| blob | b2cac71ce9f4fa4402826af7263aa0b37a19a2d0 |
1 ------------------------------------------------------------------------------
2 -- --
3 -- GNAT COMPILER COMPONENTS --
4 -- --
5 -- S E M _ R E S --
6 -- --
7 -- B o d y --
8 -- --
9 -- Copyright (C) 1992-2018, Free Software Foundation, Inc. --
10 -- --
11 -- GNAT is free software; you can redistribute it and/or modify it under --
12 -- terms of the GNU General Public License as published by the Free Soft- --
13 -- ware Foundation; either version 3, or (at your option) any later ver- --
14 -- sion. GNAT is distributed in the hope that it will be useful, but WITH- --
15 -- OUT ANY WARRANTY; without even the implied warranty of MERCHANTABILITY --
16 -- or FITNESS FOR A PARTICULAR PURPOSE. See the GNU General Public License --
17 -- for more details. You should have received a copy of the GNU General --
18 -- Public License distributed with GNAT; see file COPYING3. If not, go to --
19 -- http://www.gnu.org/licenses for a complete copy of the license. --
20 -- --
21 -- GNAT was originally developed by the GNAT team at New York University. --
22 -- Extensive contributions were provided by Ada Core Technologies Inc. --
23 -- --
24 ------------------------------------------------------------------------------
86 -----------------------
87 -- Local Subprograms --
88 -----------------------
90 -- Second pass (top-down) type checking and overload resolution procedures
91 -- Typ is the type required by context. These procedures propagate the
92 -- type information recursively to the descendants of N. If the node is not
93 -- overloaded, its Etype is established in the first pass. If overloaded,
94 -- the Resolve routines set the correct type. For arithmetic operators, the
95 -- Etype is the base type of the context.
97 -- Note that Resolve_Attribute is separated off in Sem_Attr
100 -- Enforce the restrictions on the use of discriminants when constraining
101 -- a component of a discriminated type (record or concurrent type).
104 -- Given a node for an operator associated with type T, check that the
105 -- operator is visible. Operators all of whose operands are universal must
106 -- be checked for visibility during resolution because their type is not
107 -- determinable based on their operands.
109 procedure Check_Fully_Declared_Prefix
112 -- Check that the type of the prefix of a dereference is not incomplete
115 -- Given a call node, N, which is known to occur immediately within the
116 -- subprogram being called, determines whether it is a detectable case of
117 -- an infinite recursion, and if so, outputs appropriate messages. Returns
118 -- True if an infinite recursion is detected, and False otherwise.
121 -- N is the node for a logical operator. If the operator is predefined, and
122 -- the root type of the operands is Standard.Boolean, then a check is made
123 -- for restriction No_Direct_Boolean_Operators. This procedure also handles
124 -- the style check for Style_Check_Boolean_And_Or.
127 -- N is either an indexed component or a selected component. This function
128 -- returns true if the prefix refers to an object that has an address
129 -- clause (the case in which we may want to issue a warning).
132 -- Determine whether E is an access type declared by an access declaration,
133 -- and not an (anonymous) allocator type.
136 -- Utility to check whether the entity for an operator is a predefined
137 -- operator, in which case the expression is left as an operator in the
138 -- tree (else it is rewritten into a call). An instance of an intrinsic
139 -- conversion operation may be given an operator name, but is not treated
140 -- like an operator. Note that an operator that is an imported back-end
141 -- builtin has convention Intrinsic, but is expected to be rewritten into
142 -- a call, so such an operator is not treated as predefined by this
143 -- predicate.
146 -- If a default expression in entry call N depends on the discriminants
147 -- of the task, it must be replaced with a reference to the discriminant
148 -- of the task being called.
150 procedure Resolve_Op_Concat_Arg
155 -- Internal procedure for Resolve_Op_Concat to resolve one operand of
156 -- concatenation operator. The operand is either of the array type or of
157 -- the component type. If the operand is an aggregate, and the component
158 -- type is composite, this is ambiguous if component type has aggregates.
161 -- Does the first part of the work of Resolve_Op_Concat
164 -- Does the "rest" of the work of Resolve_Op_Concat, after the left operand
165 -- has been resolved. See Resolve_Op_Concat for details.
204 function Operator_Kind
207 -- Utility to map the name of an operator into the corresponding Node. Used
208 -- by other node rewriting procedures.
211 -- Resolve actuals of call, and add default expressions for missing ones.
212 -- N is the Node_Id for the subprogram call, and Nam is the entity of the
213 -- called subprogram.
216 -- Called from Resolve_Call, when the prefix denotes an entry or element
217 -- of entry family. Actuals are resolved as for subprograms, and the node
218 -- is rebuilt as an entry call. Also called for protected operations. Typ
219 -- is the context type, which is used when the operation is a protected
220 -- function with no arguments, and the return value is indexed.
223 -- A call to a user-defined intrinsic operator is rewritten as a call to
224 -- the corresponding predefined operator, with suitable conversions. Note
225 -- that this applies only for intrinsic operators that denote predefined
226 -- operators, not ones that are intrinsic imports of back-end builtins.
229 -- Ditto, for arithmetic unary operators
232 -- If an operator node resolves to a call to a user-defined operator,
233 -- rewrite the node as a function call.
235 procedure Make_Call_Into_Operator
239 -- Inverse transformation: if an operator is given in functional notation,
240 -- then after resolving the node, transform into an operator node, so that
241 -- operands are resolved properly. Recall that predefined operators do not
242 -- have a full signature and special resolution rules apply.
244 procedure Rewrite_Renamed_Operator
248 -- An operator can rename another, e.g. in an instantiation. In that
249 -- case, the proper operator node must be constructed and resolved.
252 -- The String_Literal_Subtype is built for all strings that are not
253 -- operands of a static concatenation operation. If the argument is not
254 -- a N_String_Literal node, then the call has no effect.
257 -- Build subtype of array type, with the range specified by the slice
260 -- Called after N has been resolved and evaluated, but before range checks
261 -- have been applied. Currently simplifies a combination of floating-point
262 -- to integer conversion and Rounding or Truncation attribute.
265 -- A universal_fixed expression in an universal context is unambiguous if
266 -- there is only one applicable fixed point type. Determining whether there
267 -- is only one requires a search over all visible entities, and happens
268 -- only in very pathological cases (see 6115-006).
270 -------------------------
271 -- Ambiguous_Character --
272 -------------------------
277 begin
281 -- First the ones in Standard
286 -- Include Wide_Wide_Character in Ada 2005 mode
292 -- Now any other types that match
302 -------------------------
303 -- Analyze_And_Resolve --
304 -------------------------
307 begin
313 begin
318 -- Versions with check(s) suppressed
320 procedure Analyze_And_Resolve
324 is
327 begin
329 declare
331 begin
337 else
338 declare
340 begin
348 and then Scope_Is_Transient
349 then
350 -- This can only happen if a transient scope was created for an inner
351 -- expression, which will be removed upon completion of the analysis
352 -- of an enclosing construct. The transient scope must have the
353 -- suppress status of the enclosing environment, not of this Analyze
354 -- call.
357 Scope_Suppress;
361 procedure Analyze_And_Resolve
364 is
367 begin
369 declare
371 begin
377 else
378 declare
380 begin
389 Scope_Suppress;
393 ----------------------------
394 -- Check_Discriminant_Use --
395 ----------------------------
403 begin
404 -- Any use in a spec-expression is legal
411 -- Discriminant cannot be used to constrain a scalar type
418 then
423 -- The following check catches the unusual case where a
424 -- discriminant appears within an index constraint that is part
425 -- of a larger expression within a constraint on a component,
426 -- e.g. "C : Int range 1 .. F (new A(1 .. D))". For now we only
427 -- check case of record components, and note that a similar check
428 -- should also apply in the case of discriminant constraints
429 -- below. ???
431 -- Note that the check for N_Subtype_Declaration below is to
432 -- detect the valid use of discriminants in the constraints of a
433 -- subtype declaration when this subtype declaration appears
434 -- inside the scope of a record type (which is syntactically
435 -- illegal, but which may be created as part of derived type
436 -- processing for records). See Sem_Ch3.Build_Derived_Record_Type
437 -- for more info.
441 and then not
443 and then
445 N_Subtype_Declaration)
447 then
448 Error_Msg_N
453 -- Detect a common error:
455 -- type R (D : Positive := 100) is record
456 -- Name : String (1 .. D);
457 -- end record;
459 -- The default value causes an object of type R to be allocated
460 -- with room for Positive'Last characters. The RM does not mandate
461 -- the allocation of the maximum size, but that is what GNAT does
462 -- so we should warn the programmer that there is a problem.
471 -- Return True if type T has a large enough range that any
472 -- array whose index type covered the whole range of the type
473 -- would likely raise Storage_Error.
475 ------------------------
476 -- Large_Storage_Type --
477 ------------------------
480 begin
481 -- The type is considered large if its bounds are known at
482 -- compile time and if it requires at least as many bits as
483 -- a Positive to store the possible values.
487 and then
492 -- Start of processing for Check_Large
494 begin
495 -- Check that the Disc has a large range
501 -- If the enclosing type is limited, we allocate only the
502 -- default value, not the maximum, and there is no need for
503 -- a warning.
509 -- Check that it is the high bound
513 then
517 -- Check the array allows a large range at this bound. First
518 -- find the array
532 -- Next, find the dimension
540 loop
549 -- Now, check the dimension has a large range
555 -- Warn about the danger
557 Error_Msg_N
567 -- Legal case is in index or discriminant constraint
570 N_Discriminant_Association)
571 then
573 Error_Msg_N
578 then
579 Error_Msg_N
585 -- Otherwise, context is an expression. It should not be within (i.e. a
586 -- subexpression of) a constraint for a component.
588 else
592 N_Subtype_Indication,
593 N_Entry_Declaration)
594 loop
600 -- If the discriminant is used in an expression that is a bound of a
601 -- scalar type, an Itype is created and the bounds are attached to
602 -- its range, not to the original subtype indication. Such use is of
603 -- course a double fault.
607 N_Derived_Type_Definition)
610 -- The constraint itself may be given by a subtype indication,
611 -- rather than by a more common discrete range.
614 and then
618 then
619 Error_Msg_N
625 --------------------------------
626 -- Check_For_Visible_Operator --
627 --------------------------------
630 begin
632 Error_Msg_NE -- CODEFIX
634 Error_Msg_N -- CODEFIX
639 ----------------------------------
640 -- Check_Fully_Declared_Prefix --
641 ----------------------------------
643 procedure Check_Fully_Declared_Prefix
646 is
647 begin
648 -- Check that the designated type of the prefix of a dereference is
649 -- not an incomplete type. This cannot be done unconditionally, because
650 -- dereferences of private types are legal in default expressions. This
651 -- case is taken care of in Check_Fully_Declared, called below. There
652 -- are also 2005 cases where it is legal for the prefix to be unfrozen.
654 -- This consideration also applies to similar checks for allocators,
655 -- qualified expressions, and type conversions.
657 -- An additional exception concerns other per-object expressions that
658 -- are not directly related to component declarations, in particular
659 -- representation pragmas for tasks. These will be per-object
660 -- expressions if they depend on discriminants or some global entity.
661 -- If the task has access discriminants, the designated type may be
662 -- incomplete at the point the expression is resolved. This resolution
663 -- takes place within the body of the initialization procedure, where
664 -- the discriminant is replaced by its discriminal.
668 then
671 -- Ada 2005 (AI-326): Tagged incomplete types allowed. The wrong usages
672 -- are handled by Analyze_Access_Attribute, Analyze_Assignment,
673 -- Analyze_Object_Renaming, and Freeze_Entity.
679 E_Incomplete_Type
681 then
683 else
688 ------------------------------
689 -- Check_Infinite_Recursion --
690 ------------------------------
697 -- Check whether list of actuals is identical to list of formals of
698 -- called function (which is also the enclosing scope).
700 ------------------------
701 -- Same_Argument_List --
702 ------------------------
709 begin
712 else
730 -- Start of processing for Check_Infinite_Recursion
732 begin
733 -- Special case, if this is a procedure call and is a call to the
734 -- current procedure with the same argument list, then this is for
735 -- sure an infinite recursion and we insert a call to raise SE.
739 and then Same_Argument_List
740 then
741 declare
743 begin
747 then
759 -- If not that special case, search up tree, quitting if we reach a
760 -- construct (e.g. a conditional) that tells us that this is not a
761 -- case for an infinite recursion warning.
764 loop
767 -- If no parent, then we were not inside a subprogram, this can for
768 -- example happen when processing certain pragmas in a spec. Just
769 -- return False in this case.
775 -- Done if we get to subprogram body, this is definitely an infinite
776 -- recursion case if we did not find anything to stop us.
780 -- If appearing in conditional, result is false
783 N_And_Then,
784 N_Case_Expression,
785 N_Case_Statement,
786 N_If_Expression,
787 N_If_Statement)
788 then
793 then
794 -- If the call is the expression of a return statement and the
795 -- actuals are identical to the formals, it's worth a warning.
796 -- However, we skip this if there is an immediately preceding
797 -- raise statement, since the call is never executed.
799 -- Furthermore, this corresponds to a common idiom:
801 -- function F (L : Thing) return Boolean is
802 -- begin
803 -- raise Program_Error;
804 -- return F (L);
805 -- end F;
807 -- for generating a stub function
810 and then Same_Argument_List
811 then
814 -- OK, return statement is in a statement list, look for raise
816 declare
819 begin
820 -- Skip past N_Freeze_Entity nodes generated by expansion
825 loop
829 -- If no raise statement, give warning. We look at the
830 -- original node, because in the case of "raise ... with
831 -- ...", the node has been transformed into a call.
834 and then
842 else
854 ---------------------------------------
855 -- Check_No_Direct_Boolean_Operators --
856 ---------------------------------------
859 begin
862 then
863 -- Restriction only applies to original source code
870 -- Do style check (but skip if in instance, error is on template)
879 ------------------------------
880 -- Check_Parameterless_Call --
881 ------------------------------
887 -- If the prefix is of an access_to_subprogram type, the node must be
888 -- rewritten as a call. Ditto if the prefix is overloaded and all its
889 -- interpretations are access to subprograms.
891 ---------------------------
892 -- Prefix_Is_Access_Subp --
893 ---------------------------
899 begin
900 -- If the context is an attribute reference that can apply to
901 -- functions, this is never a parameterless call (RM 4.1.4(6)).
905 Name_Code_Address,
906 Name_Access)
907 then
912 return
915 else
920 then
931 -- Start of processing for Check_Parameterless_Call
933 begin
934 -- Defend against junk stuff if errors already detected
941 then
948 -- If the context expects a value, and the name is a procedure, this is
949 -- most likely a missing 'Access. Don't try to resolve the parameterless
950 -- call, error will be caught when the outer call is analyzed.
955 and then
957 N_Function_Call,
958 N_Procedure_Call_Statement)
959 then
963 -- Rewrite as call if overloadable entity that is (or could be, in the
964 -- overloaded case) a function call. If we know for sure that the entity
965 -- is an enumeration literal, we do not rewrite it.
967 -- If the entity is the name of an operator, it cannot be a call because
968 -- operators cannot have default parameters. In this case, this must be
969 -- a string whose contents coincide with an operator name. Set the kind
970 -- of the node appropriately.
978 -- Rewrite as call if it is an explicit dereference of an expression of
979 -- a subprogram access type, and the subprogram type is not that of a
980 -- procedure or entry.
982 or else
985 -- Rewrite as call if it is a selected component which is a function,
986 -- this is the case of a call to a protected function (which may be
987 -- overloaded with other protected operations).
989 or else
992 or else
994 E_Procedure)
997 -- If one of the above three conditions is met, rewrite as call. Apply
998 -- the rewriting only once.
1000 then
1003 then
1005 -- This may be a prefixed call that was not fully analyzed, e.g.
1006 -- an actual in an instance.
1011 then
1019 -- The node is the name of the parameterless call. Preserve its
1020 -- descendants, which may be complex expressions.
1024 -- If overloaded, overload set belongs to new copy
1028 -- Change node to parameterless function call (note that the
1029 -- Parameter_Associations associations field is left set to Empty,
1030 -- its normal default value since there are no parameters)
1048 --------------------------------
1049 -- Is_Atomic_Ref_With_Address --
1050 --------------------------------
1055 begin
1059 else
1060 declare
1063 begin
1071 -----------------------------
1072 -- Is_Definite_Access_Type --
1073 -----------------------------
1077 begin
1083 ----------------------
1084 -- Is_Predefined_Op --
1085 ----------------------
1088 begin
1089 -- Predefined operators are intrinsic subprograms
1095 -- A call to a back-end builtin is never a predefined operator
1106 -----------------------------
1107 -- Make_Call_Into_Operator --
1108 -----------------------------
1110 procedure Make_Call_Into_Operator
1114 is
1129 -- If the operand is not universal, and the operator is given by an
1130 -- expanded name, verify that the operand has an interpretation with a
1131 -- type defined in the given scope of the operator.
1134 -- Find a type of the given class in package Pack that contains the
1135 -- operator.
1137 ---------------------------
1138 -- Operand_Type_In_Scope --
1139 ---------------------------
1146 begin
1150 else
1164 ---------------
1165 -- Type_In_P --
1166 ---------------
1172 -- Verify that node is not part of the type declaration for the
1173 -- candidate type, which would otherwise be invisible.
1175 -------------
1176 -- In_Decl --
1177 -------------
1183 begin
1192 else
1195 loop
1198 else
1207 -- Start of processing for Type_In_P
1209 begin
1210 -- If the context type is declared in the prefix package, this is the
1211 -- desired base type.
1216 else
1230 -- Start of processing for Make_Call_Into_Operator
1232 begin
1235 -- Ensure that the corresponding operator has the same parent as the
1236 -- original call. This guarantees that parent traversals performed by
1237 -- the ABE mechanism succeed.
1241 -- Binary operator
1251 -- Unary operator
1253 else
1259 -- If the operator is denoted by an expanded name, and the prefix is
1260 -- not Standard, but the operator is a predefined one whose scope is
1261 -- Standard, then this is an implicit_operator, inserted as an
1262 -- interpretation by the procedure of the same name. This procedure
1263 -- overestimates the presence of implicit operators, because it does
1264 -- not examine the type of the operands. Verify now that the operand
1265 -- type appears in the given scope. If right operand is universal,
1266 -- check the other operand. In the case of concatenation, either
1267 -- argument can be the component type, so check the type of the result.
1268 -- If both arguments are literals, look for a type of the right kind
1269 -- defined in the given scope. This elaborate nonsense is brought to
1270 -- you courtesy of b33302a. The type itself must be frozen, so we must
1271 -- find the type of the proper class in the given scope.
1273 -- A final wrinkle is the multiplication operator for fixed point types,
1274 -- which is defined in Standard only, and not in the scope of the
1275 -- fixed point type itself.
1280 -- If this is a package renaming, get renamed entity, which will be
1281 -- the scope of the operands if operaton is type-correct.
1287 -- If the entity being called is defined in the given package, it is
1288 -- a renaming of a predefined operator, and known to be legal.
1292 then
1295 -- Visibility does not need to be checked in an instance: if the
1296 -- operator was not visible in the generic it has been diagnosed
1297 -- already, else there is an implicit copy of it in the instance.
1305 then
1310 -- Ada 2005 AI-420: Predefined equality on Universal_Access is
1311 -- available.
1316 then
1319 else
1328 and then Is_Binary
1330 then
1336 -- Verify that the scope contains a type that corresponds to
1337 -- the given literal. Optimize the case where Pack is Standard.
1358 else
1367 else
1376 then
1379 -- If the type is defined elsewhere, and the operator is not
1380 -- defined in the given scope (by a renaming declaration, e.g.)
1381 -- then this is an error as well. If an extension of System is
1382 -- present, and the type may be defined there, Pack must be
1383 -- System itself.
1390 then
1393 else
1399 then
1406 Error_Msg_NE
1411 -- Detect a mismatch between the context type and the result type
1412 -- in the named package, which is otherwise not detected if the
1413 -- operands are universal. Check is only needed if source entity is
1414 -- an operator, not a function that renames an operator.
1421 and then not In_Instance
1422 then
1425 then
1426 -- Already checked above
1430 -- Operator may be defined in an extension of System
1435 then
1438 else
1439 -- Could we use Wrong_Type here??? (this would require setting
1440 -- Etype (N) to the actual type found where Typ was expected).
1451 else
1455 -- If this is a call to a function that renames a predefined equality,
1456 -- the renaming declaration provides a type that must be used to
1457 -- resolve the operands. This must be done now because resolution of
1458 -- the equality node will not resolve any remaining ambiguity, and it
1459 -- assumes that the first operand is not overloaded.
1464 then
1472 -- Do rewrite setting Comes_From_Source on the result if the original
1473 -- call came from source. Although it is not strictly the case that the
1474 -- operator as such comes from the source, logically it corresponds
1475 -- exactly to the function call in the source, so it should be marked
1476 -- this way (e.g. to make sure that validity checks work fine).
1478 declare
1480 begin
1485 -- If this is an arithmetic operator and the result type is private,
1486 -- the operands and the result must be wrapped in conversion to
1487 -- expose the underlying numeric type and expand the proper checks,
1488 -- e.g. on division.
1492 when N_Op_Add
1493 | N_Op_Divide
1494 | N_Op_Expon
1495 | N_Op_Mod
1496 | N_Op_Multiply
1497 | N_Op_Rem
1498 | N_Op_Subtract
1499 =>
1502 when N_Op_Abs
1503 | N_Op_Minus
1504 | N_Op_Plus
1505 =>
1511 else
1515 -- If in ASIS_Mode, propagate operand types to original actuals of
1516 -- function call, which would otherwise not be fully resolved. If
1517 -- the call has already been constant-folded, nothing to do. We
1518 -- relocate the operand nodes rather than copy them, to preserve
1519 -- original_node pointers, given that the operands themselves may
1520 -- have been rewritten. If the call was itself a rewriting of an
1521 -- operator node, nothing to do.
1523 if ASIS_Mode
1526 then
1527 declare
1535 begin
1540 -- If the original call has named associations, replace the
1541 -- explicit actual parameter in the association with the proper
1542 -- resolved operand.
1547 then
1550 else
1555 else
1562 then
1565 else
1570 else
1574 else
1578 else
1588 -------------------
1589 -- Operator_Kind --
1590 -------------------
1592 function Operator_Kind
1595 is
1598 begin
1599 -- Use CASE statement or array???
1636 else
1640 -- Unary operators
1642 else
1651 else
1659 ----------------------------
1660 -- Preanalyze_And_Resolve --
1661 ----------------------------
1666 begin
1670 -- Normally, we suppress all checks for this preanalysis. There is no
1671 -- point in processing them now, since they will be applied properly
1672 -- and in the proper location when the default expressions reanalyzed
1673 -- and reexpanded later on. We will also have more information at that
1674 -- point for possible suppression of individual checks.
1676 -- However, in SPARK mode, most expansion is suppressed, and this
1677 -- later reanalysis and reexpansion may not occur. SPARK mode does
1678 -- require the setting of checking flags for proof purposes, so we
1679 -- do the SPARK preanalysis without suppressing checks.
1681 -- This special handling for SPARK mode is required for example in the
1682 -- case of Ada 2012 constructs such as quantified expressions, which are
1683 -- expanded in two separate steps.
1687 else
1691 Expander_Mode_Restore;
1695 -- Version without context type
1700 begin
1707 Expander_Mode_Restore;
1711 ----------------------------------
1712 -- Replace_Actual_Discriminants --
1713 ----------------------------------
1720 -- Comment needed???
1722 -------------------
1723 -- Process_Discr --
1724 -------------------
1729 begin
1735 then
1751 -- Start of processing for Replace_Actual_Discriminants
1753 begin
1757 -- Allow the replacement of concurrent discriminants in GNATprove even
1758 -- though this is a light expansion activity. Note that generic units
1759 -- are not modified.
1764 else
1780 -------------
1781 -- Resolve --
1782 -------------
1798 -- Determine whether a node comes from a predefined library unit or
1799 -- Standard.
1802 -- Try and fix up a literal so that it matches its expected type. New
1803 -- literals are manufactured if necessary to avoid cascaded errors.
1806 -- Additional diagnostics when an ambiguous call has an ambiguous
1807 -- argument (typically a controlling actual).
1810 -- Called when attempt at resolving current expression fails
1812 ------------------------------------
1813 -- Comes_From_Predefined_Lib_Unit --
1814 -------------------------------------
1817 begin
1818 return
1822 --------------------
1823 -- Patch_Up_Value --
1824 --------------------
1827 begin
1844 then
1849 Char_Literal_Value =>
1865 -------------------------------
1866 -- Report_Ambiguous_Argument --
1867 -------------------------------
1874 begin
1878 then
1881 -- Could use comments on what is going on here???
1889 else
1898 -----------------------
1899 -- Resolution_Failed --
1900 -----------------------
1903 begin
1906 -- Set the type to the desired one to minimize cascaded errors. Note
1907 -- that this is an approximation and does not work in all cases.
1914 -- The caller will return without calling the expander, so we need
1915 -- to set the analyzed flag. Note that it is fine to set Analyzed
1916 -- to True even if we are in the middle of a shallow analysis,
1917 -- (see the spec of sem for more details) since this is an error
1918 -- situation anyway, and there is no point in repeating the
1919 -- analysis later (indeed it won't work to repeat it later, since
1920 -- we haven't got a clear resolution of which entity is being
1921 -- referenced.)
1927 -- Start of processing for Resolve
1929 begin
1934 -- Access attribute on remote subprogram cannot be used for a non-remote
1935 -- access-to-subprogram type.
1939 Name_Unrestricted_Access,
1940 Name_Unchecked_Access)
1946 then
1947 Error_Msg_N
1953 -- If the context is a Remote_Access_To_Subprogram, access attributes
1954 -- must be resolved with the corresponding fat pointer. There is no need
1955 -- to check for the attribute name since the return type of an
1956 -- attribute is never a remote type.
1961 then
1962 declare
1970 begin
1971 -- Check that Typ is a remote access-to-subprogram type
1975 -- Prefix (N) must statically denote a remote subprogram
1976 -- declared in a package specification.
1980 Attr = Attribute_Unrestricted_Access
1981 then
1996 or else
1998 then
2000 Error_Msg_N
2002 N);
2005 -- If we are generating code in distributed mode, perform
2006 -- semantic checks against corresponding remote entities.
2008 if Expander_Active
2010 then
2011 Check_Subtype_Conformant
2013 Old_Id => Designated_Type
2039 then
2044 -- Return if already analyzed
2051 -- Any case of Any_Type as the Etype value means that we had a
2052 -- previous error.
2061 -- The resolution of an Expression_With_Actions is determined by
2062 -- its Expression.
2070 -- If not overloaded, then we know the type, and all that needs doing
2071 -- is to check that this type is compatible with the context.
2077 -- In the overloaded case, we must select the interpretation that
2078 -- is compatible with the context (i.e. the type passed to Resolve)
2080 else
2081 -- Loop through possible interpretations
2090 -- We are only interested in interpretations that are compatible
2091 -- with the expected type, any other interpretations are ignored.
2096 Write_Eol;
2099 else
2100 -- Skip the current interpretation if it is disabled by an
2101 -- abstract operator. This action is performed only when the
2102 -- type against which we are resolving is the same as the
2103 -- type of the interpretation.
2110 then
2118 -- First matching interpretation
2126 -- Matching interpretation that is not the first, maybe an
2127 -- error, but there are some cases where preference rules are
2128 -- used to choose between the two possibilities. These and
2129 -- some more obscure cases are handled in Disambiguate.
2131 else
2132 -- If the current statement is part of a predefined library
2133 -- unit, then all interpretations which come from user level
2134 -- packages should not be considered. Check previous and
2135 -- current one.
2143 -- Previous interpretation must be discarded
2153 -- Otherwise apply further disambiguation steps
2158 -- Disambiguation has succeeded. Skip the remaining
2159 -- interpretations.
2169 else
2170 -- Before we issue an ambiguity complaint, check for the
2171 -- case of a subprogram call where at least one of the
2172 -- arguments is Any_Type, and if so suppress the message,
2173 -- since it is a cascaded error. This can also happen for
2174 -- a generalized indexing operation.
2179 then
2180 declare
2184 begin
2200 Write_Eol;
2213 then
2218 then
2222 -- Not that special case, so issue message using the flag
2223 -- Ambiguous to control printing of the header message
2224 -- only at the start of an ambiguous set.
2229 then
2230 Error_Msg_N
2233 else
2234 Error_Msg_NE -- CODEFIX
2242 Error_Msg_N
2244 else
2245 Error_Msg_N -- CODEFIX
2251 then
2252 Report_Ambiguous_Argument;
2258 -- By default, the error message refers to the candidate
2259 -- interpretation. But if it is a predefined operator, it
2260 -- is implicitly declared at the declaration of the type
2261 -- of the operand. Recover the sloc of that declaration
2262 -- for the error message.
2268 Standard_Standard
2269 then
2274 then
2282 Standard_Standard
2283 then
2288 then
2292 -- If this is an indirect call, use the subprogram_type
2293 -- in the message, to have a meaningful location. Also
2294 -- indicate if this is an inherited operation, created
2295 -- by a type declaration.
2300 then
2302 Error_Msg_Sloc :=
2304 else
2311 then
2312 -- Special-case the message for universal_fixed
2313 -- operators, which are not declared with the type
2314 -- of the operand, but appear forever in Standard.
2318 then
2319 Error_Msg_N
2322 else
2323 Error_Msg_N
2327 elsif
2329 then
2330 Error_Msg_N
2332 else
2333 Error_Msg_N -- CODEFIX
2340 -- We have a matching interpretation, Expr_Type is the type
2341 -- from this interpretation, and Seen is the entity.
2343 -- For an operator, just set the entity name. The type will be
2344 -- set by the specific operator resolution routine.
2351 N_Character_Literal,
2352 N_Delta_Aggregate,
2353 N_If_Expression)
2354 then
2357 -- AI05-0139-2: Expression is overloaded because type has
2358 -- implicit dereference. If type matches context, no implicit
2359 -- dereference is involved. If the expression is an entity,
2360 -- generate a reference to it, as this is not done for an
2361 -- overloaded construct during analysis.
2377 then
2378 -- If the node is a general indexing, the dereference is
2379 -- is inserted when resolving the rewritten form, else
2380 -- insert it now.
2384 then
2388 -- For an explicit dereference, attribute reference, range,
2389 -- short-circuit form (which is not an operator node), or call
2390 -- with a name that is an explicit dereference, there is
2391 -- nothing to be done at this point.
2394 N_And_Then,
2395 N_Explicit_Dereference,
2396 N_Identifier,
2397 N_Indexed_Component,
2398 N_Or_Else,
2399 N_Range,
2400 N_Selected_Component,
2401 N_Slice)
2403 then
2406 -- For procedure or function calls, set the type of the name,
2407 -- and also the entity pointer for the prefix.
2411 then
2418 then
2423 -- For all other cases, just set the type of the Name
2425 else
2433 -- Move to next interpretation
2441 -- At this stage Found indicates whether or not an acceptable
2442 -- interpretation exists. If not, then we have an error, except that if
2443 -- the context is Any_Type as a result of some other error, then we
2444 -- suppress the error report.
2449 -- If type we are looking for is Void, then this is the procedure
2450 -- call case, and the error is simply that what we gave is not a
2451 -- procedure name (we think of procedure calls as expressions with
2452 -- types internally, but the user doesn't think of them this way).
2456 -- Special case message if function used as a procedure
2461 then
2462 Error_Msg_NE
2465 Error_Msg_N
2469 -- Otherwise give general message (not clear what cases this
2470 -- covers, but no harm in providing for them).
2472 else
2478 -- Otherwise we do have a subexpression with the wrong type
2480 -- Check for the case of an allocator which uses an access type
2481 -- instead of the designated type. This is a common error and we
2482 -- specialize the message, posting an error on the operand of the
2483 -- allocator, complaining that we expected the designated type of
2484 -- the allocator.
2490 then
2494 -- Check for view mismatch on Null in instances, for which the
2495 -- view-swapping mechanism has no identifier.
2501 then
2506 -- Check for an aggregate. Sometimes we can get bogus aggregates
2507 -- from misuse of parentheses, and we are about to complain about
2508 -- the aggregate without even looking inside it.
2510 -- Instead, if we have an aggregate of type Any_Composite, then
2511 -- analyze and resolve the component fields, and then only issue
2512 -- another message if we get no errors doing this (otherwise
2513 -- assume that the errors in the aggregate caused the problem).
2517 then
2518 -- Disable expansion in any case. If there is a type mismatch
2519 -- it may be fatal to try to expand the aggregate. The flag
2520 -- would otherwise be set to false when the error is posted.
2524 declare
2526 -- Check one aggregate, and set Found to True if we have a
2527 -- definite error in any of its elements
2530 -- Check one element of aggregate and set Found to True if
2531 -- we definitely have an error in the element.
2533 ----------------
2534 -- Check_Aggr --
2535 ----------------
2540 begin
2553 -- If this is a default-initialized component, then
2554 -- there is nothing to check. The box will be
2555 -- replaced by the appropriate call during late
2556 -- expansion.
2560 then
2569 ----------------
2570 -- Check_Elmt --
2571 ----------------
2574 begin
2575 -- If we have a nested aggregate, go inside it (to
2576 -- attempt a naked analyze-resolve of the aggregate can
2577 -- cause undesirable cascaded errors). Do not resolve
2578 -- expression if it needs a type from context, as for
2579 -- integer * fixed expression.
2584 else
2589 then
2599 begin
2604 -- Looks like we have a type error, but check for special case
2605 -- of Address wanted, integer found, with the configuration pragma
2606 -- Allow_Integer_Address active. If we have this case, introduce
2607 -- an unchecked conversion to allow the integer expression to be
2608 -- treated as an Address. The reverse case of integer wanted,
2609 -- Address found, is treated in an analogous manner.
2616 -- Under relaxed RM semantics silently replace occurrences of null
2617 -- by System.Address_Null.
2625 -- That special Allow_Integer_Address check did not apply, so we
2626 -- have a real type error. If an error message was issued already,
2627 -- Found got reset to True, so if it's still False, issue standard
2628 -- Wrong_Type message.
2632 declare
2635 begin
2641 -- Protected operation: retrieve operation name
2645 else
2650 Error_Msg_NE
2656 declare
2660 begin
2667 Error_Msg_NE
2673 else
2677 else
2683 Resolution_Failed;
2686 -- Test if we have more than one interpretation for the context
2689 Resolution_Failed;
2692 -- Only one intepretation
2694 else
2695 -- In Ada 2005, if we have something like "X : T := 2 + 2;", where
2696 -- the "+" on T is abstract, and the operands are of universal type,
2697 -- the above code will have (incorrectly) resolved the "+" to the
2698 -- universal one in Standard. Therefore check for this case and give
2699 -- an error. We can't do this earlier, because it would cause legal
2700 -- cases to get errors (when some other type has an abstract "+").
2706 then
2711 then
2712 Error_Msg_NE
2721 -- Here we have an acceptable interpretation for the context
2723 -- Propagate type information and normalize tree for various
2724 -- predefined operations. If the context only imposes a class of
2725 -- types, rather than a specific type, propagate the actual type
2726 -- downward.
2732 Typ = Any_Discrete
2733 then
2736 -- Any_Fixed is legal in a real context only if a specific fixed-
2737 -- point type is imposed. If Norman Cohen can be confused by this,
2738 -- it deserves a separate message.
2742 then
2749 -- A user-defined operator is transformed into a function call at
2750 -- this point, so that further processing knows that operators are
2751 -- really operators (i.e. are predefined operators). User-defined
2752 -- operators that are intrinsic are just renamings of the predefined
2753 -- ones, and need not be turned into calls either, but if they rename
2754 -- a different operator, we must transform the node accordingly.
2755 -- Instantiations of Unchecked_Conversion are intrinsic but are
2756 -- treated as functions, even if given an operator designator.
2761 then
2766 and then
2768 N_Subprogram_Renaming_Declaration
2769 then
2772 -- If the node is rewritten, it will be fully resolved in
2773 -- Rewrite_Renamed_Operator.
2782 when N_Aggregate =>
2783 Resolve_Aggregate (N, Ctx_Type);
2785 when N_Allocator =>
2786 Resolve_Allocator (N, Ctx_Type);
2788 when N_Short_Circuit =>
2789 Resolve_Short_Circuit (N, Ctx_Type);
2791 when N_Attribute_Reference =>
2792 Resolve_Attribute (N, Ctx_Type);
2794 when N_Case_Expression =>
2795 Resolve_Case_Expression (N, Ctx_Type);
2797 when N_Character_Literal =>
2798 Resolve_Character_Literal (N, Ctx_Type);
2800 when N_Delta_Aggregate =>
2801 Resolve_Delta_Aggregate (N, Ctx_Type);
2803 when N_Expanded_Name =>
2804 Resolve_Entity_Name (N, Ctx_Type);
2806 when N_Explicit_Dereference =>
2807 Resolve_Explicit_Dereference (N, Ctx_Type);
2809 when N_Expression_With_Actions =>
2810 Resolve_Expression_With_Actions (N, Ctx_Type);
2812 when N_Extension_Aggregate =>
2813 Resolve_Extension_Aggregate (N, Ctx_Type);
2815 when N_Function_Call =>
2816 Resolve_Call (N, Ctx_Type);
2818 when N_Identifier =>
2819 Resolve_Entity_Name (N, Ctx_Type);
2821 when N_If_Expression =>
2822 Resolve_If_Expression (N, Ctx_Type);
2824 when N_Indexed_Component =>
2825 Resolve_Indexed_Component (N, Ctx_Type);
2827 when N_Integer_Literal =>
2828 Resolve_Integer_Literal (N, Ctx_Type);
2830 when N_Membership_Test =>
2831 Resolve_Membership_Op (N, Ctx_Type);
2833 when N_Null =>
2834 Resolve_Null (N, Ctx_Type);
2836 when N_Op_And
2837 | N_Op_Or
2838 | N_Op_Xor
2839 =>
2840 Resolve_Logical_Op (N, Ctx_Type);
2842 when N_Op_Eq
2843 | N_Op_Ne
2844 =>
2845 Resolve_Equality_Op (N, Ctx_Type);
2847 when N_Op_Ge
2848 | N_Op_Gt
2849 | N_Op_Le
2850 | N_Op_Lt
2851 =>
2852 Resolve_Comparison_Op (N, Ctx_Type);
2854 when N_Op_Not =>
2855 Resolve_Op_Not (N, Ctx_Type);
2857 when N_Op_Add
2858 | N_Op_Divide
2859 | N_Op_Mod
2860 | N_Op_Multiply
2861 | N_Op_Rem
2862 | N_Op_Subtract
2863 =>
2864 Resolve_Arithmetic_Op (N, Ctx_Type);
2866 when N_Op_Concat =>
2867 Resolve_Op_Concat (N, Ctx_Type);
2869 when N_Op_Expon =>
2870 Resolve_Op_Expon (N, Ctx_Type);
2872 when N_Op_Abs
2873 | N_Op_Minus
2874 | N_Op_Plus
2875 =>
2876 Resolve_Unary_Op (N, Ctx_Type);
2878 when N_Op_Shift =>
2879 Resolve_Shift (N, Ctx_Type);
2881 when N_Procedure_Call_Statement =>
2882 Resolve_Call (N, Ctx_Type);
2884 when N_Operator_Symbol =>
2885 Resolve_Operator_Symbol (N, Ctx_Type);
2887 when N_Qualified_Expression =>
2888 Resolve_Qualified_Expression (N, Ctx_Type);
2890 -- Why is the following null, needs a comment ???
2892 when N_Quantified_Expression =>
2893 null;
2895 when N_Raise_Expression =>
2896 Resolve_Raise_Expression (N, Ctx_Type);
2898 when N_Raise_xxx_Error =>
2899 Set_Etype (N, Ctx_Type);
2901 when N_Range =>
2902 Resolve_Range (N, Ctx_Type);
2904 when N_Real_Literal =>
2905 Resolve_Real_Literal (N, Ctx_Type);
2907 when N_Reference =>
2908 Resolve_Reference (N, Ctx_Type);
2910 when N_Selected_Component =>
2911 Resolve_Selected_Component (N, Ctx_Type);
2913 when N_Slice =>
2914 Resolve_Slice (N, Ctx_Type);
2916 when N_String_Literal =>
2917 Resolve_String_Literal (N, Ctx_Type);
2919 when N_Target_Name =>
2920 Resolve_Target_Name (N, Ctx_Type);
2922 when N_Type_Conversion =>
2923 Resolve_Type_Conversion (N, Ctx_Type);
2925 when N_Unchecked_Expression =>
2926 Resolve_Unchecked_Expression (N, Ctx_Type);
2928 when N_Unchecked_Type_Conversion =>
2929 Resolve_Unchecked_Type_Conversion (N, Ctx_Type);
2930 end case;
2932 -- Mark relevant use-type and use-package clauses as effective using
2933 -- the original node because constant folding may have occured and
2934 -- removed references that need to be examined.
2936 if Nkind (Original_Node (N)) in N_Op then
2937 Mark_Use_Clauses (Original_Node (N));
2938 end if;
2940 -- Ada 2012 (AI05-0149): Apply an (implicit) conversion to an
2941 -- expression of an anonymous access type that occurs in the context
2942 -- of a named general access type, except when the expression is that
2943 -- of a membership test. This ensures proper legality checking in
2944 -- terms of allowed conversions (expressions that would be illegal to
2945 -- convert implicitly are allowed in membership tests).
2947 if Ada_Version >= Ada_2012
2948 and then Ekind (Ctx_Type) = E_General_Access_Type
2949 and then Ekind (Etype (N)) = E_Anonymous_Access_Type
2950 and then Nkind (Parent (N)) not in N_Membership_Test
2951 then
2952 Rewrite (N, Convert_To (Ctx_Type, Relocate_Node (N)));
2953 Analyze_And_Resolve (N, Ctx_Type);
2954 end if;
2956 -- If the subexpression was replaced by a non-subexpression, then
2957 -- all we do is to expand it. The only legitimate case we know of
2958 -- is converting procedure call statement to entry call statements,
2959 -- but there may be others, so we are making this test general.
2961 if Nkind (N) not in N_Subexpr then
2962 Debug_A_Exit ("resolving ", N, " (done)");
2963 Expand (N);
2964 return;
2965 end if;
2967 -- The expression is definitely NOT overloaded at this point, so
2968 -- we reset the Is_Overloaded flag to avoid any confusion when
2969 -- reanalyzing the node.
2971 Set_Is_Overloaded (N, False);
2973 -- Freeze expression type, entity if it is a name, and designated
2974 -- type if it is an allocator (RM 13.14(10,11,13)).
2976 -- Now that the resolution of the type of the node is complete, and
2977 -- we did not detect an error, we can expand this node. We skip the
2978 -- expand call if we are in a default expression, see section
2979 -- "Handling of Default Expressions" in Sem spec.
2981 Debug_A_Exit ("resolving ", N, " (done)");
2983 -- We unconditionally freeze the expression, even if we are in
2984 -- default expression mode (the Freeze_Expression routine tests this
2985 -- flag and only freezes static types if it is set).
2987 -- Ada 2012 (AI05-177): The declaration of an expression function
2988 -- does not cause freezing, but we never reach here in that case.
2989 -- Here we are resolving the corresponding expanded body, so we do
2990 -- need to perform normal freezing.
2992 -- As elsewhere we do not emit freeze node within a generic. We make
2993 -- an exception for entities that are expressions, only to detect
2994 -- misuses of deferred constants and preserve the output of various
2995 -- tests.
2997 if not Inside_A_Generic or else Is_Entity_Name (N) then
2998 Freeze_Expression (N);
2999 end if;
3001 -- Now we can do the expansion
3003 Expand (N);
3004 end if;
3005 end Resolve;
3007 -------------
3008 -- Resolve --
3009 -------------
3011 -- Version with check(s) suppressed
3013 procedure Resolve (N : Node_Id; Typ : Entity_Id; Suppress : Check_Id) is
3014 begin
3015 if Suppress = All_Checks then
3016 declare
3017 Sva : constant Suppress_Array := Scope_Suppress.Suppress;
3018 begin
3019 Scope_Suppress.Suppress := (others => True);
3020 Resolve (N, Typ);
3021 Scope_Suppress.Suppress := Sva;
3022 end;
3024 else
3025 declare
3026 Svg : constant Boolean := Scope_Suppress.Suppress (Suppress);
3027 begin
3028 Scope_Suppress.Suppress (Suppress) := True;
3029 Resolve (N, Typ);
3030 Scope_Suppress.Suppress (Suppress) := Svg;
3031 end;
3032 end if;
3033 end Resolve;
3035 -------------
3036 -- Resolve --
3037 -------------
3039 -- Version with implicit type
3041 procedure Resolve (N : Node_Id) is
3042 begin
3043 Resolve (N, Etype (N));
3044 end Resolve;
3046 ---------------------
3047 -- Resolve_Actuals --
3048 ---------------------
3050 procedure Resolve_Actuals (N : Node_Id; Nam : Entity_Id) is
3051 Loc : constant Source_Ptr := Sloc (N);
3052 A : Node_Id;
3053 A_Id : Entity_Id;
3054 A_Typ : Entity_Id := Empty; -- init to avoid warning
3055 F : Entity_Id;
3056 F_Typ : Entity_Id;
3057 Prev : Node_Id := Empty;
3058 Orig_A : Node_Id;
3059 Real_F : Entity_Id := Empty; -- init to avoid warning
3061 Real_Subp : Entity_Id;
3062 -- If the subprogram being called is an inherited operation for
3063 -- a formal derived type in an instance, Real_Subp is the subprogram
3064 -- that will be called. It may have different formal names than the
3065 -- operation of the formal in the generic, so after actual is resolved
3066 -- the name of the actual in a named association must carry the name
3067 -- of the actual of the subprogram being called.
3069 procedure Check_Aliased_Parameter;
3070 -- Check rules on aliased parameters and related accessibility rules
3071 -- in (RM 3.10.2 (10.2-10.4)).
3073 procedure Check_Argument_Order;
3074 -- Performs a check for the case where the actuals are all simple
3075 -- identifiers that correspond to the formal names, but in the wrong
3076 -- order, which is considered suspicious and cause for a warning.
3078 procedure Check_Prefixed_Call;
3079 -- If the original node is an overloaded call in prefix notation,
3081 -- Try_Object_Operation has already verified that there is a valid
3082 -- interpretation, but the form of the actual can only be determined
3083 -- once the primitive operation is identified.
3086 -- Emit an error concerning the illegal usage of an effectively volatile
3087 -- object in interfering context (SPARK RM 7.13(12)).
3090 -- If the actual is missing in a call, insert in the actuals list
3091 -- an instance of the default expression. The insertion is always
3092 -- a named association.
3095 -- Check whether T1 and T2, or their full views, are derived from a
3096 -- common type. Used to enforce the restrictions on array conversions
3097 -- of AI95-00246.
3100 -- Predicate to determine whether an actual that is a concatenation
3101 -- will be evaluated statically and does not need a transient scope.
3102 -- This must be determined before the actual is resolved and expanded
3103 -- because if needed the transient scope must be introduced earlier.
3105 -----------------------------
3106 -- Check_Aliased_Parameter --
3107 -----------------------------
3112 begin
3120 else
3127 -- In a generic body assume the worst for generic formals:
3128 -- they can have a constrained partial view (AI05-041).
3134 then
3137 else
3142 else
3154 then
3162 then
3163 Error_Msg_N
3169 --------------------------
3170 -- Check_Argument_Order --
3171 --------------------------
3174 begin
3175 -- Nothing to do if no parameters, or original node is neither a
3176 -- function call nor a procedure call statement (happens in the
3177 -- operator-transformed-to-function call case), or the call does
3178 -- not come from source, or this warning is off.
3180 if not Warn_On_Parameter_Order
3184 then
3188 declare
3191 begin
3192 -- Nothing to do if only one parameter
3198 -- Here if at least two arguments
3200 declare
3206 -- Set True if an out of order case is found
3208 begin
3209 -- Collect identifier names of actuals, fail if any actual is
3210 -- not a simple identifier, and record max length of name.
3216 else
3222 -- If we got this far, all actuals are identifiers and the list
3223 -- of their names is stored in the Actuals array.
3228 -- If we ran out of formals, that's odd, probably an error
3229 -- which will be detected elsewhere, but abandon the search.
3235 -- If name matches and is in order OK
3240 else
3241 -- If no match, see if it is elsewhere in list and if so
3242 -- flag potential wrong order if type is compatible.
3246 and then
3248 then
3254 -- No match
3262 -- If Formals left over, also probably an error, skip warning
3268 -- Here we give the warning if something was out of order
3271 Error_Msg_N
3278 -------------------------
3279 -- Check_Prefixed_Call --
3280 -------------------------
3289 begin
3290 -- Check whether the call is a prefixed call, with or without
3291 -- additional actuals.
3294 or else
3300 then
3303 then
3304 -- Introduce dereference on object in prefix
3306 New_A :=
3314 then
3315 -- Introduce an implicit 'Access in prefix
3318 Error_Msg_NE
3334 ---------------------------------------
3335 -- Flag_Effectively_Volatile_Objects --
3336 ---------------------------------------
3340 -- Determine whether arbitrary node N denotes an effectively volatile
3341 -- object and if it does, emit an error.
3343 -----------------
3344 -- Flag_Object --
3345 -----------------
3350 begin
3351 -- Do not consider nested function calls because they have already
3352 -- been processed during their own resolution.
3364 then
3365 Error_Msg_N
3377 -- Start of processing for Flag_Effectively_Volatile_Objects
3379 begin
3383 --------------------
3384 -- Insert_Default --
3385 --------------------
3391 begin
3392 -- Missing argument in call, nothing to insert
3397 else
3398 -- Note that we do a full New_Copy_Tree, so that any associated
3399 -- Itypes are properly copied. This may not be needed any more,
3400 -- but it does no harm as a safety measure. Defaults of a generic
3401 -- formal may be out of bounds of the corresponding actual (see
3402 -- cc1311b) and an additional check may be required.
3404 Actval :=
3405 New_Copy_Tree
3410 -- Propagate dimension information, if any.
3416 then
3422 then
3425 then
3426 -- If default is a real literal, do not introduce a
3427 -- conversion whose effect may depend on the run-time
3428 -- size of universal real.
3432 else
3443 -- Resolve aggregates with their base type, to avoid scope
3444 -- anomalies: the subtype was first built in the subprogram
3445 -- declaration, and the current call may be nested.
3449 else
3453 else
3456 -- See note above concerning aggregates
3460 then
3463 -- Resolve entities with their own type, which may differ from
3464 -- the type of a reference in a generic context (the view
3465 -- swapping mechanism did not anticipate the re-analysis of
3466 -- default values in calls).
3471 else
3476 -- If default is a tag indeterminate function call, propagate tag
3477 -- to obtain proper dispatching.
3481 then
3486 -- If the default expression raises constraint error, then just
3487 -- silently replace it with an N_Raise_Constraint_Error node, since
3488 -- we already gave the warning on the subprogram spec. If node is
3489 -- already a Raise_Constraint_Error leave as is, to prevent loops in
3490 -- the warnings removal machinery.
3494 then
3503 Assoc :=
3508 -- Case of insertion is first named actual
3512 then
3519 else
3523 else
3527 -- Case of insertion is not first named actual
3529 else
3530 Set_Next_Named_Actual
3542 -------------------
3543 -- Same_Ancestor --
3544 -------------------
3550 begin
3553 then
3559 then
3566 --------------------------
3567 -- Static_Concatenation --
3568 --------------------------
3571 begin
3578 -- Concatenation is static when both operands are static and
3579 -- the concatenation operator is a predefined one.
3582 and then
3584 and then
3589 declare
3591 begin
3594 and then
3598 else
3604 -- Start of processing for Resolve_Actuals
3606 begin
3607 Check_Argument_Order;
3611 and then In_Instance
3614 then
3616 else
3621 Check_Prefixed_Call;
3635 -- If we have an error in any actual or formal, indicated by a type
3636 -- of Any_Type, then abandon resolution attempt, and set result type
3637 -- to Any_Type. Skip this if the actual is a Raise_Expression, whose
3638 -- type is imposed from context.
3642 then
3649 -- Case where actual is present
3651 -- If the actual is an entity, generate a reference to it now. We
3652 -- do this before the actual is resolved, because a formal of some
3653 -- protected subprogram, or a task discriminant, will be rewritten
3654 -- during expansion, and the source entity reference may be lost.
3659 then
3660 -- Annotate the tree by creating a variable reference marker when
3661 -- the actual denotes a variable reference, in case the reference
3662 -- is folded or optimized away. The variable reference marker is
3663 -- automatically saved for later examination by the ABE Processing
3664 -- phase. The status of the reference is set as follows:
3666 -- status mode
3667 -- read IN, IN OUT
3668 -- write IN OUT, OUT
3670 if Needs_Variable_Reference_Marker
3673 then
3674 Build_Variable_Reference_Marker
3685 then
3692 -- RM 6.4.1(12): For an out parameter that is passed by
3693 -- copy, the formal parameter object is created, and:
3695 -- * For an access type, the formal parameter is initialized
3696 -- from the value of the actual, without checking that the
3697 -- value satisfies any constraint, any predicate, or any
3698 -- exclusion of the null value.
3700 -- * For a scalar type that has the Default_Value aspect
3701 -- specified, the formal parameter is initialized from the
3702 -- value of the actual, without checking that the value
3703 -- satisfies any constraint or any predicate.
3704 -- I do not understand why this case is included??? this is
3705 -- not a case where an OUT parameter is treated as IN OUT.
3707 -- * For a composite type with discriminants or that has
3708 -- implicit initial values for any subcomponents, the
3709 -- behavior is as for an in out parameter passed by copy.
3711 -- Hence for these cases we generate the read reference now
3712 -- (the write reference will be generated later by
3713 -- Note_Possible_Modification).
3716 and then
3718 or else
3720 and then
3722 or else
3725 or else Is_Partially_Initialized_Type
3727 then
3737 then
3738 -- If style checking mode on, check match of formal name
3746 -- If the formal is Out or In_Out, do not resolve and expand the
3747 -- conversion, because it is subsequently expanded into explicit
3748 -- temporaries and assignments. However, the object of the
3749 -- conversion can be resolved. An exception is the case of tagged
3750 -- type conversion with a class-wide actual. In that case we want
3751 -- the tag check to occur and no temporary will be needed (no
3752 -- representation change can occur) and the parameter is passed by
3753 -- reference, so we go ahead and resolve the type conversion.
3754 -- Another exception is the case of reference to component or
3755 -- subcomponent of a bit-packed array, in which case we want to
3756 -- defer expansion to the point the in and out assignments are
3757 -- performed.
3762 then
3765 then
3766 -- In a view conversion, the conversion must be legal in
3767 -- both directions, and thus both component types must be
3768 -- aliased, or neither (4.6 (8)).
3770 -- The extra rule in 4.6 (24.9.2) seems unduly restrictive:
3771 -- the privacy requirement should not apply to generic
3772 -- types, and should be checked in an instance. ARG query
3773 -- is in order ???
3777 then
3778 Error_Msg_N
3782 -- Comment here??? what set of cases???
3784 elsif
3786 then
3787 -- Check view conv between unrelated by ref array types
3791 then
3792 Error_Msg_N
3796 -- In Ada 2005 mode, check view conversion component
3797 -- type cannot be private, tagged, or volatile. Note
3798 -- that we only apply this to source conversions. The
3799 -- generated code can contain conversions which are
3800 -- not subject to this test, and we cannot extract the
3801 -- component type in such cases since it is not present.
3805 then
3806 declare
3808 Component_Type
3810 begin
3815 then
3816 Error_Msg_N
3827 -- Resolve expression if conversion is all OK
3832 then
3836 -- If the actual is a function call that returns a limited
3837 -- unconstrained object that needs finalization, create a
3838 -- transient scope for it, so that it can receive the proper
3839 -- finalization list.
3841 elsif Expander_Active
3847 then
3851 -- A small optimization: if one of the actuals is a concatenation
3852 -- create a block around a procedure call to recover stack space.
3853 -- This alleviates stack usage when several procedure calls in
3854 -- the same statement list use concatenation. We do not perform
3855 -- this wrapping for code statements, where the argument is a
3856 -- static string, and we want to preserve warnings involving
3857 -- sequences of such statements.
3859 elsif Expander_Active
3865 then
3869 else
3873 and then
3876 then
3877 Error_Msg_N
3890 -- (Ada 2005: AI-251): If the actual is an allocator whose
3891 -- directly designated type is a class-wide interface, we build
3892 -- an anonymous access type to use it as the type of the
3893 -- allocator. Later, when the subprogram call is expanded, if
3894 -- the interface has a secondary dispatch table the expander
3895 -- will add a type conversion to force the correct displacement
3896 -- of the pointer.
3899 declare
3905 begin
3908 then
3911 Set_Directly_Designated_Type
3916 -- Ada 2005, AI-162:If the actual is an allocator, the
3917 -- innermost enclosing statement is the master of the
3918 -- created object. This needs to be done with expansion
3919 -- enabled only, otherwise the transient scope will not
3920 -- be removed in the expansion of the wrapped construct.
3922 if Expander_Active
3925 then
3926 Establish_Transient_Scope
3936 -- (Ada 2005): The call may be to a primitive operation of a
3937 -- tagged synchronized type, declared outside of the type. In
3938 -- this case the controlling actual must be converted to its
3939 -- corresponding record type, which is the formal type. The
3940 -- actual may be a subtype, either because of a constraint or
3941 -- because it is a generic actual, so use base type to locate
3942 -- concurrent type.
3949 then
3950 -- If the actual is overloaded, look for an interpretation
3951 -- that has a synchronized type.
3956 else
3957 declare
3961 begin
3966 then
3976 declare
3979 begin
3982 else
3986 -- Tagged synchronized type (case 1): the actual is a
3987 -- concurrent type.
3991 then
3993 Unchecked_Convert_To
3997 -- Tagged synchronized type (case 2): the formal is a
3998 -- concurrent type.
4001 and then Present
4006 then
4009 -- Common case
4011 else
4016 -- Not a synchronized operation
4018 else
4026 -- An actual cannot be an untagged formal incomplete type
4031 then
4032 Error_Msg_N
4038 N_Procedure_Call_Statement)
4039 then
4040 -- In formal mode, check that actual parameters matching
4041 -- formals of tagged types are objects (or ancestor type
4042 -- conversions of objects), not general expressions.
4049 declare
4054 begin
4056 Check_SPARK_05_Restriction
4059 -- In formal mode, the only view conversions are those
4060 -- involving ancestor conversion of an extended type.
4062 elsif not
4068 then
4069 if Ekind_In
4071 then
4072 Check_SPARK_05_Restriction
4075 else
4076 Check_SPARK_05_Restriction
4080 else
4085 else
4089 -- In formal mode, the only view conversions are those
4090 -- involving ancestor conversion of an extended type.
4094 then
4095 Check_SPARK_05_Restriction
4101 -- has warnings suppressed, then we reset Never_Set_In_Source for
4102 -- the calling entity. The reason for this is to catch cases like
4103 -- GNAT.Spitbol.Patterns.Vstring_Var where the called subprogram
4104 -- uses trickery to modify an IN parameter.
4111 then
4115 -- Perform error checks for IN and IN OUT parameters
4119 -- Check unset reference. For scalar parameters, it is clearly
4120 -- wrong to pass an uninitialized value as either an IN or
4121 -- IN-OUT parameter. For composites, it is also clearly an
4122 -- error to pass a completely uninitialized value as an IN
4123 -- parameter, but the case of IN OUT is trickier. We prefer
4124 -- not to give a warning here. For example, suppose there is
4125 -- a routine that sets some component of a record to False.
4126 -- It is perfectly reasonable to make this IN-OUT and allow
4127 -- either initialized or uninitialized records to be passed
4128 -- in this case.
4130 -- For partially initialized composite values, we also avoid
4131 -- warnings, since it is quite likely that we are passing a
4132 -- partially initialized value and only the initialized fields
4133 -- will in fact be read in the subprogram.
4138 then
4142 -- In Ada 83 we cannot pass an OUT parameter as an IN or IN OUT
4143 -- actual to a nested call, since this constitutes a reading of
4144 -- the parameter, which is not allowed.
4149 then
4154 -- In -gnatd.q mode, forget that a given array is constant when
4155 -- it is passed as an IN parameter to a foreign-convention
4156 -- subprogram. This is in case the subprogram evilly modifies the
4157 -- object. Of course, correct code would use IN OUT.
4159 if Debug_Flag_Dot_Q
4165 then
4169 -- Case of OUT or IN OUT parameter
4173 -- For an Out parameter, check for useless assignment. Note
4174 -- that we can't set Last_Assignment this early, because we may
4175 -- kill current values in Resolve_Call, and that call would
4176 -- clobber the Last_Assignment field.
4178 -- Note: call Warn_On_Useless_Assignment before doing the check
4179 -- below for Is_OK_Variable_For_Out_Formal so that the setting
4180 -- of Referenced_As_LHS/Referenced_As_Out_Formal properly
4181 -- reflects the last assignment, not this one.
4188 then
4193 -- Validate the form of the actual. Note that the call to
4194 -- Is_OK_Variable_For_Out_Formal generates the required
4195 -- reference in this case.
4197 -- A call to an initialization procedure for an aggregate
4198 -- component may initialize a nested component of a constant
4199 -- designated object. In this context the object is variable.
4203 then
4209 then
4210 Error_Msg_N
4215 Error_Msg_N
4222 -- What's the following about???
4226 else
4227 Kill_All_Checks;
4236 -- Apply appropriate constraint/predicate checks for IN [OUT] case
4240 -- Apply predicate tests except in certain special cases. Note
4241 -- that it might be more consistent to apply these only when
4242 -- expansion is active (in Exp_Ch6.Expand_Actuals), as we do
4243 -- for the outbound predicate tests ??? In any case indicate
4244 -- the function being called, for better warnings if the call
4245 -- leads to an infinite recursion.
4251 -- Apply required constraint checks
4253 -- Gigi looks at the check flag and uses the appropriate types.
4254 -- For now since one flag is used there is an optimization
4255 -- which might not be done in the IN OUT case since Gigi does
4256 -- not do any analysis. More thought required about this ???
4258 -- In fact is this comment obsolete??? doesn't the expander now
4259 -- generate all these tests anyway???
4272 then
4275 -- For view conversions of a discriminated object, apply
4276 -- check to object itself, the conversion alreay has the
4277 -- proper type.
4281 then
4288 then
4294 then
4297 else
4301 -- Ada 2005 (AI-231): Note that the controlling parameter case
4302 -- already existed in Ada 95, which is partially checked
4303 -- elsewhere (see Checks), and we don't want the warning
4304 -- message to differ.
4309 then
4311 Apply_Compile_Time_Constraint_Error
4317 Apply_Compile_Time_Constraint_Error
4326 -- Checks for OUT parameters and IN OUT parameters
4330 -- If there is a type conversion, make sure the return value
4331 -- meets the constraints of the variable before the conversion.
4335 Apply_Scalar_Range_Check
4338 -- In addition, the returned value of the parameter must
4339 -- satisfy the bounds of the object type (see comment
4340 -- below).
4344 else
4345 Apply_Range_Check
4349 -- If no conversion, apply scalar range checks and length check
4350 -- based on the subtype of the actual (NOT that of the formal).
4351 -- This indicates that the check takes place on return from the
4352 -- call. During expansion the required constraint checks are
4353 -- inserted. In GNATprove mode, in the absence of expansion,
4354 -- the flag indicates that the returned value is valid.
4356 else
4362 then
4364 else
4369 -- Note: we do not apply the predicate checks for the case of
4370 -- OUT and IN OUT parameters. They are instead applied in the
4371 -- Expand_Actuals routine in Exp_Ch6.
4374 -- An actual associated with an access parameter is implicitly
4375 -- converted to the anonymous access type of the formal and must
4376 -- satisfy the legality checks for access conversions.
4380 Error_Msg_N
4384 -- If the actual is an access selected component of a variable,
4385 -- the call may modify its designated object. It is reasonable
4386 -- to treat this as a potential modification of the enclosing
4387 -- record, to prevent spurious warnings that it should be
4388 -- declared as a constant, because intuitively programmers
4389 -- regard the designated subcomponent as part of the record.
4394 then
4399 -- Check bad case of atomic/volatile argument (RM C.6(12))
4403 then
4406 then
4407 Error_Msg_NE
4413 then
4414 Error_Msg_NE
4420 -- Check that subprograms don't have improper controlling
4421 -- arguments (RM 3.9.2 (9)).
4423 -- A primitive operation may have an access parameter of an
4424 -- incomplete tagged type, but a dispatching call is illegal
4425 -- if the type is still incomplete.
4431 declare
4433 begin
4437 then
4438 Error_Msg_NE
4449 -- Apply legality rule 3.9.2 (9/1)
4454 and then not In_Instance
4455 then
4460 Error_Msg_NE
4464 -- Apply the checks described in 3.10.2(27): if the context is a
4465 -- specific access-to-object, the actual cannot be class-wide.
4466 -- Use base type to exclude access_to_subprogram cases.
4473 and then
4478 -- Disable these checks for call to imported C++ subprograms
4480 and then not
4484 then
4485 Error_Msg_N
4490 Error_Msg_NE
4495 Check_Aliased_Parameter;
4499 -- If it is a named association, treat the selector_name as a
4500 -- proper identifier, and mark the corresponding entity.
4504 -- Ignore reference in SPARK mode, as it refers to an entity not
4505 -- in scope at the point of reference, so the reference should
4506 -- be ignored for computing effects of subprograms.
4508 and then not GNATprove_Mode
4509 then
4510 -- If subprogram is overridden, use name of formal that
4511 -- is being called.
4517 else
4531 -- The following checks are only relevant when SPARK_Mode is on as
4532 -- they are not standard Ada legality rule. Internally generated
4533 -- temporaries are ignored.
4537 -- An effectively volatile object may act as an actual when the
4538 -- corresponding formal is of a non-scalar effectively volatile
4539 -- type (SPARK RM 7.1.3(11)).
4543 then
4546 -- An effectively volatile object may act as an actual in a
4547 -- call to an instance of Unchecked_Conversion.
4548 -- (SPARK RM 7.1.3(11)).
4553 -- The actual denotes an object
4556 Error_Msg_N
4560 -- Otherwise the actual denotes an expression. Inspect the
4561 -- expression and flag each effectively volatile object with
4562 -- enabled property Async_Writers or Effective_Reads as illegal
4563 -- because it apprears within an interfering context. Note that
4564 -- this is usually done in Resolve_Entity_Name, but when the
4565 -- effectively volatile object appears as an actual in a call,
4566 -- the call must be resolved first.
4568 else
4572 -- An effectively volatile variable cannot act as an actual
4573 -- parameter in a procedure call when the variable has enabled
4574 -- property Effective_Reads and the corresponding formal is of
4575 -- mode IN (SPARK RM 7.1.3(10)).
4580 then
4586 then
4587 Error_Msg_NE
4598 -- A formal parameter of a specific tagged type whose related
4599 -- subprogram is subject to pragma Extensions_Visible with value
4600 -- "False" cannot act as an actual in a subprogram with value
4601 -- "True" (SPARK RM 6.1.7(3)).
4605 Extensions_Visible_True
4606 then
4607 Error_Msg_N
4610 Error_Msg_NE
4614 -- The actual parameter of a Ghost subprogram whose formal is of
4615 -- mode IN OUT or OUT must be a Ghost variable (SPARK RM 6.9(12)).
4623 then
4624 Error_Msg_NE
4630 else
4637 -- Case where actual is not present
4639 else
4640 Insert_Default;
4651 -----------------------
4652 -- Resolve_Allocator --
4653 -----------------------
4665 procedure Check_Allocator_Discrim_Accessibility
4668 -- Check that accessibility level associated with an access discriminant
4669 -- initialized in an allocator by the expression Disc_Exp is not deeper
4670 -- than the level of the allocator type Alloc_Typ. An error message is
4671 -- issued if this condition is violated. Specialized checks are done for
4672 -- the cases of a constraint expression which is an access attribute or
4673 -- an access discriminant.
4676 -- If the allocator is an actual in a call, it is allowed to be class-
4677 -- wide when the context is not because it is a controlling actual.
4679 -------------------------------------------
4680 -- Check_Allocator_Discrim_Accessibility --
4681 -------------------------------------------
4683 procedure Check_Allocator_Discrim_Accessibility
4686 is
4687 begin
4690 then
4691 Error_Msg_N
4694 -- When the expression is an Access attribute the level of the prefix
4695 -- object must not be deeper than that of the allocator's type.
4699 Attribute_Access
4702 then
4703 Error_Msg_N
4705 Disc_Exp);
4707 -- When the expression is an access discriminant the check is against
4708 -- the level of the prefix object.
4714 then
4715 Error_Msg_N
4717 Disc_Exp);
4719 -- All other cases are legal
4721 else
4726 ----------------------------
4727 -- In_Dispatching_Context --
4728 ----------------------------
4733 begin
4739 -- Start of processing for Resolve_Allocator
4741 begin
4742 -- Replace general access with specific type
4752 -- For qualified expression, resolve the expression using the given
4753 -- subtype (nothing to do for type mark, subtype indication)
4758 and then not In_Dispatching_Context
4759 then
4760 Error_Msg_N
4768 -- Allocators generated by the build-in-place expansion mechanism
4769 -- are explicitly marked as coming from source but do not need to be
4770 -- checked for limited initialization. To exclude this case, ensure
4771 -- that the parent of the allocator is a source node.
4772 -- The return statement constructed for an Expression_Function does
4773 -- not come from source but requires a limited check.
4777 and then
4779 or else
4783 and then not In_Instance_Body
4784 then
4787 Error_Msg_N
4790 else
4791 Error_Msg_N
4799 -- A qualified expression requires an exact match of the type. Class-
4800 -- wide matching is not allowed.
4805 then
4809 -- Calls to build-in-place functions are not currently supported in
4810 -- allocators for access types associated with a simple storage pool.
4811 -- Supporting such allocators may require passing additional implicit
4812 -- parameters to build-in-place functions (or a significant revision
4813 -- of the current b-i-p implementation to unify the handling for
4814 -- multiple kinds of storage pools). ???
4818 then
4819 declare
4822 begin
4824 and then
4825 Present (Get_Rep_Pragma
4827 then
4828 Error_Msg_N
4835 -- A special accessibility check is needed for allocators that
4836 -- constrain access discriminants. The level of the type of the
4837 -- expression used to constrain an access discriminant cannot be
4838 -- deeper than the type of the allocator (in contrast to access
4839 -- parameters, where the level of the actual can be arbitrary).
4841 -- We can't use Valid_Conversion to perform this check because in
4842 -- general the type of the allocator is unrelated to the type of
4843 -- the access discriminant.
4847 then
4854 then
4857 -- Get the first component expression of the aggregate
4867 else
4871 else
4890 else
4895 else
4904 -- For a subtype mark or subtype indication, freeze the subtype
4906 else
4910 Error_Msg_N
4914 -- A special accessibility check is needed for allocators that
4915 -- constrain access discriminants. The level of the type of the
4916 -- expression used to constrain an access discriminant cannot be
4917 -- deeper than the type of the allocator (in contrast to access
4918 -- parameters, where the level of the actual can be arbitrary).
4919 -- We can't use Valid_Conversion to perform this check because
4920 -- in general the type of the allocator is unrelated to the type
4921 -- of the access discriminant.
4926 then
4936 else
4950 -- Ada 2005 (AI-344): A class-wide allocator requires an accessibility
4951 -- check that the level of the type of the created object is not deeper
4952 -- than the level of the allocator's access type, since extensions can
4953 -- now occur at deeper levels than their ancestor types. This is a
4954 -- static accessibility level check; a run-time check is also needed in
4955 -- the case of an initialized allocator with a class-wide argument (see
4956 -- Expand_Allocator_Expression).
4960 then
4961 declare
4964 begin
4969 else
4975 then
4978 Error_Msg_N
4987 -- Do not apply Ada 2005 accessibility checks on a class-wide
4988 -- allocator if the type given in the allocator is a formal
4989 -- type. A run-time check will be performed in the instance.
4999 -- Check for allocation from an empty storage pool
5004 -- If the context is an unchecked conversion, as may happen within an
5005 -- inlined subprogram, the allocator is being resolved with its own
5006 -- anonymous type. In that case, if the target type has a specific
5007 -- storage pool, it must be inherited explicitly by the allocator type.
5011 then
5012 Set_Associated_Storage_Pool
5020 -- Check that an allocator with task parts isn't for a nested access
5021 -- type when restriction No_Task_Hierarchy applies.
5025 then
5029 -- An illegal allocator may be rewritten as a raise Program_Error
5030 -- statement.
5034 -- Avoid coextension processing for an allocator that is the
5035 -- expansion of a build-in-place function call.
5039 N_Qualified_Expression
5041 N_Function_Call
5042 and then Is_Expanded_Build_In_Place_Call
5044 then
5047 else
5048 -- An anonymous access discriminant is the definition of a
5049 -- coextension.
5053 N_Discriminant_Specification
5054 then
5055 declare
5059 begin
5062 -- Ada 2012 AI05-0052: If the designated type of the
5063 -- allocator is limited, then the allocator shall not
5064 -- be used to define the value of an access discriminant
5065 -- unless the discriminated type is immutably limited.
5070 then
5071 Error_Msg_N
5074 N);
5078 -- Avoid marking an allocator as a dynamic coextension if it is
5079 -- within a static construct.
5084 -- Finalization and deallocation of coextensions utilizes an
5085 -- approximate implementation which does not directly adhere
5086 -- to the semantic rules. Warn on potential issues involving
5087 -- coextensions.
5090 Error_Msg_N
5093 else
5094 Error_Msg_N
5100 -- Cleanup for potential static coextensions
5102 else
5106 -- Anonymous access-to-controlled objects are not finalized on
5107 -- time because this involves run-time ownership and currently
5108 -- this property is not available. In rare cases the object may
5109 -- not be finalized at all. Warn on potential issues involving
5110 -- anonymous access-to-controlled objects.
5114 then
5115 Error_Msg_N
5125 -- Report a simple error: if the designated object is a local task,
5126 -- its body has not been seen yet, and its activation will fail an
5127 -- elaboration check.
5134 then
5140 -- Ada 2012 (AI05-0111-3): Detect an attempt to allocate a task or a
5141 -- type with a task component on a subpool. This action must raise
5142 -- Program_Error at runtime.
5148 then
5160 ---------------------------
5161 -- Resolve_Arithmetic_Op --
5162 ---------------------------
5164 -- Used for resolving all arithmetic operators except exponentiation
5175 -- We do the resolution using the base type, because intermediate values
5176 -- in expressions always are of the base type, not a subtype of it.
5179 -- Returns True if N is in a context that expects "any real type"
5182 -- Return True iff given type is Integer or universal real/integer
5185 -- Choose type of integer literal in fixed-point operation to conform
5186 -- to available fixed-point type. T is the type of the other operand,
5187 -- which is needed to determine the expected type of N.
5190 -- Set operand type to T if universal
5192 -------------------------------
5193 -- Expected_Type_Is_Any_Real --
5194 -------------------------------
5197 begin
5198 -- N is the expression after "delta" in a fixed_point_definition;
5199 -- see RM-3.5.9(6):
5202 N_Decimal_Fixed_Point_Definition,
5204 -- N is one of the bounds in a real_range_specification;
5205 -- see RM-3.5.7(5):
5207 N_Real_Range_Specification,
5209 -- N is the expression of a delta_constraint;
5210 -- see RM-J.3(3):
5212 N_Delta_Constraint);
5215 -----------------------------
5216 -- Is_Integer_Or_Universal --
5217 -----------------------------
5224 begin
5230 else
5236 then
5247 ----------------------------
5248 -- Set_Mixed_Mode_Operand --
5249 ----------------------------
5255 begin
5258 -- A universal integer literal is resolved as standard integer
5259 -- except in the case of a fixed-point result, where we leave it
5260 -- as universal (to be handled by Exp_Fixd later on)
5264 else
5272 then
5273 -- A universal real can appear in a fixed-type context. We resolve
5274 -- the literal with that context, even though this might raise an
5275 -- exception prematurely (the other operand may be zero).
5282 then
5283 -- Integer arg in mixed-mode operation. Resolve with universal
5284 -- type, in case preference rule must be applied.
5290 -- If the operand is part of a fixed multiplication operation,
5291 -- a conversion will be applied to each operand, so resolve it
5292 -- with its own type.
5297 else
5298 -- Not a mixed-mode operation, resolve with context
5305 -- N may itself be a mixed-mode operation, so use context type
5312 then
5313 -- Must be (fixed * fixed) operation, operand must have one
5314 -- compatible interpretation.
5321 then
5322 -- C * F(X) in a fixed context, where C is a real literal or a
5323 -- fixed-point expression. F must have either a fixed type
5324 -- interpretation or an integer interpretation, but not both.
5331 else
5338 else
5346 -- Reanalyze the literal with the fixed type of the context. If
5347 -- context is Universal_Fixed, we are within a conversion, leave
5348 -- the literal as a universal real because there is no usable
5349 -- fixed type, and the target of the conversion plays no role in
5350 -- the resolution.
5352 declare
5356 begin
5359 else
5365 then
5367 else
5375 -- A universal real conditional expression can appear in a fixed-type
5376 -- context and must be resolved with that context to facilitate the
5377 -- code generation in the back end.
5382 then
5385 else
5390 ----------------------
5391 -- Set_Operand_Type --
5392 ----------------------
5395 begin
5398 then
5403 -- Start of processing for Resolve_Arithmetic_Op
5405 begin
5410 then
5414 -- Special-case for mixed-mode universal expressions or fixed point type
5415 -- operation: each argument is resolved separately. The same treatment
5416 -- is required if one of the operands of a fixed point operation is
5417 -- universal real, since in this case we don't do a conversion to a
5418 -- specific fixed-point type (instead the expander handles the case).
5420 -- Set the type of the node to its universal interpretation because
5421 -- legality checks on an exponentiation operand need the context.
5426 then
5437 or else
5440 then
5445 -- If context is a fixed type and one operand is integer, the other
5446 -- is resolved with the type of the context.
5451 then
5458 then
5462 -- If both operands are universal and the context is a floating
5463 -- point type, the operands are resolved to the type of the context.
5469 else
5474 -- Check the rule in RM05-4.5.5(19.1/2) disallowing universal_fixed
5475 -- multiplying operators from being used when the expected type is
5476 -- also universal_fixed. Note that B_Typ will be Universal_Fixed in
5477 -- some cases where the expected type is actually Any_Real;
5478 -- Expected_Type_Is_Any_Real takes care of that case.
5482 then
5486 N_Unchecked_Type_Conversion)
5487 then
5494 else
5497 and then not
5499 N_Unchecked_Type_Conversion)
5500 then
5501 Error_Msg_N
5506 -- The expected type is "any real type" in contexts like
5508 -- type T is delta <universal_fixed-expression> ...
5510 -- in which case we need to set the type to Universal_Real
5511 -- so that static expression evaluation will work properly.
5515 else
5525 then
5530 -- If no previous errors, this is only possible if one operand is
5531 -- overloaded and the context is universal. Resolve as such.
5536 else
5538 and then
5540 then
5544 -- If the context is Universal_Fixed and the operands are also
5545 -- universal fixed, this is an error, unless there is only one
5546 -- applicable fixed_point type (usually Duration).
5554 else
5559 else
5564 -- If one of the arguments was resolved to a non-universal type.
5565 -- label the result of the operation itself with the same type.
5566 -- Do the same for the universal argument, if any.
5578 -- In SPARK, a multiplication or division with operands of fixed point
5579 -- types must be qualified or explicitly converted to identify the
5580 -- result type.
5585 and then
5587 then
5588 Check_SPARK_05_Restriction
5592 -- Set overflow and division checking bit
5599 -- Give warning if explicit division by zero
5603 then
5609 or else
5612 then
5613 -- Specialize the warning message according to the operation.
5614 -- When SPARK_Mode is On, force a warning instead of an error
5615 -- in that case, as this likely corresponds to deactivated
5616 -- code. The following warnings are for the case
5621 -- For division, we have two cases, for float division
5622 -- of an unconstrained float type, on a machine where
5623 -- Machine_Overflows is false, we don't get an exception
5624 -- at run-time, but rather an infinity or Nan. The Nan
5625 -- case is pretty obscure, so just warn about infinities.
5629 and then not Machine_Overflows_On_Target
5630 then
5631 Error_Msg_N
5635 -- For all other cases, we get a Constraint_Error
5637 else
5638 Apply_Compile_Time_Constraint_Error
5645 Apply_Compile_Time_Constraint_Error
5651 Apply_Compile_Time_Constraint_Error
5656 -- Division by zero can only happen with division, rem,
5657 -- and mod operations.
5663 -- In GNATprove mode, we enable the division check so that
5664 -- GNATprove will issue a message if it cannot be proved.
5670 -- Otherwise just set the flag to check at run time
5672 else
5677 -- If Restriction No_Implicit_Conditionals is active, then it is
5678 -- violated if either operand can be negative for mod, or for rem
5679 -- if both operands can be negative.
5683 then
5684 declare
5691 -- Set if corresponding operand might be negative
5693 begin
5694 Determine_Range
5698 Determine_Range
5702 -- Check if we will be generating conditionals. There are two
5703 -- cases where that can happen, first for REM, the only case
5704 -- is largest negative integer mod -1, where the division can
5705 -- overflow, but we still have to give the right result. The
5706 -- front end generates a test for this annoying case. Here we
5707 -- just test if both operands can be negative (that's what the
5708 -- expander does, so we match its logic here).
5710 -- The second case is mod where either operand can be negative.
5711 -- In this case, the back end has to generate additional tests.
5714 or else
5716 then
5727 ------------------
5728 -- Resolve_Call --
5729 ------------------
5732 function Same_Or_Aliased_Subprograms
5735 -- Returns True if the subprogram entity S is the same as E or else
5736 -- S is an alias of E.
5738 ---------------------------------
5739 -- Same_Or_Aliased_Subprograms --
5740 ---------------------------------
5742 function Same_Or_Aliased_Subprograms
5745 is
5747 begin
5751 -- Local variables
5765 -- Start of processing for Resolve_Call
5767 begin
5768 -- Preserve relevant elaboration-related attributes of the context which
5769 -- are no longer available or very expensive to recompute once analysis,
5770 -- resolution, and expansion are over.
5772 Mark_Elaboration_Attributes
5778 -- The context imposes a unique interpretation with type Typ on a
5779 -- procedure or function call. Find the entity of the subprogram that
5780 -- yields the expected type, and propagate the corresponding formal
5781 -- constraints on the actuals. The caller has established that an
5782 -- interpretation exists, and emitted an error if not unique.
5784 -- First deal with the case of a call to an access-to-subprogram,
5785 -- dereference made explicit in Analyze_Call.
5791 else
5792 -- Find the interpretation whose type (a subprogram type) has a
5793 -- return type that is compatible with the context. Analysis of
5794 -- the node has established that one exists.
5813 -- If the prefix is not an entity, then resolve it
5819 -- For an indirect call, we always invalidate checks, since we do not
5820 -- know whether the subprogram is local or global. Yes we could do
5821 -- better here, e.g. by knowing that there are no local subprograms,
5822 -- but it does not seem worth the effort. Similarly, we kill all
5823 -- knowledge of current constant values.
5825 Kill_Current_Values;
5827 -- If this is a procedure call which is really an entry call, do
5828 -- the conversion of the procedure call to an entry call. Protected
5829 -- operations use the same circuitry because the name in the call
5830 -- can be an arbitrary expression with special resolution rules.
5835 then
5842 -- Annotate the tree by creating a call marker in case the original
5843 -- call is transformed by expansion. The call marker is automatically
5844 -- saved for later examination by the ABE Processing phase.
5848 -- Kill checks and constant values, as above for indirect case
5849 -- Who knows what happens when another task is activated?
5851 Kill_Current_Values;
5854 -- Normal subprogram call with name established in Resolve
5860 -- Otherwise we must have the case of an overloaded call
5862 else
5865 -- Initialize Nam to prevent warning (we know it will be assigned
5866 -- in the loop below, but the compiler does not know that).
5886 then
5887 -- The prefix is a parameterless function call that returns an access
5888 -- to subprogram. If parameters are present in the current call, add
5889 -- add an explicit dereference. We use the base type here because
5890 -- within an instance these may be subtypes.
5892 -- The dereference is added either in Analyze_Call or here. Should
5893 -- be consolidated ???
5902 -- Check that a call to Current_Task does not occur in an entry body
5905 declare
5908 begin
5910 loop
5913 -- Exclude calls that occur within the default of a formal
5914 -- parameter of the entry, since those are evaluated outside
5915 -- of the body.
5922 then
5925 Error_Msg_NE
5939 -- Check that a procedure call does not occur in the context of the
5940 -- entry call statement of a conditional or timed entry call. Note that
5941 -- the case of a call to a subprogram renaming of an entry will also be
5942 -- rejected. The test for N not being an N_Entry_Call_Statement is
5943 -- defensive, covering the possibility that the processing of entry
5944 -- calls might reach this point due to later modifications of the code
5945 -- above.
5950 then
5954 -- Ada 2005 (AI-345): If a procedure_call_statement is used
5955 -- for a procedure_or_entry_call, the procedure_name or
5956 -- procedure_prefix of the procedure_call_statement shall denote
5957 -- an entry renamed by a procedure, or (a view of) a primitive
5958 -- subprogram of a limited interface whose first parameter is
5959 -- a controlling parameter.
5964 then
5965 Error_Msg_N
5970 -- If the SPARK_05 restriction is active, we are not allowed
5971 -- to have a call to a subprogram before we see its completion.
5976 -- Don't flag strange internal calls
5981 -- Only flag calls in extended main source
5986 -- Exclude enumeration literals from this processing
5989 then
5990 Check_SPARK_05_Restriction
5994 -- Check that this is not a call to a protected procedure or entry from
5995 -- within a protected function.
5999 -- Freeze the subprogram name if not in a spec-expression. Note that
6000 -- we freeze procedure calls as well as function calls. Procedure calls
6001 -- are not frozen according to the rules (RM 13.14(14)) because it is
6002 -- impossible to have a procedure call to a non-frozen procedure in
6003 -- pure Ada, but in the code that we generate in the expander, this
6004 -- rule needs extending because we can generate procedure calls that
6005 -- need freezing.
6007 -- In Ada 2012, expression functions may be called within pre/post
6008 -- conditions of subsequent functions or expression functions. Such
6009 -- calls do not freeze when they appear within generated bodies,
6010 -- (including the body of another expression function) which would
6011 -- place the freeze node in the wrong scope. An expression function
6012 -- is frozen in the usual fashion, by the appearance of a real body,
6013 -- or at the end of a declarative part.
6016 and then not In_Spec_Expression
6018 and then
6021 then
6025 -- For a predefined operator, the type of the result is the type imposed
6026 -- by context, except for a predefined operation on universal fixed.
6027 -- Otherwise The type of the call is the type returned by the subprogram
6028 -- being called.
6035 -- If the subprogram returns an array type, and the context requires the
6036 -- component type of that array type, the node is really an indexing of
6037 -- the parameterless call. Resolve as such. A pathological case occurs
6038 -- when the type of the component is an access to the array type. In
6039 -- this case the call is truly ambiguous. If the call is to an intrinsic
6040 -- subprogram, it can't be an indexed component. This check is necessary
6041 -- because if it's Unchecked_Conversion, and we have "type T_Ptr is
6042 -- access T;" and "type T is array (...) of T_Ptr;" (i.e. an array of
6043 -- pointers to the same array), the compiler gets confused and does an
6044 -- infinite recursion.
6047 and then
6050 or else
6053 and then
6056 then
6057 declare
6062 begin
6065 then
6066 Error_Msg_N
6068 else
6071 -- The called entity may be an explicit dereference, in which
6072 -- case there is no entity to set.
6080 or else
6082 and then
6084 then
6087 -- Indexed call to a parameterless function
6089 Index_Node :=
6091 Prefix =>
6094 else
6095 -- An Ada 2005 prefixed call to a primitive operation
6096 -- whose first parameter is the prefix. This prefix was
6097 -- prepended to the parameter list, which is actually a
6098 -- list of indexes. Remove the prefix in order to build
6099 -- the proper indexed component.
6101 Index_Node :=
6103 Prefix =>
6106 Parameter_Associations =>
6107 New_List
6112 -- Preserve the parenthesis count of the node
6116 -- Since we are correcting a node classification error made
6117 -- by the parser, we call Replace rather than Rewrite.
6129 -- Annotate the tree by creating a call marker in case
6130 -- the original call is transformed by expansion. The call
6131 -- marker is automatically saved for later examination by
6132 -- the ABE Processing phase.
6141 else
6142 -- If the called function is not declared in the main unit and it
6143 -- returns the limited view of type then use the available view (as
6144 -- is done in Try_Object_Operation) to prevent back-end confusion;
6145 -- for the function entity itself. The call must appear in a context
6146 -- where the nonlimited view is available. If the function entity is
6147 -- in the extended main unit then no action is needed, because the
6148 -- back end handles this case. In either case the type of the call
6149 -- is the nonlimited view.
6153 then
6160 else
6165 -- In the case where the call is to an overloaded subprogram, Analyze
6166 -- calls Normalize_Actuals once per overloaded subprogram. Therefore in
6167 -- such a case Normalize_Actuals needs to be called once more to order
6168 -- the actuals correctly. Otherwise the call will have the ordering
6169 -- given by the last overloaded subprogram whether this is the correct
6170 -- one being called or not.
6177 -- In any case, call is fully resolved now. Reset Overload flag, to
6178 -- prevent subsequent overload resolution if node is analyzed again
6183 -- A Ghost entity must appear in a specific context
6189 -- If we are calling the current subprogram from immediately within its
6190 -- body, then that is the case where we can sometimes detect cases of
6191 -- infinite recursion statically. Do not try this in case restriction
6192 -- No_Recursion is in effect anyway, and do it only for source calls.
6197 -- Check violation of SPARK_05 restriction which does not permit
6198 -- a subprogram body to contain a call to the subprogram directly.
6202 then
6203 Check_SPARK_05_Restriction
6207 -- Issue warning for possible infinite recursion in the absence
6208 -- of the No_Recursion restriction.
6213 then
6214 -- Here we detected and flagged an infinite recursion, so we do
6215 -- not need to test the case below for further warnings. Also we
6216 -- are all done if we now have a raise SE node.
6222 -- If call is to immediately containing subprogram, then check for
6223 -- the case of a possible run-time detectable infinite recursion.
6225 else
6229 -- Although in general case, recursion is not statically
6230 -- checkable, the case of calling an immediately containing
6231 -- subprogram is easy to catch.
6235 -- If the recursive call is to a parameterless subprogram,
6236 -- then even if we can't statically detect infinite
6237 -- recursion, this is pretty suspicious, and we output a
6238 -- warning. Furthermore, we will try later to detect some
6239 -- cases here at run time by expanding checking code (see
6240 -- Detect_Infinite_Recursion in package Exp_Ch6).
6242 -- If the recursive call is within a handler, do not emit a
6243 -- warning, because this is a common idiom: loop until input
6244 -- is correct, catch illegal input in handler and restart.
6250 then
6251 -- For the case of a procedure call. We give the message
6252 -- only if the call is the first statement in a sequence
6253 -- of statements, or if all previous statements are
6254 -- simple assignments. This is simply a heuristic to
6255 -- decrease false positives, without losing too many good
6256 -- warnings. The idea is that these previous statements
6257 -- may affect global variables the procedure depends on.
6258 -- We also exclude raise statements, that may arise from
6259 -- constraint checks and are probably unrelated to the
6260 -- intended control flow.
6264 then
6265 declare
6267 begin
6271 N_Raise_Constraint_Error)
6272 then
6281 -- Do not give warning if we are in a conditional context
6283 declare
6285 begin
6291 then
6296 -- Here warning is to be issued
6312 -- Check obsolescent reference to Ada.Characters.Handling subprogram
6316 -- If subprogram name is a predefined operator, it was given in
6317 -- functional notation. Replace call node with operator node, so
6318 -- that actuals can be resolved appropriately.
6326 then
6332 -- Create a transient scope if the resulting type requires it
6334 -- There are several notable exceptions:
6336 -- a) In init procs, the transient scope overhead is not needed, and is
6337 -- even incorrect when the call is a nested initialization call for a
6338 -- component whose expansion may generate adjust calls. However, if the
6339 -- call is some other procedure call within an initialization procedure
6340 -- (for example a call to Create_Task in the init_proc of the task
6341 -- run-time record) a transient scope must be created around this call.
6343 -- b) Enumeration literal pseudo-calls need no transient scope
6345 -- c) Intrinsic subprograms (Unchecked_Conversion and source info
6346 -- functions) do not use the secondary stack even though the return
6347 -- type may be unconstrained.
6349 -- d) Calls to a build-in-place function, since such functions may
6350 -- allocate their result directly in a target object, and cases where
6351 -- the result does get allocated in the secondary stack are checked for
6352 -- within the specialized Exp_Ch6 procedures for expanding those
6353 -- build-in-place calls.
6355 -- e) Calls to inlinable expression functions do not use the secondary
6356 -- stack (since the call will be replaced by its returned object).
6358 -- f) If the subprogram is marked Inline_Always, then even if it returns
6359 -- an unconstrained type the call does not require use of the secondary
6360 -- stack. However, inlining will only take place if the body to inline
6361 -- is already present. It may not be available if e.g. the subprogram is
6362 -- declared in a child instance.
6368 then
6375 then
6378 elsif Expander_Active
6381 then
6384 -- If the call appears within the bounds of a loop, it will be
6385 -- rewritten and reanalyzed, nothing left to do here.
6392 -- A protected function cannot be called within the definition of the
6393 -- enclosing protected type, unless it is part of a pre/postcondition
6394 -- on another protected operation. This may appear in the entry wrapper
6395 -- created for an entry with preconditions.
6400 and then not In_Spec_Expression
6402 then
6403 Error_Msg_NE
6407 -- Propagate interpretation to actuals, and add default expressions
6408 -- where needed.
6413 -- Overloaded literals are rewritten as function calls, for purpose of
6414 -- resolution. After resolution, we can replace the call with the
6415 -- literal itself.
6421 -- Avoid validation, since it is a static function call
6427 -- If the subprogram is not global, then kill all saved values and
6428 -- checks. This is a bit conservative, since in many cases we could do
6429 -- better, but it is not worth the effort. Similarly, we kill constant
6430 -- values. However we do not need to do this for internal entities
6431 -- (unless they are inherited user-defined subprograms), since they
6432 -- are not in the business of molesting local values.
6434 -- If the flag Suppress_Value_Tracking_On_Calls is set, then we also
6435 -- kill all checks and values for calls to global subprograms. This
6436 -- takes care of the case where an access to a local subprogram is
6437 -- taken, and could be passed directly or indirectly and then called
6438 -- from almost any context.
6440 -- Note: we do not do this step till after resolving the actuals. That
6441 -- way we still take advantage of the current value information while
6442 -- scanning the actuals.
6444 -- We suppress killing values if we are processing the nodes associated
6445 -- with N_Freeze_Entity nodes. Otherwise the declaration of a tagged
6446 -- type kills all the values as part of analyzing the code that
6447 -- initializes the dispatch tables.
6451 or else Suppress_Value_Tracking_On_Call
6456 then
6457 Kill_Current_Values;
6460 -- If we are warning about unread OUT parameters, this is the place to
6461 -- set Last_Assignment for OUT and IN OUT parameters. We have to do this
6462 -- after the above call to Kill_Current_Values (since that call clears
6463 -- the Last_Assignment field of all local variables).
6468 then
6469 declare
6473 begin
6483 then
6493 -- If the subprogram is a primitive operation, check whether or not
6494 -- it is a correct dispatching call.
6498 then
6503 and then not In_Instance
6504 then
6508 -- If this is a dispatching call, generate the appropriate reference,
6509 -- for better source navigation in GPS.
6513 then
6516 -- Normal case, not a dispatching call: generate a call reference
6518 else
6526 -- Check for violation of restriction No_Specific_Termination_Handlers
6527 -- and warn on a potentially blocking call to Abort_Task.
6531 or else
6533 then
6540 -- A call to Ada.Real_Time.Timing_Events.Set_Handler to set a relative
6541 -- timing event violates restriction No_Relative_Delay (AI-0211). We
6542 -- need to check the second argument to determine whether it is an
6543 -- absolute or relative timing event.
6548 then
6552 -- Issue an error for a call to an eliminated subprogram. This routine
6553 -- will not perform the check if the call appears within a default
6554 -- expression.
6558 -- In formal mode, the primitive operations of a tagged type or type
6559 -- extension do not include functions that return the tagged type.
6565 then
6569 -- Implement rule in 12.5.1 (23.3/2): In an instance, if the actual is
6570 -- class-wide and the call dispatches on result in a context that does
6571 -- not provide a tag, the call raises Program_Error.
6574 and then In_Instance
6579 then
6580 -- Verify that none of the formals are controlling
6582 declare
6586 begin
6608 -- Check for calling a function with OUT or IN OUT parameter when the
6609 -- calling context (us right now) is not Ada 2012, so does not allow
6610 -- OUT or IN OUT parameters in function calls. Functions declared in
6611 -- a predefined unit are OK, as they may be called indirectly from a
6612 -- user-declared instantiation.
6618 then
6623 -- Check the dimensions of the actuals in the call. For function calls,
6624 -- propagate the dimensions from the returned type to N.
6628 -- All done, evaluate call and deal with elaboration issues
6636 -- Annotate the tree by creating a call marker in case the original call
6637 -- is transformed by expansion. The call marker is automatically saved
6638 -- for later examination by the ABE Processing phase.
6642 -- In GNATprove mode, expansion is disabled, but we want to inline some
6643 -- subprograms to facilitate formal verification. Indirect calls through
6644 -- a subprogram type or within a generic cannot be inlined. Inlining is
6645 -- performed only for calls subject to SPARK_Mode on.
6647 if GNATprove_Mode
6650 and then not Inside_A_Generic
6651 then
6658 -- Nothing to do if the subprogram is not eligible for inlining in
6659 -- GNATprove mode, or inlining is disabled with switch -gnatdm
6663 or else Debug_Flag_M
6664 then
6667 -- Calls cannot be inlined inside assertions, as GNATprove treats
6668 -- assertions as logic expressions. Only issue a message when the
6669 -- body has been seen, otherwise this leads to spurious messages
6670 -- on expression functions.
6674 Cannot_Inline
6678 -- Calls cannot be inlined inside default expressions
6681 Cannot_Inline
6684 -- Inlining should not be performed during preanalysis
6688 -- Do not inline calls inside expression functions or functions
6689 -- generated by the front end for subtype predicates, as this
6690 -- would prevent interpreting them as logical formulas in
6691 -- GNATprove. Only issue a message when the body has been seen,
6692 -- otherwise this leads to spurious messages on callees that
6693 -- are themselves expression functions.
6696 and then
6701 then
6704 then
6706 Cannot_Inline
6711 Cannot_Inline
6716 Cannot_Inline
6720 else
6721 Cannot_Inline
6727 -- With the one-pass inlining technique, a call cannot be
6728 -- inlined if the corresponding body has not been seen yet.
6731 Cannot_Inline
6734 -- Nothing to do if there is no body to inline, indicating that
6735 -- the subprogram is not suitable for inlining in GNATprove
6736 -- mode.
6741 -- Calls cannot be inlined inside potentially unevaluated
6742 -- expressions, as this would create complex actions inside
6743 -- expressions, that are not handled by GNATprove.
6746 Cannot_Inline
6750 -- Do not inline calls which would possibly lead to missing a
6751 -- type conversion check on an input parameter.
6754 Cannot_Inline
6758 -- Otherwise, inline the call
6760 else
6772 -----------------------------
6773 -- Resolve_Case_Expression --
6774 -----------------------------
6782 begin
6794 -- When the expression is of a scalar subtype different from the
6795 -- result subtype, then insert a conversion to ensure the generation
6796 -- of a constraint check.
6806 -- Apply RM 4.5.7 (17/3): whether the expression is statically or
6807 -- dynamically tagged must be known statically.
6815 Error_Msg_N
6829 -------------------------------
6830 -- Resolve_Character_Literal --
6831 -------------------------------
6837 begin
6838 -- Verify that the character does belong to the type of the context
6843 -- Wide_Wide_Character literals must always be defined, since the set
6844 -- of wide wide character literals is complete, i.e. if a character
6845 -- literal is accepted by the parser, then it is OK for wide wide
6846 -- character (out of range character literals are rejected).
6851 -- Always accept character literal for type Any_Character, which
6852 -- occurs in error situations and in comparisons of literals, both
6853 -- of which should accept all literals.
6858 -- For Standard.Character or a type derived from it, check that the
6859 -- literal is in range.
6866 -- For Standard.Wide_Character or a type derived from it, check that the
6867 -- literal is in range.
6874 -- If the entity is already set, this has already been resolved in a
6875 -- generic context, or comes from expansion. Nothing else to do.
6880 -- Otherwise we have a user defined character type, and we can use the
6881 -- standard visibility mechanisms to locate the referenced entity.
6883 else
6896 -- If we fall through, then the literal does not match any of the
6897 -- entries of the enumeration type. This isn't just a constraint error
6898 -- situation, it is an illegality (see RM 4.2).
6900 Error_Msg_NE
6904 ---------------------------
6905 -- Resolve_Comparison_Op --
6906 ---------------------------
6908 -- Context requires a boolean type, and plays no role in resolution.
6909 -- Processing identical to that for equality operators. The result type is
6910 -- the base type, which matters when pathological subtypes of booleans with
6911 -- limited ranges are used.
6918 begin
6919 -- If this is an intrinsic operation which is not predefined, use the
6920 -- types of its declared arguments to resolve the possibly overloaded
6921 -- operands. Otherwise the operands are unambiguous and specify the
6922 -- expected type.
6927 else
6938 -- Skip remaining processing if already set to Any_Type
6944 -- Deal with other error cases
6948 T = Any_Character
6949 then
6952 else
6960 -- Resolve the operands if types OK
6969 -- In SPARK, ordering operators <, <=, >, >= are not defined for Boolean
6970 -- types or array types except String.
6973 Check_SPARK_05_Restriction
6978 then
6979 Check_SPARK_05_Restriction
6983 -- Check comparison on unordered enumeration
6987 Error_Msg_NE
6994 -- Evaluate the relation (note we do this after the above check since
6995 -- this Eval call may change N to True/False. Skip this evaluation
6996 -- inside assertions, in order to keep assertions as written by users
6997 -- for tools that rely on these, e.g. GNATprove for loop invariants.
6998 -- Except evaluation is still performed even inside assertions for
6999 -- comparisons between values of universal type, which are useless
7000 -- for static analysis tools, and not supported even by GNATprove.
7004 and then
7006 then
7011 -----------------------------------------
7012 -- Resolve_Discrete_Subtype_Indication --
7013 -----------------------------------------
7015 procedure Resolve_Discrete_Subtype_Indication
7018 is
7022 begin
7030 else
7040 Error_Msg_NE
7043 -- Rewrite the constraint as a range of Typ
7044 -- to allow compilation to proceed further.
7056 else
7060 -- Additionally, we must check that the bounds are compatible
7061 -- with the given subtype, which might be different from the
7062 -- type of the context.
7066 -- ??? If the above check statically detects a Constraint_Error
7067 -- it replaces the offending bound(s) of the range R with a
7068 -- Constraint_Error node. When the itype which uses these bounds
7069 -- is frozen the resulting call to Duplicate_Subexpr generates
7070 -- a new temporary for the bounds.
7072 -- Unfortunately there are other itypes that are also made depend
7073 -- on these bounds, so when Duplicate_Subexpr is called they get
7074 -- a forward reference to the newly created temporaries and Gigi
7075 -- aborts on such forward references. This is probably sign of a
7076 -- more fundamental problem somewhere else in either the order of
7077 -- itype freezing or the way certain itypes are constructed.
7079 -- To get around this problem we call Remove_Side_Effects right
7080 -- away if either bounds of R are a Constraint_Error.
7082 declare
7086 begin
7102 -------------------------
7103 -- Resolve_Entity_Name --
7104 -------------------------
7106 -- Used to resolve identifiers and expanded names
7109 function Is_Assignment_Or_Object_Expression
7112 -- Determine whether node Context denotes an assignment statement or an
7113 -- object declaration whose expression is node Expr.
7115 ----------------------------------------
7116 -- Is_Assignment_Or_Object_Expression --
7117 ----------------------------------------
7119 function Is_Assignment_Or_Object_Expression
7122 is
7123 begin
7125 N_Object_Declaration)
7127 then
7130 -- Check whether a construct that yields a name is the expression of
7131 -- an assignment statement or an object declaration.
7134 N_Explicit_Dereference,
7135 N_Indexed_Component,
7136 N_Selected_Component,
7137 N_Slice)
7139 or else
7141 N_Unchecked_Type_Conversion)
7143 then
7144 return
7145 Is_Assignment_Or_Object_Expression
7149 -- Otherwise the context is not an assignment statement or an object
7150 -- declaration.
7152 else
7157 -- Local variables
7162 -- Start of processing for Resolve_Entity_Name
7164 begin
7165 -- If garbage from errors, set to Any_Type and return
7172 -- Replace named numbers by corresponding literals. Note that this is
7173 -- the one case where Resolve_Entity_Name must reset the Etype, since
7174 -- it is currently marked as universal.
7184 -- For enumeration literals, we need to make sure that a proper style
7185 -- check is done, since such literals are overloaded, and thus we did
7186 -- not do a style check during the first phase of analysis.
7192 -- Case of (sub)type name appearing in a context where an expression
7193 -- is expected. This is legal if occurrence is a current instance.
7194 -- See RM 8.6 (17/3).
7200 -- Any other use is an error
7202 else
7203 Error_Msg_N
7207 -- Check discriminant use if entity is discriminant in current scope,
7208 -- i.e. discriminant of record or concurrent type currently being
7209 -- analyzed. Uses in corresponding body are unrestricted.
7214 then
7217 -- A parameterless generic function cannot appear in a context that
7218 -- requires resolution.
7223 -- In Ada 83 an OUT parameter cannot be read, but attributes of
7224 -- array types (i.e. bounds and length) are legal.
7232 or else Is_Assignment_Or_Object_Expression
7235 then
7240 -- In all other cases, just do the possible static evaluation
7242 else
7243 -- A deferred constant that appears in an expression must have a
7244 -- completion, unless it has been removed by in-place expansion of
7245 -- an aggregate. A constant that is a renaming does not need
7246 -- initialization.
7252 and then not In_Spec_Expression
7255 then
7259 then
7261 else
7262 Error_Msg_N
7272 -- When the entity appears in a parameter association, retrieve the
7273 -- related subprogram call.
7281 -- The following checks are only relevant when SPARK_Mode is on as
7282 -- they are not standard Ada legality rules.
7286 -- An effectively volatile object subject to enabled properties
7287 -- Async_Writers or Effective_Reads must appear in non-interfering
7288 -- context (SPARK RM 7.1.3(12)).
7295 then
7296 SPARK_Msg_N
7301 -- Check for possible elaboration issues with respect to reads of
7302 -- variables. The act of renaming the variable is not considered a
7303 -- read as it simply establishes an alias.
7305 if Legacy_Elaboration_Checks
7307 and then Dynamic_Elaboration_Checks
7309 then
7314 -- The variable may eventually become a constituent of a single
7315 -- protected/task type. Record the reference now and verify its
7316 -- legality when analyzing the contract of the variable
7317 -- (SPARK RM 9.3).
7323 -- A Ghost entity must appear in a specific context
7330 -- We may be resolving an entity within expanded code, so a reference to
7331 -- an entity should be ignored when calculating effective use clauses to
7332 -- avoid inappropriate marking.
7339 -------------------
7340 -- Resolve_Entry --
7341 -------------------
7353 -- If the bounds of the entry family being called depend on task
7354 -- discriminants, build a new index subtype where a discriminant is
7355 -- replaced with the value of the discriminant of the target task.
7356 -- The target task is the prefix of the entry name in the call.
7358 -----------------------
7359 -- Actual_Index_Type --
7360 -----------------------
7370 -- If the bound is given by a discriminant, replace with a reference
7371 -- to the discriminant of the same name in the target task. If the
7372 -- entry name is the target of a requeue statement and the entry is
7373 -- in the current protected object, the bound to be used is the
7374 -- discriminal of the object (see Apply_Range_Checks for details of
7375 -- the transformation).
7377 -----------------------------
7378 -- Actual_Discriminant_Ref --
7379 -----------------------------
7385 begin
7390 then
7396 then
7397 -- Note: here Bound denotes a discriminant of the corresponding
7398 -- record type tskV, whose discriminal is a formal of the
7399 -- init-proc tskVIP. What we want is the body discriminal,
7400 -- which is associated to the discriminant of the original
7401 -- concurrent type tsk.
7403 return New_Occurrence_Of
7406 else
7407 Ref :=
7417 -- Start of processing for Actual_Index_Type
7419 begin
7422 then
7425 else
7439 -- Start of processing for Resolve_Entry
7441 begin
7442 -- Find name of entry being called, and resolve prefix of name with its
7443 -- own type. The prefix can be overloaded, and the name and signature of
7444 -- the entry must be taken into account.
7448 -- Case of dealing with entry family within the current tasks
7452 else
7458 -- Entry call to an entry (or entry family) in the current task. This
7459 -- is legal even though the task will deadlock. Rewrite as call to
7460 -- current task.
7462 -- This can also be a call to an entry in an enclosing task. If this
7463 -- is a single task, we have to retrieve its name, because the scope
7464 -- of the entry is the task type, not the object. If the enclosing
7465 -- task is a task type, the identity of the task is given by its own
7466 -- self variable.
7468 -- Finally this can be a requeue on an entry of the same task or
7469 -- protected object.
7476 then
7477 -- S is an enclosing task or protected object. The concurrent
7478 -- declaration has been converted into a type declaration, and
7479 -- the object itself has an object declaration that follows
7480 -- the type in the same declarative part.
7492 -- Call to current task. Will be transformed into call to Self
7499 New_N :=
7502 Selector_Name =>
7509 then
7510 -- Use the entry name (which must be unique at this point) to find
7511 -- the prefix that returns the corresponding task/protected type.
7513 declare
7519 begin
7541 -- Generate a reference for the index when it denotes an entity
7547 -- Up to this point the expression could have been the actual in a
7548 -- simple entry call, and be given by a named association.
7552 else
7558 ------------------------
7559 -- Resolve_Entry_Call --
7560 ------------------------
7571 begin
7572 -- We kill all checks here, because it does not seem worth the effort to
7573 -- do anything better, an entry call is a big operation.
7575 Kill_All_Checks;
7577 -- Processing of the name is similar for entry calls and protected
7578 -- operation calls. Once the entity is determined, we can complete
7579 -- the resolution of the actuals.
7581 -- The selector may be overloaded, in the case of a protected object
7582 -- with overloaded functions. The type of the context is used for
7583 -- resolution.
7588 then
7589 declare
7593 begin
7612 -- Simple entry or protected operation call
7625 -- Call to member of entry family
7632 -- We cannot in general check the maximum depth of protected entry calls
7633 -- at compile time. But we can tell that any protected entry call at all
7634 -- violates a specified nesting depth of zero.
7640 -- Use context type to disambiguate a protected function that can be
7641 -- called without actuals and that returns an array type, and where the
7642 -- argument list may be an indexing of the returned value.
7647 and then
7653 and then
7654 Covers
7657 then
7658 declare
7661 begin
7662 Index_Node :=
7664 Prefix =>
7668 -- Since we are correcting a node classification error made by the
7669 -- parser, we call Replace rather than Rewrite.
7682 then
7683 -- Note the entity being called before rewriting the call, so that
7684 -- it appears used at this point.
7688 -- Rewrite as call to the precondition wrapper, adding the task
7689 -- object to the list of actuals. If the call is to a member of an
7690 -- entry family, include the index as well.
7692 declare
7696 begin
7705 New_Call :=
7707 Name =>
7712 -- Preanalyze and resolve new call. Current procedure is called
7713 -- from Resolve_Call, after which expansion will take place.
7720 -- The operation name may have been overloaded. Order the actuals
7721 -- according to the formals of the resolved entity, and set the return
7722 -- type to that of the operation.
7729 -- Reset the Is_Overloaded flag, since resolution is now completed
7731 -- Simple entry call
7736 -- Call to a member of an entry family
7746 -- Create a call reference to the entry
7754 -- Verify that a procedure call cannot masquerade as an entry
7755 -- call where an entry call is expected.
7760 then
7765 then
7770 then
7775 -- After resolution, entry calls and protected procedure calls are
7776 -- changed into entry calls, for expansion. The structure of the node
7777 -- does not change, so it can safely be done in place. Protected
7778 -- function calls must keep their structure because they are
7779 -- subexpressions.
7783 -- A protected operation that is not a function may modify the
7784 -- corresponding object, and cannot apply to a constant. If this
7785 -- is an internal call, the prefix is the type itself.
7791 then
7792 Error_Msg_N
7794 Entry_Name);
7797 declare
7800 begin
7801 Entry_Call :=
7806 -- Inherit relevant attributes from the original call
7808 Set_First_Named_Actual
7811 Set_Is_Elaboration_Checks_OK_Node
7814 Set_Is_Elaboration_Warnings_OK_Node
7817 Set_Is_SPARK_Mode_On_Node
7824 -- Protected functions can return on the secondary stack, in which case
7825 -- we must trigger the transient scope mechanism.
7827 elsif Expander_Active
7829 then
7834 -------------------------
7835 -- Resolve_Equality_Op --
7836 -------------------------
7838 -- Both arguments must have the same type, and the boolean context does
7839 -- not participate in the resolution. The first pass verifies that the
7840 -- interpretation is not ambiguous, and the type of the left argument is
7841 -- correctly set, or is Any_Type in case of ambiguity. If both arguments
7842 -- are strings or aggregates, allocators, or Null, they are ambiguous even
7843 -- though they carry a single (universal) type. Diagnose this case here.
7851 -- The resolution rule for if expressions requires that each such must
7852 -- have a unique type. This means that if several dependent expressions
7853 -- are of a non-null anonymous access type, and the context does not
7854 -- impose an expected type (as can be the case in an equality operation)
7855 -- the expression must be rejected.
7858 -- Attempt to explain the nature of a redundant comparison with True. If
7859 -- the expression N is too complex, this routine issues a general error
7860 -- message.
7863 -- In the case of allocators and access attributes, the context must
7864 -- provide an indication of the specific access type to be used. If
7865 -- one operand is of such a "generic" access type, check whether there
7866 -- is a specific visible access type that has the same designated type.
7867 -- This is semantically dubious, and of no interest to any real code,
7868 -- but c48008a makes it all worthwhile.
7870 -------------------------
7871 -- Check_If_Expression --
7872 -------------------------
7878 begin
7885 then
7891 ------------------------
7892 -- Explain_Redundancy --
7893 ------------------------
7900 begin
7903 -- Strip the operand down to an entity
7905 loop
7908 else
7913 -- The construct denotes an entity
7918 -- Do not generate an error message when the comparison is done
7919 -- against the enumeration literal Standard.True.
7923 -- Build a customized error message
7950 -- The construct is too complex to disect, issue a general message
7952 else
7957 -----------------------------
7958 -- Find_Unique_Access_Type --
7959 -----------------------------
7966 begin
7968 E_Access_Attribute_Type)
7969 then
7973 E_Access_Attribute_Type)
7974 then
7976 else
7988 then
8001 -- Start of processing for Resolve_Equality_Op
8003 begin
8014 T = Any_Character
8015 then
8018 else
8027 then
8036 -- If expressions must have a single type, and if the context does
8037 -- not impose one the dependent expressions cannot be anonymous
8038 -- access types.
8040 -- Why no similar processing for case expressions???
8044 E_Anonymous_Access_Subprogram_Type)
8046 E_Anonymous_Access_Subprogram_Type)
8047 then
8055 -- In SPARK, equality operators = and /= for array types other than
8056 -- String are only defined when, for each index position, the
8057 -- operands have equal static bounds.
8061 -- Protect call to Matching_Static_Array_Bounds to avoid costly
8062 -- operation if not needed.
8070 then
8071 Check_SPARK_05_Restriction
8076 -- If the unique type is a class-wide type then it will be expanded
8077 -- into a dispatching call to the predefined primitive. Therefore we
8078 -- check here for potential violation of such restriction.
8084 -- Only warn for redundant equality comparison to True for objects
8085 -- (e.g. "X = True") and operations (e.g. "(X < Y) = True"). For
8086 -- other expressions, it may be a matter of preference to write
8087 -- "Expr = True" or "Expr".
8089 if Warn_On_Redundant_Constructs
8094 and then
8096 or else
8098 then
8099 Error_Msg_N -- CODEFIX
8109 -- If this is an inequality, it may be the implicit inequality
8110 -- created for a user-defined operation, in which case the corres-
8111 -- ponding equality operation is not intrinsic, and the operation
8112 -- cannot be constant-folded. Else fold.
8117 or else Is_Intrinsic_Subprogram
8119 then
8125 then
8129 -- Ada 2005: If one operand is an anonymous access type, convert the
8130 -- other operand to it, to ensure that the underlying types match in
8131 -- the back-end. Same for access_to_subprogram, and the conversion
8132 -- verifies that the types are subtype conformant.
8134 -- We apply the same conversion in the case one of the operands is a
8135 -- private subtype of the type of the other.
8137 -- Why the Expander_Active test here ???
8139 if Expander_Active
8140 and then
8142 E_Anonymous_Access_Subprogram_Type)
8144 then
8164 ----------------------------------
8165 -- Resolve_Explicit_Dereference --
8166 ----------------------------------
8174 -- The candidate prefix type, if overloaded
8179 begin
8183 -- A useful optimization: check whether the dereference denotes an
8184 -- element of a container, and if so rewrite it as a call to the
8185 -- corresponding Element function.
8187 -- Disabled for now, on advice of ARG. A more restricted form of the
8188 -- predicate might be acceptable ???
8190 -- if Is_Container_Element (N) then
8191 -- return;
8192 -- end if;
8196 -- Use the context type to select the prefix that has the correct
8197 -- designated type. Keep the first match, which will be the inner-
8198 -- most.
8205 then
8210 -- Remove access types that do not match, but preserve access
8211 -- to subprogram interpretations, in case a further dereference
8212 -- is needed (see below).
8225 else
8226 -- If no interpretation covers the designated type of the prefix,
8227 -- this is the pathological case where not all implementations of
8228 -- the prefix allow the interpretation of the node as a call. Now
8229 -- that the expected type is known, Remove other interpretations
8230 -- from prefix, rewrite it as a call, and resolve again, so that
8231 -- the proper call node is generated.
8242 New_N :=
8244 Name =>
8255 -- If not overloaded, resolve P with its own type
8257 else
8261 -- If the prefix might be null, add an access check
8265 then
8269 -- If the designated type is a packed unconstrained array type, and the
8270 -- explicit dereference is not in the context of an attribute reference,
8271 -- then we must compute and set the actual subtype, since it is needed
8272 -- by Gigi. The reason we exclude the attribute case is that this is
8273 -- handled fine by Gigi, and in fact we use such attributes to build the
8274 -- actual subtype. We also exclude generated code (which builds actual
8275 -- subtypes directly if they are needed).
8282 then
8288 -- Note: No Eval processing is required for an explicit dereference,
8289 -- because such a name can never be static.
8293 -------------------------------------
8294 -- Resolve_Expression_With_Actions --
8295 -------------------------------------
8298 begin
8301 -- If N has no actions, and its expression has been constant folded,
8302 -- then rewrite N as just its expression. Note, we can't do this in
8303 -- the general case of Is_Empty_List (Actions (N)) as this would cause
8304 -- Expression (N) to be expanded again.
8308 then
8313 ----------------------------------
8314 -- Resolve_Generalized_Indexing --
8315 ----------------------------------
8323 begin
8324 -- In ASIS mode, propagate the information about the indexes back to
8325 -- to the original indexing node. The generalized indexing is either
8326 -- a function call, or a dereference of one. The actuals include the
8327 -- prefix of the original node, which is the container expression.
8336 loop
8345 -- If expression is to be reanalyzed, reset Generalized_Indexing
8346 -- to recreate call node, as is the case when the expression is
8347 -- part of an expression function.
8356 else
8362 ---------------------------
8363 -- Resolve_If_Expression --
8364 ---------------------------
8373 begin
8374 -- Defend against malformed expressions
8392 -- When the "then" expression is of a scalar subtype different from the
8393 -- result subtype, then insert a conversion to ensure the generation of
8394 -- a constraint check. The same is done for the else part below, again
8395 -- comparing subtypes rather than base types.
8402 -- If ELSE expression present, just resolve using the determined type
8403 -- If type is universal, resolve to any member of the class.
8412 else
8422 -- Apply RM 4.5.7 (17/3): whether the expression is statically or
8423 -- dynamically tagged must be known statically.
8428 then
8434 -- If no ELSE expression is present, root type must be Standard.Boolean
8435 -- and we provide a Standard.True result converted to the appropriate
8436 -- Boolean type (in case it is a derived boolean type).
8439 Else_Expr :=
8444 else
8458 -------------------------------
8459 -- Resolve_Indexed_Component --
8460 -------------------------------
8468 begin
8476 -- Use the context type to select the prefix that yields the correct
8477 -- component type.
8479 declare
8486 begin
8493 and then
8494 Covers
8497 then
8506 else
8512 else
8523 else
8530 -- If prefix is access type, dereference to get real array type.
8531 -- Note: we do not apply an access check because the expander always
8532 -- introduces an explicit dereference, and the check will happen there.
8538 -- If name was overloaded, set component type correctly now
8539 -- If a misplaced call to an entry family (which has no index types)
8540 -- return. Error will be diagnosed from calling context.
8544 else
8551 -- The prefix may have resolved to a string literal, in which case its
8552 -- etype has a special representation. This is only possible currently
8553 -- if the prefix is a static concatenation, written in functional
8554 -- notation.
8559 else
8566 else
8577 -- Do not generate the warning on suspicious index if we are analyzing
8578 -- package Ada.Tags; otherwise we will report the warning with the
8579 -- Prims_Ptr field of the dispatch table.
8582 or else not
8584 Ada_Tags)
8585 then
8590 -- If the array type is atomic, and the component is not atomic, then
8591 -- this is worth a warning, since we have a situation where the access
8592 -- to the component may cause extra read/writes of the atomic array
8593 -- object, or partial word accesses, which could be unexpected.
8599 and then Has_Atomic_Components
8602 then
8603 Error_Msg_N
8605 Error_Msg_N
8610 -----------------------------
8611 -- Resolve_Integer_Literal --
8612 -----------------------------
8615 begin
8620 --------------------------------
8621 -- Resolve_Intrinsic_Operator --
8622 --------------------------------
8631 -- If the operand is a literal, it cannot be the expression in a
8632 -- conversion. Use a qualified expression instead.
8634 ---------------------
8635 -- Convert_Operand --
8636 ---------------------
8642 begin
8644 Res :=
8650 else
8657 -- Start of processing for Resolve_Intrinsic_Operator
8659 begin
8660 -- We must preserve the original entity in a generic setting, so that
8661 -- the legality of the operation can be verified in an instance.
8676 -- If the result or operand types are private, rewrite with unchecked
8677 -- conversions on the operands and the result, to expose the proper
8678 -- underlying numeric type.
8683 then
8688 else
8709 then
8710 -- Add explicit conversion where needed, and save interpretations in
8711 -- case operands are overloaded.
8718 else
8724 else
8734 else
8739 --------------------------------------
8740 -- Resolve_Intrinsic_Unary_Operator --
8741 --------------------------------------
8743 procedure Resolve_Intrinsic_Unary_Operator
8746 is
8751 begin
8770 else
8775 ------------------------
8776 -- Resolve_Logical_Op --
8777 ------------------------
8782 begin
8785 -- Predefined operations on scalar types yield the base type. On the
8786 -- other hand, logical operations on arrays yield the type of the
8787 -- arguments (and the context).
8791 else
8795 -- The following test is required because the operands of the operation
8796 -- may be literals, in which case the resulting type appears to be
8797 -- compatible with a signed integer type, when in fact it is compatible
8798 -- only with modular types. If the context itself is universal, the
8799 -- operation is illegal.
8807 Error_Msg_N
8815 then
8819 -- Replace AND by AND THEN, or OR by OR ELSE, if Short_Circuit_And_Or
8820 -- is active and the result type is standard Boolean (do not mess with
8821 -- ops that return a nonstandard Boolean type, because something strange
8822 -- is going on).
8824 -- Note: you might expect this replacement to be done during expansion,
8825 -- but that doesn't work, because when the pragma Short_Circuit_And_Or
8826 -- is used, no part of the right operand of an "and" or "or" operator
8827 -- should be executed if the left operand would short-circuit the
8828 -- evaluation of the corresponding "and then" or "or else". If we left
8829 -- the replacement to expansion time, then run-time checks associated
8830 -- with such operands would be evaluated unconditionally, due to being
8831 -- before the condition prior to the rewriting as short-circuit forms
8832 -- during expansion.
8834 if Short_Circuit_And_Or
8837 then
8838 -- Mark the corresponding putative SCO operator as truly a logical
8839 -- (and short-circuit) operator.
8852 -- Case of OR changed to OR ELSE
8854 else
8862 -- Return now, since analysis of the rewritten ops will take care of
8863 -- other reference bookkeeping and expression folding.
8878 -- In SPARK, logical operations AND, OR and XOR for arrays are defined
8879 -- only when both operands have same static lower and higher bounds. Of
8880 -- course the types have to match, so only check if operands are
8881 -- compatible and the node itself has no errors.
8885 then
8886 declare
8890 begin
8891 -- Protect call to Matching_Static_Array_Bounds to avoid costly
8892 -- operation if not needed.
8899 then
8900 Check_SPARK_05_Restriction
8907 ---------------------------
8908 -- Resolve_Membership_Op --
8909 ---------------------------
8911 -- The context can only be a boolean type, and does not determine the
8912 -- arguments. Arguments should be unambiguous, but the preference rule for
8913 -- universal types applies.
8923 -- Analysis has determined a unique type for the left operand. Use it to
8924 -- resolve the disjuncts.
8926 ----------------------------
8927 -- Resolve_Set_Membership --
8928 ----------------------------
8934 begin
8935 -- If the left operand is overloaded, find type compatible with not
8936 -- overloaded alternative of the right operand.
8945 else
8950 -- Unclear how to resolve expression if all alternatives are also
8951 -- overloaded.
8957 else
8966 -- Alternative is an expression, a range
8967 -- or a subtype mark.
8971 then
8978 -- Check for duplicates for discrete case
8981 declare
8990 begin
8991 -- Loop checking duplicates. This is quadratic, but giant sets
8992 -- are unlikely in this context so it's a reasonable choice.
8999 N_Character_Literal)
9001 then
9018 -- RM 4.5.2 (28.1/3) specifies that for types other than records or
9019 -- limited types, evaluation of a membership test uses the predefined
9020 -- equality for the type. This may be confusing to users, and the
9021 -- following warning appears useful for the most common case.
9025 then
9026 Error_Msg_NE
9028 Error_Msg_N
9033 -- Start of processing for Resolve_Membership_Op
9035 begin
9041 Resolve_Set_Membership;
9045 and then
9047 or else
9050 then
9053 -- Ada 2005 (AI-251): Support the following case:
9055 -- type I is interface;
9056 -- type T is tagged ...
9058 -- function Test (O : I'Class) is
9059 -- begin
9060 -- return O in T'Class.
9061 -- end Test;
9063 -- In this case we have nothing else to do. The membership test will be
9064 -- done at run time.
9070 then
9072 else
9076 -- If mixed-mode operations are present and operands are all literal,
9077 -- the only interpretation involves Duration, which is probably not
9078 -- the intention of the programmer.
9093 then
9099 else
9104 -- Here after resolving membership operation
9111 ------------------
9112 -- Resolve_Null --
9113 ------------------
9118 begin
9119 -- Handle restriction against anonymous null access values This
9120 -- restriction can be turned off using -gnatdj.
9122 -- Ada 2005 (AI-231): Remove restriction
9125 and then not Debug_Flag_J
9128 then
9129 -- In the common case of a call which uses an explicitly null value
9130 -- for an access parameter, give specialized error message.
9133 Error_Msg_N
9136 -- Standard message for all other cases (are there any?)
9138 else
9139 Error_Msg_N
9144 -- Ada 2005 (AI-231): Generate the null-excluding check in case of
9145 -- assignment to a null-excluding object.
9150 then
9153 -- Decide whether to generate an if_statement around our
9154 -- null-excluding check to avoid them on certain internal object
9155 -- declarations by looking at the type the current Init_Proc
9156 -- belongs to.
9158 -- Generate:
9159 -- if T1b_skip_null_excluding_check then
9160 -- [constraint_error "access check failed"]
9161 -- end if;
9163 if Needs_Conditional_Null_Excluding_Check
9165 then
9168 Condition =>
9170 New_External_Name
9172 Then_Statements =>
9173 New_List (
9177 -- Otherwise, simply create the check
9179 else
9184 else
9185 Insert_Action
9193 -- In a distributed context, null for a remote access to subprogram may
9194 -- need to be replaced with a special record aggregate. In this case,
9195 -- return after having done the transformation.
9200 then
9204 -- The null literal takes its type from the context
9209 -----------------------
9210 -- Resolve_Op_Concat --
9211 -----------------------
9215 -- We wish to avoid deep recursion, because concatenations are often
9216 -- deeply nested, as in A&B&...&Z. Therefore, we walk down the left
9217 -- operands nonrecursively until we find something that is not a simple
9218 -- concatenation (A in this case). We resolve that, and then walk back
9219 -- up the tree following Parent pointers, calling Resolve_Op_Concat_Rest
9220 -- to do the rest of the work at each level. The Parent pointers allow
9221 -- us to avoid recursion, and thus avoid running out of memory. See also
9222 -- Sem_Ch4.Analyze_Concatenation, where a similar approach is used.
9227 begin
9228 -- The following code is equivalent to:
9230 -- Resolve_Op_Concat_First (NN, Typ);
9231 -- Resolve_Op_Concat_Arg (N, ...);
9232 -- Resolve_Op_Concat_Rest (N, Typ);
9234 -- where the Resolve_Op_Concat_Arg call recurses back here if the left
9235 -- operand is a concatenation.
9237 -- Walk down left operands
9239 loop
9248 -- Now (given the above example) NN is A&B and Op1 is A
9250 -- First resolve Op1 ...
9254 -- ... then walk NN back up until we reach N (where we started), calling
9255 -- Resolve_Op_Concat_Rest along the way.
9257 loop
9264 Check_SPARK_05_Restriction
9269 ---------------------------
9270 -- Resolve_Op_Concat_Arg --
9271 ---------------------------
9273 procedure Resolve_Op_Concat_Arg
9278 is
9282 begin
9284 if Is_Comp
9288 then
9290 else
9294 -- If both Array & Array and Array & Component are visible, there is a
9295 -- potential ambiguity that must be reported.
9300 then
9304 -- If the operation is user-defined and not overloaded use its
9305 -- profile. The operation may be a renaming, in which case it has
9306 -- been rewritten, and we want the original profile.
9311 then
9313 Etype
9317 -- Otherwise an aggregate may match both the array type and the
9318 -- component type.
9320 else
9325 else
9329 then
9330 declare
9335 begin
9340 -- Special-case the error message when the overloading is
9341 -- caused by a function that yields an array and can be
9342 -- called without parameters.
9347 Error_Msg_NE
9349 Error_Msg_NE
9353 else
9361 or else
9363 then
9364 Error_Msg_N -- CODEFIX
9382 else
9387 else
9391 -- Concatenation is restricted in SPARK: each operand must be either a
9392 -- string literal, the name of a string constant, a static character or
9393 -- string expression, or another concatenation. Arg cannot be a
9394 -- concatenation here as callers of Resolve_Op_Concat_Arg call it
9395 -- separately on each final operand, past concatenation operations.
9399 Check_SPARK_05_Restriction
9407 then
9408 Check_SPARK_05_Restriction
9412 -- Do not issue error on an operand that is neither a character nor a
9413 -- string, as the error is issued in Resolve_Op_Concat.
9415 else
9422 -----------------------------
9423 -- Resolve_Op_Concat_First --
9424 -----------------------------
9431 begin
9432 -- The parser folds an enormous sequence of concatenations of string
9433 -- literals into "" & "...", where the Is_Folded_In_Parser flag is set
9434 -- in the right operand. If the expression resolves to a predefined "&"
9435 -- operator, all is well. Otherwise, the parser's folding is wrong, so
9436 -- we give an error. See P_Simple_Expression in Par.Ch4.
9441 then
9456 ----------------------------
9457 -- Resolve_Op_Concat_Rest --
9458 ----------------------------
9464 begin
9473 -- If this is not a static concatenation, but the result is a string
9474 -- type (and not an array of strings) ensure that static string operands
9475 -- have their subtypes properly constructed.
9479 then
9485 ----------------------
9486 -- Resolve_Op_Expon --
9487 ----------------------
9492 begin
9493 -- Catch attempts to do fixed-point exponentiation with universal
9494 -- operands, which is a case where the illegality is not caught during
9495 -- normal operator analysis. This is not done in preanalysis mode
9496 -- since the tree is not fully decorated during preanalysis.
9508 then
9518 then
9525 then
9529 -- We do the resolution using the base type, because intermediate values
9530 -- in expressions are always of the base type, not a subtype of it.
9535 -- For integer types, right argument must be in Natural range
9550 -- Evaluate the exponentiation operator for dimensioned type
9553 else
9557 -- Set overflow checking bit. Much cleverer code needed here eventually
9558 -- and perhaps the Resolve routines should be separated for the various
9559 -- arithmetic operations, since they will need different processing. ???
9568 --------------------
9569 -- Resolve_Op_Not --
9570 --------------------
9576 -- This function determines if the parent node is a boolean operator or
9577 -- operation (comparison op, membership test, or short circuit form) and
9578 -- the not in question is the left operand of this operation. Note that
9579 -- if the not is in parens, then false is returned.
9581 -----------------------
9582 -- Parent_Is_Boolean --
9583 -----------------------
9586 begin
9590 else
9592 when N_And_Then
9593 | N_In
9594 | N_Not_In
9595 | N_Op_And
9596 | N_Op_Eq
9597 | N_Op_Ge
9598 | N_Op_Gt
9599 | N_Op_Le
9600 | N_Op_Lt
9601 | N_Op_Ne
9602 | N_Op_Or
9603 | N_Op_Xor
9604 | N_Or_Else
9605 =>
9614 -- Start of processing for Resolve_Op_Not
9616 begin
9617 -- Predefined operations on scalar types yield the base type. On the
9618 -- other hand, logical operations on arrays yield the type of the
9619 -- arguments (and the context).
9623 else
9627 -- Straightforward case of incorrect arguments
9634 -- Special case of probable missing parens
9638 Error_Msg_N
9641 else
9642 Error_Msg_N
9649 -- OK resolution of NOT
9651 else
9652 -- Warn if non-boolean types involved. This is a case like not a < b
9653 -- where a and b are modular, where we will get (not a) < b and most
9654 -- likely not (a < b) was intended.
9656 if Warn_On_Questionable_Missing_Parens
9658 and then Parent_Is_Boolean
9659 then
9663 -- Warn on double negation if checking redundant constructs
9665 if Warn_On_Redundant_Constructs
9670 then
9674 -- Complete resolution and evaluation of NOT
9684 -----------------------------
9685 -- Resolve_Operator_Symbol --
9686 -----------------------------
9688 -- Nothing to be done, all resolved already
9694 begin
9698 ----------------------------------
9699 -- Resolve_Qualified_Expression --
9700 ----------------------------------
9708 begin
9711 -- Protect call to Matching_Static_Array_Bounds to avoid costly
9712 -- operation if not needed.
9719 then
9720 Check_SPARK_05_Restriction
9724 -- A qualified expression requires an exact match of the type, class-
9725 -- wide matching is not allowed. However, if the qualifying type is
9726 -- specific and the expression has a class-wide type, it may still be
9727 -- okay, since it can be the result of the expansion of a call to a
9728 -- dispatching function, so we also have to check class-wideness of the
9729 -- type of the expression's original node.
9732 or else
9736 then
9740 -- If the target type is unconstrained, then we reset the type of the
9741 -- result from the type of the expression. For other cases, the actual
9742 -- subtype of the expression is the target type.
9746 then
9753 -- If we still have a qualified expression after the static evaluation,
9754 -- then apply a scalar range check if needed. The reason that we do this
9755 -- after the Eval call is that otherwise, the application of the range
9756 -- check may convert an illegal static expression and result in warning
9757 -- rather than giving an error (e.g Integer'(Integer'Last + 1)).
9763 -- Finally, check whether a predicate applies to the target type. This
9764 -- comes from AI12-0100. As for type conversions, check the enclosing
9765 -- context to prevent an infinite expansion.
9771 or else
9773 then
9776 -- In the case of a qualified expression in an allocator, the check
9777 -- is applied when expanding the allocator, so avoid redundant check.
9781 then
9787 ------------------------------
9788 -- Resolve_Raise_Expression --
9789 ------------------------------
9792 begin
9796 else
9801 -------------------
9802 -- Resolve_Range --
9803 -------------------
9810 -- Returns True if N is of the form X'First .. X'Last where X is the
9811 -- same entity for both attributes.
9813 --------------------
9814 -- First_Last_Ref --
9815 --------------------
9821 begin
9826 then
9827 declare
9830 begin
9834 then
9843 -- Start of processing for Resolve_Range
9845 begin
9848 -- The lower bound should be in Typ. The higher bound can be in Typ's
9849 -- base type if the range is null. It may still be invalid if it is
9850 -- higher than the lower bound. This is checked later in the context in
9851 -- which the range appears.
9856 -- Reanalyze the lower bound after both bounds have been analyzed, so
9857 -- that the range is known to be static or not by now. This may trigger
9858 -- more compile-time evaluation, which is useful for static analysis
9859 -- with GNATprove. This is not needed for compilation or static analysis
9860 -- with CodePeer, as full expansion does that evaluation then.
9867 -- Check for inappropriate range on unordered enumeration type
9871 -- Exclude X'First .. X'Last if X is the same entity for both
9873 and then not First_Last_Ref
9874 then
9876 Error_Msg_NE
9883 -- We have to check the bounds for being within the base range as
9884 -- required for a non-static context. Normally this is automatic and
9885 -- done as part of evaluating expressions, but the N_Range node is an
9886 -- exception, since in GNAT we consider this node to be a subexpression,
9887 -- even though in Ada it is not. The circuit in Sem_Eval could check for
9888 -- this, but that would put the test on the main evaluation path for
9889 -- expressions.
9894 -- Check for an ambiguous range over character literals. This will
9895 -- happen with a membership test involving only literals.
9903 -- If bounds are static, constant-fold them, so size computations are
9904 -- identical between front-end and back-end. Do not perform this
9905 -- transformation while analyzing generic units, as type information
9906 -- would be lost when reanalyzing the constant node in the instance.
9919 --------------------------
9920 -- Resolve_Real_Literal --
9921 --------------------------
9926 begin
9927 -- Special processing for fixed-point literals to make sure that the
9928 -- value is an exact multiple of small where this is required. We skip
9929 -- this for the universal real case, and also for generic types.
9935 then
9936 declare
9943 begin
9944 -- Case of literal is not an exact multiple of the Small
9948 -- For a source program literal for a decimal fixed-point type,
9949 -- this is statically illegal (RM 4.9(36)).
9954 then
9958 -- Generate a warning if literal from source
9961 and then Warn_On_Bad_Fixed_Value
9962 then
9963 Error_Msg_N
9965 N);
9968 -- Replace literal by a value that is the exact representation
9969 -- of a value of the type, i.e. a multiple of the small value,
9970 -- by truncation, since Machine_Rounds is false for all GNAT
9971 -- fixed-point types (RM 4.9(38)).
9981 -- In all cases, set the corresponding integer field
9987 -- Now replace the actual type by the expected type as usual
9993 -----------------------
9994 -- Resolve_Reference --
9995 -----------------------
10000 begin
10001 -- Replace general access with specific type
10009 -- If we are taking the reference of a volatile entity, then treat it as
10010 -- a potential modification of this entity. This is too conservative,
10011 -- but necessary because remove side effects can cause transformations
10012 -- of normal assignments into reference sequences that otherwise fail to
10013 -- notice the modification.
10020 --------------------------------
10021 -- Resolve_Selected_Component --
10022 --------------------------------
10037 -- Check whether this is the initialization of a component within an
10038 -- init proc (by assignment or call to another init proc). If true,
10039 -- there is no need for a discriminant check.
10041 --------------------
10042 -- Init_Component --
10043 --------------------
10046 begin
10047 return Inside_Init_Proc
10053 -- Start of processing for Resolve_Selected_Component
10055 begin
10058 -- Use the context type to select the prefix that has a selector
10059 -- of the correct name and type.
10067 else
10071 -- Locate selected component. For a private prefix the selector
10072 -- can denote a discriminant.
10076 -- The visible components of a class-wide type are those of
10077 -- the root type.
10087 then
10094 else
10098 Error_Msg_N
10103 else
10106 -- There may be an implicit dereference. Retrieve
10107 -- designated record type.
10111 else
10117 -- Resolution chooses the new interpretation.
10118 -- Find the component with the right name.
10123 loop
10140 -- There must be a legal interpretation at this point
10147 -- The type of the context and that of the component are
10148 -- compatible and in general identical, but if they are anonymous
10149 -- access-to-subprogram types, the relevant type is that of the
10150 -- component. This matters in Unnest_Subprograms mode, where the
10151 -- relevant context is the one in which the type is declared, not
10152 -- the point of use. This determines what activation record to use.
10158 else
10159 -- Resolve prefix with its type
10164 -- Generate cross-reference. We needed to wait until full overloading
10165 -- resolution was complete to do this, since otherwise we can't tell if
10166 -- we are an lvalue or not.
10170 else
10174 -- If prefix is an access type, the node will be transformed into an
10175 -- explicit dereference during expansion. The type of the node is the
10176 -- designated type of that of the prefix.
10181 else
10185 -- Set flag for expander if discriminant check required on a component
10186 -- appearing within a variant.
10192 and then
10195 and then not Init_Component
10196 then
10204 -- If the prefix is a record conversion, this may be a renamed
10205 -- discriminant whose bounds differ from those of the original
10206 -- one, so we must ensure that a range check is performed.
10211 then
10215 -- Note: No Eval processing is required, because the prefix is of a
10216 -- record type, or protected type, and neither can possibly be static.
10218 -- If the record type is atomic, and the component is non-atomic, then
10219 -- this is worth a warning, since we have a situation where the access
10220 -- to the component may cause extra read/writes of the atomic array
10221 -- object, or partial word accesses, both of which may be unexpected.
10227 then
10228 Error_Msg_N
10231 Error_Msg_N
10239 -------------------
10240 -- Resolve_Shift --
10241 -------------------
10248 begin
10249 -- We do the resolution using the base type, because intermediate values
10250 -- in expressions always are of the base type, not a subtype of it.
10263 ---------------------------
10264 -- Resolve_Short_Circuit --
10265 ---------------------------
10272 begin
10273 -- Ensure all actions associated with the left operand (e.g.
10274 -- finalization of transient objects) are fully evaluated locally within
10275 -- an expression with actions. This is particularly helpful for coverage
10276 -- analysis. However this should not happen in generics or if option
10277 -- Minimize_Expression_With_Actions is set.
10280 declare
10282 begin
10290 -- Set Comes_From_Source on L to preserve warnings for unset
10291 -- reference.
10300 -- Check for issuing warning for always False assert/check, this happens
10301 -- when assertions are turned off, in which case the pragma Assert/Check
10302 -- was transformed into:
10304 -- if False and then <condition> then ...
10306 -- and we detect this pattern
10308 if Warn_On_Assertion_Failure
10315 then
10316 declare
10319 begin
10320 -- Special handling of Asssert pragma
10324 then
10325 declare
10327 Original_Node
10328 (Expression
10331 begin
10332 -- Don't warn if original condition is explicit False,
10333 -- since obviously the failure is expected in this case.
10337 then
10340 -- Issue warning. We do not want the deletion of the
10341 -- IF/AND-THEN to take this message with it. We achieve this
10342 -- by making sure that the expanded code points to the Sloc
10343 -- of the expression, not the original pragma.
10345 else
10346 -- Note: Use Error_Msg_F here rather than Error_Msg_N.
10347 -- The source location of the expression is not usually
10348 -- the best choice here. For example, it gets located on
10349 -- the last AND keyword in a chain of boolean expressiond
10350 -- AND'ed together. It is best to put the message on the
10351 -- first character of the assertion, which is the effect
10352 -- of the First_Node call here.
10354 Error_Msg_F
10356 Expression
10361 -- Similar processing for Check pragma
10365 then
10366 -- Don't want to warn if original condition is explicit False
10368 declare
10370 Original_Node
10371 (Expression
10373 begin
10376 then
10379 -- Post warning
10381 else
10382 -- Again use Error_Msg_F rather than Error_Msg_N, see
10383 -- comment above for an explanation of why we do this.
10385 Error_Msg_F
10387 Expression
10395 -- Continue with processing of short circuit
10404 -------------------
10405 -- Resolve_Slice --
10406 -------------------
10415 begin
10418 -- Use the context type to select the prefix that yields the correct
10419 -- array type.
10421 declare
10428 begin
10436 then
10444 else
10449 else
10460 else
10470 -- If the prefix is an access to an unconstrained array, we must use
10471 -- the actual subtype of the object to perform the index checks. The
10472 -- object denoted by the prefix is implicit in the node, so we build
10473 -- an explicit representation for it in order to compute the actual
10474 -- subtype.
10479 declare
10483 begin
10494 then
10497 -- If the name is a selected component that depends on discriminants,
10498 -- build an actual subtype for it. This can happen only when the name
10499 -- itself is overloaded; otherwise the actual subtype is created when
10500 -- the selected component is analyzed.
10503 and then Full_Analysis
10505 then
10506 declare
10509 begin
10514 -- Maybe this should just be "else", instead of checking for the
10515 -- specific case of slice??? This is needed for the case where the
10516 -- prefix is an Image attribute, which gets expanded to a slice, and so
10517 -- has a constrained subtype which we want to use for the slice range
10518 -- check applied below (the range check won't get done if the
10519 -- unconstrained subtype of the 'Image is used).
10525 -- Obtain the type of the array index
10529 else
10533 -- If name was overloaded, set slice type correctly now
10537 -- Handle the generation of a range check that compares the array index
10538 -- against the discrete_range. The check is not applied to internally
10539 -- built nodes associated with the expansion of dispatch tables. Check
10540 -- that Ada.Tags has already been loaded to avoid extra dependencies on
10541 -- the unit.
10543 if Tagged_Type_Expansion
10549 then
10552 -- The discrete_range is specified by a subtype indication. Create a
10553 -- shallow copy and inherit the type, parent and source location from
10554 -- the discrete_range. This ensures that the range check is inserted
10555 -- relative to the slice and that the runtime exception points to the
10556 -- proper construct.
10565 -- The discrete_range is a regular range. Resolve the bounds and remove
10566 -- their side effects.
10568 else
10585 -- Check bad use of type with predicates
10587 declare
10590 begin
10593 then
10595 else
10600 Bad_Predicated_Subtype_Use
10605 -- Otherwise here is where we check suspicious indexes
10616 ----------------------------
10617 -- Resolve_String_Literal --
10618 ----------------------------
10629 begin
10630 -- For a string appearing in a concatenation, defer creation of the
10631 -- string_literal_subtype until the end of the resolution of the
10632 -- concatenation, because the literal may be constant-folded away. This
10633 -- is a useful optimization for long concatenation expressions.
10635 -- If the string is an aggregate built for a single character (which
10636 -- happens in a non-static context) or a is null string to which special
10637 -- checks may apply, we build the subtype. Wide strings must also get a
10638 -- string subtype if they come from a one character aggregate. Strings
10639 -- generated by attributes might be static, but it is often hard to
10640 -- determine whether the enclosing context is static, so we generate
10641 -- subtypes for them as well, thus losing some rarer optimizations ???
10642 -- Same for strings that come from a static conversion.
10644 Need_Check :=
10653 -- If the resolving type is itself a string literal subtype, we can just
10654 -- reuse it, since there is no point in creating another.
10660 and then not Need_Check
10662 N_Attribute_Reference,
10663 N_Qualified_Expression,
10664 N_Type_Conversion)
10665 then
10668 -- Do not generate a string literal subtype for the default expression
10669 -- of a formal parameter in GNATprove mode. This is because the string
10670 -- subtype is associated with the freezing actions of the subprogram,
10671 -- however freezing is disabled in GNATprove mode and as a result the
10672 -- subtype is unavailable.
10674 elsif GNATprove_Mode
10676 then
10679 -- Otherwise we must create a string literal subtype. Note that the
10680 -- whole idea of string literal subtypes is simply to avoid the need
10681 -- for building a full fledged array subtype for each literal.
10683 else
10689 or else Need_Check
10690 then
10697 then
10703 -- The validity of a null string has been checked in the call to
10704 -- Eval_String_Literal.
10709 -- Always accept string literal with component type Any_Character, which
10710 -- occurs in error situations and in comparisons of literals, both of
10711 -- which should accept all literals.
10716 -- If the type is bit-packed, then we always transform the string
10717 -- literal into a full fledged aggregate.
10722 -- Deal with cases of Wide_Wide_String, Wide_String, and String
10724 else
10725 -- For Standard.Wide_Wide_String, or any other type whose component
10726 -- type is Standard.Wide_Wide_Character, we know that all the
10727 -- characters in the string must be acceptable, since the parser
10728 -- accepted the characters as valid character literals.
10733 -- For the case of Standard.String, or any other type whose component
10734 -- type is Standard.Character, we must make sure that there are no
10735 -- wide characters in the string, i.e. that it is entirely composed
10736 -- of characters in range of type Character.
10738 -- If the string literal is the result of a static concatenation, the
10739 -- test has already been performed on the components, and need not be
10740 -- repeated.
10744 then
10748 -- If we are out of range, post error. This is one of the
10749 -- very few places that we place the flag in the middle of
10750 -- a token, right under the offending wide character. Not
10751 -- quite clear if this is right wrt wide character encoding
10752 -- sequences, but it's only an error message.
10754 Error_Msg
10761 -- For the case of Standard.Wide_String, or any other type whose
10762 -- component type is Standard.Wide_Character, we must make sure that
10763 -- there are no wide characters in the string, i.e. that it is
10764 -- entirely composed of characters in range of type Wide_Character.
10766 -- If the string literal is the result of a static concatenation,
10767 -- the test has already been performed on the components, and need
10768 -- not be repeated.
10772 then
10776 -- If we are out of range, post error. This is one of the
10777 -- very few places that we place the flag in the middle of
10778 -- a token, right under the offending wide character.
10780 -- This is not quite right, because characters in general
10781 -- will take more than one character position ???
10783 Error_Msg
10790 -- If the root type is not a standard character, then we will convert
10791 -- the string into an aggregate and will let the aggregate code do
10792 -- the checking. Standard Wide_Wide_Character is also OK here.
10794 else
10798 -- See if the component type of the array corresponding to the string
10799 -- has compile time known bounds. If yes we can directly check
10800 -- whether the evaluation of the string will raise constraint error.
10801 -- Otherwise we need to transform the string literal into the
10802 -- corresponding character aggregate and let the aggregate code do
10803 -- the checking. We use the same transformation if the component
10804 -- type has a static predicate, which will be applied to each
10805 -- character when the aggregate is resolved.
10809 -- Check for the case of full range, where we are definitely OK
10815 -- Here the range is not the complete base type range, so check
10817 declare
10825 begin
10828 then
10834 then
10835 Apply_Compile_Time_Constraint_Error
10837 CE_Range_Check_Failed,
10850 -- If we got here we meed to transform the string literal into the
10851 -- equivalent qualified positional array aggregate. This is rather
10852 -- heavy artillery for this situation, but it is hard work to avoid.
10854 declare
10859 begin
10860 -- Build the character literals, we give them source locations that
10861 -- correspond to the string positions, which is a bit tricky given
10862 -- the possible presence of wide character escape sequences.
10876 -- Should we have a call to Skip_Wide here ???
10878 -- ??? else
10879 -- Skip_Wide (P);
10887 Expression =>
10894 -------------------------
10895 -- Resolve_Target_Name --
10896 -------------------------
10899 begin
10903 -----------------------------
10904 -- Resolve_Type_Conversion --
10905 -----------------------------
10917 -- Set to False to suppress cases where we want to suppress the test
10918 -- for redundancy to avoid possible false positives on this warning.
10920 begin
10921 if not Conv_OK
10923 then
10927 -- If the Operand Etype is Universal_Fixed, then the conversion is
10928 -- never redundant. We need this check because by the time we have
10929 -- finished the rather complex transformation, the conversion looks
10930 -- redundant when it is not.
10935 -- If the operand is marked as Any_Fixed, then special processing is
10936 -- required. This is also a case where we suppress the test for a
10937 -- redundant conversion, since most certainly it is not redundant.
10942 -- Mixed-mode operation involving a literal. Context must be a fixed
10943 -- type which is applied to the literal subsequently.
10945 -- Multiplication and division involving two fixed type operands must
10946 -- yield a universal real because the result is computed in arbitrary
10947 -- precision.
10953 then
10959 or else
10961 then
10962 -- Return if expression is ambiguous
10967 -- If nothing else, the available fixed type is Duration
10969 else
10973 -- Resolve the real operand with largest available precision
10977 else
10983 -- If the operand is a literal (it could be a non-static and
10984 -- illegal exponentiation) check whether the use of Duration
10985 -- is potentially inaccurate.
10990 then
10991 Error_Msg_N
10994 Error_Msg_N
11001 then
11004 else
11013 -- In SPARK, a type conversion between array types should be restricted
11014 -- to types which have matching static bounds.
11016 -- Protect call to Matching_Static_Array_Bounds to avoid costly
11017 -- operation if not needed.
11024 then
11025 Check_SPARK_05_Restriction
11029 -- In formal mode, the operand of an ancestor type conversion must be an
11030 -- object (not an expression).
11038 then
11044 -- Note: we do the Eval_Type_Conversion call before applying the
11045 -- required checks for a subtype conversion. This is important, since
11046 -- both are prepared under certain circumstances to change the type
11047 -- conversion to a constraint error node, but in the case of
11048 -- Eval_Type_Conversion this may reflect an illegality in the static
11049 -- case, and we would miss the illegality (getting only a warning
11050 -- message), if we applied the type conversion checks first.
11054 -- Even when evaluation is not possible, we may be able to simplify the
11055 -- conversion or its expression. This needs to be done before applying
11056 -- checks, since otherwise the checks may use the original expression
11057 -- and defeat the simplifications. This is specifically the case for
11058 -- elimination of the floating-point Truncation attribute in
11059 -- float-to-int conversions.
11063 -- If after evaluation we still have a type conversion, then we may need
11064 -- to apply checks required for a subtype conversion.
11066 -- Skip these type conversion checks if universal fixed operands
11067 -- operands involved, since range checks are handled separately for
11068 -- these cases (in the appropriate Expand routines in unit Exp_Fixd).
11074 then
11078 -- Issue warning for conversion of simple object to its own type. We
11079 -- have to test the original nodes, since they may have been rewritten
11080 -- by various optimizations.
11084 -- Here we test for a redundant conversion if the warning mode is
11085 -- active (and was not locally reset), and we have a type conversion
11086 -- from source not appearing in a generic instance.
11088 if Test_Redundant
11091 and then not In_Instance
11092 then
11096 -- If the node is part of a larger expression, the Target_Type
11097 -- may not be the original type of the node if the context is a
11098 -- condition. Recover original type to see if conversion is needed.
11102 then
11106 -- If we have an entity name, then give the warning if the entity
11107 -- is the right type, or if it is a loop parameter covered by the
11108 -- original type (that's needed because loop parameters have an
11109 -- odd subtype coming from the bounds).
11112 and then
11114 or else
11118 -- If not an entity, then type of expression must match
11121 then
11122 -- One more check, do not give warning if the analyzed conversion
11123 -- has an expression with non-static bounds, and the bounds of the
11124 -- target are static. This avoids junk warnings in cases where the
11125 -- conversion is necessary to establish staticness, for example in
11126 -- a case statement.
11130 then
11133 -- Finally, if this type conversion occurs in a context requiring
11134 -- a prefix, and the expression is a qualified expression then the
11135 -- type conversion is not redundant, since a qualified expression
11136 -- is not a prefix, whereas a type conversion is. For example, "X
11137 -- := T'(Funx(...)).Y;" is illegal because a selected component
11138 -- requires a prefix, but a type conversion makes it legal: "X :=
11139 -- T(T'(Funx(...))).Y;"
11141 -- In Ada 2012, a qualified expression is a name, so this idiom is
11142 -- no longer needed, but we still suppress the warning because it
11143 -- seems unfriendly for warnings to pop up when you switch to the
11144 -- newer language version.
11148 N_Indexed_Component,
11149 N_Selected_Component,
11150 N_Slice,
11151 N_Explicit_Dereference)
11152 then
11155 -- Never warn on conversion to Long_Long_Integer'Base since
11156 -- that is most likely an artifact of the extended overflow
11157 -- checking and comes from complex expanded code.
11162 -- Here we give the redundant conversion warning. If it is an
11163 -- entity, give the name of the entity in the message. If not,
11164 -- just mention the expression.
11166 -- Shoudn't we test Warn_On_Redundant_Constructs here ???
11168 else
11171 Error_Msg_NE -- CODEFIX
11174 else
11175 Error_Msg_NE
11183 -- Ada 2005 (AI-251): Handle class-wide interface type conversions.
11184 -- No need to perform any interface conversion if the type of the
11185 -- expression coincides with the target type.
11188 and then Expander_Active
11190 then
11191 declare
11195 begin
11196 -- If the type of the operand is a limited view, use nonlimited
11197 -- view when available. If it is a class-wide type, recover the
11198 -- class-wide type of the nonlimited view.
11202 then
11218 -- Conversion from interface type
11222 -- Ada 2005 (AI-217): Handle entities from limited views
11226 Error_Msg_NE -- CODEFIX
11229 Error_Msg_N
11231 N);
11234 and then not
11237 then
11239 Error_Msg_NE -- CODEFIX
11242 Error_Msg_N
11244 N);
11246 else
11250 -- Conversion to interface type
11254 -- Handle subtypes
11261 or else Interface_Present_In_Ancestor
11264 then
11266 else
11269 Error_Msg_N
11277 -- Ada 2012: once the type conversion is resolved, check whether the
11278 -- operand statisfies the static predicate of the target type.
11284 -- If at this stage we have a real to integer conversion, make sure that
11285 -- the Do_Range_Check flag is set, because such conversions in general
11286 -- need a range check. We only need this if expansion is off.
11287 -- In GNATprove mode, we only do that when converting from fixed-point
11288 -- (as floating-point to integer conversions are now handled in
11289 -- GNATprove mode).
11292 and then not Expander_Active
11297 then
11301 -- Generating C code a type conversion of an access to constrained
11302 -- array type to access to unconstrained array type involves building
11303 -- a fat pointer which in general cannot be generated on the fly. We
11304 -- remove side effects in order to store the result of the conversion
11305 -- into a temporary.
11307 if Modify_Tree_For_C
11314 then
11319 ----------------------
11320 -- Resolve_Unary_Op --
11321 ----------------------
11330 begin
11333 Check_SPARK_05_Restriction
11337 -- Deal with intrinsic unary operators
11343 then
11348 -- Deal with universal cases
11351 or else
11353 then
11360 -- Generate warning for expressions like abs (x mod 2)
11362 if Warn_On_Redundant_Constructs
11364 then
11368 Error_Msg_N -- CODEFIX
11373 -- Deal with reference generation
11380 -- Set overflow checking bit. Much cleverer code needed here eventually
11381 -- and perhaps the Resolve routines should be separated for the various
11382 -- arithmetic operations, since they will need different processing ???
11390 -- Generate warning for expressions like -5 mod 3 for integers. No need
11391 -- to worry in the floating-point case, since parens do not affect the
11392 -- result so there is no point in giving in a warning.
11394 declare
11403 begin
11404 if Warn_On_Questionable_Missing_Parens
11408 then
11411 -- We are looking for cases where the right operand is not
11412 -- parenthesized, and is a binary operator, multiply, divide, or
11413 -- mod. These are the cases where the grouping can affect results.
11417 then
11418 -- For mod, we always give the warning, since the value is
11419 -- affected by the parenthesization (e.g. (-5) mod 315 /=
11420 -- -(5 mod 315)). But for the other cases, the only concern is
11421 -- overflow, e.g. for the case of 8 big signed (-(2 * 64)
11422 -- overflows, but (-2) * 64 does not). So we try to give the
11423 -- message only when overflow is possible.
11427 then
11432 else
11438 else
11442 -- Note that the test below is deliberately excluding the
11443 -- largest negative number, since that is a potentially
11444 -- troublesome case (e.g. -2 * x, where the result is the
11445 -- largest negative integer has an overflow with 2 * x).
11452 -- For the multiplication case, the only case we have to worry
11453 -- about is when (-a)*b is exactly the largest negative number
11454 -- so that -(a*b) can cause overflow. This can only happen if
11455 -- a is a power of 2, and more generally if any operand is a
11456 -- constant that is not a power of 2, then the parentheses
11457 -- cannot affect whether overflow occurs. We only bother to
11458 -- test the left most operand
11460 -- Loop looking at left operands for one that has known value
11467 -- Operand value of 0 or 1 skips warning
11472 -- Otherwise check power of 2, if power of 2, warn, if
11473 -- anything else, skip warning.
11475 else
11479 else
11488 -- Keep looking at left operands
11493 -- For rem or "/" we can only have a problematic situation
11494 -- if the divisor has a value of minus one or one. Otherwise
11495 -- overflow is impossible (divisor > 1) or we have a case of
11496 -- division by zero in any case.
11501 then
11505 -- If we fall through warning should be issued
11507 -- Shouldn't we test Warn_On_Questionable_Missing_Parens ???
11509 Error_Msg_N
11516 ----------------------------------
11517 -- Resolve_Unchecked_Expression --
11518 ----------------------------------
11520 procedure Resolve_Unchecked_Expression
11523 is
11524 begin
11529 ---------------------------------------
11530 -- Resolve_Unchecked_Type_Conversion --
11531 ---------------------------------------
11533 procedure Resolve_Unchecked_Type_Conversion
11536 is
11542 begin
11543 -- Resolve operand using its own type
11547 -- In an inlined context, the unchecked conversion may be applied
11548 -- to a literal, in which case its type is the type of the context.
11549 -- (In other contexts conversions cannot apply to literals).
11551 if In_Inlined_Body
11555 then
11563 ------------------------------
11564 -- Rewrite_Operator_As_Call --
11565 ------------------------------
11572 begin
11579 New_N :=
11590 ------------------------------
11591 -- Rewrite_Renamed_Operator --
11592 ------------------------------
11594 procedure Rewrite_Renamed_Operator
11598 is
11603 begin
11604 -- Do not perform this transformation within a pre/postcondition,
11605 -- because the expression will be reanalyzed, and the transformation
11606 -- might affect the visibility of the operator, e.g. in an instance.
11607 -- Note that fully analyzed and expanded pre/postconditions appear as
11608 -- pragma Check equivalents.
11614 -- Likewise when an expression function is being preanalyzed, since the
11615 -- expression will be reanalyzed as part of the generated body.
11618 declare
11620 begin
11623 N_Expression_Function
11624 then
11630 -- Rewrite the operator node using the real operator, not its renaming.
11631 -- Exclude user-defined intrinsic operations of the same name, which are
11632 -- treated separately and rewritten as calls.
11641 -- Indicate that both the original entity and its renaming are
11642 -- referenced at this point.
11653 -- If the context type is private, add the appropriate conversions so
11654 -- that the operator is applied to the full view. This is done in the
11655 -- routines that resolve intrinsic operators.
11659 when N_Op_Add
11660 | N_Op_Divide
11661 | N_Op_Expon
11662 | N_Op_Mod
11663 | N_Op_Multiply
11664 | N_Op_Rem
11665 | N_Op_Subtract
11666 =>
11669 when N_Op_Abs
11670 | N_Op_Minus
11671 | N_Op_Plus
11672 =>
11682 -- Operator renames a user-defined operator of the same name. Use the
11683 -- original operator in the node, which is the one Gigi knows about.
11690 -----------------------
11691 -- Set_Slice_Subtype --
11692 -----------------------
11694 -- Build an implicit subtype declaration to represent the type delivered by
11695 -- the slice. This is an abbreviated version of an array subtype. We define
11696 -- an index subtype for the slice, using either the subtype name or the
11697 -- discrete range of the slice. To be consistent with index usage elsewhere
11698 -- we create a list header to hold the single index. This list is not
11699 -- otherwise attached to the syntax tree.
11710 begin
11716 else
11717 -- We force the evaluation of a range. This is definitely needed in
11718 -- the renamed case, and seems safer to do unconditionally. Note in
11719 -- any case that since we will create and insert an Itype referring
11720 -- to this range, we must make sure any side effect removal actions
11721 -- are inserted before the Itype definition.
11727 -- If the discrete range is given by a subtype indication, the
11728 -- type of the slice is the base of the subtype mark.
11731 declare
11733 begin
11742 -- Take a new copy of Drange (where bounds have been rewritten to
11743 -- reference side-effect-free names). Using a separate tree ensures
11744 -- that further expansion (e.g. while rewriting a slice assignment
11745 -- into a FOR loop) does not attempt to remove side effects on the
11746 -- bounds again (which would cause the bounds in the index subtype
11747 -- definition to refer to temporaries before they are defined) (the
11748 -- reason is that some names are considered side effect free here
11749 -- for the subtype, but not in the context of a loop iteration
11750 -- scheme).
11771 -- The Etype of the existing Slice node is reset to this slice subtype.
11772 -- Its bounds are obtained from its first index.
11776 -- For bit-packed slice subtypes, freeze immediately (except in the case
11777 -- of being in a "spec expression" where we never freeze when we first
11778 -- see the expression).
11783 -- For all other cases insert an itype reference in the slice's actions
11784 -- so that the itype is frozen at the proper place in the tree (i.e. at
11785 -- the point where actions for the slice are analyzed). Note that this
11786 -- is different from freezing the itype immediately, which might be
11787 -- premature (e.g. if the slice is within a transient scope). This needs
11788 -- to be done only if expansion is enabled.
11795 --------------------------------
11796 -- Set_String_Literal_Subtype --
11797 --------------------------------
11805 begin
11817 -- The low bound is set from the low bound of the corresponding index
11818 -- type. Note that we do not store the high bound in the string literal
11819 -- subtype, but it can be deduced if necessary from the length and the
11820 -- low bound.
11825 -- If the lower bound is not static we create a range for the string
11826 -- literal, using the index type and the known length of the literal.
11827 -- The index type is not necessarily Positive, so the upper bound is
11828 -- computed as T'Val (T'Pos (Low_Bound) + L - 1).
11830 else
11831 declare
11837 Prefix =>
11841 Left_Opnd =>
11844 Prefix =>
11846 Expressions =>
11848 Right_Opnd =>
11857 begin
11859 Set_String_Literal_Low_Bound
11862 else
11863 -- If the index type is an enumeration type, build bounds
11864 -- expression with attributes.
11866 Set_String_Literal_Low_Bound
11870 Prefix =>
11877 -- Build bona fide subtype for the string, and wrap it in an
11878 -- unchecked conversion, because the back end expects the
11879 -- String_Literal_Subtype to have a static lower bound.
11881 Index_Subtype :=
11888 -- In this context, the Index_Type may already have a constraint,
11889 -- so use common base type on string subtype. The base type may
11890 -- be used when generating attributes of the string, for example
11891 -- in the context of a slice assignment.
11916 ------------------------------
11917 -- Simplify_Type_Conversion --
11918 ------------------------------
11921 begin
11923 declare
11928 begin
11929 -- Special processing if the conversion is the expression of a
11930 -- Rounding or Truncation attribute reference. In this case we
11931 -- replace:
11933 -- ityp (ftyp'Rounding (x)) or ityp (ftyp'Truncation (x))
11935 -- by
11937 -- ityp (x)
11939 -- with the Float_Truncate flag set to False or True respectively,
11940 -- which is more efficient.
11943 and then
11949 Name_Truncation)
11950 then
11951 declare
11954 begin
11964 -----------------------------
11965 -- Unique_Fixed_Point_Type --
11966 -----------------------------
11970 -- Give error messages for true ambiguity. Messages are posted on node
11971 -- N, and entities T1, T2 are the possible interpretations.
11973 -----------------------
11974 -- Fixed_Point_Error --
11975 -----------------------
11978 begin
11984 -- Local variables
11992 -- Start of processing for Unique_Fixed_Point_Type
11994 begin
11995 -- The operations on Duration are visible, so Duration is always a
11996 -- possible interpretation.
12000 -- Look for fixed-point types in enclosing scopes
12009 then
12013 else
12024 -- Look for visible fixed type declarations in the context
12035 then
12039 else
12054 else
12055 -- When the context is a type conversion, issue the warning on the
12056 -- expression of the conversion because it is the actual operation.
12060 else
12064 Error_Msg_NE
12071 ----------------------
12072 -- Valid_Conversion --
12073 ----------------------
12075 function Valid_Conversion
12080 is
12085 function Conversion_Check
12088 -- Little routine to post Msg if Valid is False, returns Valid value
12091 -- If Report_Errs, then calls Errout.Error_Msg_N with its arguments
12093 procedure Conversion_Error_NE
12097 -- If Report_Errs, then calls Errout.Error_Msg_NE with its arguments
12100 -- Return True if expression is within an instance but is not in one of
12101 -- the actuals of the instantiation. Type conversions within an instance
12102 -- are not rechecked because type visbility may lead to spurious errors,
12103 -- but conversions in an actual for a formal object must be checked.
12105 function Valid_Tagged_Conversion
12108 -- Specifically test for validity of tagged conversions
12111 -- Check index and component conformance, and accessibility levels if
12112 -- the component types are anonymous access types (Ada 2005).
12114 ----------------------
12115 -- Conversion_Check --
12116 ----------------------
12118 function Conversion_Check
12121 is
12122 begin
12123 if not Valid
12125 -- A generic unit has already been analyzed and we have verified
12126 -- that a particular conversion is OK in that context. Since the
12127 -- instance is reanalyzed without relying on the relationships
12128 -- established during the analysis of the generic, it is possible
12129 -- to end up with inconsistent views of private types. Do not emit
12130 -- the error message in such cases. The rest of the machinery in
12131 -- Valid_Conversion still ensures the proper compatibility of
12132 -- target and operand types.
12134 and then not In_Instance_Code
12135 then
12142 ------------------------
12143 -- Conversion_Error_N --
12144 ------------------------
12147 begin
12153 -------------------------
12154 -- Conversion_Error_NE --
12155 -------------------------
12157 procedure Conversion_Error_NE
12161 is
12162 begin
12168 ----------------------
12169 -- In_Instance_Code --
12170 ----------------------
12175 begin
12179 else
12183 -- The expression is part of an actual object if it appears in
12184 -- the generated object declaration in the instance.
12188 then
12191 else
12192 exit when
12201 -- Otherwise the expression appears within the instantiated unit
12207 ----------------------------
12208 -- Valid_Array_Conversion --
12209 ----------------------------
12226 begin
12227 -- Error if wrong number of dimensions
12229 if
12231 then
12232 Conversion_Error_N
12236 -- Number of dimensions matches
12238 else
12239 -- Loop through indexes of the two arrays
12247 -- Error if index types are incompatible
12253 then
12254 Conversion_Error_N
12256 Operand);
12264 -- If component types have same base type, all set
12269 -- Here if base types of components are not the same. The only
12270 -- time this is allowed is if we have anonymous access types.
12272 -- The conversion of arrays of anonymous access types can lead
12273 -- to dangling pointers. AI-392 formalizes the accessibility
12274 -- checks that must be applied to such conversions to prevent
12275 -- out-of-scope references.
12277 elsif Ekind_In
12279 E_Anonymous_Access_Subprogram_Type)
12281 and then
12283 then
12286 then
12289 Conversion_Error_N
12299 -- Conversion not allowed because of accessibility levels
12301 else
12302 Conversion_Error_N
12308 else
12312 -- All other cases where component base types do not match
12314 else
12315 Conversion_Error_N
12317 Operand);
12321 -- Check that component subtypes statically match. For numeric
12322 -- types this means that both must be either constrained or
12323 -- unconstrained. For enumeration types the bounds must match.
12324 -- All of this is checked in Subtypes_Statically_Match.
12326 if not Subtypes_Statically_Match
12328 then
12329 Conversion_Error_N
12338 -----------------------------
12339 -- Valid_Tagged_Conversion --
12340 -----------------------------
12342 function Valid_Tagged_Conversion
12345 is
12346 begin
12347 -- Upward conversions are allowed (RM 4.6(22))
12351 then
12354 -- Downward conversion are allowed if the operand is class-wide
12355 -- (RM 4.6(23)).
12359 then
12364 then
12365 return
12369 -- Ada 2005 (AI-251): The conversion to/from interface types is
12370 -- always valid. The types involved may be class-wide (sub)types.
12374 then
12377 -- If the operand is a class-wide type obtained through a limited_
12378 -- with clause, and the context includes the nonlimited view, use
12379 -- it to determine whether the conversion is legal.
12385 then
12390 then
12393 else
12394 Conversion_Error_NE
12401 -- Start of processing for Valid_Conversion
12403 begin
12407 declare
12415 begin
12416 -- Remove procedure calls, which syntactically cannot appear in
12417 -- this context, but which cannot be removed by type checking,
12418 -- because the context does not impose a type.
12420 -- The node may be labelled overloaded, but still contain only one
12421 -- interpretation because others were discarded earlier. If this
12422 -- is the case, retain the single interpretation if legal.
12430 then
12431 -- More than one candidate interpretation is available
12439 -- When compiling for a system where Address is of a visible
12440 -- integer type, spurious ambiguities can be produced when
12441 -- arithmetic operations have a literal operand and return
12442 -- System.Address or a descendant of it. These ambiguities
12443 -- are usually resolved by the context, but for conversions
12444 -- there is no context type and the removal of the spurious
12445 -- operations must be done explicitly here.
12447 if not Address_Is_Private
12449 then
12474 Conversion_Error_N
12477 -- If the interpretation involves a standard operator, use
12478 -- the location of the type, which may be user-defined.
12482 else
12486 Conversion_Error_N -- CODEFIX
12491 else
12495 Conversion_Error_N -- CODEFIX
12507 -- Deal with conversion of integer type to address if the pragma
12508 -- Allow_Integer_Address is in effect. We convert the conversion to
12509 -- an unchecked conversion in this case and we are all done.
12517 -- If we are within a child unit, check whether the type of the
12518 -- expression has an ancestor in a parent unit, in which case it
12519 -- belongs to its derivation class even if the ancestor is private.
12520 -- See RM 7.3.1 (5.2/3).
12524 -- Numeric types
12528 -- A universal fixed expression can be converted to any numeric type
12533 -- Also no need to check when in an instance or inlined body, because
12534 -- the legality has been established when the template was analyzed.
12535 -- Furthermore, numeric conversions may occur where only a private
12536 -- view of the operand type is visible at the instantiation point.
12537 -- This results in a spurious error if we check that the operand type
12538 -- is a numeric type.
12540 -- Note: in a previous version of this unit, the following tests were
12541 -- applied only for generated code (Comes_From_Source set to False),
12542 -- but in fact the test is required for source code as well, since
12543 -- this situation can arise in source code.
12548 -- Otherwise we need the conversion check
12550 else
12551 return Conversion_Check
12553 or else
12559 -- Array types
12565 then
12566 Conversion_Error_N
12570 else
12574 -- Ada 2005 (AI-251): Internally generated conversions of access to
12575 -- interface types added to force the displacement of the pointer to
12576 -- reference the corresponding dispatch table.
12581 then
12584 -- Ada 2005 (AI-251): Anonymous access types where target references an
12585 -- interface type.
12589 E_Anonymous_Access_Type)
12591 then
12592 -- Check the static accessibility rule of 4.6(17). Note that the
12593 -- check is not enforced when within an instance body, since the
12594 -- RM requires such cases to be caught at run time.
12596 -- If the operand is a rewriting of an allocator no check is needed
12597 -- because there are no accessibility issues.
12605 then
12606 -- In an instance, this is a run-time check, but one we know
12607 -- will fail, so generate an appropriate warning. The raise
12608 -- will be generated by Expand_N_Type_Conversion.
12612 Conversion_Error_N
12614 Operand);
12617 else
12618 Conversion_Error_N
12620 Operand);
12624 -- Special accessibility checks are needed in the case of access
12625 -- discriminants declared for a limited type.
12629 then
12630 -- When the operand is a selected access discriminant the check
12631 -- needs to be made against the level of the object denoted by
12632 -- the prefix of the selected name (Object_Access_Level handles
12633 -- checking the prefix of the operand for this case).
12638 then
12639 -- In an instance, this is a run-time check, but one we know
12640 -- will fail, so generate an appropriate warning. The raise
12641 -- will be generated by Expand_N_Type_Conversion.
12645 Conversion_Error_N
12650 -- Real error if not in instance body
12652 else
12653 Conversion_Error_N
12660 -- The case of a reference to an access discriminant from
12661 -- within a limited type declaration (which will appear as
12662 -- a discriminal) is always illegal because the level of the
12663 -- discriminant is considered to be deeper than any (nameable)
12664 -- access type.
12668 and then
12671 then
12672 Conversion_Error_N
12674 Operand);
12682 -- General and anonymous access types
12685 E_Anonymous_Access_Type)
12686 and then
12687 Conversion_Check
12689 and then not
12691 E_Access_Protected_Subprogram_Type),
12693 then
12696 then
12697 Conversion_Error_N
12702 -- Check the static accessibility rule of 4.6(17). Note that the
12703 -- check is not enforced when within an instance body, since the RM
12704 -- requires such cases to be caught at run time.
12709 N_Object_Declaration
12710 then
12711 -- Ada 2012 (AI05-0149): Perform legality checking on implicit
12712 -- conversions from an anonymous access type to a named general
12713 -- access type. Such conversions are not allowed in the case of
12714 -- access parameters and stand-alone objects of an anonymous
12715 -- access type. The implicit conversion case is recognized by
12716 -- testing that Comes_From_Source is False and that it's been
12717 -- rewritten. The Comes_From_Source test isn't sufficient because
12718 -- nodes in inlined calls to predefined library routines can have
12719 -- Comes_From_Source set to False. (Is there a better way to test
12720 -- for implicit conversions???)
12727 then
12730 -- Implicit conversions aren't allowed for objects of an
12731 -- anonymous access type, since such objects have nonstatic
12732 -- levels in Ada 2012.
12735 N_Object_Declaration
12736 then
12737 Conversion_Error_N
12742 -- Implicit conversions aren't allowed for anonymous access
12743 -- parameters. The "not Is_Local_Anonymous_Access_Type" test
12744 -- is done to exclude anonymous access results.
12748 N_Function_Specification,
12749 N_Procedure_Specification)
12750 then
12751 Conversion_Error_N
12756 -- This is a case where there's an enclosing object whose
12757 -- to which the "statically deeper than" relationship does
12758 -- not apply (such as an access discriminant selected from
12759 -- a dereference of an access parameter).
12763 then
12764 Conversion_Error_N
12769 -- In other cases, the level of the operand's type must be
12770 -- statically less deep than that of the target type, else
12771 -- implicit conversion is disallowed (by RM12-8.6(27.1/3)).
12775 then
12776 Conversion_Error_N
12785 then
12786 -- In an instance, this is a run-time check, but one we know
12787 -- will fail, so generate an appropriate warning. The raise
12788 -- will be generated by Expand_N_Type_Conversion.
12792 Conversion_Error_N
12794 Operand);
12797 -- If not in an instance body, this is a real error
12799 else
12800 -- Avoid generation of spurious error message
12803 Conversion_Error_N
12805 Operand);
12811 -- Special accessibility checks are needed in the case of access
12812 -- discriminants declared for a limited type.
12816 then
12817 -- When the operand is a selected access discriminant the check
12818 -- needs to be made against the level of the object denoted by
12819 -- the prefix of the selected name (Object_Access_Level handles
12820 -- checking the prefix of the operand for this case).
12825 then
12826 -- In an instance, this is a run-time check, but one we know
12827 -- will fail, so generate an appropriate warning. The raise
12828 -- will be generated by Expand_N_Type_Conversion.
12832 Conversion_Error_N
12837 -- If not in an instance body, this is a real error
12839 else
12840 Conversion_Error_N
12847 -- The case of a reference to an access discriminant from
12848 -- within a limited type declaration (which will appear as
12849 -- a discriminal) is always illegal because the level of the
12850 -- discriminant is considered to be deeper than any (nameable)
12851 -- access type.
12854 and then
12857 then
12858 Conversion_Error_N
12860 Operand);
12866 -- In the presence of limited_with clauses we have to use nonlimited
12867 -- views, if available.
12871 -- Helper function to handle limited views
12873 --------------------------
12874 -- Full_Designated_Type --
12875 --------------------------
12880 begin
12881 -- Handle the limited view of a type
12885 then
12887 else
12892 -- Local Declarations
12900 -- Start of processing for Check_Limited
12902 begin
12906 else
12908 Conversion_Error_NE
12913 -- Ada 2005 AI-384: legality rule is symmetric in both
12914 -- designated types. The conversion is legal (with possible
12915 -- constraint check) if either designated type is
12916 -- unconstrained.
12919 or else
12921 and then
12924 then
12925 -- Special case, if Value_Size has been used to make the
12926 -- sizes different, the conversion is not allowed even
12927 -- though the subtypes statically match.
12932 then
12933 Conversion_Error_NE
12936 Conversion_Error_NE
12941 -- Normal case where conversion is allowed
12943 else
12947 else
12948 Error_Msg_NE
12956 -- Access to subprogram types. If the operand is an access parameter,
12957 -- the type has a deeper accessibility that any master, and cannot be
12958 -- assigned. We must make an exception if the conversion is part of an
12959 -- assignment and the target is the return object of an extended return
12960 -- statement, because in that case the accessibility check takes place
12961 -- after the return.
12965 -- Note: this test of Opnd_Type is there to prevent entering this
12966 -- branch in the case of a remote access to subprogram type, which
12967 -- is internally represented as an E_Record_Type.
12970 then
12974 and then
12978 then
12979 Conversion_Error_N
12981 Operand);
12982 Conversion_Error_N
12985 Operand);
12987 Error_Msg_NE
12992 -- Check that the designated types are subtype conformant
12998 -- Check the static accessibility rule of 4.6(20)
13002 then
13003 Conversion_Error_N
13005 Operand);
13007 -- Check that if the operand type is declared in a generic body,
13008 -- then the target type must be declared within that same body
13009 -- (enforces last sentence of 4.6(20)).
13012 declare
13018 begin
13025 Conversion_Error_N
13034 -- Remote access to subprogram types
13038 then
13039 -- It is valid to convert from one RAS type to another provided
13040 -- that their specification statically match.
13042 -- Note: at this point, remote access to subprogram types have been
13043 -- expanded to their E_Record_Type representation, and we need to
13044 -- go back to the original access type definition using the
13045 -- Corresponding_Remote_Type attribute in order to check that the
13046 -- designated profiles match.
13051 Check_Subtype_Conformant
13054 Old_Id =>
13056 Err_Loc =>
13057 N);
13060 -- If it was legal in the generic, it's legal in the instance
13065 -- If both are tagged types, check legality of view conversions
13068 and then
13070 then
13073 -- Types derived from the same root type are convertible
13078 -- In an instance or an inlined body, there may be inconsistent views of
13079 -- the same type, or of types derived from a common root.
13082 and then
13085 then
13088 -- Special check for common access type error case
13092 then
13094 Conversion_Error_NE -- CODEFIX
13098 -- Here we have a real conversion error
13100 else
13101 -- Check for missing regular with_clause when only a limited view of
13102 -- target is available.
13105 Conversion_Error_NE
13108 Conversion_Error_NE
13113 and then In_Package_Body
13114 then
13115 Conversion_Error_NE
13117 Conversion_Error_NE
13121 else
13122 Conversion_Error_NE