| blob | 23a95a46c8e64e1185c2e1dc98e27a1ca244d042 |
1 ------------------------------------------------------------------------------
2 -- --
3 -- GNAT COMPILER COMPONENTS --
4 -- --
5 -- S E M _ R E S --
6 -- --
7 -- B o d y --
8 -- --
9 -- Copyright (C) 1992-2017, 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 -- If the type of the object being initialized uses the secondary stack
122 -- directly or indirectly, create a transient scope for the call to the
123 -- init proc. This is because we do not create transient scopes for the
124 -- initialization of individual components within the init proc itself.
125 -- Could be optimized away perhaps?
128 -- N is the node for a logical operator. If the operator is predefined, and
129 -- the root type of the operands is Standard.Boolean, then a check is made
130 -- for restriction No_Direct_Boolean_Operators. This procedure also handles
131 -- the style check for Style_Check_Boolean_And_Or.
134 -- N is either an indexed component or a selected component. This function
135 -- returns true if the prefix refers to an object that has an address
136 -- clause (the case in which we may want to issue a warning).
139 -- Determine whether E is an access type declared by an access declaration,
140 -- and not an (anonymous) allocator type.
143 -- Utility to check whether the entity for an operator is a predefined
144 -- operator, in which case the expression is left as an operator in the
145 -- tree (else it is rewritten into a call). An instance of an intrinsic
146 -- conversion operation may be given an operator name, but is not treated
147 -- like an operator. Note that an operator that is an imported back-end
148 -- builtin has convention Intrinsic, but is expected to be rewritten into
149 -- a call, so such an operator is not treated as predefined by this
150 -- predicate.
153 -- If a default expression in entry call N depends on the discriminants
154 -- of the task, it must be replaced with a reference to the discriminant
155 -- of the task being called.
157 procedure Resolve_Op_Concat_Arg
162 -- Internal procedure for Resolve_Op_Concat to resolve one operand of
163 -- concatenation operator. The operand is either of the array type or of
164 -- the component type. If the operand is an aggregate, and the component
165 -- type is composite, this is ambiguous if component type has aggregates.
168 -- Does the first part of the work of Resolve_Op_Concat
171 -- Does the "rest" of the work of Resolve_Op_Concat, after the left operand
172 -- has been resolved. See Resolve_Op_Concat for details.
211 function Operator_Kind
214 -- Utility to map the name of an operator into the corresponding Node. Used
215 -- by other node rewriting procedures.
218 -- Resolve actuals of call, and add default expressions for missing ones.
219 -- N is the Node_Id for the subprogram call, and Nam is the entity of the
220 -- called subprogram.
223 -- Called from Resolve_Call, when the prefix denotes an entry or element
224 -- of entry family. Actuals are resolved as for subprograms, and the node
225 -- is rebuilt as an entry call. Also called for protected operations. Typ
226 -- is the context type, which is used when the operation is a protected
227 -- function with no arguments, and the return value is indexed.
230 -- A call to a user-defined intrinsic operator is rewritten as a call to
231 -- the corresponding predefined operator, with suitable conversions. Note
232 -- that this applies only for intrinsic operators that denote predefined
233 -- operators, not ones that are intrinsic imports of back-end builtins.
236 -- Ditto, for arithmetic unary operators
239 -- If an operator node resolves to a call to a user-defined operator,
240 -- rewrite the node as a function call.
242 procedure Make_Call_Into_Operator
246 -- Inverse transformation: if an operator is given in functional notation,
247 -- then after resolving the node, transform into an operator node, so that
248 -- operands are resolved properly. Recall that predefined operators do not
249 -- have a full signature and special resolution rules apply.
251 procedure Rewrite_Renamed_Operator
255 -- An operator can rename another, e.g. in an instantiation. In that
256 -- case, the proper operator node must be constructed and resolved.
259 -- The String_Literal_Subtype is built for all strings that are not
260 -- operands of a static concatenation operation. If the argument is not
261 -- a N_String_Literal node, then the call has no effect.
264 -- Build subtype of array type, with the range specified by the slice
267 -- Called after N has been resolved and evaluated, but before range checks
268 -- have been applied. Currently simplifies a combination of floating-point
269 -- to integer conversion and Rounding or Truncation attribute.
272 -- A universal_fixed expression in an universal context is unambiguous if
273 -- there is only one applicable fixed point type. Determining whether there
274 -- is only one requires a search over all visible entities, and happens
275 -- only in very pathological cases (see 6115-006).
277 -------------------------
278 -- Ambiguous_Character --
279 -------------------------
284 begin
288 -- First the ones in Standard
293 -- Include Wide_Wide_Character in Ada 2005 mode
299 -- Now any other types that match
309 -------------------------
310 -- Analyze_And_Resolve --
311 -------------------------
314 begin
320 begin
325 -- Versions with check(s) suppressed
327 procedure Analyze_And_Resolve
331 is
334 begin
336 declare
338 begin
344 else
345 declare
347 begin
355 and then Scope_Is_Transient
356 then
357 -- This can only happen if a transient scope was created for an inner
358 -- expression, which will be removed upon completion of the analysis
359 -- of an enclosing construct. The transient scope must have the
360 -- suppress status of the enclosing environment, not of this Analyze
361 -- call.
364 Scope_Suppress;
368 procedure Analyze_And_Resolve
371 is
374 begin
376 declare
378 begin
384 else
385 declare
387 begin
396 Scope_Suppress;
400 ----------------------------
401 -- Check_Discriminant_Use --
402 ----------------------------
410 begin
411 -- Any use in a spec-expression is legal
418 -- Discriminant cannot be used to constrain a scalar type
425 then
430 -- The following check catches the unusual case where a
431 -- discriminant appears within an index constraint that is part
432 -- of a larger expression within a constraint on a component,
433 -- e.g. "C : Int range 1 .. F (new A(1 .. D))". For now we only
434 -- check case of record components, and note that a similar check
435 -- should also apply in the case of discriminant constraints
436 -- below. ???
438 -- Note that the check for N_Subtype_Declaration below is to
439 -- detect the valid use of discriminants in the constraints of a
440 -- subtype declaration when this subtype declaration appears
441 -- inside the scope of a record type (which is syntactically
442 -- illegal, but which may be created as part of derived type
443 -- processing for records). See Sem_Ch3.Build_Derived_Record_Type
444 -- for more info.
448 and then not
450 and then
452 N_Subtype_Declaration)
454 then
455 Error_Msg_N
460 -- Detect a common error:
462 -- type R (D : Positive := 100) is record
463 -- Name : String (1 .. D);
464 -- end record;
466 -- The default value causes an object of type R to be allocated
467 -- with room for Positive'Last characters. The RM does not mandate
468 -- the allocation of the maximum size, but that is what GNAT does
469 -- so we should warn the programmer that there is a problem.
478 -- Return True if type T has a large enough range that any
479 -- array whose index type covered the whole range of the type
480 -- would likely raise Storage_Error.
482 ------------------------
483 -- Large_Storage_Type --
484 ------------------------
487 begin
488 -- The type is considered large if its bounds are known at
489 -- compile time and if it requires at least as many bits as
490 -- a Positive to store the possible values.
494 and then
499 -- Start of processing for Check_Large
501 begin
502 -- Check that the Disc has a large range
508 -- If the enclosing type is limited, we allocate only the
509 -- default value, not the maximum, and there is no need for
510 -- a warning.
516 -- Check that it is the high bound
520 then
524 -- Check the array allows a large range at this bound. First
525 -- find the array
539 -- Next, find the dimension
547 loop
556 -- Now, check the dimension has a large range
562 -- Warn about the danger
564 Error_Msg_N
574 -- Legal case is in index or discriminant constraint
577 N_Discriminant_Association)
578 then
580 Error_Msg_N
585 then
586 Error_Msg_N
592 -- Otherwise, context is an expression. It should not be within (i.e. a
593 -- subexpression of) a constraint for a component.
595 else
599 N_Subtype_Indication,
600 N_Entry_Declaration)
601 loop
607 -- If the discriminant is used in an expression that is a bound of a
608 -- scalar type, an Itype is created and the bounds are attached to
609 -- its range, not to the original subtype indication. Such use is of
610 -- course a double fault.
614 N_Derived_Type_Definition)
617 -- The constraint itself may be given by a subtype indication,
618 -- rather than by a more common discrete range.
621 and then
625 then
626 Error_Msg_N
632 --------------------------------
633 -- Check_For_Visible_Operator --
634 --------------------------------
637 begin
639 Error_Msg_NE -- CODEFIX
641 Error_Msg_N -- CODEFIX
646 ----------------------------------
647 -- Check_Fully_Declared_Prefix --
648 ----------------------------------
650 procedure Check_Fully_Declared_Prefix
653 is
654 begin
655 -- Check that the designated type of the prefix of a dereference is
656 -- not an incomplete type. This cannot be done unconditionally, because
657 -- dereferences of private types are legal in default expressions. This
658 -- case is taken care of in Check_Fully_Declared, called below. There
659 -- are also 2005 cases where it is legal for the prefix to be unfrozen.
661 -- This consideration also applies to similar checks for allocators,
662 -- qualified expressions, and type conversions.
664 -- An additional exception concerns other per-object expressions that
665 -- are not directly related to component declarations, in particular
666 -- representation pragmas for tasks. These will be per-object
667 -- expressions if they depend on discriminants or some global entity.
668 -- If the task has access discriminants, the designated type may be
669 -- incomplete at the point the expression is resolved. This resolution
670 -- takes place within the body of the initialization procedure, where
671 -- the discriminant is replaced by its discriminal.
675 then
678 -- Ada 2005 (AI-326): Tagged incomplete types allowed. The wrong usages
679 -- are handled by Analyze_Access_Attribute, Analyze_Assignment,
680 -- Analyze_Object_Renaming, and Freeze_Entity.
686 E_Incomplete_Type
688 then
690 else
695 ------------------------------
696 -- Check_Infinite_Recursion --
697 ------------------------------
704 -- Check whether list of actuals is identical to list of formals of
705 -- called function (which is also the enclosing scope).
707 ------------------------
708 -- Same_Argument_List --
709 ------------------------
716 begin
719 else
737 -- Start of processing for Check_Infinite_Recursion
739 begin
740 -- Special case, if this is a procedure call and is a call to the
741 -- current procedure with the same argument list, then this is for
742 -- sure an infinite recursion and we insert a call to raise SE.
746 and then Same_Argument_List
747 then
748 declare
750 begin
754 then
766 -- If not that special case, search up tree, quitting if we reach a
767 -- construct (e.g. a conditional) that tells us that this is not a
768 -- case for an infinite recursion warning.
771 loop
774 -- If no parent, then we were not inside a subprogram, this can for
775 -- example happen when processing certain pragmas in a spec. Just
776 -- return False in this case.
782 -- Done if we get to subprogram body, this is definitely an infinite
783 -- recursion case if we did not find anything to stop us.
787 -- If appearing in conditional, result is false
790 N_And_Then,
791 N_Case_Expression,
792 N_Case_Statement,
793 N_If_Expression,
794 N_If_Statement)
795 then
800 then
801 -- If the call is the expression of a return statement and the
802 -- actuals are identical to the formals, it's worth a warning.
803 -- However, we skip this if there is an immediately preceding
804 -- raise statement, since the call is never executed.
806 -- Furthermore, this corresponds to a common idiom:
808 -- function F (L : Thing) return Boolean is
809 -- begin
810 -- raise Program_Error;
811 -- return F (L);
812 -- end F;
814 -- for generating a stub function
817 and then Same_Argument_List
818 then
821 -- OK, return statement is in a statement list, look for raise
823 declare
826 begin
827 -- Skip past N_Freeze_Entity nodes generated by expansion
832 loop
836 -- If no raise statement, give warning. We look at the
837 -- original node, because in the case of "raise ... with
838 -- ...", the node has been transformed into a call.
841 and then
849 else
861 -------------------------------
862 -- Check_Initialization_Call --
863 -------------------------------
869 -- Check whether the creation of an object of the type will involve
870 -- use of the secondary stack. If T is a record type, this is true
871 -- if the expression for some component uses the secondary stack, e.g.
872 -- through a call to a function that returns an unconstrained value.
873 -- False if T is controlled, because cleanups occur elsewhere.
875 -------------
876 -- Uses_SS --
877 -------------
884 begin
885 -- Normally we want to use the underlying type, but if it's not set
886 -- then continue with T.
903 then
904 -- The expression for a dynamic component may be rewritten
905 -- as a dereference, so retrieve original node.
909 -- Return True if the expression is a call to a function
910 -- (including an attribute function such as Image, or a
911 -- user-defined operator) with a result that requires a
912 -- transient scope.
919 then
932 else
937 -- Start of processing for Check_Initialization_Call
939 begin
940 -- Establish a transient scope if the type needs it
947 ---------------------------------------
948 -- Check_No_Direct_Boolean_Operators --
949 ---------------------------------------
952 begin
955 then
956 -- Restriction only applies to original source code
963 -- Do style check (but skip if in instance, error is on template)
972 ------------------------------
973 -- Check_Parameterless_Call --
974 ------------------------------
980 -- If the prefix is of an access_to_subprogram type, the node must be
981 -- rewritten as a call. Ditto if the prefix is overloaded and all its
982 -- interpretations are access to subprograms.
984 ---------------------------
985 -- Prefix_Is_Access_Subp --
986 ---------------------------
992 begin
993 -- If the context is an attribute reference that can apply to
994 -- functions, this is never a parameterless call (RM 4.1.4(6)).
998 Name_Code_Address,
999 Name_Access)
1000 then
1005 return
1008 else
1013 then
1024 -- Start of processing for Check_Parameterless_Call
1026 begin
1027 -- Defend against junk stuff if errors already detected
1034 then
1041 -- If the context expects a value, and the name is a procedure, this is
1042 -- most likely a missing 'Access. Don't try to resolve the parameterless
1043 -- call, error will be caught when the outer call is analyzed.
1048 and then
1050 N_Function_Call,
1051 N_Procedure_Call_Statement)
1052 then
1056 -- Rewrite as call if overloadable entity that is (or could be, in the
1057 -- overloaded case) a function call. If we know for sure that the entity
1058 -- is an enumeration literal, we do not rewrite it.
1060 -- If the entity is the name of an operator, it cannot be a call because
1061 -- operators cannot have default parameters. In this case, this must be
1062 -- a string whose contents coincide with an operator name. Set the kind
1063 -- of the node appropriately.
1071 -- Rewrite as call if it is an explicit dereference of an expression of
1072 -- a subprogram access type, and the subprogram type is not that of a
1073 -- procedure or entry.
1075 or else
1078 -- Rewrite as call if it is a selected component which is a function,
1079 -- this is the case of a call to a protected function (which may be
1080 -- overloaded with other protected operations).
1082 or else
1085 or else
1087 E_Procedure)
1090 -- If one of the above three conditions is met, rewrite as call. Apply
1091 -- the rewriting only once.
1093 then
1096 then
1098 -- This may be a prefixed call that was not fully analyzed, e.g.
1099 -- an actual in an instance.
1104 then
1112 -- The node is the name of the parameterless call. Preserve its
1113 -- descendants, which may be complex expressions.
1117 -- If overloaded, overload set belongs to new copy
1121 -- Change node to parameterless function call (note that the
1122 -- Parameter_Associations associations field is left set to Empty,
1123 -- its normal default value since there are no parameters)
1141 --------------------------------
1142 -- Is_Atomic_Ref_With_Address --
1143 --------------------------------
1148 begin
1152 else
1153 declare
1156 begin
1164 -----------------------------
1165 -- Is_Definite_Access_Type --
1166 -----------------------------
1170 begin
1176 ----------------------
1177 -- Is_Predefined_Op --
1178 ----------------------
1181 begin
1182 -- Predefined operators are intrinsic subprograms
1188 -- A call to a back-end builtin is never a predefined operator
1199 -----------------------------
1200 -- Make_Call_Into_Operator --
1201 -----------------------------
1203 procedure Make_Call_Into_Operator
1207 is
1222 -- If the operand is not universal, and the operator is given by an
1223 -- expanded name, verify that the operand has an interpretation with a
1224 -- type defined in the given scope of the operator.
1227 -- Find a type of the given class in package Pack that contains the
1228 -- operator.
1230 ---------------------------
1231 -- Operand_Type_In_Scope --
1232 ---------------------------
1239 begin
1243 else
1257 ---------------
1258 -- Type_In_P --
1259 ---------------
1265 -- Verify that node is not part of the type declaration for the
1266 -- candidate type, which would otherwise be invisible.
1268 -------------
1269 -- In_Decl --
1270 -------------
1276 begin
1285 else
1288 loop
1291 else
1300 -- Start of processing for Type_In_P
1302 begin
1303 -- If the context type is declared in the prefix package, this is the
1304 -- desired base type.
1309 else
1323 -- Start of processing for Make_Call_Into_Operator
1325 begin
1328 -- Ensure that the corresponding operator has the same parent as the
1329 -- original call. This guarantees that parent traversals performed by
1330 -- the ABE mechanism succeed.
1334 -- Binary operator
1344 -- Unary operator
1346 else
1352 -- If the operator is denoted by an expanded name, and the prefix is
1353 -- not Standard, but the operator is a predefined one whose scope is
1354 -- Standard, then this is an implicit_operator, inserted as an
1355 -- interpretation by the procedure of the same name. This procedure
1356 -- overestimates the presence of implicit operators, because it does
1357 -- not examine the type of the operands. Verify now that the operand
1358 -- type appears in the given scope. If right operand is universal,
1359 -- check the other operand. In the case of concatenation, either
1360 -- argument can be the component type, so check the type of the result.
1361 -- If both arguments are literals, look for a type of the right kind
1362 -- defined in the given scope. This elaborate nonsense is brought to
1363 -- you courtesy of b33302a. The type itself must be frozen, so we must
1364 -- find the type of the proper class in the given scope.
1366 -- A final wrinkle is the multiplication operator for fixed point types,
1367 -- which is defined in Standard only, and not in the scope of the
1368 -- fixed point type itself.
1373 -- If this is a package renaming, get renamed entity, which will be
1374 -- the scope of the operands if operaton is type-correct.
1380 -- If the entity being called is defined in the given package, it is
1381 -- a renaming of a predefined operator, and known to be legal.
1385 then
1388 -- Visibility does not need to be checked in an instance: if the
1389 -- operator was not visible in the generic it has been diagnosed
1390 -- already, else there is an implicit copy of it in the instance.
1398 then
1403 -- Ada 2005 AI-420: Predefined equality on Universal_Access is
1404 -- available.
1409 then
1412 else
1421 and then Is_Binary
1423 then
1429 -- Verify that the scope contains a type that corresponds to
1430 -- the given literal. Optimize the case where Pack is Standard.
1451 else
1460 else
1469 then
1472 -- If the type is defined elsewhere, and the operator is not
1473 -- defined in the given scope (by a renaming declaration, e.g.)
1474 -- then this is an error as well. If an extension of System is
1475 -- present, and the type may be defined there, Pack must be
1476 -- System itself.
1483 then
1486 else
1492 then
1499 Error_Msg_NE
1504 -- Detect a mismatch between the context type and the result type
1505 -- in the named package, which is otherwise not detected if the
1506 -- operands are universal. Check is only needed if source entity is
1507 -- an operator, not a function that renames an operator.
1514 and then not In_Instance
1515 then
1518 then
1519 -- Already checked above
1523 -- Operator may be defined in an extension of System
1528 then
1531 else
1532 -- Could we use Wrong_Type here??? (this would require setting
1533 -- Etype (N) to the actual type found where Typ was expected).
1544 else
1548 -- If this is a call to a function that renames a predefined equality,
1549 -- the renaming declaration provides a type that must be used to
1550 -- resolve the operands. This must be done now because resolution of
1551 -- the equality node will not resolve any remaining ambiguity, and it
1552 -- assumes that the first operand is not overloaded.
1557 then
1565 -- Do rewrite setting Comes_From_Source on the result if the original
1566 -- call came from source. Although it is not strictly the case that the
1567 -- operator as such comes from the source, logically it corresponds
1568 -- exactly to the function call in the source, so it should be marked
1569 -- this way (e.g. to make sure that validity checks work fine).
1571 declare
1573 begin
1578 -- If this is an arithmetic operator and the result type is private,
1579 -- the operands and the result must be wrapped in conversion to
1580 -- expose the underlying numeric type and expand the proper checks,
1581 -- e.g. on division.
1585 when N_Op_Add
1586 | N_Op_Divide
1587 | N_Op_Expon
1588 | N_Op_Mod
1589 | N_Op_Multiply
1590 | N_Op_Rem
1591 | N_Op_Subtract
1592 =>
1595 when N_Op_Abs
1596 | N_Op_Minus
1597 | N_Op_Plus
1598 =>
1604 else
1608 -- If in ASIS_Mode, propagate operand types to original actuals of
1609 -- function call, which would otherwise not be fully resolved. If
1610 -- the call has already been constant-folded, nothing to do. We
1611 -- relocate the operand nodes rather than copy them, to preserve
1612 -- original_node pointers, given that the operands themselves may
1613 -- have been rewritten. If the call was itself a rewriting of an
1614 -- operator node, nothing to do.
1616 if ASIS_Mode
1619 then
1620 declare
1628 begin
1633 -- If the original call has named associations, replace the
1634 -- explicit actual parameter in the association with the proper
1635 -- resolved operand.
1640 then
1643 else
1648 else
1655 then
1658 else
1663 else
1667 else
1671 else
1681 -------------------
1682 -- Operator_Kind --
1683 -------------------
1685 function Operator_Kind
1688 is
1691 begin
1692 -- Use CASE statement or array???
1729 else
1733 -- Unary operators
1735 else
1744 else
1752 ----------------------------
1753 -- Preanalyze_And_Resolve --
1754 ----------------------------
1759 begin
1763 -- Normally, we suppress all checks for this preanalysis. There is no
1764 -- point in processing them now, since they will be applied properly
1765 -- and in the proper location when the default expressions reanalyzed
1766 -- and reexpanded later on. We will also have more information at that
1767 -- point for possible suppression of individual checks.
1769 -- However, in SPARK mode, most expansion is suppressed, and this
1770 -- later reanalysis and reexpansion may not occur. SPARK mode does
1771 -- require the setting of checking flags for proof purposes, so we
1772 -- do the SPARK preanalysis without suppressing checks.
1774 -- This special handling for SPARK mode is required for example in the
1775 -- case of Ada 2012 constructs such as quantified expressions, which are
1776 -- expanded in two separate steps.
1780 else
1784 Expander_Mode_Restore;
1788 -- Version without context type
1793 begin
1800 Expander_Mode_Restore;
1804 ----------------------------------
1805 -- Replace_Actual_Discriminants --
1806 ----------------------------------
1813 -- Comment needed???
1815 -------------------
1816 -- Process_Discr --
1817 -------------------
1822 begin
1828 then
1844 -- Start of processing for Replace_Actual_Discriminants
1846 begin
1850 -- Allow the replacement of concurrent discriminants in GNATprove even
1851 -- though this is a light expansion activity. Note that generic units
1852 -- are not modified.
1857 else
1873 -------------
1874 -- Resolve --
1875 -------------
1891 -- Determine whether a node comes from a predefined library unit or
1892 -- Standard.
1895 -- Try and fix up a literal so that it matches its expected type. New
1896 -- literals are manufactured if necessary to avoid cascaded errors.
1899 -- Additional diagnostics when an ambiguous call has an ambiguous
1900 -- argument (typically a controlling actual).
1903 -- Called when attempt at resolving current expression fails
1905 ------------------------------------
1906 -- Comes_From_Predefined_Lib_Unit --
1907 -------------------------------------
1910 begin
1911 return
1915 --------------------
1916 -- Patch_Up_Value --
1917 --------------------
1920 begin
1937 then
1942 Char_Literal_Value =>
1958 -------------------------------
1959 -- Report_Ambiguous_Argument --
1960 -------------------------------
1967 begin
1971 then
1974 -- Could use comments on what is going on here???
1982 else
1991 -----------------------
1992 -- Resolution_Failed --
1993 -----------------------
1996 begin
1999 -- Set the type to the desired one to minimize cascaded errors. Note
2000 -- that this is an approximation and does not work in all cases.
2007 -- The caller will return without calling the expander, so we need
2008 -- to set the analyzed flag. Note that it is fine to set Analyzed
2009 -- to True even if we are in the middle of a shallow analysis,
2010 -- (see the spec of sem for more details) since this is an error
2011 -- situation anyway, and there is no point in repeating the
2012 -- analysis later (indeed it won't work to repeat it later, since
2013 -- we haven't got a clear resolution of which entity is being
2014 -- referenced.)
2020 -- Start of processing for Resolve
2022 begin
2027 -- Access attribute on remote subprogram cannot be used for a non-remote
2028 -- access-to-subprogram type.
2032 Name_Unrestricted_Access,
2033 Name_Unchecked_Access)
2039 then
2040 Error_Msg_N
2046 -- If the context is a Remote_Access_To_Subprogram, access attributes
2047 -- must be resolved with the corresponding fat pointer. There is no need
2048 -- to check for the attribute name since the return type of an
2049 -- attribute is never a remote type.
2054 then
2055 declare
2063 begin
2064 -- Check that Typ is a remote access-to-subprogram type
2068 -- Prefix (N) must statically denote a remote subprogram
2069 -- declared in a package specification.
2073 Attr = Attribute_Unrestricted_Access
2074 then
2089 or else
2091 then
2093 Error_Msg_N
2095 N);
2098 -- If we are generating code in distributed mode, perform
2099 -- semantic checks against corresponding remote entities.
2101 if Expander_Active
2103 then
2104 Check_Subtype_Conformant
2106 Old_Id => Designated_Type
2132 then
2137 -- Return if already analyzed
2144 -- Any case of Any_Type as the Etype value means that we had a
2145 -- previous error.
2154 -- The resolution of an Expression_With_Actions is determined by
2155 -- its Expression.
2163 -- If not overloaded, then we know the type, and all that needs doing
2164 -- is to check that this type is compatible with the context.
2170 -- In the overloaded case, we must select the interpretation that
2171 -- is compatible with the context (i.e. the type passed to Resolve)
2173 else
2174 -- Loop through possible interpretations
2183 -- We are only interested in interpretations that are compatible
2184 -- with the expected type, any other interpretations are ignored.
2189 Write_Eol;
2192 else
2193 -- Skip the current interpretation if it is disabled by an
2194 -- abstract operator. This action is performed only when the
2195 -- type against which we are resolving is the same as the
2196 -- type of the interpretation.
2203 then
2211 -- First matching interpretation
2219 -- Matching interpretation that is not the first, maybe an
2220 -- error, but there are some cases where preference rules are
2221 -- used to choose between the two possibilities. These and
2222 -- some more obscure cases are handled in Disambiguate.
2224 else
2225 -- If the current statement is part of a predefined library
2226 -- unit, then all interpretations which come from user level
2227 -- packages should not be considered. Check previous and
2228 -- current one.
2236 -- Previous interpretation must be discarded
2246 -- Otherwise apply further disambiguation steps
2251 -- Disambiguation has succeeded. Skip the remaining
2252 -- interpretations.
2262 else
2263 -- Before we issue an ambiguity complaint, check for the
2264 -- case of a subprogram call where at least one of the
2265 -- arguments is Any_Type, and if so suppress the message,
2266 -- since it is a cascaded error. This can also happen for
2267 -- a generalized indexing operation.
2272 then
2273 declare
2277 begin
2293 Write_Eol;
2306 then
2311 then
2315 -- Not that special case, so issue message using the flag
2316 -- Ambiguous to control printing of the header message
2317 -- only at the start of an ambiguous set.
2322 then
2323 Error_Msg_N
2326 else
2327 Error_Msg_NE -- CODEFIX
2335 Error_Msg_N
2337 else
2338 Error_Msg_N -- CODEFIX
2344 then
2345 Report_Ambiguous_Argument;
2351 -- By default, the error message refers to the candidate
2352 -- interpretation. But if it is a predefined operator, it
2353 -- is implicitly declared at the declaration of the type
2354 -- of the operand. Recover the sloc of that declaration
2355 -- for the error message.
2361 Standard_Standard
2362 then
2367 then
2375 Standard_Standard
2376 then
2381 then
2385 -- If this is an indirect call, use the subprogram_type
2386 -- in the message, to have a meaningful location. Also
2387 -- indicate if this is an inherited operation, created
2388 -- by a type declaration.
2393 then
2395 Error_Msg_Sloc :=
2397 else
2404 then
2405 -- Special-case the message for universal_fixed
2406 -- operators, which are not declared with the type
2407 -- of the operand, but appear forever in Standard.
2411 then
2412 Error_Msg_N
2415 else
2416 Error_Msg_N
2420 elsif
2422 then
2423 Error_Msg_N
2425 else
2426 Error_Msg_N -- CODEFIX
2433 -- We have a matching interpretation, Expr_Type is the type
2434 -- from this interpretation, and Seen is the entity.
2436 -- For an operator, just set the entity name. The type will be
2437 -- set by the specific operator resolution routine.
2444 N_Character_Literal,
2445 N_Delta_Aggregate,
2446 N_If_Expression)
2447 then
2450 -- AI05-0139-2: Expression is overloaded because type has
2451 -- implicit dereference. If type matches context, no implicit
2452 -- dereference is involved. If the expression is an entity,
2453 -- generate a reference to it, as this is not done for an
2454 -- overloaded construct during analysis.
2470 then
2471 -- If the node is a general indexing, the dereference is
2472 -- is inserted when resolving the rewritten form, else
2473 -- insert it now.
2477 then
2481 -- For an explicit dereference, attribute reference, range,
2482 -- short-circuit form (which is not an operator node), or call
2483 -- with a name that is an explicit dereference, there is
2484 -- nothing to be done at this point.
2487 N_And_Then,
2488 N_Explicit_Dereference,
2489 N_Identifier,
2490 N_Indexed_Component,
2491 N_Or_Else,
2492 N_Range,
2493 N_Selected_Component,
2494 N_Slice)
2496 then
2499 -- For procedure or function calls, set the type of the name,
2500 -- and also the entity pointer for the prefix.
2504 then
2511 then
2516 -- For all other cases, just set the type of the Name
2518 else
2526 -- Move to next interpretation
2534 -- At this stage Found indicates whether or not an acceptable
2535 -- interpretation exists. If not, then we have an error, except that if
2536 -- the context is Any_Type as a result of some other error, then we
2537 -- suppress the error report.
2542 -- If type we are looking for is Void, then this is the procedure
2543 -- call case, and the error is simply that what we gave is not a
2544 -- procedure name (we think of procedure calls as expressions with
2545 -- types internally, but the user doesn't think of them this way).
2549 -- Special case message if function used as a procedure
2554 then
2555 Error_Msg_NE
2558 Error_Msg_N
2562 -- Otherwise give general message (not clear what cases this
2563 -- covers, but no harm in providing for them).
2565 else
2571 -- Otherwise we do have a subexpression with the wrong type
2573 -- Check for the case of an allocator which uses an access type
2574 -- instead of the designated type. This is a common error and we
2575 -- specialize the message, posting an error on the operand of the
2576 -- allocator, complaining that we expected the designated type of
2577 -- the allocator.
2583 then
2587 -- Check for view mismatch on Null in instances, for which the
2588 -- view-swapping mechanism has no identifier.
2594 then
2599 -- Check for an aggregate. Sometimes we can get bogus aggregates
2600 -- from misuse of parentheses, and we are about to complain about
2601 -- the aggregate without even looking inside it.
2603 -- Instead, if we have an aggregate of type Any_Composite, then
2604 -- analyze and resolve the component fields, and then only issue
2605 -- another message if we get no errors doing this (otherwise
2606 -- assume that the errors in the aggregate caused the problem).
2610 then
2611 -- Disable expansion in any case. If there is a type mismatch
2612 -- it may be fatal to try to expand the aggregate. The flag
2613 -- would otherwise be set to false when the error is posted.
2617 declare
2619 -- Check one aggregate, and set Found to True if we have a
2620 -- definite error in any of its elements
2623 -- Check one element of aggregate and set Found to True if
2624 -- we definitely have an error in the element.
2626 ----------------
2627 -- Check_Aggr --
2628 ----------------
2633 begin
2646 -- If this is a default-initialized component, then
2647 -- there is nothing to check. The box will be
2648 -- replaced by the appropriate call during late
2649 -- expansion.
2653 then
2662 ----------------
2663 -- Check_Elmt --
2664 ----------------
2667 begin
2668 -- If we have a nested aggregate, go inside it (to
2669 -- attempt a naked analyze-resolve of the aggregate can
2670 -- cause undesirable cascaded errors). Do not resolve
2671 -- expression if it needs a type from context, as for
2672 -- integer * fixed expression.
2677 else
2682 then
2692 begin
2697 -- Looks like we have a type error, but check for special case
2698 -- of Address wanted, integer found, with the configuration pragma
2699 -- Allow_Integer_Address active. If we have this case, introduce
2700 -- an unchecked conversion to allow the integer expression to be
2701 -- treated as an Address. The reverse case of integer wanted,
2702 -- Address found, is treated in an analogous manner.
2709 -- Under relaxed RM semantics silently replace occurrences of null
2710 -- by System.Address_Null.
2718 -- That special Allow_Integer_Address check did not apply, so we
2719 -- have a real type error. If an error message was issued already,
2720 -- Found got reset to True, so if it's still False, issue standard
2721 -- Wrong_Type message.
2725 declare
2728 begin
2734 -- Protected operation: retrieve operation name
2738 else
2743 Error_Msg_NE
2749 declare
2753 begin
2760 Error_Msg_NE
2766 else
2770 else
2776 Resolution_Failed;
2779 -- Test if we have more than one interpretation for the context
2782 Resolution_Failed;
2785 -- Only one intepretation
2787 else
2788 -- In Ada 2005, if we have something like "X : T := 2 + 2;", where
2789 -- the "+" on T is abstract, and the operands are of universal type,
2790 -- the above code will have (incorrectly) resolved the "+" to the
2791 -- universal one in Standard. Therefore check for this case and give
2792 -- an error. We can't do this earlier, because it would cause legal
2793 -- cases to get errors (when some other type has an abstract "+").
2799 then
2804 then
2805 Error_Msg_NE
2814 -- Here we have an acceptable interpretation for the context
2816 -- Propagate type information and normalize tree for various
2817 -- predefined operations. If the context only imposes a class of
2818 -- types, rather than a specific type, propagate the actual type
2819 -- downward.
2825 Typ = Any_Discrete
2826 then
2829 -- Any_Fixed is legal in a real context only if a specific fixed-
2830 -- point type is imposed. If Norman Cohen can be confused by this,
2831 -- it deserves a separate message.
2835 then
2842 -- A user-defined operator is transformed into a function call at
2843 -- this point, so that further processing knows that operators are
2844 -- really operators (i.e. are predefined operators). User-defined
2845 -- operators that are intrinsic are just renamings of the predefined
2846 -- ones, and need not be turned into calls either, but if they rename
2847 -- a different operator, we must transform the node accordingly.
2848 -- Instantiations of Unchecked_Conversion are intrinsic but are
2849 -- treated as functions, even if given an operator designator.
2854 then
2859 and then
2861 N_Subprogram_Renaming_Declaration
2862 then
2865 -- If the node is rewritten, it will be fully resolved in
2866 -- Rewrite_Renamed_Operator.
2875 when N_Aggregate =>
2876 Resolve_Aggregate (N, Ctx_Type);
2878 when N_Allocator =>
2879 Resolve_Allocator (N, Ctx_Type);
2881 when N_Short_Circuit =>
2882 Resolve_Short_Circuit (N, Ctx_Type);
2884 when N_Attribute_Reference =>
2885 Resolve_Attribute (N, Ctx_Type);
2887 when N_Case_Expression =>
2888 Resolve_Case_Expression (N, Ctx_Type);
2890 when N_Character_Literal =>
2891 Resolve_Character_Literal (N, Ctx_Type);
2893 when N_Delta_Aggregate =>
2894 Resolve_Delta_Aggregate (N, Ctx_Type);
2896 when N_Expanded_Name =>
2897 Resolve_Entity_Name (N, Ctx_Type);
2899 when N_Explicit_Dereference =>
2900 Resolve_Explicit_Dereference (N, Ctx_Type);
2902 when N_Expression_With_Actions =>
2903 Resolve_Expression_With_Actions (N, Ctx_Type);
2905 when N_Extension_Aggregate =>
2906 Resolve_Extension_Aggregate (N, Ctx_Type);
2908 when N_Function_Call =>
2909 Resolve_Call (N, Ctx_Type);
2911 when N_Identifier =>
2912 Resolve_Entity_Name (N, Ctx_Type);
2914 when N_If_Expression =>
2915 Resolve_If_Expression (N, Ctx_Type);
2917 when N_Indexed_Component =>
2918 Resolve_Indexed_Component (N, Ctx_Type);
2920 when N_Integer_Literal =>
2921 Resolve_Integer_Literal (N, Ctx_Type);
2923 when N_Membership_Test =>
2924 Resolve_Membership_Op (N, Ctx_Type);
2926 when N_Null =>
2927 Resolve_Null (N, Ctx_Type);
2929 when N_Op_And
2930 | N_Op_Or
2931 | N_Op_Xor
2932 =>
2933 Resolve_Logical_Op (N, Ctx_Type);
2935 when N_Op_Eq
2936 | N_Op_Ne
2937 =>
2938 Resolve_Equality_Op (N, Ctx_Type);
2940 when N_Op_Ge
2941 | N_Op_Gt
2942 | N_Op_Le
2943 | N_Op_Lt
2944 =>
2945 Resolve_Comparison_Op (N, Ctx_Type);
2947 when N_Op_Not =>
2948 Resolve_Op_Not (N, Ctx_Type);
2950 when N_Op_Add
2951 | N_Op_Divide
2952 | N_Op_Mod
2953 | N_Op_Multiply
2954 | N_Op_Rem
2955 | N_Op_Subtract
2956 =>
2957 Resolve_Arithmetic_Op (N, Ctx_Type);
2959 when N_Op_Concat =>
2960 Resolve_Op_Concat (N, Ctx_Type);
2962 when N_Op_Expon =>
2963 Resolve_Op_Expon (N, Ctx_Type);
2965 when N_Op_Abs
2966 | N_Op_Minus
2967 | N_Op_Plus
2968 =>
2969 Resolve_Unary_Op (N, Ctx_Type);
2971 when N_Op_Shift =>
2972 Resolve_Shift (N, Ctx_Type);
2974 when N_Procedure_Call_Statement =>
2975 Resolve_Call (N, Ctx_Type);
2977 when N_Operator_Symbol =>
2978 Resolve_Operator_Symbol (N, Ctx_Type);
2980 when N_Qualified_Expression =>
2981 Resolve_Qualified_Expression (N, Ctx_Type);
2983 -- Why is the following null, needs a comment ???
2985 when N_Quantified_Expression =>
2986 null;
2988 when N_Raise_Expression =>
2989 Resolve_Raise_Expression (N, Ctx_Type);
2991 when N_Raise_xxx_Error =>
2992 Set_Etype (N, Ctx_Type);
2994 when N_Range =>
2995 Resolve_Range (N, Ctx_Type);
2997 when N_Real_Literal =>
2998 Resolve_Real_Literal (N, Ctx_Type);
3000 when N_Reduction_Expression =>
3001 null;
3002 -- Resolve (Expression (N), Ctx_Type);
3004 when N_Reduction_Expression_Parameter =>
3005 null;
3007 when N_Reference =>
3008 Resolve_Reference (N, Ctx_Type);
3010 when N_Selected_Component =>
3011 Resolve_Selected_Component (N, Ctx_Type);
3013 when N_Slice =>
3014 Resolve_Slice (N, Ctx_Type);
3016 when N_String_Literal =>
3017 Resolve_String_Literal (N, Ctx_Type);
3019 when N_Target_Name =>
3020 Resolve_Target_Name (N, Ctx_Type);
3022 when N_Type_Conversion =>
3023 Resolve_Type_Conversion (N, Ctx_Type);
3025 when N_Unchecked_Expression =>
3026 Resolve_Unchecked_Expression (N, Ctx_Type);
3028 when N_Unchecked_Type_Conversion =>
3029 Resolve_Unchecked_Type_Conversion (N, Ctx_Type);
3030 end case;
3032 -- Mark relevant use-type and use-package clauses as effective using
3033 -- the original node because constant folding may have occured and
3034 -- removed references that need to be examined.
3036 if Nkind (Original_Node (N)) in N_Op then
3037 Mark_Use_Clauses (Original_Node (N));
3038 end if;
3040 -- Ada 2012 (AI05-0149): Apply an (implicit) conversion to an
3041 -- expression of an anonymous access type that occurs in the context
3042 -- of a named general access type, except when the expression is that
3043 -- of a membership test. This ensures proper legality checking in
3044 -- terms of allowed conversions (expressions that would be illegal to
3045 -- convert implicitly are allowed in membership tests).
3047 if Ada_Version >= Ada_2012
3048 and then Ekind (Ctx_Type) = E_General_Access_Type
3049 and then Ekind (Etype (N)) = E_Anonymous_Access_Type
3050 and then Nkind (Parent (N)) not in N_Membership_Test
3051 then
3052 Rewrite (N, Convert_To (Ctx_Type, Relocate_Node (N)));
3053 Analyze_And_Resolve (N, Ctx_Type);
3054 end if;
3056 -- If the subexpression was replaced by a non-subexpression, then
3057 -- all we do is to expand it. The only legitimate case we know of
3058 -- is converting procedure call statement to entry call statements,
3059 -- but there may be others, so we are making this test general.
3061 if Nkind (N) not in N_Subexpr then
3062 Debug_A_Exit ("resolving ", N, " (done)");
3063 Expand (N);
3064 return;
3065 end if;
3067 -- The expression is definitely NOT overloaded at this point, so
3068 -- we reset the Is_Overloaded flag to avoid any confusion when
3069 -- reanalyzing the node.
3071 Set_Is_Overloaded (N, False);
3073 -- Freeze expression type, entity if it is a name, and designated
3074 -- type if it is an allocator (RM 13.14(10,11,13)).
3076 -- Now that the resolution of the type of the node is complete, and
3077 -- we did not detect an error, we can expand this node. We skip the
3078 -- expand call if we are in a default expression, see section
3079 -- "Handling of Default Expressions" in Sem spec.
3081 Debug_A_Exit ("resolving ", N, " (done)");
3083 -- We unconditionally freeze the expression, even if we are in
3084 -- default expression mode (the Freeze_Expression routine tests this
3085 -- flag and only freezes static types if it is set).
3087 -- Ada 2012 (AI05-177): The declaration of an expression function
3088 -- does not cause freezing, but we never reach here in that case.
3089 -- Here we are resolving the corresponding expanded body, so we do
3090 -- need to perform normal freezing.
3092 -- As elsewhere we do not emit freeze node within a generic. We make
3093 -- an exception for entities that are expressions, only to detect
3094 -- misuses of deferred constants and preserve the output of various
3095 -- tests.
3097 if not Inside_A_Generic or else Is_Entity_Name (N) then
3098 Freeze_Expression (N);
3099 end if;
3101 -- Now we can do the expansion
3103 Expand (N);
3104 end if;
3105 end Resolve;
3107 -------------
3108 -- Resolve --
3109 -------------
3111 -- Version with check(s) suppressed
3113 procedure Resolve (N : Node_Id; Typ : Entity_Id; Suppress : Check_Id) is
3114 begin
3115 if Suppress = All_Checks then
3116 declare
3117 Sva : constant Suppress_Array := Scope_Suppress.Suppress;
3118 begin
3119 Scope_Suppress.Suppress := (others => True);
3120 Resolve (N, Typ);
3121 Scope_Suppress.Suppress := Sva;
3122 end;
3124 else
3125 declare
3126 Svg : constant Boolean := Scope_Suppress.Suppress (Suppress);
3127 begin
3128 Scope_Suppress.Suppress (Suppress) := True;
3129 Resolve (N, Typ);
3130 Scope_Suppress.Suppress (Suppress) := Svg;
3131 end;
3132 end if;
3133 end Resolve;
3135 -------------
3136 -- Resolve --
3137 -------------
3139 -- Version with implicit type
3141 procedure Resolve (N : Node_Id) is
3142 begin
3143 Resolve (N, Etype (N));
3144 end Resolve;
3146 ---------------------
3147 -- Resolve_Actuals --
3148 ---------------------
3150 procedure Resolve_Actuals (N : Node_Id; Nam : Entity_Id) is
3151 Loc : constant Source_Ptr := Sloc (N);
3152 A : Node_Id;
3153 A_Id : Entity_Id;
3154 A_Typ : Entity_Id := Empty; -- init to avoid warning
3155 F : Entity_Id;
3156 F_Typ : Entity_Id;
3157 Prev : Node_Id := Empty;
3158 Orig_A : Node_Id;
3159 Real_F : Entity_Id := Empty; -- init to avoid warning
3161 Real_Subp : Entity_Id;
3162 -- If the subprogram being called is an inherited operation for
3163 -- a formal derived type in an instance, Real_Subp is the subprogram
3164 -- that will be called. It may have different formal names than the
3165 -- operation of the formal in the generic, so after actual is resolved
3166 -- the name of the actual in a named association must carry the name
3167 -- of the actual of the subprogram being called.
3169 procedure Check_Aliased_Parameter;
3170 -- Check rules on aliased parameters and related accessibility rules
3171 -- in (RM 3.10.2 (10.2-10.4)).
3173 procedure Check_Argument_Order;
3174 -- Performs a check for the case where the actuals are all simple
3175 -- identifiers that correspond to the formal names, but in the wrong
3176 -- order, which is considered suspicious and cause for a warning.
3178 procedure Check_Prefixed_Call;
3179 -- If the original node is an overloaded call in prefix notation,
3181 -- Try_Object_Operation has already verified that there is a valid
3182 -- interpretation, but the form of the actual can only be determined
3183 -- once the primitive operation is identified.
3186 -- Emit an error concerning the illegal usage of an effectively volatile
3187 -- object in interfering context (SPARK RM 7.13(12)).
3190 -- If the actual is missing in a call, insert in the actuals list
3191 -- an instance of the default expression. The insertion is always
3192 -- a named association.
3195 -- Check whether T1 and T2, or their full views, are derived from a
3196 -- common type. Used to enforce the restrictions on array conversions
3197 -- of AI95-00246.
3200 -- Predicate to determine whether an actual that is a concatenation
3201 -- will be evaluated statically and does not need a transient scope.
3202 -- This must be determined before the actual is resolved and expanded
3203 -- because if needed the transient scope must be introduced earlier.
3205 -----------------------------
3206 -- Check_Aliased_Parameter --
3207 -----------------------------
3212 begin
3220 else
3227 -- In a generic body assume the worst for generic formals:
3228 -- they can have a constrained partial view (AI05-041).
3234 then
3237 else
3242 else
3254 then
3262 then
3263 Error_Msg_N
3269 --------------------------
3270 -- Check_Argument_Order --
3271 --------------------------
3274 begin
3275 -- Nothing to do if no parameters, or original node is neither a
3276 -- function call nor a procedure call statement (happens in the
3277 -- operator-transformed-to-function call case), or the call does
3278 -- not come from source, or this warning is off.
3280 if not Warn_On_Parameter_Order
3284 then
3288 declare
3291 begin
3292 -- Nothing to do if only one parameter
3298 -- Here if at least two arguments
3300 declare
3306 -- Set True if an out of order case is found
3308 begin
3309 -- Collect identifier names of actuals, fail if any actual is
3310 -- not a simple identifier, and record max length of name.
3316 else
3322 -- If we got this far, all actuals are identifiers and the list
3323 -- of their names is stored in the Actuals array.
3328 -- If we ran out of formals, that's odd, probably an error
3329 -- which will be detected elsewhere, but abandon the search.
3335 -- If name matches and is in order OK
3340 else
3341 -- If no match, see if it is elsewhere in list and if so
3342 -- flag potential wrong order if type is compatible.
3346 and then
3348 then
3354 -- No match
3362 -- If Formals left over, also probably an error, skip warning
3368 -- Here we give the warning if something was out of order
3371 Error_Msg_N
3378 -------------------------
3379 -- Check_Prefixed_Call --
3380 -------------------------
3389 begin
3390 -- Check whether the call is a prefixed call, with or without
3391 -- additional actuals.
3394 or else
3400 then
3403 then
3404 -- Introduce dereference on object in prefix
3406 New_A :=
3414 then
3415 -- Introduce an implicit 'Access in prefix
3418 Error_Msg_NE
3434 ---------------------------------------
3435 -- Flag_Effectively_Volatile_Objects --
3436 ---------------------------------------
3440 -- Determine whether arbitrary node N denotes an effectively volatile
3441 -- object and if it does, emit an error.
3443 -----------------
3444 -- Flag_Object --
3445 -----------------
3450 begin
3451 -- Do not consider nested function calls because they have already
3452 -- been processed during their own resolution.
3464 then
3465 Error_Msg_N
3477 -- Start of processing for Flag_Effectively_Volatile_Objects
3479 begin
3483 --------------------
3484 -- Insert_Default --
3485 --------------------
3491 begin
3492 -- Missing argument in call, nothing to insert
3497 else
3498 -- Note that we do a full New_Copy_Tree, so that any associated
3499 -- Itypes are properly copied. This may not be needed any more,
3500 -- but it does no harm as a safety measure. Defaults of a generic
3501 -- formal may be out of bounds of the corresponding actual (see
3502 -- cc1311b) and an additional check may be required.
3504 Actval :=
3505 New_Copy_Tree
3510 -- Propagate dimension information, if any.
3516 then
3522 then
3525 then
3526 -- If default is a real literal, do not introduce a
3527 -- conversion whose effect may depend on the run-time
3528 -- size of universal real.
3532 else
3543 -- Resolve aggregates with their base type, to avoid scope
3544 -- anomalies: the subtype was first built in the subprogram
3545 -- declaration, and the current call may be nested.
3549 else
3553 else
3556 -- See note above concerning aggregates
3560 then
3563 -- Resolve entities with their own type, which may differ from
3564 -- the type of a reference in a generic context (the view
3565 -- swapping mechanism did not anticipate the re-analysis of
3566 -- default values in calls).
3571 else
3576 -- If default is a tag indeterminate function call, propagate tag
3577 -- to obtain proper dispatching.
3581 then
3586 -- If the default expression raises constraint error, then just
3587 -- silently replace it with an N_Raise_Constraint_Error node, since
3588 -- we already gave the warning on the subprogram spec. If node is
3589 -- already a Raise_Constraint_Error leave as is, to prevent loops in
3590 -- the warnings removal machinery.
3594 then
3603 Assoc :=
3608 -- Case of insertion is first named actual
3612 then
3619 else
3623 else
3627 -- Case of insertion is not first named actual
3629 else
3630 Set_Next_Named_Actual
3642 -------------------
3643 -- Same_Ancestor --
3644 -------------------
3650 begin
3653 then
3659 then
3666 --------------------------
3667 -- Static_Concatenation --
3668 --------------------------
3671 begin
3678 -- Concatenation is static when both operands are static and
3679 -- the concatenation operator is a predefined one.
3682 and then
3684 and then
3689 declare
3691 begin
3694 and then
3698 else
3704 -- Start of processing for Resolve_Actuals
3706 begin
3707 Check_Argument_Order;
3711 and then In_Instance
3714 then
3716 else
3721 Check_Prefixed_Call;
3735 -- If we have an error in any actual or formal, indicated by a type
3736 -- of Any_Type, then abandon resolution attempt, and set result type
3737 -- to Any_Type. Skip this if the actual is a Raise_Expression, whose
3738 -- type is imposed from context.
3742 then
3749 -- Case where actual is present
3751 -- If the actual is an entity, generate a reference to it now. We
3752 -- do this before the actual is resolved, because a formal of some
3753 -- protected subprogram, or a task discriminant, will be rewritten
3754 -- during expansion, and the source entity reference may be lost.
3759 then
3760 -- Annotate the tree by creating a variable reference marker when
3761 -- the actual denotes a variable reference, in case the reference
3762 -- is folded or optimized away. The variable reference marker is
3763 -- automatically saved for later examination by the ABE Processing
3764 -- phase. The status of the reference is set as follows:
3766 -- status mode
3767 -- read IN, IN OUT
3768 -- write IN OUT, OUT
3770 Build_Variable_Reference_Marker
3780 then
3787 -- RM 6.4.1(12): For an out parameter that is passed by
3788 -- copy, the formal parameter object is created, and:
3790 -- * For an access type, the formal parameter is initialized
3791 -- from the value of the actual, without checking that the
3792 -- value satisfies any constraint, any predicate, or any
3793 -- exclusion of the null value.
3795 -- * For a scalar type that has the Default_Value aspect
3796 -- specified, the formal parameter is initialized from the
3797 -- value of the actual, without checking that the value
3798 -- satisfies any constraint or any predicate.
3799 -- I do not understand why this case is included??? this is
3800 -- not a case where an OUT parameter is treated as IN OUT.
3802 -- * For a composite type with discriminants or that has
3803 -- implicit initial values for any subcomponents, the
3804 -- behavior is as for an in out parameter passed by copy.
3806 -- Hence for these cases we generate the read reference now
3807 -- (the write reference will be generated later by
3808 -- Note_Possible_Modification).
3811 and then
3813 or else
3815 and then
3817 or else
3820 or else Is_Partially_Initialized_Type
3822 then
3832 then
3833 -- If style checking mode on, check match of formal name
3841 -- If the formal is Out or In_Out, do not resolve and expand the
3842 -- conversion, because it is subsequently expanded into explicit
3843 -- temporaries and assignments. However, the object of the
3844 -- conversion can be resolved. An exception is the case of tagged
3845 -- type conversion with a class-wide actual. In that case we want
3846 -- the tag check to occur and no temporary will be needed (no
3847 -- representation change can occur) and the parameter is passed by
3848 -- reference, so we go ahead and resolve the type conversion.
3849 -- Another exception is the case of reference to component or
3850 -- subcomponent of a bit-packed array, in which case we want to
3851 -- defer expansion to the point the in and out assignments are
3852 -- performed.
3857 then
3860 then
3861 -- In a view conversion, the conversion must be legal in
3862 -- both directions, and thus both component types must be
3863 -- aliased, or neither (4.6 (8)).
3865 -- The extra rule in 4.6 (24.9.2) seems unduly restrictive:
3866 -- the privacy requirement should not apply to generic
3867 -- types, and should be checked in an instance. ARG query
3868 -- is in order ???
3872 then
3873 Error_Msg_N
3877 -- Comment here??? what set of cases???
3879 elsif
3881 then
3882 -- Check view conv between unrelated by ref array types
3886 then
3887 Error_Msg_N
3891 -- In Ada 2005 mode, check view conversion component
3892 -- type cannot be private, tagged, or volatile. Note
3893 -- that we only apply this to source conversions. The
3894 -- generated code can contain conversions which are
3895 -- not subject to this test, and we cannot extract the
3896 -- component type in such cases since it is not present.
3900 then
3901 declare
3903 Component_Type
3905 begin
3910 then
3911 Error_Msg_N
3922 -- Resolve expression if conversion is all OK
3927 then
3931 -- If the actual is a function call that returns a limited
3932 -- unconstrained object that needs finalization, create a
3933 -- transient scope for it, so that it can receive the proper
3934 -- finalization list.
3939 and then Expander_Active
3941 then
3945 -- A small optimization: if one of the actuals is a concatenation
3946 -- create a block around a procedure call to recover stack space.
3947 -- This alleviates stack usage when several procedure calls in
3948 -- the same statement list use concatenation. We do not perform
3949 -- this wrapping for code statements, where the argument is a
3950 -- static string, and we want to preserve warnings involving
3951 -- sequences of such statements.
3955 and then Expander_Active
3956 and then
3960 then
3964 else
3968 and then
3971 then
3972 Error_Msg_N
3985 -- (Ada 2005: AI-251): If the actual is an allocator whose
3986 -- directly designated type is a class-wide interface, we build
3987 -- an anonymous access type to use it as the type of the
3988 -- allocator. Later, when the subprogram call is expanded, if
3989 -- the interface has a secondary dispatch table the expander
3990 -- will add a type conversion to force the correct displacement
3991 -- of the pointer.
3994 declare
4000 begin
4003 then
4006 Set_Directly_Designated_Type
4011 -- Ada 2005, AI-162:If the actual is an allocator, the
4012 -- innermost enclosing statement is the master of the
4013 -- created object. This needs to be done with expansion
4014 -- enabled only, otherwise the transient scope will not
4015 -- be removed in the expansion of the wrapped construct.
4018 and then Expander_Active
4019 then
4029 -- (Ada 2005): The call may be to a primitive operation of a
4030 -- tagged synchronized type, declared outside of the type. In
4031 -- this case the controlling actual must be converted to its
4032 -- corresponding record type, which is the formal type. The
4033 -- actual may be a subtype, either because of a constraint or
4034 -- because it is a generic actual, so use base type to locate
4035 -- concurrent type.
4042 then
4043 -- If the actual is overloaded, look for an interpretation
4044 -- that has a synchronized type.
4049 else
4050 declare
4054 begin
4059 then
4069 declare
4072 begin
4075 else
4079 -- Tagged synchronized type (case 1): the actual is a
4080 -- concurrent type.
4084 then
4086 Unchecked_Convert_To
4090 -- Tagged synchronized type (case 2): the formal is a
4091 -- concurrent type.
4094 and then Present
4099 then
4102 -- Common case
4104 else
4109 -- Not a synchronized operation
4111 else
4119 -- An actual cannot be an untagged formal incomplete type
4124 then
4125 Error_Msg_N
4131 N_Procedure_Call_Statement)
4132 then
4133 -- In formal mode, check that actual parameters matching
4134 -- formals of tagged types are objects (or ancestor type
4135 -- conversions of objects), not general expressions.
4142 declare
4147 begin
4149 Check_SPARK_05_Restriction
4152 -- In formal mode, the only view conversions are those
4153 -- involving ancestor conversion of an extended type.
4155 elsif not
4161 then
4162 if Ekind_In
4164 then
4165 Check_SPARK_05_Restriction
4168 else
4169 Check_SPARK_05_Restriction
4173 else
4178 else
4182 -- In formal mode, the only view conversions are those
4183 -- involving ancestor conversion of an extended type.
4187 then
4188 Check_SPARK_05_Restriction
4194 -- has warnings suppressed, then we reset Never_Set_In_Source for
4195 -- the calling entity. The reason for this is to catch cases like
4196 -- GNAT.Spitbol.Patterns.Vstring_Var where the called subprogram
4197 -- uses trickery to modify an IN parameter.
4204 then
4208 -- Perform error checks for IN and IN OUT parameters
4212 -- Check unset reference. For scalar parameters, it is clearly
4213 -- wrong to pass an uninitialized value as either an IN or
4214 -- IN-OUT parameter. For composites, it is also clearly an
4215 -- error to pass a completely uninitialized value as an IN
4216 -- parameter, but the case of IN OUT is trickier. We prefer
4217 -- not to give a warning here. For example, suppose there is
4218 -- a routine that sets some component of a record to False.
4219 -- It is perfectly reasonable to make this IN-OUT and allow
4220 -- either initialized or uninitialized records to be passed
4221 -- in this case.
4223 -- For partially initialized composite values, we also avoid
4224 -- warnings, since it is quite likely that we are passing a
4225 -- partially initialized value and only the initialized fields
4226 -- will in fact be read in the subprogram.
4231 then
4235 -- In Ada 83 we cannot pass an OUT parameter as an IN or IN OUT
4236 -- actual to a nested call, since this constitutes a reading of
4237 -- the parameter, which is not allowed.
4242 then
4247 -- In -gnatd.q mode, forget that a given array is constant when
4248 -- it is passed as an IN parameter to a foreign-convention
4249 -- subprogram. This is in case the subprogram evilly modifies the
4250 -- object. Of course, correct code would use IN OUT.
4252 if Debug_Flag_Dot_Q
4258 then
4262 -- Case of OUT or IN OUT parameter
4266 -- For an Out parameter, check for useless assignment. Note
4267 -- that we can't set Last_Assignment this early, because we may
4268 -- kill current values in Resolve_Call, and that call would
4269 -- clobber the Last_Assignment field.
4271 -- Note: call Warn_On_Useless_Assignment before doing the check
4272 -- below for Is_OK_Variable_For_Out_Formal so that the setting
4273 -- of Referenced_As_LHS/Referenced_As_Out_Formal properly
4274 -- reflects the last assignment, not this one.
4281 then
4286 -- Validate the form of the actual. Note that the call to
4287 -- Is_OK_Variable_For_Out_Formal generates the required
4288 -- reference in this case.
4290 -- A call to an initialization procedure for an aggregate
4291 -- component may initialize a nested component of a constant
4292 -- designated object. In this context the object is variable.
4296 then
4302 then
4303 Error_Msg_N
4308 Error_Msg_N
4315 -- What's the following about???
4319 else
4320 Kill_All_Checks;
4329 -- Apply appropriate constraint/predicate checks for IN [OUT] case
4333 -- Apply predicate tests except in certain special cases. Note
4334 -- that it might be more consistent to apply these only when
4335 -- expansion is active (in Exp_Ch6.Expand_Actuals), as we do
4336 -- for the outbound predicate tests ??? In any case indicate
4337 -- the function being called, for better warnings if the call
4338 -- leads to an infinite recursion.
4344 -- Apply required constraint checks
4346 -- Gigi looks at the check flag and uses the appropriate types.
4347 -- For now since one flag is used there is an optimization
4348 -- which might not be done in the IN OUT case since Gigi does
4349 -- not do any analysis. More thought required about this ???
4351 -- In fact is this comment obsolete??? doesn't the expander now
4352 -- generate all these tests anyway???
4365 then
4368 -- For view conversions of a discriminated object, apply
4369 -- check to object itself, the conversion alreay has the
4370 -- proper type.
4374 then
4381 then
4387 then
4390 else
4394 -- Ada 2005 (AI-231): Note that the controlling parameter case
4395 -- already existed in Ada 95, which is partially checked
4396 -- elsewhere (see Checks), and we don't want the warning
4397 -- message to differ.
4402 then
4404 Apply_Compile_Time_Constraint_Error
4410 Apply_Compile_Time_Constraint_Error
4419 -- Checks for OUT parameters and IN OUT parameters
4423 -- If there is a type conversion, make sure the return value
4424 -- meets the constraints of the variable before the conversion.
4428 Apply_Scalar_Range_Check
4431 -- In addition, the returned value of the parameter must
4432 -- satisfy the bounds of the object type (see comment
4433 -- below).
4437 else
4438 Apply_Range_Check
4442 -- If no conversion, apply scalar range checks and length check
4443 -- based on the subtype of the actual (NOT that of the formal).
4444 -- This indicates that the check takes place on return from the
4445 -- call. During expansion the required constraint checks are
4446 -- inserted. In GNATprove mode, in the absence of expansion,
4447 -- the flag indicates that the returned value is valid.
4449 else
4455 then
4457 else
4462 -- Note: we do not apply the predicate checks for the case of
4463 -- OUT and IN OUT parameters. They are instead applied in the
4464 -- Expand_Actuals routine in Exp_Ch6.
4467 -- An actual associated with an access parameter is implicitly
4468 -- converted to the anonymous access type of the formal and must
4469 -- satisfy the legality checks for access conversions.
4473 Error_Msg_N
4477 -- If the actual is an access selected component of a variable,
4478 -- the call may modify its designated object. It is reasonable
4479 -- to treat this as a potential modification of the enclosing
4480 -- record, to prevent spurious warnings that it should be
4481 -- declared as a constant, because intuitively programmers
4482 -- regard the designated subcomponent as part of the record.
4487 then
4492 -- Check bad case of atomic/volatile argument (RM C.6(12))
4496 then
4499 then
4500 Error_Msg_NE
4506 then
4507 Error_Msg_NE
4513 -- Check that subprograms don't have improper controlling
4514 -- arguments (RM 3.9.2 (9)).
4516 -- A primitive operation may have an access parameter of an
4517 -- incomplete tagged type, but a dispatching call is illegal
4518 -- if the type is still incomplete.
4524 declare
4526 begin
4530 then
4531 Error_Msg_NE
4542 -- Apply legality rule 3.9.2 (9/1)
4547 and then not In_Instance
4548 then
4553 Error_Msg_NE
4557 -- Apply the checks described in 3.10.2(27): if the context is a
4558 -- specific access-to-object, the actual cannot be class-wide.
4559 -- Use base type to exclude access_to_subprogram cases.
4566 and then
4571 -- Disable these checks for call to imported C++ subprograms
4573 and then not
4577 then
4578 Error_Msg_N
4583 Error_Msg_NE
4588 Check_Aliased_Parameter;
4592 -- If it is a named association, treat the selector_name as a
4593 -- proper identifier, and mark the corresponding entity.
4597 -- Ignore reference in SPARK mode, as it refers to an entity not
4598 -- in scope at the point of reference, so the reference should
4599 -- be ignored for computing effects of subprograms.
4601 and then not GNATprove_Mode
4602 then
4603 -- If subprogram is overridden, use name of formal that
4604 -- is being called.
4610 else
4624 -- The following checks are only relevant when SPARK_Mode is on as
4625 -- they are not standard Ada legality rule. Internally generated
4626 -- temporaries are ignored.
4630 -- An effectively volatile object may act as an actual when the
4631 -- corresponding formal is of a non-scalar effectively volatile
4632 -- type (SPARK RM 7.1.3(11)).
4636 then
4639 -- An effectively volatile object may act as an actual in a
4640 -- call to an instance of Unchecked_Conversion.
4641 -- (SPARK RM 7.1.3(11)).
4646 -- The actual denotes an object
4649 Error_Msg_N
4653 -- Otherwise the actual denotes an expression. Inspect the
4654 -- expression and flag each effectively volatile object with
4655 -- enabled property Async_Writers or Effective_Reads as illegal
4656 -- because it apprears within an interfering context. Note that
4657 -- this is usually done in Resolve_Entity_Name, but when the
4658 -- effectively volatile object appears as an actual in a call,
4659 -- the call must be resolved first.
4661 else
4665 -- An effectively volatile variable cannot act as an actual
4666 -- parameter in a procedure call when the variable has enabled
4667 -- property Effective_Reads and the corresponding formal is of
4668 -- mode IN (SPARK RM 7.1.3(10)).
4673 then
4679 then
4680 Error_Msg_NE
4691 -- A formal parameter of a specific tagged type whose related
4692 -- subprogram is subject to pragma Extensions_Visible with value
4693 -- "False" cannot act as an actual in a subprogram with value
4694 -- "True" (SPARK RM 6.1.7(3)).
4698 Extensions_Visible_True
4699 then
4700 Error_Msg_N
4703 Error_Msg_NE
4707 -- The actual parameter of a Ghost subprogram whose formal is of
4708 -- mode IN OUT or OUT must be a Ghost variable (SPARK RM 6.9(12)).
4716 then
4717 Error_Msg_NE
4723 else
4730 -- Case where actual is not present
4732 else
4733 Insert_Default;
4744 -----------------------
4745 -- Resolve_Allocator --
4746 -----------------------
4758 procedure Check_Allocator_Discrim_Accessibility
4761 -- Check that accessibility level associated with an access discriminant
4762 -- initialized in an allocator by the expression Disc_Exp is not deeper
4763 -- than the level of the allocator type Alloc_Typ. An error message is
4764 -- issued if this condition is violated. Specialized checks are done for
4765 -- the cases of a constraint expression which is an access attribute or
4766 -- an access discriminant.
4769 -- If the allocator is an actual in a call, it is allowed to be class-
4770 -- wide when the context is not because it is a controlling actual.
4772 -------------------------------------------
4773 -- Check_Allocator_Discrim_Accessibility --
4774 -------------------------------------------
4776 procedure Check_Allocator_Discrim_Accessibility
4779 is
4780 begin
4783 then
4784 Error_Msg_N
4787 -- When the expression is an Access attribute the level of the prefix
4788 -- object must not be deeper than that of the allocator's type.
4792 Attribute_Access
4795 then
4796 Error_Msg_N
4798 Disc_Exp);
4800 -- When the expression is an access discriminant the check is against
4801 -- the level of the prefix object.
4807 then
4808 Error_Msg_N
4810 Disc_Exp);
4812 -- All other cases are legal
4814 else
4819 ----------------------------
4820 -- In_Dispatching_Context --
4821 ----------------------------
4826 begin
4832 -- Start of processing for Resolve_Allocator
4834 begin
4835 -- Replace general access with specific type
4845 -- For qualified expression, resolve the expression using the given
4846 -- subtype (nothing to do for type mark, subtype indication)
4851 and then not In_Dispatching_Context
4852 then
4853 Error_Msg_N
4861 -- Allocators generated by the build-in-place expansion mechanism
4862 -- are explicitly marked as coming from source but do not need to be
4863 -- checked for limited initialization. To exclude this case, ensure
4864 -- that the parent of the allocator is a source node.
4865 -- The return statement constructed for an Expression_Function does
4866 -- not come from source but requires a limited check.
4870 and then
4872 or else
4876 and then not In_Instance_Body
4877 then
4880 Error_Msg_N
4883 else
4884 Error_Msg_N
4892 -- A qualified expression requires an exact match of the type. Class-
4893 -- wide matching is not allowed.
4898 then
4902 -- Calls to build-in-place functions are not currently supported in
4903 -- allocators for access types associated with a simple storage pool.
4904 -- Supporting such allocators may require passing additional implicit
4905 -- parameters to build-in-place functions (or a significant revision
4906 -- of the current b-i-p implementation to unify the handling for
4907 -- multiple kinds of storage pools). ???
4911 then
4912 declare
4915 begin
4917 and then
4918 Present (Get_Rep_Pragma
4920 then
4921 Error_Msg_N
4928 -- A special accessibility check is needed for allocators that
4929 -- constrain access discriminants. The level of the type of the
4930 -- expression used to constrain an access discriminant cannot be
4931 -- deeper than the type of the allocator (in contrast to access
4932 -- parameters, where the level of the actual can be arbitrary).
4934 -- We can't use Valid_Conversion to perform this check because in
4935 -- general the type of the allocator is unrelated to the type of
4936 -- the access discriminant.
4940 then
4947 then
4950 -- Get the first component expression of the aggregate
4960 else
4964 else
4983 else
4988 else
4997 -- For a subtype mark or subtype indication, freeze the subtype
4999 else
5003 Error_Msg_N
5007 -- A special accessibility check is needed for allocators that
5008 -- constrain access discriminants. The level of the type of the
5009 -- expression used to constrain an access discriminant cannot be
5010 -- deeper than the type of the allocator (in contrast to access
5011 -- parameters, where the level of the actual can be arbitrary).
5012 -- We can't use Valid_Conversion to perform this check because
5013 -- in general the type of the allocator is unrelated to the type
5014 -- of the access discriminant.
5019 then
5029 else
5043 -- Ada 2005 (AI-344): A class-wide allocator requires an accessibility
5044 -- check that the level of the type of the created object is not deeper
5045 -- than the level of the allocator's access type, since extensions can
5046 -- now occur at deeper levels than their ancestor types. This is a
5047 -- static accessibility level check; a run-time check is also needed in
5048 -- the case of an initialized allocator with a class-wide argument (see
5049 -- Expand_Allocator_Expression).
5053 then
5054 declare
5057 begin
5062 else
5068 then
5071 Error_Msg_N
5080 -- Do not apply Ada 2005 accessibility checks on a class-wide
5081 -- allocator if the type given in the allocator is a formal
5082 -- type. A run-time check will be performed in the instance.
5092 -- Check for allocation from an empty storage pool
5097 -- If the context is an unchecked conversion, as may happen within an
5098 -- inlined subprogram, the allocator is being resolved with its own
5099 -- anonymous type. In that case, if the target type has a specific
5100 -- storage pool, it must be inherited explicitly by the allocator type.
5104 then
5105 Set_Associated_Storage_Pool
5113 -- Check that an allocator with task parts isn't for a nested access
5114 -- type when restriction No_Task_Hierarchy applies.
5118 then
5122 -- An illegal allocator may be rewritten as a raise Program_Error
5123 -- statement.
5127 -- Avoid coextension processing for an allocator that is the
5128 -- expansion of a build-in-place function call.
5132 N_Qualified_Expression
5134 N_Function_Call
5135 and then Is_Expanded_Build_In_Place_Call
5137 then
5140 else
5141 -- An anonymous access discriminant is the definition of a
5142 -- coextension.
5146 N_Discriminant_Specification
5147 then
5148 declare
5152 begin
5155 -- Ada 2012 AI05-0052: If the designated type of the
5156 -- allocator is limited, then the allocator shall not
5157 -- be used to define the value of an access discriminant
5158 -- unless the discriminated type is immutably limited.
5163 then
5164 Error_Msg_N
5167 N);
5171 -- Avoid marking an allocator as a dynamic coextension if it is
5172 -- within a static construct.
5177 -- Finalization and deallocation of coextensions utilizes an
5178 -- approximate implementation which does not directly adhere
5179 -- to the semantic rules. Warn on potential issues involving
5180 -- coextensions.
5183 Error_Msg_N
5186 else
5187 Error_Msg_N
5193 -- Cleanup for potential static coextensions
5195 else
5199 -- Anonymous access-to-controlled objects are not finalized on
5200 -- time because this involves run-time ownership and currently
5201 -- this property is not available. In rare cases the object may
5202 -- not be finalized at all. Warn on potential issues involving
5203 -- anonymous access-to-controlled objects.
5207 then
5208 Error_Msg_N
5218 -- Report a simple error: if the designated object is a local task,
5219 -- its body has not been seen yet, and its activation will fail an
5220 -- elaboration check.
5227 then
5233 -- Ada 2012 (AI05-0111-3): Detect an attempt to allocate a task or a
5234 -- type with a task component on a subpool. This action must raise
5235 -- Program_Error at runtime.
5241 then
5253 ---------------------------
5254 -- Resolve_Arithmetic_Op --
5255 ---------------------------
5257 -- Used for resolving all arithmetic operators except exponentiation
5268 -- We do the resolution using the base type, because intermediate values
5269 -- in expressions always are of the base type, not a subtype of it.
5272 -- Returns True if N is in a context that expects "any real type"
5275 -- Return True iff given type is Integer or universal real/integer
5278 -- Choose type of integer literal in fixed-point operation to conform
5279 -- to available fixed-point type. T is the type of the other operand,
5280 -- which is needed to determine the expected type of N.
5283 -- Set operand type to T if universal
5285 -------------------------------
5286 -- Expected_Type_Is_Any_Real --
5287 -------------------------------
5290 begin
5291 -- N is the expression after "delta" in a fixed_point_definition;
5292 -- see RM-3.5.9(6):
5295 N_Decimal_Fixed_Point_Definition,
5297 -- N is one of the bounds in a real_range_specification;
5298 -- see RM-3.5.7(5):
5300 N_Real_Range_Specification,
5302 -- N is the expression of a delta_constraint;
5303 -- see RM-J.3(3):
5305 N_Delta_Constraint);
5308 -----------------------------
5309 -- Is_Integer_Or_Universal --
5310 -----------------------------
5317 begin
5323 else
5329 then
5340 ----------------------------
5341 -- Set_Mixed_Mode_Operand --
5342 ----------------------------
5348 begin
5351 -- A universal integer literal is resolved as standard integer
5352 -- except in the case of a fixed-point result, where we leave it
5353 -- as universal (to be handled by Exp_Fixd later on)
5357 else
5365 then
5366 -- A universal real can appear in a fixed-type context. We resolve
5367 -- the literal with that context, even though this might raise an
5368 -- exception prematurely (the other operand may be zero).
5375 then
5376 -- Integer arg in mixed-mode operation. Resolve with universal
5377 -- type, in case preference rule must be applied.
5383 then
5384 -- Not a mixed-mode operation, resolve with context
5390 -- N may itself be a mixed-mode operation, so use context type
5397 then
5398 -- Must be (fixed * fixed) operation, operand must have one
5399 -- compatible interpretation.
5406 then
5407 -- C * F(X) in a fixed context, where C is a real literal or a
5408 -- fixed-point expression. F must have either a fixed type
5409 -- interpretation or an integer interpretation, but not both.
5416 else
5423 else
5431 -- Reanalyze the literal with the fixed type of the context. If
5432 -- context is Universal_Fixed, we are within a conversion, leave
5433 -- the literal as a universal real because there is no usable
5434 -- fixed type, and the target of the conversion plays no role in
5435 -- the resolution.
5437 declare
5441 begin
5444 else
5450 then
5452 else
5460 -- A universal real conditional expression can appear in a fixed-type
5461 -- context and must be resolved with that context to facilitate the
5462 -- code generation to the backend.
5467 then
5470 else
5475 ----------------------
5476 -- Set_Operand_Type --
5477 ----------------------
5480 begin
5483 then
5488 -- Start of processing for Resolve_Arithmetic_Op
5490 begin
5495 then
5499 -- Special-case for mixed-mode universal expressions or fixed point type
5500 -- operation: each argument is resolved separately. The same treatment
5501 -- is required if one of the operands of a fixed point operation is
5502 -- universal real, since in this case we don't do a conversion to a
5503 -- specific fixed-point type (instead the expander handles the case).
5505 -- Set the type of the node to its universal interpretation because
5506 -- legality checks on an exponentiation operand need the context.
5511 then
5522 or else
5525 then
5530 -- If context is a fixed type and one operand is integer, the other
5531 -- is resolved with the type of the context.
5536 then
5543 then
5547 -- If both operands are universal and the context is a floating
5548 -- point type, the operands are resolved to the type of the context.
5554 else
5559 -- Check the rule in RM05-4.5.5(19.1/2) disallowing universal_fixed
5560 -- multiplying operators from being used when the expected type is
5561 -- also universal_fixed. Note that B_Typ will be Universal_Fixed in
5562 -- some cases where the expected type is actually Any_Real;
5563 -- Expected_Type_Is_Any_Real takes care of that case.
5567 then
5571 N_Unchecked_Type_Conversion)
5572 then
5579 else
5582 and then not
5584 N_Unchecked_Type_Conversion)
5585 then
5586 Error_Msg_N
5591 -- The expected type is "any real type" in contexts like
5593 -- type T is delta <universal_fixed-expression> ...
5595 -- in which case we need to set the type to Universal_Real
5596 -- so that static expression evaluation will work properly.
5600 else
5610 then
5615 -- If no previous errors, this is only possible if one operand is
5616 -- overloaded and the context is universal. Resolve as such.
5621 else
5623 and then
5625 then
5629 -- If the context is Universal_Fixed and the operands are also
5630 -- universal fixed, this is an error, unless there is only one
5631 -- applicable fixed_point type (usually Duration).
5639 else
5644 else
5649 -- If one of the arguments was resolved to a non-universal type.
5650 -- label the result of the operation itself with the same type.
5651 -- Do the same for the universal argument, if any.
5663 -- In SPARK, a multiplication or division with operands of fixed point
5664 -- types must be qualified or explicitly converted to identify the
5665 -- result type.
5670 and then
5672 then
5673 Check_SPARK_05_Restriction
5677 -- Set overflow and division checking bit
5684 -- Give warning if explicit division by zero
5688 then
5694 or else
5697 then
5698 -- Specialize the warning message according to the operation.
5699 -- When SPARK_Mode is On, force a warning instead of an error
5700 -- in that case, as this likely corresponds to deactivated
5701 -- code. The following warnings are for the case
5706 -- For division, we have two cases, for float division
5707 -- of an unconstrained float type, on a machine where
5708 -- Machine_Overflows is false, we don't get an exception
5709 -- at run-time, but rather an infinity or Nan. The Nan
5710 -- case is pretty obscure, so just warn about infinities.
5714 and then not Machine_Overflows_On_Target
5715 then
5716 Error_Msg_N
5720 -- For all other cases, we get a Constraint_Error
5722 else
5723 Apply_Compile_Time_Constraint_Error
5730 Apply_Compile_Time_Constraint_Error
5736 Apply_Compile_Time_Constraint_Error
5741 -- Division by zero can only happen with division, rem,
5742 -- and mod operations.
5748 -- In GNATprove mode, we enable the division check so that
5749 -- GNATprove will issue a message if it cannot be proved.
5755 -- Otherwise just set the flag to check at run time
5757 else
5762 -- If Restriction No_Implicit_Conditionals is active, then it is
5763 -- violated if either operand can be negative for mod, or for rem
5764 -- if both operands can be negative.
5768 then
5769 declare
5776 -- Set if corresponding operand might be negative
5778 begin
5779 Determine_Range
5783 Determine_Range
5787 -- Check if we will be generating conditionals. There are two
5788 -- cases where that can happen, first for REM, the only case
5789 -- is largest negative integer mod -1, where the division can
5790 -- overflow, but we still have to give the right result. The
5791 -- front end generates a test for this annoying case. Here we
5792 -- just test if both operands can be negative (that's what the
5793 -- expander does, so we match its logic here).
5795 -- The second case is mod where either operand can be negative.
5796 -- In this case, the back end has to generate additional tests.
5799 or else
5801 then
5812 ------------------
5813 -- Resolve_Call --
5814 ------------------
5817 function Same_Or_Aliased_Subprograms
5820 -- Returns True if the subprogram entity S is the same as E or else
5821 -- S is an alias of E.
5823 ---------------------------------
5824 -- Same_Or_Aliased_Subprograms --
5825 ---------------------------------
5827 function Same_Or_Aliased_Subprograms
5830 is
5832 begin
5836 -- Local variables
5850 -- Start of processing for Resolve_Call
5852 begin
5853 -- Preserve relevant elaboration-related attributes of the context which
5854 -- are no longer available or very expensive to recompute once analysis,
5855 -- resolution, and expansion are over.
5857 Mark_Elaboration_Attributes
5863 -- The context imposes a unique interpretation with type Typ on a
5864 -- procedure or function call. Find the entity of the subprogram that
5865 -- yields the expected type, and propagate the corresponding formal
5866 -- constraints on the actuals. The caller has established that an
5867 -- interpretation exists, and emitted an error if not unique.
5869 -- First deal with the case of a call to an access-to-subprogram,
5870 -- dereference made explicit in Analyze_Call.
5876 else
5877 -- Find the interpretation whose type (a subprogram type) has a
5878 -- return type that is compatible with the context. Analysis of
5879 -- the node has established that one exists.
5898 -- If the prefix is not an entity, then resolve it
5904 -- For an indirect call, we always invalidate checks, since we do not
5905 -- know whether the subprogram is local or global. Yes we could do
5906 -- better here, e.g. by knowing that there are no local subprograms,
5907 -- but it does not seem worth the effort. Similarly, we kill all
5908 -- knowledge of current constant values.
5910 Kill_Current_Values;
5912 -- If this is a procedure call which is really an entry call, do
5913 -- the conversion of the procedure call to an entry call. Protected
5914 -- operations use the same circuitry because the name in the call
5915 -- can be an arbitrary expression with special resolution rules.
5920 then
5927 -- Annotate the tree by creating a call marker in case the original
5928 -- call is transformed by expansion. The call marker is automatically
5929 -- saved for later examination by the ABE Processing phase.
5933 -- Kill checks and constant values, as above for indirect case
5934 -- Who knows what happens when another task is activated?
5936 Kill_Current_Values;
5939 -- Normal subprogram call with name established in Resolve
5945 -- Otherwise we must have the case of an overloaded call
5947 else
5950 -- Initialize Nam to prevent warning (we know it will be assigned
5951 -- in the loop below, but the compiler does not know that).
5971 then
5972 -- The prefix is a parameterless function call that returns an access
5973 -- to subprogram. If parameters are present in the current call, add
5974 -- add an explicit dereference. We use the base type here because
5975 -- within an instance these may be subtypes.
5977 -- The dereference is added either in Analyze_Call or here. Should
5978 -- be consolidated ???
5987 -- Check that a call to Current_Task does not occur in an entry body
5990 declare
5993 begin
5995 loop
5998 -- Exclude calls that occur within the default of a formal
5999 -- parameter of the entry, since those are evaluated outside
6000 -- of the body.
6007 then
6010 Error_Msg_NE
6024 -- Check that a procedure call does not occur in the context of the
6025 -- entry call statement of a conditional or timed entry call. Note that
6026 -- the case of a call to a subprogram renaming of an entry will also be
6027 -- rejected. The test for N not being an N_Entry_Call_Statement is
6028 -- defensive, covering the possibility that the processing of entry
6029 -- calls might reach this point due to later modifications of the code
6030 -- above.
6035 then
6039 -- Ada 2005 (AI-345): If a procedure_call_statement is used
6040 -- for a procedure_or_entry_call, the procedure_name or
6041 -- procedure_prefix of the procedure_call_statement shall denote
6042 -- an entry renamed by a procedure, or (a view of) a primitive
6043 -- subprogram of a limited interface whose first parameter is
6044 -- a controlling parameter.
6049 then
6050 Error_Msg_N
6055 -- If the SPARK_05 restriction is active, we are not allowed
6056 -- to have a call to a subprogram before we see its completion.
6061 -- Don't flag strange internal calls
6066 -- Only flag calls in extended main source
6071 -- Exclude enumeration literals from this processing
6074 then
6075 Check_SPARK_05_Restriction
6079 -- Check that this is not a call to a protected procedure or entry from
6080 -- within a protected function.
6084 -- Freeze the subprogram name if not in a spec-expression. Note that
6085 -- we freeze procedure calls as well as function calls. Procedure calls
6086 -- are not frozen according to the rules (RM 13.14(14)) because it is
6087 -- impossible to have a procedure call to a non-frozen procedure in
6088 -- pure Ada, but in the code that we generate in the expander, this
6089 -- rule needs extending because we can generate procedure calls that
6090 -- need freezing.
6092 -- In Ada 2012, expression functions may be called within pre/post
6093 -- conditions of subsequent functions or expression functions. Such
6094 -- calls do not freeze when they appear within generated bodies,
6095 -- (including the body of another expression function) which would
6096 -- place the freeze node in the wrong scope. An expression function
6097 -- is frozen in the usual fashion, by the appearance of a real body,
6098 -- or at the end of a declarative part.
6101 and then not In_Spec_Expression
6103 and then
6106 then
6110 -- For a predefined operator, the type of the result is the type imposed
6111 -- by context, except for a predefined operation on universal fixed.
6112 -- Otherwise The type of the call is the type returned by the subprogram
6113 -- being called.
6120 -- If the subprogram returns an array type, and the context requires the
6121 -- component type of that array type, the node is really an indexing of
6122 -- the parameterless call. Resolve as such. A pathological case occurs
6123 -- when the type of the component is an access to the array type. In
6124 -- this case the call is truly ambiguous. If the call is to an intrinsic
6125 -- subprogram, it can't be an indexed component. This check is necessary
6126 -- because if it's Unchecked_Conversion, and we have "type T_Ptr is
6127 -- access T;" and "type T is array (...) of T_Ptr;" (i.e. an array of
6128 -- pointers to the same array), the compiler gets confused and does an
6129 -- infinite recursion.
6132 and then
6135 or else
6138 and then
6141 then
6142 declare
6147 begin
6150 then
6151 Error_Msg_N
6153 else
6156 -- The called entity may be an explicit dereference, in which
6157 -- case there is no entity to set.
6165 or else
6167 and then
6169 then
6172 -- Indexed call to a parameterless function
6174 Index_Node :=
6176 Prefix =>
6179 else
6180 -- An Ada 2005 prefixed call to a primitive operation
6181 -- whose first parameter is the prefix. This prefix was
6182 -- prepended to the parameter list, which is actually a
6183 -- list of indexes. Remove the prefix in order to build
6184 -- the proper indexed component.
6186 Index_Node :=
6188 Prefix =>
6191 Parameter_Associations =>
6192 New_List
6197 -- Preserve the parenthesis count of the node
6201 -- Since we are correcting a node classification error made
6202 -- by the parser, we call Replace rather than Rewrite.
6214 -- Annotate the tree by creating a call marker in case
6215 -- the original call is transformed by expansion. The call
6216 -- marker is automatically saved for later examination by
6217 -- the ABE Processing phase.
6226 else
6227 -- If the called function is not declared in the main unit and it
6228 -- returns the limited view of type then use the available view (as
6229 -- is done in Try_Object_Operation) to prevent back-end confusion;
6230 -- for the function entity itself. The call must appear in a context
6231 -- where the nonlimited view is available. If the function entity is
6232 -- in the extended main unit then no action is needed, because the
6233 -- back end handles this case. In either case the type of the call
6234 -- is the nonlimited view.
6238 then
6245 else
6250 -- In the case where the call is to an overloaded subprogram, Analyze
6251 -- calls Normalize_Actuals once per overloaded subprogram. Therefore in
6252 -- such a case Normalize_Actuals needs to be called once more to order
6253 -- the actuals correctly. Otherwise the call will have the ordering
6254 -- given by the last overloaded subprogram whether this is the correct
6255 -- one being called or not.
6262 -- In any case, call is fully resolved now. Reset Overload flag, to
6263 -- prevent subsequent overload resolution if node is analyzed again
6268 -- A Ghost entity must appear in a specific context
6274 -- If we are calling the current subprogram from immediately within its
6275 -- body, then that is the case where we can sometimes detect cases of
6276 -- infinite recursion statically. Do not try this in case restriction
6277 -- No_Recursion is in effect anyway, and do it only for source calls.
6282 -- Check violation of SPARK_05 restriction which does not permit
6283 -- a subprogram body to contain a call to the subprogram directly.
6287 then
6288 Check_SPARK_05_Restriction
6292 -- Issue warning for possible infinite recursion in the absence
6293 -- of the No_Recursion restriction.
6298 then
6299 -- Here we detected and flagged an infinite recursion, so we do
6300 -- not need to test the case below for further warnings. Also we
6301 -- are all done if we now have a raise SE node.
6307 -- If call is to immediately containing subprogram, then check for
6308 -- the case of a possible run-time detectable infinite recursion.
6310 else
6314 -- Although in general case, recursion is not statically
6315 -- checkable, the case of calling an immediately containing
6316 -- subprogram is easy to catch.
6320 -- If the recursive call is to a parameterless subprogram,
6321 -- then even if we can't statically detect infinite
6322 -- recursion, this is pretty suspicious, and we output a
6323 -- warning. Furthermore, we will try later to detect some
6324 -- cases here at run time by expanding checking code (see
6325 -- Detect_Infinite_Recursion in package Exp_Ch6).
6327 -- If the recursive call is within a handler, do not emit a
6328 -- warning, because this is a common idiom: loop until input
6329 -- is correct, catch illegal input in handler and restart.
6335 then
6336 -- For the case of a procedure call. We give the message
6337 -- only if the call is the first statement in a sequence
6338 -- of statements, or if all previous statements are
6339 -- simple assignments. This is simply a heuristic to
6340 -- decrease false positives, without losing too many good
6341 -- warnings. The idea is that these previous statements
6342 -- may affect global variables the procedure depends on.
6343 -- We also exclude raise statements, that may arise from
6344 -- constraint checks and are probably unrelated to the
6345 -- intended control flow.
6349 then
6350 declare
6352 begin
6356 N_Raise_Constraint_Error)
6357 then
6366 -- Do not give warning if we are in a conditional context
6368 declare
6370 begin
6376 then
6381 -- Here warning is to be issued
6397 -- Check obsolescent reference to Ada.Characters.Handling subprogram
6401 -- If subprogram name is a predefined operator, it was given in
6402 -- functional notation. Replace call node with operator node, so
6403 -- that actuals can be resolved appropriately.
6411 then
6417 -- Create a transient scope if the resulting type requires it
6419 -- There are several notable exceptions:
6421 -- a) In init procs, the transient scope overhead is not needed, and is
6422 -- even incorrect when the call is a nested initialization call for a
6423 -- component whose expansion may generate adjust calls. However, if the
6424 -- call is some other procedure call within an initialization procedure
6425 -- (for example a call to Create_Task in the init_proc of the task
6426 -- run-time record) a transient scope must be created around this call.
6428 -- b) Enumeration literal pseudo-calls need no transient scope
6430 -- c) Intrinsic subprograms (Unchecked_Conversion and source info
6431 -- functions) do not use the secondary stack even though the return
6432 -- type may be unconstrained.
6434 -- d) Calls to a build-in-place function, since such functions may
6435 -- allocate their result directly in a target object, and cases where
6436 -- the result does get allocated in the secondary stack are checked for
6437 -- within the specialized Exp_Ch6 procedures for expanding those
6438 -- build-in-place calls.
6440 -- e) Calls to inlinable expression functions do not use the secondary
6441 -- stack (since the call will be replaced by its returned object).
6443 -- f) If the subprogram is marked Inline_Always, then even if it returns
6444 -- an unconstrained type the call does not require use of the secondary
6445 -- stack. However, inlining will only take place if the body to inline
6446 -- is already present. It may not be available if e.g. the subprogram is
6447 -- declared in a child instance.
6449 -- If this is an initialization call for a type whose construction
6450 -- uses the secondary stack, and it is not a nested call to initialize
6451 -- a component, we do need to create a transient scope for it. We
6452 -- check for this by traversing the type in Check_Initialization_Call.
6458 then
6465 then
6468 elsif Expander_Active
6471 and then
6473 or else
6475 then
6478 -- If the call appears within the bounds of a loop, it will
6479 -- be rewritten and reanalyzed, nothing left to do here.
6486 and then not Within_Init_Proc
6487 then
6491 -- A protected function cannot be called within the definition of the
6492 -- enclosing protected type, unless it is part of a pre/postcondition
6493 -- on another protected operation. This may appear in the entry wrapper
6494 -- created for an entry with preconditions.
6499 and then not In_Spec_Expression
6501 then
6502 Error_Msg_NE
6506 -- Propagate interpretation to actuals, and add default expressions
6507 -- where needed.
6512 -- Overloaded literals are rewritten as function calls, for purpose of
6513 -- resolution. After resolution, we can replace the call with the
6514 -- literal itself.
6520 -- Avoid validation, since it is a static function call
6526 -- If the subprogram is not global, then kill all saved values and
6527 -- checks. This is a bit conservative, since in many cases we could do
6528 -- better, but it is not worth the effort. Similarly, we kill constant
6529 -- values. However we do not need to do this for internal entities
6530 -- (unless they are inherited user-defined subprograms), since they
6531 -- are not in the business of molesting local values.
6533 -- If the flag Suppress_Value_Tracking_On_Calls is set, then we also
6534 -- kill all checks and values for calls to global subprograms. This
6535 -- takes care of the case where an access to a local subprogram is
6536 -- taken, and could be passed directly or indirectly and then called
6537 -- from almost any context.
6539 -- Note: we do not do this step till after resolving the actuals. That
6540 -- way we still take advantage of the current value information while
6541 -- scanning the actuals.
6543 -- We suppress killing values if we are processing the nodes associated
6544 -- with N_Freeze_Entity nodes. Otherwise the declaration of a tagged
6545 -- type kills all the values as part of analyzing the code that
6546 -- initializes the dispatch tables.
6550 or else Suppress_Value_Tracking_On_Call
6555 then
6556 Kill_Current_Values;
6559 -- If we are warning about unread OUT parameters, this is the place to
6560 -- set Last_Assignment for OUT and IN OUT parameters. We have to do this
6561 -- after the above call to Kill_Current_Values (since that call clears
6562 -- the Last_Assignment field of all local variables).
6567 then
6568 declare
6572 begin
6582 then
6592 -- If the subprogram is a primitive operation, check whether or not
6593 -- it is a correct dispatching call.
6597 then
6602 and then not In_Instance
6603 then
6607 -- If this is a dispatching call, generate the appropriate reference,
6608 -- for better source navigation in GPS.
6612 then
6615 -- Normal case, not a dispatching call: generate a call reference
6617 else
6625 -- Check for violation of restriction No_Specific_Termination_Handlers
6626 -- and warn on a potentially blocking call to Abort_Task.
6630 or else
6632 then
6639 -- A call to Ada.Real_Time.Timing_Events.Set_Handler to set a relative
6640 -- timing event violates restriction No_Relative_Delay (AI-0211). We
6641 -- need to check the second argument to determine whether it is an
6642 -- absolute or relative timing event.
6647 then
6651 -- Issue an error for a call to an eliminated subprogram. This routine
6652 -- will not perform the check if the call appears within a default
6653 -- expression.
6657 -- In formal mode, the primitive operations of a tagged type or type
6658 -- extension do not include functions that return the tagged type.
6664 then
6668 -- Implement rule in 12.5.1 (23.3/2): In an instance, if the actual is
6669 -- class-wide and the call dispatches on result in a context that does
6670 -- not provide a tag, the call raises Program_Error.
6673 and then In_Instance
6678 then
6679 -- Verify that none of the formals are controlling
6681 declare
6685 begin
6707 -- Check for calling a function with OUT or IN OUT parameter when the
6708 -- calling context (us right now) is not Ada 2012, so does not allow
6709 -- OUT or IN OUT parameters in function calls. Functions declared in
6710 -- a predefined unit are OK, as they may be called indirectly from a
6711 -- user-declared instantiation.
6717 then
6722 -- Check the dimensions of the actuals in the call. For function calls,
6723 -- propagate the dimensions from the returned type to N.
6727 -- All done, evaluate call and deal with elaboration issues
6735 -- Annotate the tree by creating a call marker in case the original call
6736 -- is transformed by expansion. The call marker is automatically saved
6737 -- for later examination by the ABE Processing phase.
6741 -- In GNATprove mode, expansion is disabled, but we want to inline some
6742 -- subprograms to facilitate formal verification. Indirect calls through
6743 -- a subprogram type or within a generic cannot be inlined. Inlining is
6744 -- performed only for calls subject to SPARK_Mode on.
6746 if GNATprove_Mode
6749 and then not Inside_A_Generic
6750 then
6757 -- Nothing to do if the subprogram is not eligible for inlining in
6758 -- GNATprove mode, or inlining is disabled with switch -gnatdm
6762 or else Debug_Flag_M
6763 then
6766 -- Calls cannot be inlined inside assertions, as GNATprove treats
6767 -- assertions as logic expressions. Only issue a message when the
6768 -- body has been seen, otherwise this leads to spurious messages
6769 -- on expression functions.
6773 Cannot_Inline
6777 -- Calls cannot be inlined inside default expressions
6780 Cannot_Inline
6783 -- Inlining should not be performed during pre-analysis
6787 -- Do not inline calls inside expression functions, as this
6788 -- would prevent interpreting them as logical formulas in
6789 -- GNATprove. Only issue a message when the body has been seen,
6790 -- otherwise this leads to spurious messages on callees that
6791 -- are themselves expression functions.
6794 and then Is_Expression_Function_Or_Completion
6796 then
6799 then
6800 Cannot_Inline
6805 -- With the one-pass inlining technique, a call cannot be
6806 -- inlined if the corresponding body has not been seen yet.
6809 Cannot_Inline
6812 -- Nothing to do if there is no body to inline, indicating that
6813 -- the subprogram is not suitable for inlining in GNATprove
6814 -- mode.
6819 -- Calls cannot be inlined inside potentially unevaluated
6820 -- expressions, as this would create complex actions inside
6821 -- expressions, that are not handled by GNATprove.
6824 Cannot_Inline
6828 -- Do not inline calls which would possibly lead to missing a
6829 -- type conversion check on an input parameter.
6832 Cannot_Inline
6836 -- Otherwise, inline the call
6838 else
6850 -----------------------------
6851 -- Resolve_Case_Expression --
6852 -----------------------------
6860 begin
6872 -- When the expression is of a scalar subtype different from the
6873 -- result subtype, then insert a conversion to ensure the generation
6874 -- of a constraint check.
6884 -- Apply RM 4.5.7 (17/3): whether the expression is statically or
6885 -- dynamically tagged must be known statically.
6893 Error_Msg_N
6907 -------------------------------
6908 -- Resolve_Character_Literal --
6909 -------------------------------
6915 begin
6916 -- Verify that the character does belong to the type of the context
6921 -- Wide_Wide_Character literals must always be defined, since the set
6922 -- of wide wide character literals is complete, i.e. if a character
6923 -- literal is accepted by the parser, then it is OK for wide wide
6924 -- character (out of range character literals are rejected).
6929 -- Always accept character literal for type Any_Character, which
6930 -- occurs in error situations and in comparisons of literals, both
6931 -- of which should accept all literals.
6936 -- For Standard.Character or a type derived from it, check that the
6937 -- literal is in range.
6944 -- For Standard.Wide_Character or a type derived from it, check that the
6945 -- literal is in range.
6952 -- If the entity is already set, this has already been resolved in a
6953 -- generic context, or comes from expansion. Nothing else to do.
6958 -- Otherwise we have a user defined character type, and we can use the
6959 -- standard visibility mechanisms to locate the referenced entity.
6961 else
6974 -- If we fall through, then the literal does not match any of the
6975 -- entries of the enumeration type. This isn't just a constraint error
6976 -- situation, it is an illegality (see RM 4.2).
6978 Error_Msg_NE
6982 ---------------------------
6983 -- Resolve_Comparison_Op --
6984 ---------------------------
6986 -- Context requires a boolean type, and plays no role in resolution.
6987 -- Processing identical to that for equality operators. The result type is
6988 -- the base type, which matters when pathological subtypes of booleans with
6989 -- limited ranges are used.
6996 begin
6997 -- If this is an intrinsic operation which is not predefined, use the
6998 -- types of its declared arguments to resolve the possibly overloaded
6999 -- operands. Otherwise the operands are unambiguous and specify the
7000 -- expected type.
7005 else
7016 -- Skip remaining processing if already set to Any_Type
7022 -- Deal with other error cases
7026 T = Any_Character
7027 then
7030 else
7038 -- Resolve the operands if types OK
7047 -- In SPARK, ordering operators <, <=, >, >= are not defined for Boolean
7048 -- types or array types except String.
7051 Check_SPARK_05_Restriction
7056 then
7057 Check_SPARK_05_Restriction
7061 -- Check comparison on unordered enumeration
7065 Error_Msg_NE
7072 -- Evaluate the relation (note we do this after the above check since
7073 -- this Eval call may change N to True/False. Skip this evaluation
7074 -- inside assertions, in order to keep assertions as written by users
7075 -- for tools that rely on these, e.g. GNATprove for loop invariants.
7076 -- Except evaluation is still performed even inside assertions for
7077 -- comparisons between values of universal type, which are useless
7078 -- for static analysis tools, and not supported even by GNATprove.
7082 and then
7084 then
7089 -----------------------------------------
7090 -- Resolve_Discrete_Subtype_Indication --
7091 -----------------------------------------
7093 procedure Resolve_Discrete_Subtype_Indication
7096 is
7100 begin
7108 else
7118 Error_Msg_NE
7121 -- Rewrite the constraint as a range of Typ
7122 -- to allow compilation to proceed further.
7134 else
7138 -- Additionally, we must check that the bounds are compatible
7139 -- with the given subtype, which might be different from the
7140 -- type of the context.
7144 -- ??? If the above check statically detects a Constraint_Error
7145 -- it replaces the offending bound(s) of the range R with a
7146 -- Constraint_Error node. When the itype which uses these bounds
7147 -- is frozen the resulting call to Duplicate_Subexpr generates
7148 -- a new temporary for the bounds.
7150 -- Unfortunately there are other itypes that are also made depend
7151 -- on these bounds, so when Duplicate_Subexpr is called they get
7152 -- a forward reference to the newly created temporaries and Gigi
7153 -- aborts on such forward references. This is probably sign of a
7154 -- more fundamental problem somewhere else in either the order of
7155 -- itype freezing or the way certain itypes are constructed.
7157 -- To get around this problem we call Remove_Side_Effects right
7158 -- away if either bounds of R are a Constraint_Error.
7160 declare
7164 begin
7180 -------------------------
7181 -- Resolve_Entity_Name --
7182 -------------------------
7184 -- Used to resolve identifiers and expanded names
7187 function Is_Assignment_Or_Object_Expression
7190 -- Determine whether node Context denotes an assignment statement or an
7191 -- object declaration whose expression is node Expr.
7193 ----------------------------------------
7194 -- Is_Assignment_Or_Object_Expression --
7195 ----------------------------------------
7197 function Is_Assignment_Or_Object_Expression
7200 is
7201 begin
7203 N_Object_Declaration)
7205 then
7208 -- Check whether a construct that yields a name is the expression of
7209 -- an assignment statement or an object declaration.
7212 N_Explicit_Dereference,
7213 N_Indexed_Component,
7214 N_Selected_Component,
7215 N_Slice)
7217 or else
7219 N_Unchecked_Type_Conversion)
7221 then
7222 return
7223 Is_Assignment_Or_Object_Expression
7227 -- Otherwise the context is not an assignment statement or an object
7228 -- declaration.
7230 else
7235 -- Local variables
7240 -- Start of processing for Resolve_Entity_Name
7242 begin
7243 -- If garbage from errors, set to Any_Type and return
7250 -- Replace named numbers by corresponding literals. Note that this is
7251 -- the one case where Resolve_Entity_Name must reset the Etype, since
7252 -- it is currently marked as universal.
7262 -- For enumeration literals, we need to make sure that a proper style
7263 -- check is done, since such literals are overloaded, and thus we did
7264 -- not do a style check during the first phase of analysis.
7270 -- Case of (sub)type name appearing in a context where an expression
7271 -- is expected. This is legal if occurrence is a current instance.
7272 -- See RM 8.6 (17/3).
7278 -- Any other use is an error
7280 else
7281 Error_Msg_N
7285 -- Check discriminant use if entity is discriminant in current scope,
7286 -- i.e. discriminant of record or concurrent type currently being
7287 -- analyzed. Uses in corresponding body are unrestricted.
7292 then
7295 -- A parameterless generic function cannot appear in a context that
7296 -- requires resolution.
7301 -- In Ada 83 an OUT parameter cannot be read, but attributes of
7302 -- array types (i.e. bounds and length) are legal.
7310 or else Is_Assignment_Or_Object_Expression
7313 then
7318 -- In all other cases, just do the possible static evaluation
7320 else
7321 -- A deferred constant that appears in an expression must have a
7322 -- completion, unless it has been removed by in-place expansion of
7323 -- an aggregate. A constant that is a renaming does not need
7324 -- initialization.
7330 and then not In_Spec_Expression
7333 then
7337 then
7339 else
7340 Error_Msg_N
7350 -- When the entity appears in a parameter association, retrieve the
7351 -- related subprogram call.
7359 -- The following checks are only relevant when SPARK_Mode is on as
7360 -- they are not standard Ada legality rules.
7364 -- An effectively volatile object subject to enabled properties
7365 -- Async_Writers or Effective_Reads must appear in non-interfering
7366 -- context (SPARK RM 7.1.3(12)).
7373 then
7374 SPARK_Msg_N
7379 -- Check for possible elaboration issues with respect to reads of
7380 -- variables. The act of renaming the variable is not considered a
7381 -- read as it simply establishes an alias.
7383 if Legacy_Elaboration_Checks
7385 and then Dynamic_Elaboration_Checks
7387 then
7392 -- The variable may eventually become a constituent of a single
7393 -- protected/task type. Record the reference now and verify its
7394 -- legality when analyzing the contract of the variable
7395 -- (SPARK RM 9.3).
7401 -- A Ghost entity must appear in a specific context
7411 -------------------
7412 -- Resolve_Entry --
7413 -------------------
7425 -- If the bounds of the entry family being called depend on task
7426 -- discriminants, build a new index subtype where a discriminant is
7427 -- replaced with the value of the discriminant of the target task.
7428 -- The target task is the prefix of the entry name in the call.
7430 -----------------------
7431 -- Actual_Index_Type --
7432 -----------------------
7442 -- If the bound is given by a discriminant, replace with a reference
7443 -- to the discriminant of the same name in the target task. If the
7444 -- entry name is the target of a requeue statement and the entry is
7445 -- in the current protected object, the bound to be used is the
7446 -- discriminal of the object (see Apply_Range_Checks for details of
7447 -- the transformation).
7449 -----------------------------
7450 -- Actual_Discriminant_Ref --
7451 -----------------------------
7457 begin
7462 then
7468 then
7469 -- Note: here Bound denotes a discriminant of the corresponding
7470 -- record type tskV, whose discriminal is a formal of the
7471 -- init-proc tskVIP. What we want is the body discriminal,
7472 -- which is associated to the discriminant of the original
7473 -- concurrent type tsk.
7475 return New_Occurrence_Of
7478 else
7479 Ref :=
7489 -- Start of processing for Actual_Index_Type
7491 begin
7494 then
7497 else
7511 -- Start of processing for Resolve_Entry
7513 begin
7514 -- Find name of entry being called, and resolve prefix of name with its
7515 -- own type. The prefix can be overloaded, and the name and signature of
7516 -- the entry must be taken into account.
7520 -- Case of dealing with entry family within the current tasks
7524 else
7530 -- Entry call to an entry (or entry family) in the current task. This
7531 -- is legal even though the task will deadlock. Rewrite as call to
7532 -- current task.
7534 -- This can also be a call to an entry in an enclosing task. If this
7535 -- is a single task, we have to retrieve its name, because the scope
7536 -- of the entry is the task type, not the object. If the enclosing
7537 -- task is a task type, the identity of the task is given by its own
7538 -- self variable.
7540 -- Finally this can be a requeue on an entry of the same task or
7541 -- protected object.
7548 then
7549 -- S is an enclosing task or protected object. The concurrent
7550 -- declaration has been converted into a type declaration, and
7551 -- the object itself has an object declaration that follows
7552 -- the type in the same declarative part.
7564 -- Call to current task. Will be transformed into call to Self
7571 New_N :=
7574 Selector_Name =>
7581 then
7582 -- Use the entry name (which must be unique at this point) to find
7583 -- the prefix that returns the corresponding task/protected type.
7585 declare
7591 begin
7613 -- Generate a reference for the index when it denotes an entity
7619 -- Up to this point the expression could have been the actual in a
7620 -- simple entry call, and be given by a named association.
7624 else
7630 ------------------------
7631 -- Resolve_Entry_Call --
7632 ------------------------
7643 begin
7644 -- We kill all checks here, because it does not seem worth the effort to
7645 -- do anything better, an entry call is a big operation.
7647 Kill_All_Checks;
7649 -- Processing of the name is similar for entry calls and protected
7650 -- operation calls. Once the entity is determined, we can complete
7651 -- the resolution of the actuals.
7653 -- The selector may be overloaded, in the case of a protected object
7654 -- with overloaded functions. The type of the context is used for
7655 -- resolution.
7660 then
7661 declare
7665 begin
7684 -- Simple entry or protected operation call
7697 -- Call to member of entry family
7704 -- We cannot in general check the maximum depth of protected entry calls
7705 -- at compile time. But we can tell that any protected entry call at all
7706 -- violates a specified nesting depth of zero.
7712 -- Use context type to disambiguate a protected function that can be
7713 -- called without actuals and that returns an array type, and where the
7714 -- argument list may be an indexing of the returned value.
7719 and then
7725 and then
7726 Covers
7729 then
7730 declare
7733 begin
7734 Index_Node :=
7736 Prefix =>
7740 -- Since we are correcting a node classification error made by the
7741 -- parser, we call Replace rather than Rewrite.
7754 then
7755 -- Note the entity being called before rewriting the call, so that
7756 -- it appears used at this point.
7760 -- Rewrite as call to the precondition wrapper, adding the task
7761 -- object to the list of actuals. If the call is to a member of an
7762 -- entry family, include the index as well.
7764 declare
7768 begin
7777 New_Call :=
7779 Name =>
7784 -- Preanalyze and resolve new call. Current procedure is called
7785 -- from Resolve_Call, after which expansion will take place.
7792 -- The operation name may have been overloaded. Order the actuals
7793 -- according to the formals of the resolved entity, and set the return
7794 -- type to that of the operation.
7801 -- Reset the Is_Overloaded flag, since resolution is now completed
7803 -- Simple entry call
7808 -- Call to a member of an entry family
7818 -- Create a call reference to the entry
7826 -- Verify that a procedure call cannot masquerade as an entry
7827 -- call where an entry call is expected.
7832 then
7837 then
7842 then
7847 -- After resolution, entry calls and protected procedure calls are
7848 -- changed into entry calls, for expansion. The structure of the node
7849 -- does not change, so it can safely be done in place. Protected
7850 -- function calls must keep their structure because they are
7851 -- subexpressions.
7855 -- A protected operation that is not a function may modify the
7856 -- corresponding object, and cannot apply to a constant. If this
7857 -- is an internal call, the prefix is the type itself.
7863 then
7864 Error_Msg_N
7866 Entry_Name);
7869 declare
7872 begin
7873 Entry_Call :=
7878 -- Inherit relevant attributes from the original call
7880 Set_First_Named_Actual
7883 Set_Is_Elaboration_Checks_OK_Node
7886 Set_Is_Elaboration_Warnings_OK_Node
7889 Set_Is_SPARK_Mode_On_Node
7896 -- Protected functions can return on the secondary stack, in which
7897 -- case we must trigger the transient scope mechanism.
7899 elsif Expander_Active
7901 then
7906 -------------------------
7907 -- Resolve_Equality_Op --
7908 -------------------------
7910 -- Both arguments must have the same type, and the boolean context does
7911 -- not participate in the resolution. The first pass verifies that the
7912 -- interpretation is not ambiguous, and the type of the left argument is
7913 -- correctly set, or is Any_Type in case of ambiguity. If both arguments
7914 -- are strings or aggregates, allocators, or Null, they are ambiguous even
7915 -- though they carry a single (universal) type. Diagnose this case here.
7923 -- The resolution rule for if expressions requires that each such must
7924 -- have a unique type. This means that if several dependent expressions
7925 -- are of a non-null anonymous access type, and the context does not
7926 -- impose an expected type (as can be the case in an equality operation)
7927 -- the expression must be rejected.
7930 -- Attempt to explain the nature of a redundant comparison with True. If
7931 -- the expression N is too complex, this routine issues a general error
7932 -- message.
7935 -- In the case of allocators and access attributes, the context must
7936 -- provide an indication of the specific access type to be used. If
7937 -- one operand is of such a "generic" access type, check whether there
7938 -- is a specific visible access type that has the same designated type.
7939 -- This is semantically dubious, and of no interest to any real code,
7940 -- but c48008a makes it all worthwhile.
7942 -------------------------
7943 -- Check_If_Expression --
7944 -------------------------
7950 begin
7957 then
7963 ------------------------
7964 -- Explain_Redundancy --
7965 ------------------------
7972 begin
7975 -- Strip the operand down to an entity
7977 loop
7980 else
7985 -- The construct denotes an entity
7990 -- Do not generate an error message when the comparison is done
7991 -- against the enumeration literal Standard.True.
7995 -- Build a customized error message
8022 -- The construct is too complex to disect, issue a general message
8024 else
8029 -----------------------------
8030 -- Find_Unique_Access_Type --
8031 -----------------------------
8038 begin
8040 E_Access_Attribute_Type)
8041 then
8045 E_Access_Attribute_Type)
8046 then
8048 else
8060 then
8073 -- Start of processing for Resolve_Equality_Op
8075 begin
8086 T = Any_Character
8087 then
8090 else
8099 then
8108 -- If expressions must have a single type, and if the context does
8109 -- not impose one the dependent expressions cannot be anonymous
8110 -- access types.
8112 -- Why no similar processing for case expressions???
8116 E_Anonymous_Access_Subprogram_Type)
8118 E_Anonymous_Access_Subprogram_Type)
8119 then
8127 -- In SPARK, equality operators = and /= for array types other than
8128 -- String are only defined when, for each index position, the
8129 -- operands have equal static bounds.
8133 -- Protect call to Matching_Static_Array_Bounds to avoid costly
8134 -- operation if not needed.
8142 then
8143 Check_SPARK_05_Restriction
8148 -- If the unique type is a class-wide type then it will be expanded
8149 -- into a dispatching call to the predefined primitive. Therefore we
8150 -- check here for potential violation of such restriction.
8156 -- Only warn for redundant equality comparison to True for objects
8157 -- (e.g. "X = True") and operations (e.g. "(X < Y) = True"). For
8158 -- other expressions, it may be a matter of preference to write
8159 -- "Expr = True" or "Expr".
8161 if Warn_On_Redundant_Constructs
8166 and then
8168 or else
8170 then
8171 Error_Msg_N -- CODEFIX
8181 -- If this is an inequality, it may be the implicit inequality
8182 -- created for a user-defined operation, in which case the corres-
8183 -- ponding equality operation is not intrinsic, and the operation
8184 -- cannot be constant-folded. Else fold.
8189 or else Is_Intrinsic_Subprogram
8191 then
8197 then
8201 -- Ada 2005: If one operand is an anonymous access type, convert the
8202 -- other operand to it, to ensure that the underlying types match in
8203 -- the back-end. Same for access_to_subprogram, and the conversion
8204 -- verifies that the types are subtype conformant.
8206 -- We apply the same conversion in the case one of the operands is a
8207 -- private subtype of the type of the other.
8209 -- Why the Expander_Active test here ???
8211 if Expander_Active
8212 and then
8214 E_Anonymous_Access_Subprogram_Type)
8216 then
8236 ----------------------------------
8237 -- Resolve_Explicit_Dereference --
8238 ----------------------------------
8246 -- The candidate prefix type, if overloaded
8251 begin
8255 -- A useful optimization: check whether the dereference denotes an
8256 -- element of a container, and if so rewrite it as a call to the
8257 -- corresponding Element function.
8259 -- Disabled for now, on advice of ARG. A more restricted form of the
8260 -- predicate might be acceptable ???
8262 -- if Is_Container_Element (N) then
8263 -- return;
8264 -- end if;
8268 -- Use the context type to select the prefix that has the correct
8269 -- designated type. Keep the first match, which will be the inner-
8270 -- most.
8277 then
8282 -- Remove access types that do not match, but preserve access
8283 -- to subprogram interpretations, in case a further dereference
8284 -- is needed (see below).
8297 else
8298 -- If no interpretation covers the designated type of the prefix,
8299 -- this is the pathological case where not all implementations of
8300 -- the prefix allow the interpretation of the node as a call. Now
8301 -- that the expected type is known, Remove other interpretations
8302 -- from prefix, rewrite it as a call, and resolve again, so that
8303 -- the proper call node is generated.
8314 New_N :=
8316 Name =>
8327 -- If not overloaded, resolve P with its own type
8329 else
8333 -- If the prefix might be null, add an access check
8337 then
8341 -- If the designated type is a packed unconstrained array type, and the
8342 -- explicit dereference is not in the context of an attribute reference,
8343 -- then we must compute and set the actual subtype, since it is needed
8344 -- by Gigi. The reason we exclude the attribute case is that this is
8345 -- handled fine by Gigi, and in fact we use such attributes to build the
8346 -- actual subtype. We also exclude generated code (which builds actual
8347 -- subtypes directly if they are needed).
8354 then
8360 -- Note: No Eval processing is required for an explicit dereference,
8361 -- because such a name can never be static.
8365 -------------------------------------
8366 -- Resolve_Expression_With_Actions --
8367 -------------------------------------
8370 begin
8373 -- If N has no actions, and its expression has been constant folded,
8374 -- then rewrite N as just its expression. Note, we can't do this in
8375 -- the general case of Is_Empty_List (Actions (N)) as this would cause
8376 -- Expression (N) to be expanded again.
8380 then
8385 ----------------------------------
8386 -- Resolve_Generalized_Indexing --
8387 ----------------------------------
8395 begin
8396 -- In ASIS mode, propagate the information about the indexes back to
8397 -- to the original indexing node. The generalized indexing is either
8398 -- a function call, or a dereference of one. The actuals include the
8399 -- prefix of the original node, which is the container expression.
8408 loop
8417 -- If expression is to be reanalyzed, reset Generalized_Indexing
8418 -- to recreate call node, as is the case when the expression is
8419 -- part of an expression function.
8428 else
8434 ---------------------------
8435 -- Resolve_If_Expression --
8436 ---------------------------
8445 begin
8446 -- Defend against malformed expressions
8464 -- When the "then" expression is of a scalar subtype different from the
8465 -- result subtype, then insert a conversion to ensure the generation of
8466 -- a constraint check. The same is done for the else part below, again
8467 -- comparing subtypes rather than base types.
8474 -- If ELSE expression present, just resolve using the determined type
8475 -- If type is universal, resolve to any member of the class.
8484 else
8494 -- Apply RM 4.5.7 (17/3): whether the expression is statically or
8495 -- dynamically tagged must be known statically.
8500 then
8506 -- If no ELSE expression is present, root type must be Standard.Boolean
8507 -- and we provide a Standard.True result converted to the appropriate
8508 -- Boolean type (in case it is a derived boolean type).
8511 Else_Expr :=
8516 else
8530 -------------------------------
8531 -- Resolve_Indexed_Component --
8532 -------------------------------
8540 begin
8548 -- Use the context type to select the prefix that yields the correct
8549 -- component type.
8551 declare
8558 begin
8565 and then
8566 Covers
8569 then
8578 else
8584 else
8595 else
8602 -- If prefix is access type, dereference to get real array type.
8603 -- Note: we do not apply an access check because the expander always
8604 -- introduces an explicit dereference, and the check will happen there.
8610 -- If name was overloaded, set component type correctly now
8611 -- If a misplaced call to an entry family (which has no index types)
8612 -- return. Error will be diagnosed from calling context.
8616 else
8623 -- The prefix may have resolved to a string literal, in which case its
8624 -- etype has a special representation. This is only possible currently
8625 -- if the prefix is a static concatenation, written in functional
8626 -- notation.
8631 else
8638 else
8649 -- Do not generate the warning on suspicious index if we are analyzing
8650 -- package Ada.Tags; otherwise we will report the warning with the
8651 -- Prims_Ptr field of the dispatch table.
8654 or else not
8656 Ada_Tags)
8657 then
8662 -- If the array type is atomic, and the component is not atomic, then
8663 -- this is worth a warning, since we have a situation where the access
8664 -- to the component may cause extra read/writes of the atomic array
8665 -- object, or partial word accesses, which could be unexpected.
8671 and then Has_Atomic_Components
8674 then
8675 Error_Msg_N
8677 Error_Msg_N
8682 -----------------------------
8683 -- Resolve_Integer_Literal --
8684 -----------------------------
8687 begin
8692 --------------------------------
8693 -- Resolve_Intrinsic_Operator --
8694 --------------------------------
8703 -- If the operand is a literal, it cannot be the expression in a
8704 -- conversion. Use a qualified expression instead.
8706 ---------------------
8707 -- Convert_Operand --
8708 ---------------------
8714 begin
8716 Res :=
8722 else
8729 -- Start of processing for Resolve_Intrinsic_Operator
8731 begin
8732 -- We must preserve the original entity in a generic setting, so that
8733 -- the legality of the operation can be verified in an instance.
8748 -- If the result or operand types are private, rewrite with unchecked
8749 -- conversions on the operands and the result, to expose the proper
8750 -- underlying numeric type.
8755 then
8760 else
8781 then
8782 -- Add explicit conversion where needed, and save interpretations in
8783 -- case operands are overloaded.
8790 else
8796 else
8806 else
8811 --------------------------------------
8812 -- Resolve_Intrinsic_Unary_Operator --
8813 --------------------------------------
8815 procedure Resolve_Intrinsic_Unary_Operator
8818 is
8823 begin
8842 else
8847 ------------------------
8848 -- Resolve_Logical_Op --
8849 ------------------------
8854 begin
8857 -- Predefined operations on scalar types yield the base type. On the
8858 -- other hand, logical operations on arrays yield the type of the
8859 -- arguments (and the context).
8863 else
8867 -- The following test is required because the operands of the operation
8868 -- may be literals, in which case the resulting type appears to be
8869 -- compatible with a signed integer type, when in fact it is compatible
8870 -- only with modular types. If the context itself is universal, the
8871 -- operation is illegal.
8879 Error_Msg_N
8887 then
8891 -- Replace AND by AND THEN, or OR by OR ELSE, if Short_Circuit_And_Or
8892 -- is active and the result type is standard Boolean (do not mess with
8893 -- ops that return a nonstandard Boolean type, because something strange
8894 -- is going on).
8896 -- Note: you might expect this replacement to be done during expansion,
8897 -- but that doesn't work, because when the pragma Short_Circuit_And_Or
8898 -- is used, no part of the right operand of an "and" or "or" operator
8899 -- should be executed if the left operand would short-circuit the
8900 -- evaluation of the corresponding "and then" or "or else". If we left
8901 -- the replacement to expansion time, then run-time checks associated
8902 -- with such operands would be evaluated unconditionally, due to being
8903 -- before the condition prior to the rewriting as short-circuit forms
8904 -- during expansion.
8906 if Short_Circuit_And_Or
8909 then
8910 -- Mark the corresponding putative SCO operator as truly a logical
8911 -- (and short-circuit) operator.
8924 -- Case of OR changed to OR ELSE
8926 else
8934 -- Return now, since analysis of the rewritten ops will take care of
8935 -- other reference bookkeeping and expression folding.
8950 -- In SPARK, logical operations AND, OR and XOR for arrays are defined
8951 -- only when both operands have same static lower and higher bounds. Of
8952 -- course the types have to match, so only check if operands are
8953 -- compatible and the node itself has no errors.
8957 then
8958 declare
8962 begin
8963 -- Protect call to Matching_Static_Array_Bounds to avoid costly
8964 -- operation if not needed.
8971 then
8972 Check_SPARK_05_Restriction
8979 ---------------------------
8980 -- Resolve_Membership_Op --
8981 ---------------------------
8983 -- The context can only be a boolean type, and does not determine the
8984 -- arguments. Arguments should be unambiguous, but the preference rule for
8985 -- universal types applies.
8995 -- Analysis has determined a unique type for the left operand. Use it to
8996 -- resolve the disjuncts.
8998 ----------------------------
8999 -- Resolve_Set_Membership --
9000 ----------------------------
9006 begin
9007 -- If the left operand is overloaded, find type compatible with not
9008 -- overloaded alternative of the right operand.
9017 else
9022 -- Unclear how to resolve expression if all alternatives are also
9023 -- overloaded.
9029 else
9038 -- Alternative is an expression, a range
9039 -- or a subtype mark.
9043 then
9050 -- Check for duplicates for discrete case
9053 declare
9062 begin
9063 -- Loop checking duplicates. This is quadratic, but giant sets
9064 -- are unlikely in this context so it's a reasonable choice.
9071 N_Character_Literal)
9073 then
9090 -- RM 4.5.2 (28.1/3) specifies that for types other than records or
9091 -- limited types, evaluation of a membership test uses the predefined
9092 -- equality for the type. This may be confusing to users, and the
9093 -- following warning appears useful for the most common case.
9097 then
9098 Error_Msg_NE
9100 Error_Msg_N
9106 -- Start of processing for Resolve_Membership_Op
9108 begin
9114 Resolve_Set_Membership;
9118 and then
9120 or else
9123 then
9126 -- Ada 2005 (AI-251): Support the following case:
9128 -- type I is interface;
9129 -- type T is tagged ...
9131 -- function Test (O : I'Class) is
9132 -- begin
9133 -- return O in T'Class.
9134 -- end Test;
9136 -- In this case we have nothing else to do. The membership test will be
9137 -- done at run time.
9144 then
9146 else
9150 -- If mixed-mode operations are present and operands are all literal,
9151 -- the only interpretation involves Duration, which is probably not
9152 -- the intention of the programmer.
9167 then
9173 else
9178 -- Here after resolving membership operation
9185 ------------------
9186 -- Resolve_Null --
9187 ------------------
9192 begin
9193 -- Handle restriction against anonymous null access values This
9194 -- restriction can be turned off using -gnatdj.
9196 -- Ada 2005 (AI-231): Remove restriction
9199 and then not Debug_Flag_J
9202 then
9203 -- In the common case of a call which uses an explicitly null value
9204 -- for an access parameter, give specialized error message.
9207 Error_Msg_N
9210 -- Standard message for all other cases (are there any?)
9212 else
9213 Error_Msg_N
9218 -- Ada 2005 (AI-231): Generate the null-excluding check in case of
9219 -- assignment to a null-excluding object
9224 then
9226 Insert_Action
9231 else
9238 -- In a distributed context, null for a remote access to subprogram may
9239 -- need to be replaced with a special record aggregate. In this case,
9240 -- return after having done the transformation.
9245 then
9249 -- The null literal takes its type from the context
9254 -----------------------
9255 -- Resolve_Op_Concat --
9256 -----------------------
9260 -- We wish to avoid deep recursion, because concatenations are often
9261 -- deeply nested, as in A&B&...&Z. Therefore, we walk down the left
9262 -- operands nonrecursively until we find something that is not a simple
9263 -- concatenation (A in this case). We resolve that, and then walk back
9264 -- up the tree following Parent pointers, calling Resolve_Op_Concat_Rest
9265 -- to do the rest of the work at each level. The Parent pointers allow
9266 -- us to avoid recursion, and thus avoid running out of memory. See also
9267 -- Sem_Ch4.Analyze_Concatenation, where a similar approach is used.
9272 begin
9273 -- The following code is equivalent to:
9275 -- Resolve_Op_Concat_First (NN, Typ);
9276 -- Resolve_Op_Concat_Arg (N, ...);
9277 -- Resolve_Op_Concat_Rest (N, Typ);
9279 -- where the Resolve_Op_Concat_Arg call recurses back here if the left
9280 -- operand is a concatenation.
9282 -- Walk down left operands
9284 loop
9293 -- Now (given the above example) NN is A&B and Op1 is A
9295 -- First resolve Op1 ...
9299 -- ... then walk NN back up until we reach N (where we started), calling
9300 -- Resolve_Op_Concat_Rest along the way.
9302 loop
9309 Check_SPARK_05_Restriction
9314 ---------------------------
9315 -- Resolve_Op_Concat_Arg --
9316 ---------------------------
9318 procedure Resolve_Op_Concat_Arg
9323 is
9327 begin
9329 if Is_Comp
9333 then
9335 else
9339 -- If both Array & Array and Array & Component are visible, there is a
9340 -- potential ambiguity that must be reported.
9345 then
9349 -- If the operation is user-defined and not overloaded use its
9350 -- profile. The operation may be a renaming, in which case it has
9351 -- been rewritten, and we want the original profile.
9356 then
9358 Etype
9362 -- Otherwise an aggregate may match both the array type and the
9363 -- component type.
9365 else
9370 else
9374 then
9375 declare
9380 begin
9385 -- Special-case the error message when the overloading is
9386 -- caused by a function that yields an array and can be
9387 -- called without parameters.
9392 Error_Msg_NE
9394 Error_Msg_NE
9398 else
9406 or else
9408 then
9409 Error_Msg_N -- CODEFIX
9427 else
9432 else
9436 -- Concatenation is restricted in SPARK: each operand must be either a
9437 -- string literal, the name of a string constant, a static character or
9438 -- string expression, or another concatenation. Arg cannot be a
9439 -- concatenation here as callers of Resolve_Op_Concat_Arg call it
9440 -- separately on each final operand, past concatenation operations.
9444 Check_SPARK_05_Restriction
9452 then
9453 Check_SPARK_05_Restriction
9457 -- Do not issue error on an operand that is neither a character nor a
9458 -- string, as the error is issued in Resolve_Op_Concat.
9460 else
9467 -----------------------------
9468 -- Resolve_Op_Concat_First --
9469 -----------------------------
9476 begin
9477 -- The parser folds an enormous sequence of concatenations of string
9478 -- literals into "" & "...", where the Is_Folded_In_Parser flag is set
9479 -- in the right operand. If the expression resolves to a predefined "&"
9480 -- operator, all is well. Otherwise, the parser's folding is wrong, so
9481 -- we give an error. See P_Simple_Expression in Par.Ch4.
9486 then
9501 ----------------------------
9502 -- Resolve_Op_Concat_Rest --
9503 ----------------------------
9509 begin
9518 -- If this is not a static concatenation, but the result is a string
9519 -- type (and not an array of strings) ensure that static string operands
9520 -- have their subtypes properly constructed.
9524 then
9530 ----------------------
9531 -- Resolve_Op_Expon --
9532 ----------------------
9537 begin
9538 -- Catch attempts to do fixed-point exponentiation with universal
9539 -- operands, which is a case where the illegality is not caught during
9540 -- normal operator analysis. This is not done in preanalysis mode
9541 -- since the tree is not fully decorated during preanalysis.
9553 then
9563 then
9570 then
9574 -- We do the resolution using the base type, because intermediate values
9575 -- in expressions are always of the base type, not a subtype of it.
9580 -- For integer types, right argument must be in Natural range
9595 -- Evaluate the exponentiation operator for dimensioned type
9598 else
9602 -- Set overflow checking bit. Much cleverer code needed here eventually
9603 -- and perhaps the Resolve routines should be separated for the various
9604 -- arithmetic operations, since they will need different processing. ???
9613 --------------------
9614 -- Resolve_Op_Not --
9615 --------------------
9621 -- This function determines if the parent node is a boolean operator or
9622 -- operation (comparison op, membership test, or short circuit form) and
9623 -- the not in question is the left operand of this operation. Note that
9624 -- if the not is in parens, then false is returned.
9626 -----------------------
9627 -- Parent_Is_Boolean --
9628 -----------------------
9631 begin
9635 else
9637 when N_And_Then
9638 | N_In
9639 | N_Not_In
9640 | N_Op_And
9641 | N_Op_Eq
9642 | N_Op_Ge
9643 | N_Op_Gt
9644 | N_Op_Le
9645 | N_Op_Lt
9646 | N_Op_Ne
9647 | N_Op_Or
9648 | N_Op_Xor
9649 | N_Or_Else
9650 =>
9659 -- Start of processing for Resolve_Op_Not
9661 begin
9662 -- Predefined operations on scalar types yield the base type. On the
9663 -- other hand, logical operations on arrays yield the type of the
9664 -- arguments (and the context).
9668 else
9672 -- Straightforward case of incorrect arguments
9679 -- Special case of probable missing parens
9683 Error_Msg_N
9686 else
9687 Error_Msg_N
9694 -- OK resolution of NOT
9696 else
9697 -- Warn if non-boolean types involved. This is a case like not a < b
9698 -- where a and b are modular, where we will get (not a) < b and most
9699 -- likely not (a < b) was intended.
9701 if Warn_On_Questionable_Missing_Parens
9703 and then Parent_Is_Boolean
9704 then
9708 -- Warn on double negation if checking redundant constructs
9710 if Warn_On_Redundant_Constructs
9715 then
9719 -- Complete resolution and evaluation of NOT
9729 -----------------------------
9730 -- Resolve_Operator_Symbol --
9731 -----------------------------
9733 -- Nothing to be done, all resolved already
9739 begin
9743 ----------------------------------
9744 -- Resolve_Qualified_Expression --
9745 ----------------------------------
9753 begin
9756 -- Protect call to Matching_Static_Array_Bounds to avoid costly
9757 -- operation if not needed.
9764 then
9765 Check_SPARK_05_Restriction
9769 -- A qualified expression requires an exact match of the type, class-
9770 -- wide matching is not allowed. However, if the qualifying type is
9771 -- specific and the expression has a class-wide type, it may still be
9772 -- okay, since it can be the result of the expansion of a call to a
9773 -- dispatching function, so we also have to check class-wideness of the
9774 -- type of the expression's original node.
9777 or else
9781 then
9785 -- If the target type is unconstrained, then we reset the type of the
9786 -- result from the type of the expression. For other cases, the actual
9787 -- subtype of the expression is the target type.
9791 then
9798 -- If we still have a qualified expression after the static evaluation,
9799 -- then apply a scalar range check if needed. The reason that we do this
9800 -- after the Eval call is that otherwise, the application of the range
9801 -- check may convert an illegal static expression and result in warning
9802 -- rather than giving an error (e.g Integer'(Integer'Last + 1)).
9808 -- Finally, check whether a predicate applies to the target type. This
9809 -- comes from AI12-0100. As for type conversions, check the enclosing
9810 -- context to prevent an infinite expansion.
9816 or else
9818 then
9821 -- In the case of a qualified expression in an allocator, the check
9822 -- is applied when expanding the allocator, so avoid redundant check.
9826 then
9832 ------------------------------
9833 -- Resolve_Raise_Expression --
9834 ------------------------------
9837 begin
9841 else
9846 -------------------
9847 -- Resolve_Range --
9848 -------------------
9855 -- Returns True if N is of the form X'First .. X'Last where X is the
9856 -- same entity for both attributes.
9858 --------------------
9859 -- First_Last_Ref --
9860 --------------------
9866 begin
9871 then
9872 declare
9875 begin
9879 then
9888 -- Start of processing for Resolve_Range
9890 begin
9893 -- The lower bound should be in Typ. The higher bound can be in Typ's
9894 -- base type if the range is null. It may still be invalid if it is
9895 -- higher than the lower bound. This is checked later in the context in
9896 -- which the range appears.
9901 -- Check for inappropriate range on unordered enumeration type
9905 -- Exclude X'First .. X'Last if X is the same entity for both
9907 and then not First_Last_Ref
9908 then
9910 Error_Msg_NE
9917 -- We have to check the bounds for being within the base range as
9918 -- required for a non-static context. Normally this is automatic and
9919 -- done as part of evaluating expressions, but the N_Range node is an
9920 -- exception, since in GNAT we consider this node to be a subexpression,
9921 -- even though in Ada it is not. The circuit in Sem_Eval could check for
9922 -- this, but that would put the test on the main evaluation path for
9923 -- expressions.
9928 -- Check for an ambiguous range over character literals. This will
9929 -- happen with a membership test involving only literals.
9937 -- If bounds are static, constant-fold them, so size computations are
9938 -- identical between front-end and back-end. Do not perform this
9939 -- transformation while analyzing generic units, as type information
9940 -- would be lost when reanalyzing the constant node in the instance.
9953 --------------------------
9954 -- Resolve_Real_Literal --
9955 --------------------------
9960 begin
9961 -- Special processing for fixed-point literals to make sure that the
9962 -- value is an exact multiple of small where this is required. We skip
9963 -- this for the universal real case, and also for generic types.
9969 then
9970 declare
9977 begin
9978 -- Case of literal is not an exact multiple of the Small
9982 -- For a source program literal for a decimal fixed-point type,
9983 -- this is statically illegal (RM 4.9(36)).
9988 then
9992 -- Generate a warning if literal from source
9995 and then Warn_On_Bad_Fixed_Value
9996 then
9997 Error_Msg_N
9999 N);
10002 -- Replace literal by a value that is the exact representation
10003 -- of a value of the type, i.e. a multiple of the small value,
10004 -- by truncation, since Machine_Rounds is false for all GNAT
10005 -- fixed-point types (RM 4.9(38)).
10015 -- In all cases, set the corresponding integer field
10021 -- Now replace the actual type by the expected type as usual
10027 -----------------------
10028 -- Resolve_Reference --
10029 -----------------------
10034 begin
10035 -- Replace general access with specific type
10043 -- If we are taking the reference of a volatile entity, then treat it as
10044 -- a potential modification of this entity. This is too conservative,
10045 -- but necessary because remove side effects can cause transformations
10046 -- of normal assignments into reference sequences that otherwise fail to
10047 -- notice the modification.
10054 --------------------------------
10055 -- Resolve_Selected_Component --
10056 --------------------------------
10071 -- Check whether this is the initialization of a component within an
10072 -- init proc (by assignment or call to another init proc). If true,
10073 -- there is no need for a discriminant check.
10075 --------------------
10076 -- Init_Component --
10077 --------------------
10080 begin
10081 return Inside_Init_Proc
10087 -- Start of processing for Resolve_Selected_Component
10089 begin
10092 -- Use the context type to select the prefix that has a selector
10093 -- of the correct name and type.
10101 else
10105 -- Locate selected component. For a private prefix the selector
10106 -- can denote a discriminant.
10110 -- The visible components of a class-wide type are those of
10111 -- the root type.
10121 then
10128 else
10132 Error_Msg_N
10137 else
10140 -- There may be an implicit dereference. Retrieve
10141 -- designated record type.
10145 else
10151 -- Resolution chooses the new interpretation.
10152 -- Find the component with the right name.
10157 loop
10174 -- There must be a legal interpretation at this point
10181 else
10182 -- Resolve prefix with its type
10187 -- Generate cross-reference. We needed to wait until full overloading
10188 -- resolution was complete to do this, since otherwise we can't tell if
10189 -- we are an lvalue or not.
10193 else
10197 -- If prefix is an access type, the node will be transformed into an
10198 -- explicit dereference during expansion. The type of the node is the
10199 -- designated type of that of the prefix.
10204 else
10208 -- Set flag for expander if discriminant check required on a component
10209 -- appearing within a variant.
10215 and then
10218 and then not Init_Component
10219 then
10227 -- If the prefix is a record conversion, this may be a renamed
10228 -- discriminant whose bounds differ from those of the original
10229 -- one, so we must ensure that a range check is performed.
10234 then
10238 -- Note: No Eval processing is required, because the prefix is of a
10239 -- record type, or protected type, and neither can possibly be static.
10241 -- If the record type is atomic, and the component is non-atomic, then
10242 -- this is worth a warning, since we have a situation where the access
10243 -- to the component may cause extra read/writes of the atomic array
10244 -- object, or partial word accesses, both of which may be unexpected.
10250 then
10251 Error_Msg_N
10254 Error_Msg_N
10262 -------------------
10263 -- Resolve_Shift --
10264 -------------------
10271 begin
10272 -- We do the resolution using the base type, because intermediate values
10273 -- in expressions always are of the base type, not a subtype of it.
10286 ---------------------------
10287 -- Resolve_Short_Circuit --
10288 ---------------------------
10295 begin
10296 -- Ensure all actions associated with the left operand (e.g.
10297 -- finalization of transient objects) are fully evaluated locally within
10298 -- an expression with actions. This is particularly helpful for coverage
10299 -- analysis. However this should not happen in generics or if option
10300 -- Minimize_Expression_With_Actions is set.
10303 declare
10305 begin
10313 -- Set Comes_From_Source on L to preserve warnings for unset
10314 -- reference.
10323 -- Check for issuing warning for always False assert/check, this happens
10324 -- when assertions are turned off, in which case the pragma Assert/Check
10325 -- was transformed into:
10327 -- if False and then <condition> then ...
10329 -- and we detect this pattern
10331 if Warn_On_Assertion_Failure
10338 then
10339 declare
10342 begin
10343 -- Special handling of Asssert pragma
10347 then
10348 declare
10350 Original_Node
10351 (Expression
10354 begin
10355 -- Don't warn if original condition is explicit False,
10356 -- since obviously the failure is expected in this case.
10360 then
10363 -- Issue warning. We do not want the deletion of the
10364 -- IF/AND-THEN to take this message with it. We achieve this
10365 -- by making sure that the expanded code points to the Sloc
10366 -- of the expression, not the original pragma.
10368 else
10369 -- Note: Use Error_Msg_F here rather than Error_Msg_N.
10370 -- The source location of the expression is not usually
10371 -- the best choice here. For example, it gets located on
10372 -- the last AND keyword in a chain of boolean expressiond
10373 -- AND'ed together. It is best to put the message on the
10374 -- first character of the assertion, which is the effect
10375 -- of the First_Node call here.
10377 Error_Msg_F
10379 Expression
10384 -- Similar processing for Check pragma
10388 then
10389 -- Don't want to warn if original condition is explicit False
10391 declare
10393 Original_Node
10394 (Expression
10396 begin
10399 then
10402 -- Post warning
10404 else
10405 -- Again use Error_Msg_F rather than Error_Msg_N, see
10406 -- comment above for an explanation of why we do this.
10408 Error_Msg_F
10410 Expression
10418 -- Continue with processing of short circuit
10427 -------------------
10428 -- Resolve_Slice --
10429 -------------------
10438 begin
10441 -- Use the context type to select the prefix that yields the correct
10442 -- array type.
10444 declare
10451 begin
10459 then
10467 else
10472 else
10483 else
10493 -- If the prefix is an access to an unconstrained array, we must use
10494 -- the actual subtype of the object to perform the index checks. The
10495 -- object denoted by the prefix is implicit in the node, so we build
10496 -- an explicit representation for it in order to compute the actual
10497 -- subtype.
10502 declare
10506 begin
10517 then
10520 -- If the name is a selected component that depends on discriminants,
10521 -- build an actual subtype for it. This can happen only when the name
10522 -- itself is overloaded; otherwise the actual subtype is created when
10523 -- the selected component is analyzed.
10526 and then Full_Analysis
10528 then
10529 declare
10532 begin
10537 -- Maybe this should just be "else", instead of checking for the
10538 -- specific case of slice??? This is needed for the case where the
10539 -- prefix is an Image attribute, which gets expanded to a slice, and so
10540 -- has a constrained subtype which we want to use for the slice range
10541 -- check applied below (the range check won't get done if the
10542 -- unconstrained subtype of the 'Image is used).
10548 -- Obtain the type of the array index
10552 else
10556 -- If name was overloaded, set slice type correctly now
10560 -- Handle the generation of a range check that compares the array index
10561 -- against the discrete_range. The check is not applied to internally
10562 -- built nodes associated with the expansion of dispatch tables. Check
10563 -- that Ada.Tags has already been loaded to avoid extra dependencies on
10564 -- the unit.
10566 if Tagged_Type_Expansion
10572 then
10575 -- The discrete_range is specified by a subtype indication. Create a
10576 -- shallow copy and inherit the type, parent and source location from
10577 -- the discrete_range. This ensures that the range check is inserted
10578 -- relative to the slice and that the runtime exception points to the
10579 -- proper construct.
10588 -- The discrete_range is a regular range. Resolve the bounds and remove
10589 -- their side effects.
10591 else
10608 -- Check bad use of type with predicates
10610 declare
10613 begin
10616 then
10618 else
10623 Bad_Predicated_Subtype_Use
10628 -- Otherwise here is where we check suspicious indexes
10639 ----------------------------
10640 -- Resolve_String_Literal --
10641 ----------------------------
10652 begin
10653 -- For a string appearing in a concatenation, defer creation of the
10654 -- string_literal_subtype until the end of the resolution of the
10655 -- concatenation, because the literal may be constant-folded away. This
10656 -- is a useful optimization for long concatenation expressions.
10658 -- If the string is an aggregate built for a single character (which
10659 -- happens in a non-static context) or a is null string to which special
10660 -- checks may apply, we build the subtype. Wide strings must also get a
10661 -- string subtype if they come from a one character aggregate. Strings
10662 -- generated by attributes might be static, but it is often hard to
10663 -- determine whether the enclosing context is static, so we generate
10664 -- subtypes for them as well, thus losing some rarer optimizations ???
10665 -- Same for strings that come from a static conversion.
10667 Need_Check :=
10676 -- If the resolving type is itself a string literal subtype, we can just
10677 -- reuse it, since there is no point in creating another.
10683 and then not Need_Check
10685 N_Attribute_Reference,
10686 N_Qualified_Expression,
10687 N_Type_Conversion)
10688 then
10691 -- Do not generate a string literal subtype for the default expression
10692 -- of a formal parameter in GNATprove mode. This is because the string
10693 -- subtype is associated with the freezing actions of the subprogram,
10694 -- however freezing is disabled in GNATprove mode and as a result the
10695 -- subtype is unavailable.
10697 elsif GNATprove_Mode
10699 then
10702 -- Otherwise we must create a string literal subtype. Note that the
10703 -- whole idea of string literal subtypes is simply to avoid the need
10704 -- for building a full fledged array subtype for each literal.
10706 else
10712 or else Need_Check
10713 then
10720 then
10726 -- The validity of a null string has been checked in the call to
10727 -- Eval_String_Literal.
10732 -- Always accept string literal with component type Any_Character, which
10733 -- occurs in error situations and in comparisons of literals, both of
10734 -- which should accept all literals.
10739 -- If the type is bit-packed, then we always transform the string
10740 -- literal into a full fledged aggregate.
10745 -- Deal with cases of Wide_Wide_String, Wide_String, and String
10747 else
10748 -- For Standard.Wide_Wide_String, or any other type whose component
10749 -- type is Standard.Wide_Wide_Character, we know that all the
10750 -- characters in the string must be acceptable, since the parser
10751 -- accepted the characters as valid character literals.
10756 -- For the case of Standard.String, or any other type whose component
10757 -- type is Standard.Character, we must make sure that there are no
10758 -- wide characters in the string, i.e. that it is entirely composed
10759 -- of characters in range of type Character.
10761 -- If the string literal is the result of a static concatenation, the
10762 -- test has already been performed on the components, and need not be
10763 -- repeated.
10767 then
10771 -- If we are out of range, post error. This is one of the
10772 -- very few places that we place the flag in the middle of
10773 -- a token, right under the offending wide character. Not
10774 -- quite clear if this is right wrt wide character encoding
10775 -- sequences, but it's only an error message.
10777 Error_Msg
10784 -- For the case of Standard.Wide_String, or any other type whose
10785 -- component type is Standard.Wide_Character, we must make sure that
10786 -- there are no wide characters in the string, i.e. that it is
10787 -- entirely composed of characters in range of type Wide_Character.
10789 -- If the string literal is the result of a static concatenation,
10790 -- the test has already been performed on the components, and need
10791 -- not be repeated.
10795 then
10799 -- If we are out of range, post error. This is one of the
10800 -- very few places that we place the flag in the middle of
10801 -- a token, right under the offending wide character.
10803 -- This is not quite right, because characters in general
10804 -- will take more than one character position ???
10806 Error_Msg
10813 -- If the root type is not a standard character, then we will convert
10814 -- the string into an aggregate and will let the aggregate code do
10815 -- the checking. Standard Wide_Wide_Character is also OK here.
10817 else
10821 -- See if the component type of the array corresponding to the string
10822 -- has compile time known bounds. If yes we can directly check
10823 -- whether the evaluation of the string will raise constraint error.
10824 -- Otherwise we need to transform the string literal into the
10825 -- corresponding character aggregate and let the aggregate code do
10826 -- the checking.
10830 -- Check for the case of full range, where we are definitely OK
10836 -- Here the range is not the complete base type range, so check
10838 declare
10846 begin
10849 then
10855 then
10856 Apply_Compile_Time_Constraint_Error
10858 CE_Range_Check_Failed,
10869 -- If we got here we meed to transform the string literal into the
10870 -- equivalent qualified positional array aggregate. This is rather
10871 -- heavy artillery for this situation, but it is hard work to avoid.
10873 declare
10878 begin
10879 -- Build the character literals, we give them source locations that
10880 -- correspond to the string positions, which is a bit tricky given
10881 -- the possible presence of wide character escape sequences.
10895 -- Should we have a call to Skip_Wide here ???
10897 -- ??? else
10898 -- Skip_Wide (P);
10906 Expression =>
10913 -------------------------
10914 -- Resolve_Target_Name --
10915 -------------------------
10918 begin
10922 -----------------------------
10923 -- Resolve_Type_Conversion --
10924 -----------------------------
10936 -- Set to False to suppress cases where we want to suppress the test
10937 -- for redundancy to avoid possible false positives on this warning.
10939 begin
10940 if not Conv_OK
10942 then
10946 -- If the Operand Etype is Universal_Fixed, then the conversion is
10947 -- never redundant. We need this check because by the time we have
10948 -- finished the rather complex transformation, the conversion looks
10949 -- redundant when it is not.
10954 -- If the operand is marked as Any_Fixed, then special processing is
10955 -- required. This is also a case where we suppress the test for a
10956 -- redundant conversion, since most certainly it is not redundant.
10961 -- Mixed-mode operation involving a literal. Context must be a fixed
10962 -- type which is applied to the literal subsequently.
10964 -- Multiplication and division involving two fixed type operands must
10965 -- yield a universal real because the result is computed in arbitrary
10966 -- precision.
10972 then
10978 or else
10980 then
10981 -- Return if expression is ambiguous
10986 -- If nothing else, the available fixed type is Duration
10988 else
10992 -- Resolve the real operand with largest available precision
10996 else
11002 -- If the operand is a literal (it could be a non-static and
11003 -- illegal exponentiation) check whether the use of Duration
11004 -- is potentially inaccurate.
11009 then
11010 Error_Msg_N
11013 Error_Msg_N
11020 then
11023 else
11032 -- In SPARK, a type conversion between array types should be restricted
11033 -- to types which have matching static bounds.
11035 -- Protect call to Matching_Static_Array_Bounds to avoid costly
11036 -- operation if not needed.
11043 then
11044 Check_SPARK_05_Restriction
11048 -- In formal mode, the operand of an ancestor type conversion must be an
11049 -- object (not an expression).
11057 then
11063 -- Note: we do the Eval_Type_Conversion call before applying the
11064 -- required checks for a subtype conversion. This is important, since
11065 -- both are prepared under certain circumstances to change the type
11066 -- conversion to a constraint error node, but in the case of
11067 -- Eval_Type_Conversion this may reflect an illegality in the static
11068 -- case, and we would miss the illegality (getting only a warning
11069 -- message), if we applied the type conversion checks first.
11073 -- Even when evaluation is not possible, we may be able to simplify the
11074 -- conversion or its expression. This needs to be done before applying
11075 -- checks, since otherwise the checks may use the original expression
11076 -- and defeat the simplifications. This is specifically the case for
11077 -- elimination of the floating-point Truncation attribute in
11078 -- float-to-int conversions.
11082 -- If after evaluation we still have a type conversion, then we may need
11083 -- to apply checks required for a subtype conversion.
11085 -- Skip these type conversion checks if universal fixed operands
11086 -- operands involved, since range checks are handled separately for
11087 -- these cases (in the appropriate Expand routines in unit Exp_Fixd).
11093 then
11097 -- Issue warning for conversion of simple object to its own type. We
11098 -- have to test the original nodes, since they may have been rewritten
11099 -- by various optimizations.
11103 -- Here we test for a redundant conversion if the warning mode is
11104 -- active (and was not locally reset), and we have a type conversion
11105 -- from source not appearing in a generic instance.
11107 if Test_Redundant
11110 and then not In_Instance
11111 then
11115 -- If the node is part of a larger expression, the Target_Type
11116 -- may not be the original type of the node if the context is a
11117 -- condition. Recover original type to see if conversion is needed.
11121 then
11125 -- If we have an entity name, then give the warning if the entity
11126 -- is the right type, or if it is a loop parameter covered by the
11127 -- original type (that's needed because loop parameters have an
11128 -- odd subtype coming from the bounds).
11131 and then
11133 or else
11137 -- If not an entity, then type of expression must match
11140 then
11141 -- One more check, do not give warning if the analyzed conversion
11142 -- has an expression with non-static bounds, and the bounds of the
11143 -- target are static. This avoids junk warnings in cases where the
11144 -- conversion is necessary to establish staticness, for example in
11145 -- a case statement.
11149 then
11152 -- Finally, if this type conversion occurs in a context requiring
11153 -- a prefix, and the expression is a qualified expression then the
11154 -- type conversion is not redundant, since a qualified expression
11155 -- is not a prefix, whereas a type conversion is. For example, "X
11156 -- := T'(Funx(...)).Y;" is illegal because a selected component
11157 -- requires a prefix, but a type conversion makes it legal: "X :=
11158 -- T(T'(Funx(...))).Y;"
11160 -- In Ada 2012, a qualified expression is a name, so this idiom is
11161 -- no longer needed, but we still suppress the warning because it
11162 -- seems unfriendly for warnings to pop up when you switch to the
11163 -- newer language version.
11167 N_Indexed_Component,
11168 N_Selected_Component,
11169 N_Slice,
11170 N_Explicit_Dereference)
11171 then
11174 -- Never warn on conversion to Long_Long_Integer'Base since
11175 -- that is most likely an artifact of the extended overflow
11176 -- checking and comes from complex expanded code.
11181 -- Here we give the redundant conversion warning. If it is an
11182 -- entity, give the name of the entity in the message. If not,
11183 -- just mention the expression.
11185 -- Shoudn't we test Warn_On_Redundant_Constructs here ???
11187 else
11190 Error_Msg_NE -- CODEFIX
11193 else
11194 Error_Msg_NE
11202 -- Ada 2005 (AI-251): Handle class-wide interface type conversions.
11203 -- No need to perform any interface conversion if the type of the
11204 -- expression coincides with the target type.
11207 and then Expander_Active
11209 then
11210 declare
11214 begin
11215 -- If the type of the operand is a limited view, use nonlimited
11216 -- view when available. If it is a class-wide type, recover the
11217 -- class-wide type of the nonlimited view.
11221 then
11237 -- Conversion from interface type
11241 -- Ada 2005 (AI-217): Handle entities from limited views
11245 Error_Msg_NE -- CODEFIX
11248 Error_Msg_N
11250 N);
11253 and then not
11256 then
11258 Error_Msg_NE -- CODEFIX
11261 Error_Msg_N
11263 N);
11265 else
11269 -- Conversion to interface type
11273 -- Handle subtypes
11280 or else Interface_Present_In_Ancestor
11283 then
11285 else
11288 Error_Msg_N
11296 -- Ada 2012: once the type conversion is resolved, check whether the
11297 -- operand statisfies the static predicate of the target type.
11303 -- If at this stage we have a real to integer conversion, make sure that
11304 -- the Do_Range_Check flag is set, because such conversions in general
11305 -- need a range check. We only need this if expansion is off.
11306 -- In GNATprove mode, we only do that when converting from fixed-point
11307 -- (as floating-point to integer conversions are now handled in
11308 -- GNATprove mode).
11311 and then not Expander_Active
11316 then
11320 -- Generating C code a type conversion of an access to constrained
11321 -- array type to access to unconstrained array type involves building
11322 -- a fat pointer which in general cannot be generated on the fly. We
11323 -- remove side effects in order to store the result of the conversion
11324 -- into a temporary.
11326 if Modify_Tree_For_C
11333 then
11338 ----------------------
11339 -- Resolve_Unary_Op --
11340 ----------------------
11349 begin
11352 Check_SPARK_05_Restriction
11356 -- Deal with intrinsic unary operators
11362 then
11367 -- Deal with universal cases
11370 or else
11372 then
11379 -- Generate warning for expressions like abs (x mod 2)
11381 if Warn_On_Redundant_Constructs
11383 then
11387 Error_Msg_N -- CODEFIX
11392 -- Deal with reference generation
11399 -- Set overflow checking bit. Much cleverer code needed here eventually
11400 -- and perhaps the Resolve routines should be separated for the various
11401 -- arithmetic operations, since they will need different processing ???
11409 -- Generate warning for expressions like -5 mod 3 for integers. No need
11410 -- to worry in the floating-point case, since parens do not affect the
11411 -- result so there is no point in giving in a warning.
11413 declare
11422 begin
11423 if Warn_On_Questionable_Missing_Parens
11427 then
11430 -- We are looking for cases where the right operand is not
11431 -- parenthesized, and is a binary operator, multiply, divide, or
11432 -- mod. These are the cases where the grouping can affect results.
11436 then
11437 -- For mod, we always give the warning, since the value is
11438 -- affected by the parenthesization (e.g. (-5) mod 315 /=
11439 -- -(5 mod 315)). But for the other cases, the only concern is
11440 -- overflow, e.g. for the case of 8 big signed (-(2 * 64)
11441 -- overflows, but (-2) * 64 does not). So we try to give the
11442 -- message only when overflow is possible.
11446 then
11451 else
11457 else
11461 -- Note that the test below is deliberately excluding the
11462 -- largest negative number, since that is a potentially
11463 -- troublesome case (e.g. -2 * x, where the result is the
11464 -- largest negative integer has an overflow with 2 * x).
11471 -- For the multiplication case, the only case we have to worry
11472 -- about is when (-a)*b is exactly the largest negative number
11473 -- so that -(a*b) can cause overflow. This can only happen if
11474 -- a is a power of 2, and more generally if any operand is a
11475 -- constant that is not a power of 2, then the parentheses
11476 -- cannot affect whether overflow occurs. We only bother to
11477 -- test the left most operand
11479 -- Loop looking at left operands for one that has known value
11486 -- Operand value of 0 or 1 skips warning
11491 -- Otherwise check power of 2, if power of 2, warn, if
11492 -- anything else, skip warning.
11494 else
11498 else
11507 -- Keep looking at left operands
11512 -- For rem or "/" we can only have a problematic situation
11513 -- if the divisor has a value of minus one or one. Otherwise
11514 -- overflow is impossible (divisor > 1) or we have a case of
11515 -- division by zero in any case.
11520 then
11524 -- If we fall through warning should be issued
11526 -- Shouldn't we test Warn_On_Questionable_Missing_Parens ???
11528 Error_Msg_N
11535 ----------------------------------
11536 -- Resolve_Unchecked_Expression --
11537 ----------------------------------
11539 procedure Resolve_Unchecked_Expression
11542 is
11543 begin
11548 ---------------------------------------
11549 -- Resolve_Unchecked_Type_Conversion --
11550 ---------------------------------------
11552 procedure Resolve_Unchecked_Type_Conversion
11555 is
11561 begin
11562 -- Resolve operand using its own type
11566 -- In an inlined context, the unchecked conversion may be applied
11567 -- to a literal, in which case its type is the type of the context.
11568 -- (In other contexts conversions cannot apply to literals).
11570 if In_Inlined_Body
11574 then
11582 ------------------------------
11583 -- Rewrite_Operator_As_Call --
11584 ------------------------------
11591 begin
11598 New_N :=
11609 ------------------------------
11610 -- Rewrite_Renamed_Operator --
11611 ------------------------------
11613 procedure Rewrite_Renamed_Operator
11617 is
11622 begin
11623 -- Do not perform this transformation within a pre/postcondition,
11624 -- because the expression will be reanalyzed, and the transformation
11625 -- might affect the visibility of the operator, e.g. in an instance.
11626 -- Note that fully analyzed and expanded pre/postconditions appear as
11627 -- pragma Check equivalents.
11633 -- Likewise when an expression function is being preanalyzed, since the
11634 -- expression will be reanalyzed as part of the generated body.
11637 declare
11639 begin
11642 N_Expression_Function
11643 then
11649 -- Rewrite the operator node using the real operator, not its renaming.
11650 -- Exclude user-defined intrinsic operations of the same name, which are
11651 -- treated separately and rewritten as calls.
11660 -- Indicate that both the original entity and its renaming are
11661 -- referenced at this point.
11672 -- If the context type is private, add the appropriate conversions so
11673 -- that the operator is applied to the full view. This is done in the
11674 -- routines that resolve intrinsic operators.
11678 when N_Op_Add
11679 | N_Op_Divide
11680 | N_Op_Expon
11681 | N_Op_Mod
11682 | N_Op_Multiply
11683 | N_Op_Rem
11684 | N_Op_Subtract
11685 =>
11688 when N_Op_Abs
11689 | N_Op_Minus
11690 | N_Op_Plus
11691 =>
11701 -- Operator renames a user-defined operator of the same name. Use the
11702 -- original operator in the node, which is the one Gigi knows about.
11709 -----------------------
11710 -- Set_Slice_Subtype --
11711 -----------------------
11713 -- Build an implicit subtype declaration to represent the type delivered by
11714 -- the slice. This is an abbreviated version of an array subtype. We define
11715 -- an index subtype for the slice, using either the subtype name or the
11716 -- discrete range of the slice. To be consistent with index usage elsewhere
11717 -- we create a list header to hold the single index. This list is not
11718 -- otherwise attached to the syntax tree.
11729 begin
11735 else
11736 -- We force the evaluation of a range. This is definitely needed in
11737 -- the renamed case, and seems safer to do unconditionally. Note in
11738 -- any case that since we will create and insert an Itype referring
11739 -- to this range, we must make sure any side effect removal actions
11740 -- are inserted before the Itype definition.
11746 -- If the discrete range is given by a subtype indication, the
11747 -- type of the slice is the base of the subtype mark.
11750 declare
11752 begin
11761 -- Take a new copy of Drange (where bounds have been rewritten to
11762 -- reference side-effect-free names). Using a separate tree ensures
11763 -- that further expansion (e.g. while rewriting a slice assignment
11764 -- into a FOR loop) does not attempt to remove side effects on the
11765 -- bounds again (which would cause the bounds in the index subtype
11766 -- definition to refer to temporaries before they are defined) (the
11767 -- reason is that some names are considered side effect free here
11768 -- for the subtype, but not in the context of a loop iteration
11769 -- scheme).
11790 -- The Etype of the existing Slice node is reset to this slice subtype.
11791 -- Its bounds are obtained from its first index.
11795 -- For bit-packed slice subtypes, freeze immediately (except in the case
11796 -- of being in a "spec expression" where we never freeze when we first
11797 -- see the expression).
11802 -- For all other cases insert an itype reference in the slice's actions
11803 -- so that the itype is frozen at the proper place in the tree (i.e. at
11804 -- the point where actions for the slice are analyzed). Note that this
11805 -- is different from freezing the itype immediately, which might be
11806 -- premature (e.g. if the slice is within a transient scope). This needs
11807 -- to be done only if expansion is enabled.
11814 --------------------------------
11815 -- Set_String_Literal_Subtype --
11816 --------------------------------
11824 begin
11836 -- The low bound is set from the low bound of the corresponding index
11837 -- type. Note that we do not store the high bound in the string literal
11838 -- subtype, but it can be deduced if necessary from the length and the
11839 -- low bound.
11844 -- If the lower bound is not static we create a range for the string
11845 -- literal, using the index type and the known length of the literal.
11846 -- The index type is not necessarily Positive, so the upper bound is
11847 -- computed as T'Val (T'Pos (Low_Bound) + L - 1).
11849 else
11850 declare
11856 Prefix =>
11860 Left_Opnd =>
11863 Prefix =>
11865 Expressions =>
11867 Right_Opnd =>
11876 begin
11878 Set_String_Literal_Low_Bound
11881 else
11882 -- If the index type is an enumeration type, build bounds
11883 -- expression with attributes.
11885 Set_String_Literal_Low_Bound
11889 Prefix =>
11896 -- Build bona fide subtype for the string, and wrap it in an
11897 -- unchecked conversion, because the backend expects the
11898 -- String_Literal_Subtype to have a static lower bound.
11900 Index_Subtype :=
11907 -- In the context, the Index_Type may already have a constraint,
11908 -- so use common base type on string subtype. The base type may
11909 -- be used when generating attributes of the string, for example
11910 -- in the context of a slice assignment.
11935 ------------------------------
11936 -- Simplify_Type_Conversion --
11937 ------------------------------
11940 begin
11942 declare
11947 begin
11948 -- Special processing if the conversion is the expression of a
11949 -- Rounding or Truncation attribute reference. In this case we
11950 -- replace:
11952 -- ityp (ftyp'Rounding (x)) or ityp (ftyp'Truncation (x))
11954 -- by
11956 -- ityp (x)
11958 -- with the Float_Truncate flag set to False or True respectively,
11959 -- which is more efficient.
11962 and then
11968 Name_Truncation)
11969 then
11970 declare
11973 begin
11983 -----------------------------
11984 -- Unique_Fixed_Point_Type --
11985 -----------------------------
11989 -- Give error messages for true ambiguity. Messages are posted on node
11990 -- N, and entities T1, T2 are the possible interpretations.
11992 -----------------------
11993 -- Fixed_Point_Error --
11994 -----------------------
11997 begin
12003 -- Local variables
12011 -- Start of processing for Unique_Fixed_Point_Type
12013 begin
12014 -- The operations on Duration are visible, so Duration is always a
12015 -- possible interpretation.
12019 -- Look for fixed-point types in enclosing scopes
12028 then
12032 else
12043 -- Look for visible fixed type declarations in the context
12054 then
12058 else
12073 else
12074 -- When the context is a type conversion, issue the warning on the
12075 -- expression of the conversion because it is the actual operation.
12079 else
12083 Error_Msg_NE
12090 ----------------------
12091 -- Valid_Conversion --
12092 ----------------------
12094 function Valid_Conversion
12099 is
12104 function Conversion_Check
12107 -- Little routine to post Msg if Valid is False, returns Valid value
12110 -- If Report_Errs, then calls Errout.Error_Msg_N with its arguments
12112 procedure Conversion_Error_NE
12116 -- If Report_Errs, then calls Errout.Error_Msg_NE with its arguments
12119 -- Return True if expression is within an instance but is not in one of
12120 -- the actuals of the instantiation. Type conversions within an instance
12121 -- are not rechecked because type visbility may lead to spurious errors,
12122 -- but conversions in an actual for a formal object must be checked.
12124 function Valid_Tagged_Conversion
12127 -- Specifically test for validity of tagged conversions
12130 -- Check index and component conformance, and accessibility levels if
12131 -- the component types are anonymous access types (Ada 2005).
12133 ----------------------
12134 -- Conversion_Check --
12135 ----------------------
12137 function Conversion_Check
12140 is
12141 begin
12142 if not Valid
12144 -- A generic unit has already been analyzed and we have verified
12145 -- that a particular conversion is OK in that context. Since the
12146 -- instance is reanalyzed without relying on the relationships
12147 -- established during the analysis of the generic, it is possible
12148 -- to end up with inconsistent views of private types. Do not emit
12149 -- the error message in such cases. The rest of the machinery in
12150 -- Valid_Conversion still ensures the proper compatibility of
12151 -- target and operand types.
12153 and then not In_Instance_Code
12154 then
12161 ------------------------
12162 -- Conversion_Error_N --
12163 ------------------------
12166 begin
12172 -------------------------
12173 -- Conversion_Error_NE --
12174 -------------------------
12176 procedure Conversion_Error_NE
12180 is
12181 begin
12187 ----------------------
12188 -- In_Instance_Code --
12189 ----------------------
12194 begin
12198 else
12202 -- The expression is part of an actual object if it appears in
12203 -- the generated object declaration in the instance.
12207 then
12210 else
12211 exit when
12220 -- Otherwise the expression appears within the instantiated unit
12226 ----------------------------
12227 -- Valid_Array_Conversion --
12228 ----------------------------
12245 begin
12246 -- Error if wrong number of dimensions
12248 if
12250 then
12251 Conversion_Error_N
12255 -- Number of dimensions matches
12257 else
12258 -- Loop through indexes of the two arrays
12266 -- Error if index types are incompatible
12272 then
12273 Conversion_Error_N
12275 Operand);
12283 -- If component types have same base type, all set
12288 -- Here if base types of components are not the same. The only
12289 -- time this is allowed is if we have anonymous access types.
12291 -- The conversion of arrays of anonymous access types can lead
12292 -- to dangling pointers. AI-392 formalizes the accessibility
12293 -- checks that must be applied to such conversions to prevent
12294 -- out-of-scope references.
12296 elsif Ekind_In
12298 E_Anonymous_Access_Subprogram_Type)
12300 and then
12302 then
12305 then
12308 Conversion_Error_N
12318 -- Conversion not allowed because of accessibility levels
12320 else
12321 Conversion_Error_N
12327 else
12331 -- All other cases where component base types do not match
12333 else
12334 Conversion_Error_N
12336 Operand);
12340 -- Check that component subtypes statically match. For numeric
12341 -- types this means that both must be either constrained or
12342 -- unconstrained. For enumeration types the bounds must match.
12343 -- All of this is checked in Subtypes_Statically_Match.
12345 if not Subtypes_Statically_Match
12347 then
12348 Conversion_Error_N
12357 -----------------------------
12358 -- Valid_Tagged_Conversion --
12359 -----------------------------
12361 function Valid_Tagged_Conversion
12364 is
12365 begin
12366 -- Upward conversions are allowed (RM 4.6(22))
12370 then
12373 -- Downward conversion are allowed if the operand is class-wide
12374 -- (RM 4.6(23)).
12378 then
12383 then
12384 return
12388 -- Ada 2005 (AI-251): The conversion to/from interface types is
12389 -- always valid. The types involved may be class-wide (sub)types.
12393 then
12396 -- If the operand is a class-wide type obtained through a limited_
12397 -- with clause, and the context includes the nonlimited view, use
12398 -- it to determine whether the conversion is legal.
12404 then
12409 then
12412 else
12413 Conversion_Error_NE
12420 -- Start of processing for Valid_Conversion
12422 begin
12426 declare
12434 begin
12435 -- Remove procedure calls, which syntactically cannot appear in
12436 -- this context, but which cannot be removed by type checking,
12437 -- because the context does not impose a type.
12439 -- The node may be labelled overloaded, but still contain only one
12440 -- interpretation because others were discarded earlier. If this
12441 -- is the case, retain the single interpretation if legal.
12449 then
12450 -- More than one candidate interpretation is available
12458 -- When compiling for a system where Address is of a visible
12459 -- integer type, spurious ambiguities can be produced when
12460 -- arithmetic operations have a literal operand and return
12461 -- System.Address or a descendant of it. These ambiguities
12462 -- are usually resolved by the context, but for conversions
12463 -- there is no context type and the removal of the spurious
12464 -- operations must be done explicitly here.
12466 if not Address_Is_Private
12468 then
12493 Conversion_Error_N
12496 -- If the interpretation involves a standard operator, use
12497 -- the location of the type, which may be user-defined.
12501 else
12505 Conversion_Error_N -- CODEFIX
12510 else
12514 Conversion_Error_N -- CODEFIX
12526 -- Deal with conversion of integer type to address if the pragma
12527 -- Allow_Integer_Address is in effect. We convert the conversion to
12528 -- an unchecked conversion in this case and we are all done.
12536 -- If we are within a child unit, check whether the type of the
12537 -- expression has an ancestor in a parent unit, in which case it
12538 -- belongs to its derivation class even if the ancestor is private.
12539 -- See RM 7.3.1 (5.2/3).
12543 -- Numeric types
12547 -- A universal fixed expression can be converted to any numeric type
12552 -- Also no need to check when in an instance or inlined body, because
12553 -- the legality has been established when the template was analyzed.
12554 -- Furthermore, numeric conversions may occur where only a private
12555 -- view of the operand type is visible at the instantiation point.
12556 -- This results in a spurious error if we check that the operand type
12557 -- is a numeric type.
12559 -- Note: in a previous version of this unit, the following tests were
12560 -- applied only for generated code (Comes_From_Source set to False),
12561 -- but in fact the test is required for source code as well, since
12562 -- this situation can arise in source code.
12567 -- Otherwise we need the conversion check
12569 else
12570 return Conversion_Check
12572 or else
12578 -- Array types
12584 then
12585 Conversion_Error_N
12589 else
12593 -- Ada 2005 (AI-251): Internally generated conversions of access to
12594 -- interface types added to force the displacement of the pointer to
12595 -- reference the corresponding dispatch table.
12600 then
12603 -- Ada 2005 (AI-251): Anonymous access types where target references an
12604 -- interface type.
12608 E_Anonymous_Access_Type)
12610 then
12611 -- Check the static accessibility rule of 4.6(17). Note that the
12612 -- check is not enforced when within an instance body, since the
12613 -- RM requires such cases to be caught at run time.
12615 -- If the operand is a rewriting of an allocator no check is needed
12616 -- because there are no accessibility issues.
12624 then
12625 -- In an instance, this is a run-time check, but one we know
12626 -- will fail, so generate an appropriate warning. The raise
12627 -- will be generated by Expand_N_Type_Conversion.
12631 Conversion_Error_N
12633 Operand);
12636 else
12637 Conversion_Error_N
12639 Operand);
12643 -- Special accessibility checks are needed in the case of access
12644 -- discriminants declared for a limited type.
12648 then
12649 -- When the operand is a selected access discriminant the check
12650 -- needs to be made against the level of the object denoted by
12651 -- the prefix of the selected name (Object_Access_Level handles
12652 -- checking the prefix of the operand for this case).
12657 then
12658 -- In an instance, this is a run-time check, but one we know
12659 -- will fail, so generate an appropriate warning. The raise
12660 -- will be generated by Expand_N_Type_Conversion.
12664 Conversion_Error_N
12669 -- Real error if not in instance body
12671 else
12672 Conversion_Error_N
12679 -- The case of a reference to an access discriminant from
12680 -- within a limited type declaration (which will appear as
12681 -- a discriminal) is always illegal because the level of the
12682 -- discriminant is considered to be deeper than any (nameable)
12683 -- access type.
12687 and then
12690 then
12691 Conversion_Error_N
12693 Operand);
12701 -- General and anonymous access types
12704 E_Anonymous_Access_Type)
12705 and then
12706 Conversion_Check
12708 and then not
12710 E_Access_Protected_Subprogram_Type),
12712 then
12715 then
12716 Conversion_Error_N
12721 -- Check the static accessibility rule of 4.6(17). Note that the
12722 -- check is not enforced when within an instance body, since the RM
12723 -- requires such cases to be caught at run time.
12728 N_Object_Declaration
12729 then
12730 -- Ada 2012 (AI05-0149): Perform legality checking on implicit
12731 -- conversions from an anonymous access type to a named general
12732 -- access type. Such conversions are not allowed in the case of
12733 -- access parameters and stand-alone objects of an anonymous
12734 -- access type. The implicit conversion case is recognized by
12735 -- testing that Comes_From_Source is False and that it's been
12736 -- rewritten. The Comes_From_Source test isn't sufficient because
12737 -- nodes in inlined calls to predefined library routines can have
12738 -- Comes_From_Source set to False. (Is there a better way to test
12739 -- for implicit conversions???)
12746 then
12749 -- Implicit conversions aren't allowed for objects of an
12750 -- anonymous access type, since such objects have nonstatic
12751 -- levels in Ada 2012.
12754 N_Object_Declaration
12755 then
12756 Conversion_Error_N
12761 -- Implicit conversions aren't allowed for anonymous access
12762 -- parameters. The "not Is_Local_Anonymous_Access_Type" test
12763 -- is done to exclude anonymous access results.
12767 N_Function_Specification,
12768 N_Procedure_Specification)
12769 then
12770 Conversion_Error_N
12775 -- This is a case where there's an enclosing object whose
12776 -- to which the "statically deeper than" relationship does
12777 -- not apply (such as an access discriminant selected from
12778 -- a dereference of an access parameter).
12782 then
12783 Conversion_Error_N
12788 -- In other cases, the level of the operand's type must be
12789 -- statically less deep than that of the target type, else
12790 -- implicit conversion is disallowed (by RM12-8.6(27.1/3)).
12794 then
12795 Conversion_Error_N
12804 then
12805 -- In an instance, this is a run-time check, but one we know
12806 -- will fail, so generate an appropriate warning. The raise
12807 -- will be generated by Expand_N_Type_Conversion.
12811 Conversion_Error_N
12813 Operand);
12816 -- If not in an instance body, this is a real error
12818 else
12819 -- Avoid generation of spurious error message
12822 Conversion_Error_N
12824 Operand);
12830 -- Special accessibility checks are needed in the case of access
12831 -- discriminants declared for a limited type.
12835 then
12836 -- When the operand is a selected access discriminant the check
12837 -- needs to be made against the level of the object denoted by
12838 -- the prefix of the selected name (Object_Access_Level handles
12839 -- checking the prefix of the operand for this case).
12844 then
12845 -- In an instance, this is a run-time check, but one we know
12846 -- will fail, so generate an appropriate warning. The raise
12847 -- will be generated by Expand_N_Type_Conversion.
12851 Conversion_Error_N
12856 -- If not in an instance body, this is a real error
12858 else
12859 Conversion_Error_N
12866 -- The case of a reference to an access discriminant from
12867 -- within a limited type declaration (which will appear as
12868 -- a discriminal) is always illegal because the level of the
12869 -- discriminant is considered to be deeper than any (nameable)
12870 -- access type.
12873 and then
12876 then
12877 Conversion_Error_N
12879 Operand);
12885 -- In the presence of limited_with clauses we have to use nonlimited
12886 -- views, if available.
12890 -- Helper function to handle limited views
12892 --------------------------
12893 -- Full_Designated_Type --
12894 --------------------------
12899 begin
12900 -- Handle the limited view of a type
12904 then
12906 else
12911 -- Local Declarations
12919 -- Start of processing for Check_Limited
12921 begin
12925 else
12927 Conversion_Error_NE
12932 -- Ada 2005 AI-384: legality rule is symmetric in both
12933 -- designated types. The conversion is legal (with possible
12934 -- constraint check) if either designated type is
12935 -- unconstrained.
12938 or else
12940 and then
12943 then
12944 -- Special case, if Value_Size has been used to make the
12945 -- sizes different, the conversion is not allowed even
12946 -- though the subtypes statically match.
12951 then
12952 Conversion_Error_NE
12955 Conversion_Error_NE
12960 -- Normal case where conversion is allowed
12962 else
12966 else
12967 Error_Msg_NE
12975 -- Access to subprogram types. If the operand is an access parameter,
12976 -- the type has a deeper accessibility that any master, and cannot be
12977 -- assigned. We must make an exception if the conversion is part of an
12978 -- assignment and the target is the return object of an extended return
12979 -- statement, because in that case the accessibility check takes place
12980 -- after the return.
12984 -- Note: this test of Opnd_Type is there to prevent entering this
12985 -- branch in the case of a remote access to subprogram type, which
12986 -- is internally represented as an E_Record_Type.
12989 then
12993 and then
12997 then
12998 Conversion_Error_N
13000 Operand);
13001 Conversion_Error_N
13004 Operand);
13006 Error_Msg_NE
13011 -- Check that the designated types are subtype conformant
13017 -- Check the static accessibility rule of 4.6(20)
13021 then
13022 Conversion_Error_N
13024 Operand);
13026 -- Check that if the operand type is declared in a generic body,
13027 -- then the target type must be declared within that same body
13028 -- (enforces last sentence of 4.6(20)).
13031 declare
13037 begin
13044 Conversion_Error_N
13053 -- Remote access to subprogram types
13057 then
13058 -- It is valid to convert from one RAS type to another provided
13059 -- that their specification statically match.
13061 -- Note: at this point, remote access to subprogram types have been
13062 -- expanded to their E_Record_Type representation, and we need to
13063 -- go back to the original access type definition using the
13064 -- Corresponding_Remote_Type attribute in order to check that the
13065 -- designated profiles match.
13070 Check_Subtype_Conformant
13073 Old_Id =>
13075 Err_Loc =>
13076 N);
13079 -- If it was legal in the generic, it's legal in the instance
13084 -- If both are tagged types, check legality of view conversions
13087 and then
13089 then
13092 -- Types derived from the same root type are convertible
13097 -- In an instance or an inlined body, there may be inconsistent views of
13098 -- the same type, or of types derived from a common root.
13101 and then
13104 then
13107 -- Special check for common access type error case
13111 then
13113 Conversion_Error_NE -- CODEFIX
13117 -- Here we have a real conversion error
13119 else
13120 Conversion_Error_NE