @cindex Hash tables
11936 A generic implementation of hash tables that can be used to hash arbitrary
11937 data. Provides two approaches, one a simple static approach, and the other
11938 allowing arbitrary dynamic hash tables.
11940 @node GNAT.IO (g-io.ads)
11941 @section @code{GNAT.IO} (@file{g-io.ads})
11942 @cindex @code{GNAT.IO} (@file{g-io.ads})
11944 @cindex Input/Output facilities
11947 A simple preelaborable input-output package that provides a subset of
11948 simple Text_IO functions for reading characters and strings from
11949 Standard_Input, and writing characters, strings and integers to either
11950 Standard_Output or Standard_Error.
11952 @node GNAT.IO_Aux (g-io_aux.ads)
11953 @section @code{GNAT.IO_Aux} (@file{g-io_aux.ads})
11954 @cindex @code{GNAT.IO_Aux} (@file{g-io_aux.ads})
11956 @cindex Input/Output facilities
11958 Provides some auxiliary functions for use with Text_IO, including a test
11959 for whether a file exists, and functions for reading a line of text.
11961 @node GNAT.Lock_Files (g-locfil.ads)
11962 @section @code{GNAT.Lock_Files} (@file{g-locfil.ads})
11963 @cindex @code{GNAT.Lock_Files} (@file{g-locfil.ads})
11964 @cindex File locking
11965 @cindex Locking using files
11968 Provides a general interface for using files as locks. Can be used for
11969 providing program level synchronization.
11971 @node GNAT.MD5 (g-md5.ads)
11972 @section @code{GNAT.MD5} (@file{g-md5.ads})
11973 @cindex @code{GNAT.MD5} (@file{g-md5.ads})
11974 @cindex Message Digest MD5
11977 Implements the MD5 Message-Digest Algorithm as described in RFC 1321.
11979 @node GNAT.Memory_Dump (g-memdum.ads)
11980 @section @code{GNAT.Memory_Dump} (@file{g-memdum.ads})
11981 @cindex @code{GNAT.Memory_Dump} (@file{g-memdum.ads})
11982 @cindex Dump Memory
11985 Provides a convenient routine for dumping raw memory to either the
11986 standard output or standard error files. Uses GNAT.IO for actual
11989 @node GNAT.Most_Recent_Exception (g-moreex.ads)
11990 @section @code{GNAT.Most_Recent_Exception} (@file{g-moreex.ads})
11991 @cindex @code{GNAT.Most_Recent_Exception} (@file{g-moreex.ads})
11992 @cindex Exception, obtaining most recent
11995 Provides access to the most recently raised exception. Can be used for
11996 various logging purposes, including duplicating functionality of some
11997 Ada 83 implementation dependent extensions.
11999 @node GNAT.OS_Lib (g-os_lib.ads)
12000 @section @code{GNAT.OS_Lib} (@file{g-os_lib.ads})
12001 @cindex @code{GNAT.OS_Lib} (@file{g-os_lib.ads})
12002 @cindex Operating System interface
12003 @cindex Spawn capability
12006 Provides a range of target independent operating system interface functions,
12007 including time/date management, file operations, subprocess management,
12008 including a portable spawn procedure, and access to environment variables
12009 and error return codes.
12011 @node GNAT.Perfect_Hash.Generators (g-pehage.ads)
12012 @section @code{GNAT.Perfect_Hash.Generators} (@file{g-pehage.ads})
12013 @cindex @code{GNAT.Perfect_Hash.Generators} (@file{g-pehage.ads})
12014 @cindex Hash functions
12017 Provides a generator of static minimal perfect hash functions. No
12018 collisions occur and each item can be retrieved from the table in one
12019 probe (perfect property). The hash table size corresponds to the exact
12020 size of the key set and no larger (minimal property). The key set has to
12021 be know in advance (static property). The hash functions are also order
12022 preservering. If w2 is inserted after w1 in the generator, their
12023 hashcode are in the same order. These hashing functions are very
12024 convenient for use with realtime applications.
12026 @node GNAT.Regexp (g-regexp.ads)
12027 @section @code{GNAT.Regexp} (@file{g-regexp.ads})
12028 @cindex @code{GNAT.Regexp} (@file{g-regexp.ads})
12029 @cindex Regular expressions
12030 @cindex Pattern matching
12033 A simple implementation of regular expressions, using a subset of regular
12034 expression syntax copied from familiar Unix style utilities. This is the
12035 simples of the three pattern matching packages provided, and is particularly
12036 suitable for ``file globbing'' applications.
12038 @node GNAT.Registry (g-regist.ads)
12039 @section @code{GNAT.Registry} (@file{g-regist.ads})
12040 @cindex @code{GNAT.Registry} (@file{g-regist.ads})
12041 @cindex Windows Registry
12044 This is a high level binding to the Windows registry. It is possible to
12045 do simple things like reading a key value, creating a new key. For full
12046 registry API, but at a lower level of abstraction, refer to the Win32.Winreg
12047 package provided with the Win32Ada binding
12049 @node GNAT.Regpat (g-regpat.ads)
12050 @section @code{GNAT.Regpat} (@file{g-regpat.ads})
12051 @cindex @code{GNAT.Regpat} (@file{g-regpat.ads})
12052 @cindex Regular expressions
12053 @cindex Pattern matching
12056 A complete implementation of Unix-style regular expression matching, copied
12057 from the original V7 style regular expression library written in C by
12058 Henry Spencer (and binary compatible with this C library).
12060 @node GNAT.Secondary_Stack_Info (g-sestin.ads)
12061 @section @code{GNAT.Secondary_Stack_Info} (@file{g-sestin.ads})
12062 @cindex @code{GNAT.Secondary_Stack_Info} (@file{g-sestin.ads})
12063 @cindex Secondary Stack Info
12066 Provide the capability to query the high water mark of the current task's
12069 @node GNAT.Semaphores (g-semaph.ads)
12070 @section @code{GNAT.Semaphores} (@file{g-semaph.ads})
12071 @cindex @code{GNAT.Semaphores} (@file{g-semaph.ads})
12075 Provides classic counting and binary semaphores using protected types.
12077 @node GNAT.Signals (g-signal.ads)
12078 @section @code{GNAT.Signals} (@file{g-signal.ads})
12079 @cindex @code{GNAT.Signals} (@file{g-signal.ads})
12083 Provides the ability to manipulate the blocked status of signals on supported
12086 @node GNAT.Sockets (g-socket.ads)
12087 @section @code{GNAT.Sockets} (@file{g-socket.ads})
12088 @cindex @code{GNAT.Sockets} (@file{g-socket.ads})
12092 A high level and portable interface to develop sockets based applications.
12093 This package is based on the sockets thin binding found in
12094 @code{GNAT.Sockets.Thin}. Currently @code{GNAT.Sockets} is implemented
12095 on all native GNAT ports except for OpenVMS@. It is not implemented
12096 for the LynxOS@ cross port.
12098 @node GNAT.Source_Info (g-souinf.ads)
12099 @section @code{GNAT.Source_Info} (@file{g-souinf.ads})
12100 @cindex @code{GNAT.Source_Info} (@file{g-souinf.ads})
12101 @cindex Source Information
12104 Provides subprograms that give access to source code information known at
12105 compile time, such as the current file name and line number.
12107 @node GNAT.Spell_Checker (g-speche.ads)
12108 @section @code{GNAT.Spell_Checker} (@file{g-speche.ads})
12109 @cindex @code{GNAT.Spell_Checker} (@file{g-speche.ads})
12110 @cindex Spell checking
12113 Provides a function for determining whether one string is a plausible
12114 near misspelling of another string.
12116 @node GNAT.Spitbol.Patterns (g-spipat.ads)
12117 @section @code{GNAT.Spitbol.Patterns} (@file{g-spipat.ads})
12118 @cindex @code{GNAT.Spitbol.Patterns} (@file{g-spipat.ads})
12119 @cindex SPITBOL pattern matching
12120 @cindex Pattern matching
12123 A complete implementation of SNOBOL4 style pattern matching. This is the
12124 most elaborate of the pattern matching packages provided. It fully duplicates
12125 the SNOBOL4 dynamic pattern construction and matching capabilities, using the
12126 efficient algorithm developed by Robert Dewar for the SPITBOL system.
12128 @node GNAT.Spitbol (g-spitbo.ads)
12129 @section @code{GNAT.Spitbol} (@file{g-spitbo.ads})
12130 @cindex @code{GNAT.Spitbol} (@file{g-spitbo.ads})
12131 @cindex SPITBOL interface
12134 The top level package of the collection of SPITBOL-style functionality, this
12135 package provides basic SNOBOL4 string manipulation functions, such as
12136 Pad, Reverse, Trim, Substr capability, as well as a generic table function
12137 useful for constructing arbitrary mappings from strings in the style of
12138 the SNOBOL4 TABLE function.
12140 @node GNAT.Spitbol.Table_Boolean (g-sptabo.ads)
12141 @section @code{GNAT.Spitbol.Table_Boolean} (@file{g-sptabo.ads})
12142 @cindex @code{GNAT.Spitbol.Table_Boolean} (@file{g-sptabo.ads})
12143 @cindex Sets of strings
12144 @cindex SPITBOL Tables
12147 A library level of instantiation of @code{GNAT.Spitbol.Patterns.Table}
12148 for type @code{Standard.Boolean}, giving an implementation of sets of
12151 @node GNAT.Spitbol.Table_Integer (g-sptain.ads)
12152 @section @code{GNAT.Spitbol.Table_Integer} (@file{g-sptain.ads})
12153 @cindex @code{GNAT.Spitbol.Table_Integer} (@file{g-sptain.ads})
12154 @cindex Integer maps
12156 @cindex SPITBOL Tables
12159 A library level of instantiation of @code{GNAT.Spitbol.Patterns.Table}
12160 for type @code{Standard.Integer}, giving an implementation of maps
12161 from string to integer values.
12163 @node GNAT.Spitbol.Table_VString (g-sptavs.ads)
12164 @section @code{GNAT.Spitbol.Table_VString} (@file{g-sptavs.ads})
12165 @cindex @code{GNAT.Spitbol.Table_VString} (@file{g-sptavs.ads})
12166 @cindex String maps
12168 @cindex SPITBOL Tables
12171 A library level of instantiation of @code{GNAT.Spitbol.Patterns.Table} for
12172 a variable length string type, giving an implementation of general
12173 maps from strings to strings.
12175 @node GNAT.Strings (g-string.ads)
12176 @section @code{GNAT.Strings} (@file{g-string.ads})
12177 @cindex @code{GNAT.Strings} (@file{g-string.ads})
12180 Common String access types and related subprograms. Basically it
12181 defines a string access and an array of string access types.
12183 @node GNAT.String_Split (g-strspl.ads)
12184 @section @code{GNAT.String_Split} (@file{g-strspl.ads})
12185 @cindex @code{GNAT.String_Split} (@file{g-strspl.ads})
12186 @cindex String splitter
12189 Useful string-manipulation routines: given a set of separators, split
12190 a string wherever the separators appear, and provide direct access
12191 to the resulting slices. This package is instantiated from
12192 @code{GNAT.Array_Split}.
12194 @node GNAT.Table (g-table.ads)
12195 @section @code{GNAT.Table} (@file{g-table.ads})
12196 @cindex @code{GNAT.Table} (@file{g-table.ads})
12197 @cindex Table implementation
12198 @cindex Arrays, extendable
12201 A generic package providing a single dimension array abstraction where the
12202 length of the array can be dynamically modified.
12205 This package provides a facility similar to that of @code{GNAT.Dynamic_Tables},
12206 except that this package declares a single instance of the table type,
12207 while an instantiation of @code{GNAT.Dynamic_Tables} creates a type that can be
12208 used to define dynamic instances of the table.
12210 @node GNAT.Task_Lock (g-tasloc.ads)
12211 @section @code{GNAT.Task_Lock} (@file{g-tasloc.ads})
12212 @cindex @code{GNAT.Task_Lock} (@file{g-tasloc.ads})
12213 @cindex Task synchronization
12214 @cindex Task locking
12218 A very simple facility for locking and unlocking sections of code using a
12219 single global task lock. Appropriate for use in situations where contention
12220 between tasks is very rarely expected.
12222 @node GNAT.Threads (g-thread.ads)
12223 @section @code{GNAT.Threads} (@file{g-thread.ads})
12224 @cindex @code{GNAT.Threads} (@file{g-thread.ads})
12225 @cindex Foreign threads
12226 @cindex Threads, foreign
12229 Provides facilities for creating and destroying threads with explicit calls.
12230 These threads are known to the GNAT run-time system. These subprograms are
12231 exported C-convention procedures intended to be called from foreign code.
12232 By using these primitives rather than directly calling operating systems
12233 routines, compatibility with the Ada tasking runt-time is provided.
12235 @node GNAT.Traceback (g-traceb.ads)
12236 @section @code{GNAT.Traceback} (@file{g-traceb.ads})
12237 @cindex @code{GNAT.Traceback} (@file{g-traceb.ads})
12238 @cindex Trace back facilities
12241 Provides a facility for obtaining non-symbolic traceback information, useful
12242 in various debugging situations.
12244 @node GNAT.Traceback.Symbolic (g-trasym.ads)
12245 @section @code{GNAT.Traceback.Symbolic} (@file{g-trasym.ads})
12246 @cindex @code{GNAT.Traceback.Symbolic} (@file{g-trasym.ads})
12247 @cindex Trace back facilities
12250 Provides symbolic traceback information that includes the subprogram
12251 name and line number information.
12253 @node GNAT.Wide_String_Split (g-wistsp.ads)
12254 @section @code{GNAT.Wide_String_Split} (@file{g-wistsp.ads})
12255 @cindex @code{GNAT.Wide_String_Split} (@file{g-wistsp.ads})
12256 @cindex Wide_String splitter
12259 Useful wide_string-manipulation routines: given a set of separators, split
12260 a wide_string wherever the separators appear, and provide direct access
12261 to the resulting slices. This package is instantiated from
12262 @code{GNAT.Array_Split}.
12264 @node Interfaces.C.Extensions (i-cexten.ads)
12265 @section @code{Interfaces.C.Extensions} (@file{i-cexten.ads})
12266 @cindex @code{Interfaces.C.Extensions} (@file{i-cexten.ads})
12269 This package contains additional C-related definitions, intended
12270 for use with either manually or automatically generated bindings
12273 @node Interfaces.C.Streams (i-cstrea.ads)
12274 @section @code{Interfaces.C.Streams} (@file{i-cstrea.ads})
12275 @cindex @code{Interfaces.C.Streams} (@file{i-cstrea.ads})
12276 @cindex C streams, interfacing
12279 This package is a binding for the most commonly used operations
12282 @node Interfaces.CPP (i-cpp.ads)
12283 @section @code{Interfaces.CPP} (@file{i-cpp.ads})
12284 @cindex @code{Interfaces.CPP} (@file{i-cpp.ads})
12285 @cindex C++ interfacing
12286 @cindex Interfacing, to C++
12289 This package provides facilities for use in interfacing to C++. It
12290 is primarily intended to be used in connection with automated tools
12291 for the generation of C++ interfaces.
12293 @node Interfaces.Os2lib (i-os2lib.ads)
12294 @section @code{Interfaces.Os2lib} (@file{i-os2lib.ads})
12295 @cindex @code{Interfaces.Os2lib} (@file{i-os2lib.ads})
12296 @cindex Interfacing, to OS/2
12297 @cindex OS/2 interfacing
12300 This package provides interface definitions to the OS/2 library.
12301 It is a thin binding which is a direct translation of the
12302 various @file{<bse@.h>} files.
12304 @node Interfaces.Os2lib.Errors (i-os2err.ads)
12305 @section @code{Interfaces.Os2lib.Errors} (@file{i-os2err.ads})
12306 @cindex @code{Interfaces.Os2lib.Errors} (@file{i-os2err.ads})
12307 @cindex OS/2 Error codes
12308 @cindex Interfacing, to OS/2
12309 @cindex OS/2 interfacing
12312 This package provides definitions of the OS/2 error codes.
12314 @node Interfaces.Os2lib.Synchronization (i-os2syn.ads)
12315 @section @code{Interfaces.Os2lib.Synchronization} (@file{i-os2syn.ads})
12316 @cindex @code{Interfaces.Os2lib.Synchronization} (@file{i-os2syn.ads})
12317 @cindex Interfacing, to OS/2
12318 @cindex Synchronization, OS/2
12319 @cindex OS/2 synchronization primitives
12322 This is a child package that provides definitions for interfacing
12323 to the @code{OS/2} synchronization primitives.
12325 @node Interfaces.Os2lib.Threads (i-os2thr.ads)
12326 @section @code{Interfaces.Os2lib.Threads} (@file{i-os2thr.ads})
12327 @cindex @code{Interfaces.Os2lib.Threads} (@file{i-os2thr.ads})
12328 @cindex Interfacing, to OS/2
12329 @cindex Thread control, OS/2
12330 @cindex OS/2 thread interfacing
12333 This is a child package that provides definitions for interfacing
12334 to the @code{OS/2} thread primitives.
12336 @node Interfaces.Packed_Decimal (i-pacdec.ads)
12337 @section @code{Interfaces.Packed_Decimal} (@file{i-pacdec.ads})
12338 @cindex @code{Interfaces.Packed_Decimal} (@file{i-pacdec.ads})
12339 @cindex IBM Packed Format
12340 @cindex Packed Decimal
12343 This package provides a set of routines for conversions to and
12344 from a packed decimal format compatible with that used on IBM
12347 @node Interfaces.VxWorks (i-vxwork.ads)
12348 @section @code{Interfaces.VxWorks} (@file{i-vxwork.ads})
12349 @cindex @code{Interfaces.VxWorks} (@file{i-vxwork.ads})
12350 @cindex Interfacing to VxWorks
12351 @cindex VxWorks, interfacing
12354 This package provides a limited binding to the VxWorks API.
12355 In particular, it interfaces with the
12356 VxWorks hardware interrupt facilities.
12358 @node Interfaces.VxWorks.IO (i-vxwoio.ads)
12359 @section @code{Interfaces.VxWorks.IO} (@file{i-vxwoio.ads})
12360 @cindex @code{Interfaces.VxWorks.IO} (@file{i-vxwoio.ads})
12361 @cindex Interfacing to VxWorks' I/O
12362 @cindex VxWorks, I/O interfacing
12363 @cindex VxWorks, Get_Immediate
12364 @cindex Get_Immediate, VxWorks
12367 This package provides a binding to the ioctl (IO/Control)
12368 function of VxWorks, defining a set of option values and
12369 function codes. A particular use of this package is
12370 to enable the use of Get_Immediate under VxWorks.
12372 @node System.Address_Image (s-addima.ads)
12373 @section @code{System.Address_Image} (@file{s-addima.ads})
12374 @cindex @code{System.Address_Image} (@file{s-addima.ads})
12375 @cindex Address image
12376 @cindex Image, of an address
12379 This function provides a useful debugging
12380 function that gives an (implementation dependent)
12381 string which identifies an address.
12383 @node System.Assertions (s-assert.ads)
12384 @section @code{System.Assertions} (@file{s-assert.ads})
12385 @cindex @code{System.Assertions} (@file{s-assert.ads})
12387 @cindex Assert_Failure, exception
12390 This package provides the declaration of the exception raised
12391 by an run-time assertion failure, as well as the routine that
12392 is used internally to raise this assertion.
12394 @node System.Memory (s-memory.ads)
12395 @section @code{System.Memory} (@file{s-memory.ads})
12396 @cindex @code{System.Memory} (@file{s-memory.ads})
12397 @cindex Memory allocation
12400 This package provides the interface to the low level routines used
12401 by the generated code for allocation and freeing storage for the
12402 default storage pool (analogous to the C routines malloc and free.
12403 It also provides a reallocation interface analogous to the C routine
12404 realloc. The body of this unit may be modified to provide alternative
12405 allocation mechanisms for the default pool, and in addition, direct
12406 calls to this unit may be made for low level allocation uses (for
12407 example see the body of @code{GNAT.Tables}).
12409 @node System.Partition_Interface (s-parint.ads)
12410 @section @code{System.Partition_Interface} (@file{s-parint.ads})
12411 @cindex @code{System.Partition_Interface} (@file{s-parint.ads})
12412 @cindex Partition intefacing functions
12415 This package provides facilities for partition interfacing. It
12416 is used primarily in a distribution context when using Annex E
12419 @node System.Restrictions (s-restri.ads)
12420 @section @code{System.Restrictions} (@file{s-restri.ads})
12421 @cindex @code{System.Restrictions} (@file{s-restri.ads})
12422 @cindex Run-time restrictions access
12425 This package provides facilities for accessing at run-time
12426 the status of restrictions specified at compile time for
12427 the partition. Information is available both with regard
12428 to actual restrictions specified, and with regard to
12429 compiler determined information on which restrictions
12430 are violated by one or more packages in the partition.
12432 @node System.Rident (s-rident.ads)
12433 @section @code{System.Rident} (@file{s-rident.ads})
12434 @cindex @code{System.Rident} (@file{s-rident.ads})
12435 @cindex Restrictions definitions
12438 This package provides definitions of the restrictions
12439 identifiers supported by GNAT, and also the format of
12440 the restrictions provided in package System.Restrictions.
12441 It is not normally necessary to @code{with} this generic package
12442 since the necessary instantiation is included in
12443 package System.Restrictions.
12445 @node System.Task_Info (s-tasinf.ads)
12446 @section @code{System.Task_Info} (@file{s-tasinf.ads})
12447 @cindex @code{System.Task_Info} (@file{s-tasinf.ads})
12448 @cindex Task_Info pragma
12451 This package provides target dependent functionality that is used
12452 to support the @code{Task_Info} pragma
12454 @node System.Wch_Cnv (s-wchcnv.ads)
12455 @section @code{System.Wch_Cnv} (@file{s-wchcnv.ads})
12456 @cindex @code{System.Wch_Cnv} (@file{s-wchcnv.ads})
12457 @cindex Wide Character, Representation
12458 @cindex Wide String, Conversion
12459 @cindex Representation of wide characters
12462 This package provides routines for converting between
12463 wide characters and a representation as a value of type
12464 @code{Standard.String}, using a specified wide character
12465 encoding method. It uses definitions in
12466 package @code{System.Wch_Con}.
12468 @node System.Wch_Con (s-wchcon.ads)
12469 @section @code{System.Wch_Con} (@file{s-wchcon.ads})
12470 @cindex @code{System.Wch_Con} (@file{s-wchcon.ads})
12473 This package provides definitions and descriptions of
12474 the various methods used for encoding wide characters
12475 in ordinary strings. These definitions are used by
12476 the package @code{System.Wch_Cnv}.
12478 @node Interfacing to Other Languages
12479 @chapter Interfacing to Other Languages
12481 The facilities in annex B of the Ada 95 Reference Manual are fully
12482 implemented in GNAT, and in addition, a full interface to C++ is
12486 * Interfacing to C::
12487 * Interfacing to C++::
12488 * Interfacing to COBOL::
12489 * Interfacing to Fortran::
12490 * Interfacing to non-GNAT Ada code::
12493 @node Interfacing to C
12494 @section Interfacing to C
12497 Interfacing to C with GNAT can use one of two approaches:
12501 The types in the package @code{Interfaces.C} may be used.
12503 Standard Ada types may be used directly. This may be less portable to
12504 other compilers, but will work on all GNAT compilers, which guarantee
12505 correspondence between the C and Ada types.
12509 Pragma @code{Convention C} may be applied to Ada types, but mostly has no
12510 effect, since this is the default. The following table shows the
12511 correspondence between Ada scalar types and the corresponding C types.
12516 @item Short_Integer
12518 @item Short_Short_Integer
12522 @item Long_Long_Integer
12530 @item Long_Long_Float
12531 This is the longest floating-point type supported by the hardware.
12535 Additionally, there are the following general correspondences between Ada
12539 Ada enumeration types map to C enumeration types directly if pragma
12540 @code{Convention C} is specified, which causes them to have int
12541 length. Without pragma @code{Convention C}, Ada enumeration types map to
12542 8, 16, or 32 bits (i.e.@: C types @code{signed char}, @code{short},
12543 @code{int}, respectively) depending on the number of values passed.
12544 This is the only case in which pragma @code{Convention C} affects the
12545 representation of an Ada type.
12548 Ada access types map to C pointers, except for the case of pointers to
12549 unconstrained types in Ada, which have no direct C equivalent.
12552 Ada arrays map directly to C arrays.
12555 Ada records map directly to C structures.
12558 Packed Ada records map to C structures where all members are bit fields
12559 of the length corresponding to the @code{@var{type}'Size} value in Ada.
12562 @node Interfacing to C++
12563 @section Interfacing to C++
12566 The interface to C++ makes use of the following pragmas, which are
12567 primarily intended to be constructed automatically using a binding generator
12568 tool, although it is possible to construct them by hand. Ada Core
12569 Technologies does not currently supply a suitable binding generator tool.
12571 Using these pragmas it is possible to achieve complete
12572 inter-operability between Ada tagged types and C class definitions.
12573 See @ref{Implementation Defined Pragmas}, for more details.
12576 @item pragma CPP_Class ([Entity =>] @var{local_name})
12577 The argument denotes an entity in the current declarative region that is
12578 declared as a tagged or untagged record type. It indicates that the type
12579 corresponds to an externally declared C++ class type, and is to be laid
12580 out the same way that C++ would lay out the type.
12582 @item pragma CPP_Constructor ([Entity =>] @var{local_name})
12583 This pragma identifies an imported function (imported in the usual way
12584 with pragma @code{Import}) as corresponding to a C++ constructor.
12586 @item pragma CPP_Vtable @dots{}
12587 One @code{CPP_Vtable} pragma can be present for each component of type
12588 @code{CPP.Interfaces.Vtable_Ptr} in a record to which pragma @code{CPP_Class}
12592 @node Interfacing to COBOL
12593 @section Interfacing to COBOL
12596 Interfacing to COBOL is achieved as described in section B.4 of
12597 the Ada 95 reference manual.
12599 @node Interfacing to Fortran
12600 @section Interfacing to Fortran
12603 Interfacing to Fortran is achieved as described in section B.5 of the
12604 reference manual. The pragma @code{Convention Fortran}, applied to a
12605 multi-dimensional array causes the array to be stored in column-major
12606 order as required for convenient interface to Fortran.
12608 @node Interfacing to non-GNAT Ada code
12609 @section Interfacing to non-GNAT Ada code
12611 It is possible to specify the convention @code{Ada} in a pragma
12612 @code{Import} or pragma @code{Export}. However this refers to
12613 the calling conventions used by GNAT, which may or may not be
12614 similar enough to those used by some other Ada 83 or Ada 95
12615 compiler to allow interoperation.
12617 If arguments types are kept simple, and if the foreign compiler generally
12618 follows system calling conventions, then it may be possible to integrate
12619 files compiled by other Ada compilers, provided that the elaboration
12620 issues are adequately addressed (for example by eliminating the
12621 need for any load time elaboration).
12623 In particular, GNAT running on VMS is designed to
12624 be highly compatible with the DEC Ada 83 compiler, so this is one
12625 case in which it is possible to import foreign units of this type,
12626 provided that the data items passed are restricted to simple scalar
12627 values or simple record types without variants, or simple array
12628 types with fixed bounds.
12630 @node Specialized Needs Annexes
12631 @chapter Specialized Needs Annexes
12634 Ada 95 defines a number of specialized needs annexes, which are not
12635 required in all implementations. However, as described in this chapter,
12636 GNAT implements all of these special needs annexes:
12639 @item Systems Programming (Annex C)
12640 The Systems Programming Annex is fully implemented.
12642 @item Real-Time Systems (Annex D)
12643 The Real-Time Systems Annex is fully implemented.
12645 @item Distributed Systems (Annex E)
12646 Stub generation is fully implemented in the GNAT compiler. In addition,
12647 a complete compatible PCS is available as part of the GLADE system,
12648 a separate product. When the two
12649 products are used in conjunction, this annex is fully implemented.
12651 @item Information Systems (Annex F)
12652 The Information Systems annex is fully implemented.
12654 @item Numerics (Annex G)
12655 The Numerics Annex is fully implemented.
12657 @item Safety and Security (Annex H)
12658 The Safety and Security annex is fully implemented.
12661 @node Implementation of Specific Ada Features
12662 @chapter Implementation of Specific Ada Features
12665 This chapter describes the GNAT implementation of several Ada language
12669 * Machine Code Insertions::
12670 * GNAT Implementation of Tasking::
12671 * GNAT Implementation of Shared Passive Packages::
12672 * Code Generation for Array Aggregates::
12675 @node Machine Code Insertions
12676 @section Machine Code Insertions
12679 Package @code{Machine_Code} provides machine code support as described
12680 in the Ada 95 Reference Manual in two separate forms:
12683 Machine code statements, consisting of qualified expressions that
12684 fit the requirements of RM section 13.8.
12686 An intrinsic callable procedure, providing an alternative mechanism of
12687 including machine instructions in a subprogram.
12691 The two features are similar, and both are closely related to the mechanism
12692 provided by the asm instruction in the GNU C compiler. Full understanding
12693 and use of the facilities in this package requires understanding the asm
12694 instruction as described in @cite{Using the GNU Compiler Collection (GCC)}
12695 by Richard Stallman. The relevant section is titled ``Extensions to the C
12696 Language Family'' -> ``Assembler Instructions with C Expression Operands''.
12698 Calls to the function @code{Asm} and the procedure @code{Asm} have identical
12699 semantic restrictions and effects as described below. Both are provided so
12700 that the procedure call can be used as a statement, and the function call
12701 can be used to form a code_statement.
12703 The first example given in the GCC documentation is the C @code{asm}
12706 asm ("fsinx %1 %0" : "=f" (result) : "f" (angle));
12710 The equivalent can be written for GNAT as:
12712 @smallexample @c ada
12713 Asm ("fsinx %1 %0",
12714 My_Float'Asm_Output ("=f", result),
12715 My_Float'Asm_Input ("f", angle));
12719 The first argument to @code{Asm} is the assembler template, and is
12720 identical to what is used in GNU C@. This string must be a static
12721 expression. The second argument is the output operand list. It is
12722 either a single @code{Asm_Output} attribute reference, or a list of such
12723 references enclosed in parentheses (technically an array aggregate of
12726 The @code{Asm_Output} attribute denotes a function that takes two
12727 parameters. The first is a string, the second is the name of a variable
12728 of the type designated by the attribute prefix. The first (string)
12729 argument is required to be a static expression and designates the
12730 constraint for the parameter (e.g.@: what kind of register is
12731 required). The second argument is the variable to be updated with the
12732 result. The possible values for constraint are the same as those used in
12733 the RTL, and are dependent on the configuration file used to build the
12734 GCC back end. If there are no output operands, then this argument may
12735 either be omitted, or explicitly given as @code{No_Output_Operands}.
12737 The second argument of @code{@var{my_float}'Asm_Output} functions as
12738 though it were an @code{out} parameter, which is a little curious, but
12739 all names have the form of expressions, so there is no syntactic
12740 irregularity, even though normally functions would not be permitted
12741 @code{out} parameters. The third argument is the list of input
12742 operands. It is either a single @code{Asm_Input} attribute reference, or
12743 a list of such references enclosed in parentheses (technically an array
12744 aggregate of such references).
12746 The @code{Asm_Input} attribute denotes a function that takes two
12747 parameters. The first is a string, the second is an expression of the
12748 type designated by the prefix. The first (string) argument is required
12749 to be a static expression, and is the constraint for the parameter,
12750 (e.g.@: what kind of register is required). The second argument is the
12751 value to be used as the input argument. The possible values for the
12752 constant are the same as those used in the RTL, and are dependent on
12753 the configuration file used to built the GCC back end.
12755 If there are no input operands, this argument may either be omitted, or
12756 explicitly given as @code{No_Input_Operands}. The fourth argument, not
12757 present in the above example, is a list of register names, called the
12758 @dfn{clobber} argument. This argument, if given, must be a static string
12759 expression, and is a space or comma separated list of names of registers
12760 that must be considered destroyed as a result of the @code{Asm} call. If
12761 this argument is the null string (the default value), then the code
12762 generator assumes that no additional registers are destroyed.
12764 The fifth argument, not present in the above example, called the
12765 @dfn{volatile} argument, is by default @code{False}. It can be set to
12766 the literal value @code{True} to indicate to the code generator that all
12767 optimizations with respect to the instruction specified should be
12768 suppressed, and that in particular, for an instruction that has outputs,
12769 the instruction will still be generated, even if none of the outputs are
12770 used. See the full description in the GCC manual for further details.
12772 The @code{Asm} subprograms may be used in two ways. First the procedure
12773 forms can be used anywhere a procedure call would be valid, and
12774 correspond to what the RM calls ``intrinsic'' routines. Such calls can
12775 be used to intersperse machine instructions with other Ada statements.
12776 Second, the function forms, which return a dummy value of the limited
12777 private type @code{Asm_Insn}, can be used in code statements, and indeed
12778 this is the only context where such calls are allowed. Code statements
12779 appear as aggregates of the form:
12781 @smallexample @c ada
12782 Asm_Insn'(Asm (@dots{}));
12783 Asm_Insn'(Asm_Volatile (@dots{}));
12787 In accordance with RM rules, such code statements are allowed only
12788 within subprograms whose entire body consists of such statements. It is
12789 not permissible to intermix such statements with other Ada statements.
12791 Typically the form using intrinsic procedure calls is more convenient
12792 and more flexible. The code statement form is provided to meet the RM
12793 suggestion that such a facility should be made available. The following
12794 is the exact syntax of the call to @code{Asm}. As usual, if named notation
12795 is used, the arguments may be given in arbitrary order, following the
12796 normal rules for use of positional and named arguments)
12800 [Template =>] static_string_EXPRESSION
12801 [,[Outputs =>] OUTPUT_OPERAND_LIST ]
12802 [,[Inputs =>] INPUT_OPERAND_LIST ]
12803 [,[Clobber =>] static_string_EXPRESSION ]
12804 [,[Volatile =>] static_boolean_EXPRESSION] )
12806 OUTPUT_OPERAND_LIST ::=
12807 [PREFIX.]No_Output_Operands
12808 | OUTPUT_OPERAND_ATTRIBUTE
12809 | (OUTPUT_OPERAND_ATTRIBUTE @{,OUTPUT_OPERAND_ATTRIBUTE@})
12811 OUTPUT_OPERAND_ATTRIBUTE ::=
12812 SUBTYPE_MARK'Asm_Output (static_string_EXPRESSION, NAME)
12814 INPUT_OPERAND_LIST ::=
12815 [PREFIX.]No_Input_Operands
12816 | INPUT_OPERAND_ATTRIBUTE
12817 | (INPUT_OPERAND_ATTRIBUTE @{,INPUT_OPERAND_ATTRIBUTE@})
12819 INPUT_OPERAND_ATTRIBUTE ::=
12820 SUBTYPE_MARK'Asm_Input (static_string_EXPRESSION, EXPRESSION)
12824 The identifiers @code{No_Input_Operands} and @code{No_Output_Operands}
12825 are declared in the package @code{Machine_Code} and must be referenced
12826 according to normal visibility rules. In particular if there is no
12827 @code{use} clause for this package, then appropriate package name
12828 qualification is required.
12830 @node GNAT Implementation of Tasking
12831 @section GNAT Implementation of Tasking
12834 This chapter outlines the basic GNAT approach to tasking (in particular,
12835 a multi-layered library for portability) and discusses issues related
12836 to compliance with the Real-Time Systems Annex.
12839 * Mapping Ada Tasks onto the Underlying Kernel Threads::
12840 * Ensuring Compliance with the Real-Time Annex::
12843 @node Mapping Ada Tasks onto the Underlying Kernel Threads
12844 @subsection Mapping Ada Tasks onto the Underlying Kernel Threads
12847 GNAT's run-time support comprises two layers:
12850 @item GNARL (GNAT Run-time Layer)
12851 @item GNULL (GNAT Low-level Library)
12855 In GNAT, Ada's tasking services rely on a platform and OS independent
12856 layer known as GNARL@. This code is responsible for implementing the
12857 correct semantics of Ada's task creation, rendezvous, protected
12860 GNARL decomposes Ada's tasking semantics into simpler lower level
12861 operations such as create a thread, set the priority of a thread,
12862 yield, create a lock, lock/unlock, etc. The spec for these low-level
12863 operations constitutes GNULLI, the GNULL Interface. This interface is
12864 directly inspired from the POSIX real-time API@.
12866 If the underlying executive or OS implements the POSIX standard
12867 faithfully, the GNULL Interface maps as is to the services offered by
12868 the underlying kernel. Otherwise, some target dependent glue code maps
12869 the services offered by the underlying kernel to the semantics expected
12872 Whatever the underlying OS (VxWorks, UNIX, OS/2, Windows NT, etc.) the
12873 key point is that each Ada task is mapped on a thread in the underlying
12874 kernel. For example, in the case of VxWorks, one Ada task = one VxWorks task.
12876 In addition Ada task priorities map onto the underlying thread priorities.
12877 Mapping Ada tasks onto the underlying kernel threads has several advantages:
12881 The underlying scheduler is used to schedule the Ada tasks. This
12882 makes Ada tasks as efficient as kernel threads from a scheduling
12886 Interaction with code written in C containing threads is eased
12887 since at the lowest level Ada tasks and C threads map onto the same
12888 underlying kernel concept.
12891 When an Ada task is blocked during I/O the remaining Ada tasks are
12895 On multiprocessor systems Ada tasks can execute in parallel.
12899 Some threads libraries offer a mechanism to fork a new process, with the
12900 child process duplicating the threads from the parent.
12902 support this functionality when the parent contains more than one task.
12903 @cindex Forking a new process
12905 @node Ensuring Compliance with the Real-Time Annex
12906 @subsection Ensuring Compliance with the Real-Time Annex
12907 @cindex Real-Time Systems Annex compliance
12910 Although mapping Ada tasks onto
12911 the underlying threads has significant advantages, it does create some
12912 complications when it comes to respecting the scheduling semantics
12913 specified in the real-time annex (Annex D).
12915 For instance the Annex D requirement for the @code{FIFO_Within_Priorities}
12916 scheduling policy states:
12919 @emph{When the active priority of a ready task that is not running
12920 changes, or the setting of its base priority takes effect, the
12921 task is removed from the ready queue for its old active priority
12922 and is added at the tail of the ready queue for its new active
12923 priority, except in the case where the active priority is lowered
12924 due to the loss of inherited priority, in which case the task is
12925 added at the head of the ready queue for its new active priority.}
12929 While most kernels do put tasks at the end of the priority queue when
12930 a task changes its priority, (which respects the main
12931 FIFO_Within_Priorities requirement), almost none keep a thread at the
12932 beginning of its priority queue when its priority drops from the loss
12933 of inherited priority.
12935 As a result most vendors have provided incomplete Annex D implementations.
12937 The GNAT run-time, has a nice cooperative solution to this problem
12938 which ensures that accurate FIFO_Within_Priorities semantics are
12941 The principle is as follows. When an Ada task T is about to start
12942 running, it checks whether some other Ada task R with the same
12943 priority as T has been suspended due to the loss of priority
12944 inheritance. If this is the case, T yields and is placed at the end of
12945 its priority queue. When R arrives at the front of the queue it
12948 Note that this simple scheme preserves the relative order of the tasks
12949 that were ready to execute in the priority queue where R has been
12952 @node GNAT Implementation of Shared Passive Packages
12953 @section GNAT Implementation of Shared Passive Packages
12954 @cindex Shared passive packages
12957 GNAT fully implements the pragma @code{Shared_Passive} for
12958 @cindex pragma @code{Shared_Passive}
12959 the purpose of designating shared passive packages.
12960 This allows the use of passive partitions in the
12961 context described in the Ada Reference Manual; i.e. for communication
12962 between separate partitions of a distributed application using the
12963 features in Annex E.
12965 @cindex Distribution Systems Annex
12967 However, the implementation approach used by GNAT provides for more
12968 extensive usage as follows:
12971 @item Communication between separate programs
12973 This allows separate programs to access the data in passive
12974 partitions, using protected objects for synchronization where
12975 needed. The only requirement is that the two programs have a
12976 common shared file system. It is even possible for programs
12977 running on different machines with different architectures
12978 (e.g. different endianness) to communicate via the data in
12979 a passive partition.
12981 @item Persistence between program runs
12983 The data in a passive package can persist from one run of a
12984 program to another, so that a later program sees the final
12985 values stored by a previous run of the same program.
12990 The implementation approach used is to store the data in files. A
12991 separate stream file is created for each object in the package, and
12992 an access to an object causes the corresponding file to be read or
12995 The environment variable @code{SHARED_MEMORY_DIRECTORY} should be
12996 @cindex @code{SHARED_MEMORY_DIRECTORY} environment variable
12997 set to the directory to be used for these files.
12998 The files in this directory
12999 have names that correspond to their fully qualified names. For
13000 example, if we have the package
13002 @smallexample @c ada
13004 pragma Shared_Passive (X);
13011 and the environment variable is set to @code{/stemp/}, then the files created
13012 will have the names:
13020 These files are created when a value is initially written to the object, and
13021 the files are retained until manually deleted. This provides the persistence
13022 semantics. If no file exists, it means that no partition has assigned a value
13023 to the variable; in this case the initial value declared in the package
13024 will be used. This model ensures that there are no issues in synchronizing
13025 the elaboration process, since elaboration of passive packages elaborates the
13026 initial values, but does not create the files.
13028 The files are written using normal @code{Stream_IO} access.
13029 If you want to be able
13030 to communicate between programs or partitions running on different
13031 architectures, then you should use the XDR versions of the stream attribute
13032 routines, since these are architecture independent.
13034 If active synchronization is required for access to the variables in the
13035 shared passive package, then as described in the Ada Reference Manual, the
13036 package may contain protected objects used for this purpose. In this case
13037 a lock file (whose name is @file{___lock} (three underscores)
13038 is created in the shared memory directory.
13039 @cindex @file{___lock} file (for shared passive packages)
13040 This is used to provide the required locking
13041 semantics for proper protected object synchronization.
13043 As of January 2003, GNAT supports shared passive packages on all platforms
13044 except for OpenVMS.
13046 @node Code Generation for Array Aggregates
13047 @section Code Generation for Array Aggregates
13050 * Static constant aggregates with static bounds::
13051 * Constant aggregates with an unconstrained nominal types::
13052 * Aggregates with static bounds::
13053 * Aggregates with non-static bounds::
13054 * Aggregates in assignment statements::
13058 Aggregate have a rich syntax and allow the user to specify the values of
13059 complex data structures by means of a single construct. As a result, the
13060 code generated for aggregates can be quite complex and involve loops, case
13061 statements and multiple assignments. In the simplest cases, however, the
13062 compiler will recognize aggregates whose components and constraints are
13063 fully static, and in those cases the compiler will generate little or no
13064 executable code. The following is an outline of the code that GNAT generates
13065 for various aggregate constructs. For further details, the user will find it
13066 useful to examine the output produced by the -gnatG flag to see the expanded
13067 source that is input to the code generator. The user will also want to examine
13068 the assembly code generated at various levels of optimization.
13070 The code generated for aggregates depends on the context, the component values,
13071 and the type. In the context of an object declaration the code generated is
13072 generally simpler than in the case of an assignment. As a general rule, static
13073 component values and static subtypes also lead to simpler code.
13075 @node Static constant aggregates with static bounds
13076 @subsection Static constant aggregates with static bounds
13079 For the declarations:
13080 @smallexample @c ada
13081 type One_Dim is array (1..10) of integer;
13082 ar0 : constant One_Dim := ( 1, 2, 3, 4, 5, 6, 7, 8, 9, 0);
13086 GNAT generates no executable code: the constant ar0 is placed in static memory.
13087 The same is true for constant aggregates with named associations:
13089 @smallexample @c ada
13090 Cr1 : constant One_Dim := (4 => 16, 2 => 4, 3 => 9, 1=> 1);
13091 Cr3 : constant One_Dim := (others => 7777);
13095 The same is true for multidimensional constant arrays such as:
13097 @smallexample @c ada
13098 type two_dim is array (1..3, 1..3) of integer;
13099 Unit : constant two_dim := ( (1,0,0), (0,1,0), (0,0,1));
13103 The same is true for arrays of one-dimensional arrays: the following are
13106 @smallexample @c ada
13107 type ar1b is array (1..3) of boolean;
13108 type ar_ar is array (1..3) of ar1b;
13109 None : constant ar1b := (others => false); -- fully static
13110 None2 : constant ar_ar := (1..3 => None); -- fully static
13114 However, for multidimensional aggregates with named associations, GNAT will
13115 generate assignments and loops, even if all associations are static. The
13116 following two declarations generate a loop for the first dimension, and
13117 individual component assignments for the second dimension:
13119 @smallexample @c ada
13120 Zero1: constant two_dim := (1..3 => (1..3 => 0));
13121 Zero2: constant two_dim := (others => (others => 0));
13124 @node Constant aggregates with an unconstrained nominal types
13125 @subsection Constant aggregates with an unconstrained nominal types
13128 In such cases the aggregate itself establishes the subtype, so that
13129 associations with @code{others} cannot be used. GNAT determines the
13130 bounds for the actual subtype of the aggregate, and allocates the
13131 aggregate statically as well. No code is generated for the following:
13133 @smallexample @c ada
13134 type One_Unc is array (natural range <>) of integer;
13135 Cr_Unc : constant One_Unc := (12,24,36);
13138 @node Aggregates with static bounds
13139 @subsection Aggregates with static bounds
13142 In all previous examples the aggregate was the initial (and immutable) value
13143 of a constant. If the aggregate initializes a variable, then code is generated
13144 for it as a combination of individual assignments and loops over the target
13145 object. The declarations
13147 @smallexample @c ada
13148 Cr_Var1 : One_Dim := (2, 5, 7, 11);
13149 Cr_Var2 : One_Dim := (others > -1);
13153 generate the equivalent of
13155 @smallexample @c ada
13161 for I in Cr_Var2'range loop
13162 Cr_Var2 (I) := =-1;
13166 @node Aggregates with non-static bounds
13167 @subsection Aggregates with non-static bounds
13170 If the bounds of the aggregate are not statically compatible with the bounds
13171 of the nominal subtype of the target, then constraint checks have to be
13172 generated on the bounds. For a multidimensional array, constraint checks may
13173 have to be applied to sub-arrays individually, if they do not have statically
13174 compatible subtypes.
13176 @node Aggregates in assignment statements
13177 @subsection Aggregates in assignment statements
13180 In general, aggregate assignment requires the construction of a temporary,
13181 and a copy from the temporary to the target of the assignment. This is because
13182 it is not always possible to convert the assignment into a series of individual
13183 component assignments. For example, consider the simple case:
13185 @smallexample @c ada
13190 This cannot be converted into:
13192 @smallexample @c ada
13198 So the aggregate has to be built first in a separate location, and then
13199 copied into the target. GNAT recognizes simple cases where this intermediate
13200 step is not required, and the assignments can be performed in place, directly
13201 into the target. The following sufficient criteria are applied:
13205 The bounds of the aggregate are static, and the associations are static.
13207 The components of the aggregate are static constants, names of
13208 simple variables that are not renamings, or expressions not involving
13209 indexed components whose operands obey these rules.
13213 If any of these conditions are violated, the aggregate will be built in
13214 a temporary (created either by the front-end or the code generator) and then
13215 that temporary will be copied onto the target.
13217 @node Project File Reference
13218 @chapter Project File Reference
13221 This chapter describes the syntax and semantics of project files.
13222 Project files specify the options to be used when building a system.
13223 Project files can specify global settings for all tools,
13224 as well as tool-specific settings.
13225 See the chapter on project files in the GNAT Users guide for examples of use.
13229 * Lexical Elements::
13231 * Typed string declarations::
13235 * Project Attributes::
13236 * Attribute References::
13237 * External Values::
13238 * Case Construction::
13240 * Package Renamings::
13242 * Project Extensions::
13243 * Project File Elaboration::
13246 @node Reserved Words
13247 @section Reserved Words
13250 All Ada95 reserved words are reserved in project files, and cannot be used
13251 as variable names or project names. In addition, the following are
13252 also reserved in project files:
13255 @item @code{extends}
13257 @item @code{external}
13259 @item @code{project}
13263 @node Lexical Elements
13264 @section Lexical Elements
13267 Rules for identifiers are the same as in Ada95. Identifiers
13268 are case-insensitive. Strings are case sensitive, except where noted.
13269 Comments have the same form as in Ada95.
13279 simple_name @{. simple_name@}
13283 @section Declarations
13286 Declarations introduce new entities that denote types, variables, attributes,
13287 and packages. Some declarations can only appear immediately within a project
13288 declaration. Others can appear within a project or within a package.
13292 declarative_item ::=
13293 simple_declarative_item |
13294 typed_string_declaration |
13295 package_declaration
13297 simple_declarative_item ::=
13298 variable_declaration |
13299 typed_variable_declaration |
13300 attribute_declaration |
13304 @node Typed string declarations
13305 @section Typed string declarations
13308 Typed strings are sequences of string literals. Typed strings are the only
13309 named types in project files. They are used in case constructions, where they
13310 provide support for conditional attribute definitions.
13314 typed_string_declaration ::=
13315 @b{type} <typed_string_>_simple_name @b{is}
13316 ( string_literal @{, string_literal@} );
13320 A typed string declaration can only appear immediately within a project
13323 All the string literals in a typed string declaration must be distinct.
13329 Variables denote values, and appear as constituents of expressions.
13332 typed_variable_declaration ::=
13333 <typed_variable_>simple_name : <typed_string_>name := string_expression ;
13335 variable_declaration ::=
13336 <variable_>simple_name := expression;
13340 The elaboration of a variable declaration introduces the variable and
13341 assigns to it the value of the expression. The name of the variable is
13342 available after the assignment symbol.
13345 A typed_variable can only be declare once.
13348 a non typed variable can be declared multiple times.
13351 Before the completion of its first declaration, the value of variable
13352 is the null string.
13355 @section Expressions
13358 An expression is a formula that defines a computation or retrieval of a value.
13359 In a project file the value of an expression is either a string or a list
13360 of strings. A string value in an expression is either a literal, the current
13361 value of a variable, an external value, an attribute reference, or a
13362 concatenation operation.
13375 attribute_reference
13381 ( <string_>expression @{ , <string_>expression @} )
13384 @subsection Concatenation
13386 The following concatenation functions are defined:
13388 @smallexample @c ada
13389 function "&" (X : String; Y : String) return String;
13390 function "&" (X : String_List; Y : String) return String_List;
13391 function "&" (X : String_List; Y : String_List) return String_List;
13395 @section Attributes
13398 An attribute declaration defines a property of a project or package. This
13399 property can later be queried by means of an attribute reference.
13400 Attribute values are strings or string lists.
13402 Some attributes are associative arrays. These attributes are mappings whose
13403 domain is a set of strings. These attributes are declared one association
13404 at a time, by specifying a point in the domain and the corresponding image
13405 of the attribute. They may also be declared as a full associative array,
13406 getting the same associations as the corresponding attribute in an imported
13407 or extended project.
13409 Attributes that are not associative arrays are called simple attributes.
13413 attribute_declaration ::=
13414 full_associative_array_declaration |
13415 @b{for} attribute_designator @b{use} expression ;
13417 full_associative_array_declaration ::=
13418 @b{for} <associative_array_attribute_>simple_name @b{use}
13419 <project_>simple_name [ . <package_>simple_Name ] ' <attribute_>simple_name ;
13421 attribute_designator ::=
13422 <simple_attribute_>simple_name |
13423 <associative_array_attribute_>simple_name ( string_literal )
13427 Some attributes are project-specific, and can only appear immediately within
13428 a project declaration. Others are package-specific, and can only appear within
13429 the proper package.
13431 The expression in an attribute definition must be a string or a string_list.
13432 The string literal appearing in the attribute_designator of an associative
13433 array attribute is case-insensitive.
13435 @node Project Attributes
13436 @section Project Attributes
13439 The following attributes apply to a project. All of them are simple
13444 Expression must be a path name. The attribute defines the
13445 directory in which the object files created by the build are to be placed. If
13446 not specified, object files are placed in the project directory.
13449 Expression must be a path name. The attribute defines the
13450 directory in which the executables created by the build are to be placed.
13451 If not specified, executables are placed in the object directory.
13454 Expression must be a list of path names. The attribute
13455 defines the directories in which the source files for the project are to be
13456 found. If not specified, source files are found in the project directory.
13459 Expression must be a list of file names. The attribute
13460 defines the individual files, in the project directory, which are to be used
13461 as sources for the project. File names are path_names that contain no directory
13462 information. If the project has no sources the attribute must be declared
13463 explicitly with an empty list.
13465 @item Source_List_File
13466 Expression must a single path name. The attribute
13467 defines a text file that contains a list of source file names to be used
13468 as sources for the project
13471 Expression must be a path name. The attribute defines the
13472 directory in which a library is to be built. The directory must exist, must
13473 be distinct from the project's object directory, and must be writable.
13476 Expression must be a string that is a legal file name,
13477 without extension. The attribute defines a string that is used to generate
13478 the name of the library to be built by the project.
13481 Argument must be a string value that must be one of the
13482 following @code{"static"}, @code{"dynamic"} or @code{"relocatable"}. This
13483 string is case-insensitive. If this attribute is not specified, the library is
13484 a static library. Otherwise, the library may be dynamic or relocatable. This
13485 distinction is operating-system dependent.
13487 @item Library_Version
13488 Expression must be a string value whose interpretation
13489 is platform dependent. On UNIX, it is used only for dynamic/relocatable
13490 libraries as the internal name of the library (the @code{"soname"}). If the
13491 library file name (built from the @code{Library_Name}) is different from the
13492 @code{Library_Version}, then the library file will be a symbolic link to the
13493 actual file whose name will be @code{Library_Version}.
13495 @item Library_Interface
13496 Expression must be a string list. Each element of the string list
13497 must designate a unit of the project.
13498 If this attribute is present in a Library Project File, then the project
13499 file is a Stand-alone Library_Project_File.
13501 @item Library_Auto_Init
13502 Expression must be a single string "true" or "false", case-insensitive.
13503 If this attribute is present in a Stand-alone Library Project File,
13504 it indicates if initialization is automatic when the dynamic library
13507 @item Library_Options
13508 Expression must be a string list. Indicates additional switches that
13509 are to be used when building a shared library.
13512 Expression must be a single string. Designates an alternative to "gcc"
13513 for building shared libraries.
13515 @item Library_Src_Dir
13516 Expression must be a path name. The attribute defines the
13517 directory in which the sources of the interfaces of a Stand-alone Library will
13518 be copied. The directory must exist, must be distinct from the project's
13519 object directory and source directories, and must be writable.
13522 Expression must be a list of strings that are legal file names.
13523 These file names designate existing compilation units in the source directory
13524 that are legal main subprograms.
13526 When a project file is elaborated, as part of the execution of a gnatmake
13527 command, one or several executables are built and placed in the Exec_Dir.
13528 If the gnatmake command does not include explicit file names, the executables
13529 that are built correspond to the files specified by this attribute.
13531 @item Main_Language
13532 This is a simple attribute. Its value is a string that specifies the
13533 language of the main program.
13536 Expression must be a string list. Each string designates
13537 a programming language that is known to GNAT. The strings are case-insensitive.
13539 @item Locally_Removed_Files
13540 This attribute is legal only in a project file that extends another.
13541 Expression must be a list of strings that are legal file names.
13542 Each file name must designate a source that would normally be inherited
13543 by the current project file. It cannot designate an immediate source that is
13544 not inherited. Each of the source files in the list are not considered to
13545 be sources of the project file: they are not inherited.
13548 @node Attribute References
13549 @section Attribute References
13552 Attribute references are used to retrieve the value of previously defined
13553 attribute for a package or project.
13556 attribute_reference ::=
13557 attribute_prefix ' <simple_attribute_>simple_name [ ( string_literal ) ]
13559 attribute_prefix ::=
13561 <project_simple_name | package_identifier |
13562 <project_>simple_name . package_identifier
13566 If an attribute has not been specified for a given package or project, its
13567 value is the null string or the empty list.
13569 @node External Values
13570 @section External Values
13573 An external value is an expression whose value is obtained from the command
13574 that invoked the processing of the current project file (typically a
13580 @b{external} ( string_literal [, string_literal] )
13584 The first string_literal is the string to be used on the command line or
13585 in the environment to specify the external value. The second string_literal,
13586 if present, is the default to use if there is no specification for this
13587 external value either on the command line or in the environment.
13589 @node Case Construction
13590 @section Case Construction
13593 A case construction supports attribute declarations that depend on the value of
13594 a previously declared variable.
13598 case_construction ::=
13599 @b{case} <typed_variable_>name @b{is}
13604 @b{when} discrete_choice_list =>
13605 @{case_construction | attribute_declaration@}
13607 discrete_choice_list ::=
13608 string_literal @{| string_literal@} |
13613 All choices in a choice list must be distinct. The choice lists of two
13614 distinct alternatives must be disjoint. Unlike Ada, the choice lists of all
13615 alternatives do not need to include all values of the type. An @code{others}
13616 choice must appear last in the list of alternatives.
13622 A package provides a grouping of variable declarations and attribute
13623 declarations to be used when invoking various GNAT tools. The name of
13624 the package indicates the tool(s) to which it applies.
13628 package_declaration ::=
13629 package_specification | package_renaming
13631 package_specification ::=
13632 @b{package} package_identifier @b{is}
13633 @{simple_declarative_item@}
13634 @b{end} package_identifier ;
13636 package_identifier ::=
13637 @code{Naming} | @code{Builder} | @code{Compiler} | @code{Binder} |
13638 @code{Linker} | @code{Finder} | @code{Cross_Reference} |
13639 @code{gnatls} | @code{IDE} | @code{Pretty_Printer}
13642 @subsection Package Naming
13645 The attributes of a @code{Naming} package specifies the naming conventions
13646 that apply to the source files in a project. When invoking other GNAT tools,
13647 they will use the sources in the source directories that satisfy these
13648 naming conventions.
13650 The following attributes apply to a @code{Naming} package:
13654 This is a simple attribute whose value is a string. Legal values of this
13655 string are @code{"lowercase"}, @code{"uppercase"} or @code{"mixedcase"}.
13656 These strings are themselves case insensitive.
13659 If @code{Casing} is not specified, then the default is @code{"lowercase"}.
13661 @item Dot_Replacement
13662 This is a simple attribute whose string value satisfies the following
13666 @item It must not be empty
13667 @item It cannot start or end with an alphanumeric character
13668 @item It cannot be a single underscore
13669 @item It cannot start with an underscore followed by an alphanumeric
13670 @item It cannot contain a dot @code{'.'} if longer than one character
13674 If @code{Dot_Replacement} is not specified, then the default is @code{"-"}.
13677 This is an associative array attribute, defined on language names,
13678 whose image is a string that must satisfy the following
13682 @item It must not be empty
13683 @item It cannot start with an alphanumeric character
13684 @item It cannot start with an underscore followed by an alphanumeric character
13688 For Ada, the attribute denotes the suffix used in file names that contain
13689 library unit declarations, that is to say units that are package and
13690 subprogram declarations. If @code{Spec_Suffix ("Ada")} is not
13691 specified, then the default is @code{".ads"}.
13693 For C and C++, the attribute denotes the suffix used in file names that
13694 contain prototypes.
13697 This is an associative array attribute defined on language names,
13698 whose image is a string that must satisfy the following
13702 @item It must not be empty
13703 @item It cannot start with an alphanumeric character
13704 @item It cannot start with an underscore followed by an alphanumeric character
13705 @item It cannot be a suffix of @code{Spec_Suffix}
13709 For Ada, the attribute denotes the suffix used in file names that contain
13710 library bodies, that is to say units that are package and subprogram bodies.
13711 If @code{Body_Suffix ("Ada")} is not specified, then the default is
13714 For C and C++, the attribute denotes the suffix used in file names that contain
13717 @item Separate_Suffix
13718 This is a simple attribute whose value satisfies the same conditions as
13719 @code{Body_Suffix}.
13721 This attribute is specific to Ada. It denotes the suffix used in file names
13722 that contain separate bodies. If it is not specified, then it defaults to same
13723 value as @code{Body_Suffix ("Ada")}.
13726 This is an associative array attribute, specific to Ada, defined over
13727 compilation unit names. The image is a string that is the name of the file
13728 that contains that library unit. The file name is case sensitive if the
13729 conventions of the host operating system require it.
13732 This is an associative array attribute, specific to Ada, defined over
13733 compilation unit names. The image is a string that is the name of the file
13734 that contains the library unit body for the named unit. The file name is case
13735 sensitive if the conventions of the host operating system require it.
13737 @item Specification_Exceptions
13738 This is an associative array attribute defined on language names,
13739 whose value is a list of strings.
13741 This attribute is not significant for Ada.
13743 For C and C++, each string in the list denotes the name of a file that
13744 contains prototypes, but whose suffix is not necessarily the
13745 @code{Spec_Suffix} for the language.
13747 @item Implementation_Exceptions
13748 This is an associative array attribute defined on language names,
13749 whose value is a list of strings.
13751 This attribute is not significant for Ada.
13753 For C and C++, each string in the list denotes the name of a file that
13754 contains source code, but whose suffix is not necessarily the
13755 @code{Body_Suffix} for the language.
13758 The following attributes of package @code{Naming} are obsolescent. They are
13759 kept as synonyms of other attributes for compatibility with previous versions
13760 of the Project Manager.
13763 @item Specification_Suffix
13764 This is a synonym of @code{Spec_Suffix}.
13766 @item Implementation_Suffix
13767 This is a synonym of @code{Body_Suffix}.
13769 @item Specification
13770 This is a synonym of @code{Spec}.
13772 @item Implementation
13773 This is a synonym of @code{Body}.
13776 @subsection package Compiler
13779 The attributes of the @code{Compiler} package specify the compilation options
13780 to be used by the underlying compiler.
13783 @item Default_Switches
13784 This is an associative array attribute. Its
13785 domain is a set of language names. Its range is a string list that
13786 specifies the compilation options to be used when compiling a component
13787 written in that language, for which no file-specific switches have been
13791 This is an associative array attribute. Its domain is
13792 a set of file names. Its range is a string list that specifies the
13793 compilation options to be used when compiling the named file. If a file
13794 is not specified in the Switches attribute, it is compiled with the
13795 settings specified by Default_Switches.
13797 @item Local_Configuration_Pragmas.
13798 This is a simple attribute, whose
13799 value is a path name that designates a file containing configuration pragmas
13800 to be used for all invocations of the compiler for immediate sources of the
13804 This is an associative array attribute. Its domain is
13805 a set of main source file names. Its range is a simple string that specifies
13806 the executable file name to be used when linking the specified main source.
13807 If a main source is not specified in the Executable attribute, the executable
13808 file name is deducted from the main source file name.
13811 @subsection package Builder
13814 The attributes of package @code{Builder} specify the compilation, binding, and
13815 linking options to be used when building an executable for a project. The
13816 following attributes apply to package @code{Builder}:
13819 @item Default_Switches
13825 @item Global_Configuration_Pragmas
13826 This is a simple attribute, whose
13827 value is a path name that designates a file that contains configuration pragmas
13828 to be used in every build of an executable. If both local and global
13829 configuration pragmas are specified, a compilation makes use of both sets.
13832 This is an associative array attribute, defined over
13833 compilation unit names. The image is a string that is the name of the
13834 executable file corresponding to the main source file index.
13835 This attribute has no effect if its value is the empty string.
13837 @item Executable_Suffix
13838 This is a simple attribute whose value is a suffix to be added to
13839 the executables that don't have an attribute Executable specified.
13842 @subsection package Gnatls
13845 The attributes of package @code{Gnatls} specify the tool options to be used
13846 when invoking the library browser @command{gnatls}.
13847 The following attributes apply to package @code{Gnatls}:
13854 @subsection package Binder
13857 The attributes of package @code{Binder} specify the options to be used
13858 when invoking the binder in the construction of an executable.
13859 The following attributes apply to package @code{Binder}:
13862 @item Default_Switches
13868 @subsection package Linker
13871 The attributes of package @code{Linker} specify the options to be used when
13872 invoking the linker in the construction of an executable.
13873 The following attributes apply to package @code{Linker}:
13876 @item Default_Switches
13882 @subsection package Cross_Reference
13885 The attributes of package @code{Cross_Reference} specify the tool options
13887 when invoking the library tool @command{gnatxref}.
13888 The following attributes apply to package @code{Cross_Reference}:
13891 @item Default_Switches
13897 @subsection package Finder
13900 The attributes of package @code{Finder} specify the tool options to be used
13901 when invoking the search tool @command{gnatfind}.
13902 The following attributes apply to package @code{Finder}:
13905 @item Default_Switches
13911 @subsection package Pretty_Printer
13914 The attributes of package @code{Pretty_Printer}
13915 specify the tool options to be used
13916 when invoking the formatting tool @command{gnatpp}.
13917 The following attributes apply to package @code{Pretty_Printer}:
13920 @item Default_switches
13926 @subsection package IDE
13929 The attributes of package @code{IDE} specify the options to be used when using
13930 an Integrated Development Environment such as @command{GPS}.
13934 This is a simple attribute. Its value is a string that designates the remote
13935 host in a cross-compilation environment, to be used for remote compilation and
13936 debugging. This field should not be specified when running on the local
13940 This is a simple attribute. Its value is a string that specifies the
13941 name of IP address of the embedded target in a cross-compilation environment,
13942 on which the program should execute.
13944 @item Communication_Protocol
13945 This is a simple string attribute. Its value is the name of the protocol
13946 to use to communicate with the target in a cross-compilation environment,
13947 e.g. @code{"wtx"} or @code{"vxworks"}.
13949 @item Compiler_Command
13950 This is an associative array attribute, whose domain is a language name. Its
13951 value is string that denotes the command to be used to invoke the compiler.
13952 The value of @code{Compiler_Command ("Ada")} is expected to be compatible with
13953 gnatmake, in particular in the handling of switches.
13955 @item Debugger_Command
13956 This is simple attribute, Its value is a string that specifies the name of
13957 the debugger to be used, such as gdb, powerpc-wrs-vxworks-gdb or gdb-4.
13959 @item Default_Switches
13960 This is an associative array attribute. Its indexes are the name of the
13961 external tools that the GNAT Programming System (GPS) is supporting. Its
13962 value is a list of switches to use when invoking that tool.
13965 This is a simple attribute. Its value is a string that specifies the name
13966 of the @command{gnatls} utility to be used to retrieve information about the
13967 predefined path; e.g., @code{"gnatls"}, @code{"powerpc-wrs-vxworks-gnatls"}.
13970 This is a simple atribute. Is value is a string used to specify the
13971 Version Control System (VCS) to be used for this project, e.g CVS, RCS
13972 ClearCase or Perforce.
13974 @item VCS_File_Check
13975 This is a simple attribute. Its value is a string that specifies the
13976 command used by the VCS to check the validity of a file, either
13977 when the user explicitly asks for a check, or as a sanity check before
13978 doing the check-in.
13980 @item VCS_Log_Check
13981 This is a simple attribute. Its value is a string that specifies
13982 the command used by the VCS to check the validity of a log file.
13986 @node Package Renamings
13987 @section Package Renamings
13990 A package can be defined by a renaming declaration. The new package renames
13991 a package declared in a different project file, and has the same attributes
13992 as the package it renames.
13995 package_renaming ::==
13996 @b{package} package_identifier @b{renames}
13997 <project_>simple_name.package_identifier ;
14001 The package_identifier of the renamed package must be the same as the
14002 package_identifier. The project whose name is the prefix of the renamed
14003 package must contain a package declaration with this name. This project
14004 must appear in the context_clause of the enclosing project declaration,
14005 or be the parent project of the enclosing child project.
14011 A project file specifies a set of rules for constructing a software system.
14012 A project file can be self-contained, or depend on other project files.
14013 Dependencies are expressed through a context clause that names other projects.
14019 context_clause project_declaration
14021 project_declaration ::=
14022 simple_project_declaration | project_extension
14024 simple_project_declaration ::=
14025 @b{project} <project_>simple_name @b{is}
14026 @{declarative_item@}
14027 @b{end} <project_>simple_name;
14033 [@b{limited}] @b{with} path_name @{ , path_name @} ;
14040 A path name denotes a project file. A path name can be absolute or relative.
14041 An absolute path name includes a sequence of directories, in the syntax of
14042 the host operating system, that identifies uniquely the project file in the
14043 file system. A relative path name identifies the project file, relative
14044 to the directory that contains the current project, or relative to a
14045 directory listed in the environment variable ADA_PROJECT_PATH.
14046 Path names are case sensitive if file names in the host operating system
14047 are case sensitive.
14049 The syntax of the environment variable ADA_PROJECT_PATH is a list of
14050 directory names separated by colons (semicolons on Windows).
14052 A given project name can appear only once in a context_clause.
14054 It is illegal for a project imported by a context clause to refer, directly
14055 or indirectly, to the project in which this context clause appears (the
14056 dependency graph cannot contain cycles), except when one of the with_clause
14057 in the cycle is a @code{limited with}.
14059 @node Project Extensions
14060 @section Project Extensions
14063 A project extension introduces a new project, which inherits the declarations
14064 of another project.
14068 project_extension ::=
14069 @b{project} <project_>simple_name @b{extends} path_name @b{is}
14070 @{declarative_item@}
14071 @b{end} <project_>simple_name;
14075 The project extension declares a child project. The child project inherits
14076 all the declarations and all the files of the parent project, These inherited
14077 declaration can be overridden in the child project, by means of suitable
14080 @node Project File Elaboration
14081 @section Project File Elaboration
14084 A project file is processed as part of the invocation of a gnat tool that
14085 uses the project option. Elaboration of the process file consists in the
14086 sequential elaboration of all its declarations. The computed values of
14087 attributes and variables in the project are then used to establish the
14088 environment in which the gnat tool will execute.
14091 @c GNU Free Documentation License
14093 @node Index,,GNU Free Documentation License, Top