| blob | 1d73bf462a1e18bbb8ff5f0706730c1be9dd79ea |
1 ------------------------------------------------------------------------------
2 -- --
3 -- GNAT COMPILER COMPONENTS --
4 -- --
5 -- S E M _ R E S --
6 -- --
7 -- B o d y --
8 -- --
9 -- Copyright (C) 1992-2016, 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 ------------------------------------------------------------------------------
87 -----------------------
88 -- Local Subprograms --
89 -----------------------
91 -- Second pass (top-down) type checking and overload resolution procedures
92 -- Typ is the type required by context. These procedures propagate the
93 -- type information recursively to the descendants of N. If the node is not
94 -- overloaded, its Etype is established in the first pass. If overloaded,
95 -- the Resolve routines set the correct type. For arithmetic operators, the
96 -- Etype is the base type of the context.
98 -- Note that Resolve_Attribute is separated off in Sem_Attr
101 -- Enforce the restrictions on the use of discriminants when constraining
102 -- a component of a discriminated type (record or concurrent type).
105 -- Given a node for an operator associated with type T, check that the
106 -- operator is visible. Operators all of whose operands are universal must
107 -- be checked for visibility during resolution because their type is not
108 -- determinable based on their operands.
110 procedure Check_Fully_Declared_Prefix
113 -- Check that the type of the prefix of a dereference is not incomplete
116 -- Given a call node, N, which is known to occur immediately within the
117 -- subprogram being called, determines whether it is a detectable case of
118 -- an infinite recursion, and if so, outputs appropriate messages. Returns
119 -- True if an infinite recursion is detected, and False otherwise.
122 -- If the type of the object being initialized uses the secondary stack
123 -- directly or indirectly, create a transient scope for the call to the
124 -- init proc. This is because we do not create transient scopes for the
125 -- initialization of individual components within the init proc itself.
126 -- Could be optimized away perhaps?
129 -- N is the node for a logical operator. If the operator is predefined, and
130 -- the root type of the operands is Standard.Boolean, then a check is made
131 -- for restriction No_Direct_Boolean_Operators. This procedure also handles
132 -- the style check for Style_Check_Boolean_And_Or.
135 -- N is either an indexed component or a selected component. This function
136 -- returns true if the prefix refers to an object that has an address
137 -- clause (the case in which we may want to issue a warning).
140 -- Determine whether E is an access type declared by an access declaration,
141 -- and not an (anonymous) allocator type.
144 -- Utility to check whether the entity for an operator is a predefined
145 -- operator, in which case the expression is left as an operator in the
146 -- tree (else it is rewritten into a call). An instance of an intrinsic
147 -- conversion operation may be given an operator name, but is not treated
148 -- like an operator. Note that an operator that is an imported back-end
149 -- builtin has convention Intrinsic, but is expected to be rewritten into
150 -- a call, so such an operator is not treated as predefined by this
151 -- predicate.
154 -- If a default expression in entry call N depends on the discriminants
155 -- of the task, it must be replaced with a reference to the discriminant
156 -- of the task being called.
158 procedure Resolve_Op_Concat_Arg
163 -- Internal procedure for Resolve_Op_Concat to resolve one operand of
164 -- concatenation operator. The operand is either of the array type or of
165 -- the component type. If the operand is an aggregate, and the component
166 -- type is composite, this is ambiguous if component type has aggregates.
169 -- Does the first part of the work of Resolve_Op_Concat
172 -- Does the "rest" of the work of Resolve_Op_Concat, after the left operand
173 -- 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 -- Binary operator
1338 -- Unary operator
1340 else
1346 -- If the operator is denoted by an expanded name, and the prefix is
1347 -- not Standard, but the operator is a predefined one whose scope is
1348 -- Standard, then this is an implicit_operator, inserted as an
1349 -- interpretation by the procedure of the same name. This procedure
1350 -- overestimates the presence of implicit operators, because it does
1351 -- not examine the type of the operands. Verify now that the operand
1352 -- type appears in the given scope. If right operand is universal,
1353 -- check the other operand. In the case of concatenation, either
1354 -- argument can be the component type, so check the type of the result.
1355 -- If both arguments are literals, look for a type of the right kind
1356 -- defined in the given scope. This elaborate nonsense is brought to
1357 -- you courtesy of b33302a. The type itself must be frozen, so we must
1358 -- find the type of the proper class in the given scope.
1360 -- A final wrinkle is the multiplication operator for fixed point types,
1361 -- which is defined in Standard only, and not in the scope of the
1362 -- fixed point type itself.
1367 -- If this is a package renaming, get renamed entity, which will be
1368 -- the scope of the operands if operaton is type-correct.
1374 -- If the entity being called is defined in the given package, it is
1375 -- a renaming of a predefined operator, and known to be legal.
1379 then
1382 -- Visibility does not need to be checked in an instance: if the
1383 -- operator was not visible in the generic it has been diagnosed
1384 -- already, else there is an implicit copy of it in the instance.
1392 then
1397 -- Ada 2005 AI-420: Predefined equality on Universal_Access is
1398 -- available.
1403 then
1406 else
1415 and then Is_Binary
1417 then
1423 -- Verify that the scope contains a type that corresponds to
1424 -- the given literal. Optimize the case where Pack is Standard.
1446 else
1455 else
1464 then
1467 -- If the type is defined elsewhere, and the operator is not
1468 -- defined in the given scope (by a renaming declaration, e.g.)
1469 -- then this is an error as well. If an extension of System is
1470 -- present, and the type may be defined there, Pack must be
1471 -- System itself.
1478 then
1481 else
1487 then
1494 Error_Msg_NE
1499 -- Detect a mismatch between the context type and the result type
1500 -- in the named package, which is otherwise not detected if the
1501 -- operands are universal. Check is only needed if source entity is
1502 -- an operator, not a function that renames an operator.
1509 and then not In_Instance
1510 then
1513 then
1514 -- Already checked above
1518 -- Operator may be defined in an extension of System
1522 then
1525 else
1526 -- Could we use Wrong_Type here??? (this would require setting
1527 -- Etype (N) to the actual type found where Typ was expected).
1538 else
1542 -- If this is a call to a function that renames a predefined equality,
1543 -- the renaming declaration provides a type that must be used to
1544 -- resolve the operands. This must be done now because resolution of
1545 -- the equality node will not resolve any remaining ambiguity, and it
1546 -- assumes that the first operand is not overloaded.
1551 then
1559 -- Do rewrite setting Comes_From_Source on the result if the original
1560 -- call came from source. Although it is not strictly the case that the
1561 -- operator as such comes from the source, logically it corresponds
1562 -- exactly to the function call in the source, so it should be marked
1563 -- this way (e.g. to make sure that validity checks work fine).
1565 declare
1567 begin
1572 -- If this is an arithmetic operator and the result type is private,
1573 -- the operands and the result must be wrapped in conversion to
1574 -- expose the underlying numeric type and expand the proper checks,
1575 -- e.g. on division.
1579 when N_Op_Add | N_Op_Subtract | N_Op_Multiply | N_Op_Divide |
1580 N_Op_Expon | N_Op_Mod | N_Op_Rem =>
1589 else
1593 -- If in ASIS_Mode, propagate operand types to original actuals of
1594 -- function call, which would otherwise not be fully resolved. If
1595 -- the call has already been constant-folded, nothing to do. We
1596 -- relocate the operand nodes rather than copy them, to preserve
1597 -- original_node pointers, given that the operands themselves may
1598 -- have been rewritten. If the call was itself a rewriting of an
1599 -- operator node, nothing to do.
1601 if ASIS_Mode
1604 then
1605 declare
1613 begin
1618 -- If the original call has named associations, replace the
1619 -- explicit actual parameter in the association with the proper
1620 -- resolved operand.
1625 then
1628 else
1633 else
1640 then
1643 else
1648 else
1652 else
1656 else
1666 -------------------
1667 -- Operator_Kind --
1668 -------------------
1670 function Operator_Kind
1673 is
1676 begin
1677 -- Use CASE statement or array???
1714 else
1718 -- Unary operators
1720 else
1729 else
1737 ----------------------------
1738 -- Preanalyze_And_Resolve --
1739 ----------------------------
1744 begin
1748 -- Normally, we suppress all checks for this preanalysis. There is no
1749 -- point in processing them now, since they will be applied properly
1750 -- and in the proper location when the default expressions reanalyzed
1751 -- and reexpanded later on. We will also have more information at that
1752 -- point for possible suppression of individual checks.
1754 -- However, in SPARK mode, most expansion is suppressed, and this
1755 -- later reanalysis and reexpansion may not occur. SPARK mode does
1756 -- require the setting of checking flags for proof purposes, so we
1757 -- do the SPARK preanalysis without suppressing checks.
1759 -- This special handling for SPARK mode is required for example in the
1760 -- case of Ada 2012 constructs such as quantified expressions, which are
1761 -- expanded in two separate steps.
1765 else
1769 Expander_Mode_Restore;
1773 -- Version without context type
1778 begin
1785 Expander_Mode_Restore;
1789 ----------------------------------
1790 -- Replace_Actual_Discriminants --
1791 ----------------------------------
1798 -- Comment needed???
1800 -------------------
1801 -- Process_Discr --
1802 -------------------
1807 begin
1813 then
1829 -- Start of processing for Replace_Actual_Discriminants
1831 begin
1845 else
1850 -------------
1851 -- Resolve --
1852 -------------
1868 -- Determine whether a node comes from a predefined library unit or
1869 -- Standard.
1872 -- Try and fix up a literal so that it matches its expected type. New
1873 -- literals are manufactured if necessary to avoid cascaded errors.
1876 -- Additional diagnostics when an ambiguous call has an ambiguous
1877 -- argument (typically a controlling actual).
1880 -- Called when attempt at resolving current expression fails
1882 ------------------------------------
1883 -- Comes_From_Predefined_Lib_Unit --
1884 -------------------------------------
1887 begin
1888 return
1890 or else Is_Predefined_File_Name
1894 --------------------
1895 -- Patch_Up_Value --
1896 --------------------
1899 begin
1916 then
1921 Char_Literal_Value =>
1937 -------------------------------
1938 -- Report_Ambiguous_Argument --
1939 -------------------------------
1946 begin
1950 then
1953 -- Could use comments on what is going on here???
1961 else
1970 -----------------------
1971 -- Resolution_Failed --
1972 -----------------------
1975 begin
1981 -- The caller will return without calling the expander, so we need
1982 -- to set the analyzed flag. Note that it is fine to set Analyzed
1983 -- to True even if we are in the middle of a shallow analysis,
1984 -- (see the spec of sem for more details) since this is an error
1985 -- situation anyway, and there is no point in repeating the
1986 -- analysis later (indeed it won't work to repeat it later, since
1987 -- we haven't got a clear resolution of which entity is being
1988 -- referenced.)
1994 -- Local variables
1998 -- Start of processing for Resolve
2000 begin
2005 -- A declaration may be subject to pragma Ghost. Set the mode now to
2006 -- ensure that any nodes generated during analysis and expansion are
2007 -- marked as Ghost.
2013 -- Access attribute on remote subprogram cannot be used for a non-remote
2014 -- access-to-subprogram type.
2018 Name_Unrestricted_Access,
2019 Name_Unchecked_Access)
2025 then
2026 Error_Msg_N
2032 -- If the context is a Remote_Access_To_Subprogram, access attributes
2033 -- must be resolved with the corresponding fat pointer. There is no need
2034 -- to check for the attribute name since the return type of an
2035 -- attribute is never a remote type.
2040 then
2041 declare
2049 begin
2050 -- Check that Typ is a remote access-to-subprogram type
2054 -- Prefix (N) must statically denote a remote subprogram
2055 -- declared in a package specification.
2059 Attr = Attribute_Unrestricted_Access
2060 then
2075 or else
2077 then
2079 Error_Msg_N
2081 N);
2084 -- If we are generating code in distributed mode, perform
2085 -- semantic checks against corresponding remote entities.
2087 if Expander_Active
2089 then
2090 Check_Subtype_Conformant
2092 Old_Id => Designated_Type
2118 then
2123 -- Return if already analyzed
2131 -- Any case of Any_Type as the Etype value means that we had a
2132 -- previous error.
2142 -- The resolution of an Expression_With_Actions is determined by
2143 -- its Expression.
2151 -- If not overloaded, then we know the type, and all that needs doing
2152 -- is to check that this type is compatible with the context.
2158 -- In the overloaded case, we must select the interpretation that
2159 -- is compatible with the context (i.e. the type passed to Resolve)
2161 else
2162 -- Loop through possible interpretations
2171 -- We are only interested in interpretations that are compatible
2172 -- with the expected type, any other interpretations are ignored.
2177 Write_Eol;
2180 else
2181 -- Skip the current interpretation if it is disabled by an
2182 -- abstract operator. This action is performed only when the
2183 -- type against which we are resolving is the same as the
2184 -- type of the interpretation.
2191 then
2199 -- First matching interpretation
2207 -- Matching interpretation that is not the first, maybe an
2208 -- error, but there are some cases where preference rules are
2209 -- used to choose between the two possibilities. These and
2210 -- some more obscure cases are handled in Disambiguate.
2212 else
2213 -- If the current statement is part of a predefined library
2214 -- unit, then all interpretations which come from user level
2215 -- packages should not be considered. Check previous and
2216 -- current one.
2224 -- Previous interpretation must be discarded
2234 -- Otherwise apply further disambiguation steps
2239 -- Disambiguation has succeeded. Skip the remaining
2240 -- interpretations.
2250 else
2251 -- Before we issue an ambiguity complaint, check for the
2252 -- case of a subprogram call where at least one of the
2253 -- arguments is Any_Type, and if so suppress the message,
2254 -- since it is a cascaded error. This can also happen for
2255 -- a generalized indexing operation.
2260 then
2261 declare
2265 begin
2281 Write_Eol;
2294 then
2299 then
2303 -- Not that special case, so issue message using the flag
2304 -- Ambiguous to control printing of the header message
2305 -- only at the start of an ambiguous set.
2310 then
2311 Error_Msg_N
2314 else
2315 Error_Msg_NE -- CODEFIX
2323 Error_Msg_N
2325 else
2326 Error_Msg_N -- CODEFIX
2332 then
2333 Report_Ambiguous_Argument;
2339 -- By default, the error message refers to the candidate
2340 -- interpretation. But if it is a predefined operator, it
2341 -- is implicitly declared at the declaration of the type
2342 -- of the operand. Recover the sloc of that declaration
2343 -- for the error message.
2349 Standard_Standard
2350 then
2355 then
2363 Standard_Standard
2364 then
2369 then
2373 -- If this is an indirect call, use the subprogram_type
2374 -- in the message, to have a meaningful location. Also
2375 -- indicate if this is an inherited operation, created
2376 -- by a type declaration.
2381 then
2383 Error_Msg_Sloc :=
2385 else
2392 then
2393 -- Special-case the message for universal_fixed
2394 -- operators, which are not declared with the type
2395 -- of the operand, but appear forever in Standard.
2399 then
2400 Error_Msg_N
2403 else
2404 Error_Msg_N
2408 elsif
2410 then
2411 Error_Msg_N
2413 else
2414 Error_Msg_N -- CODEFIX
2421 -- We have a matching interpretation, Expr_Type is the type
2422 -- from this interpretation, and Seen is the entity.
2424 -- For an operator, just set the entity name. The type will be
2425 -- set by the specific operator resolution routine.
2440 -- AI05-0139-2: Expression is overloaded because type has
2441 -- implicit dereference. If type matches context, no implicit
2442 -- dereference is involved.
2453 then
2454 -- If the node is a general indexing, the dereference is
2455 -- is inserted when resolving the rewritten form, else
2456 -- insert it now.
2460 then
2464 -- For an explicit dereference, attribute reference, range,
2465 -- short-circuit form (which is not an operator node), or call
2466 -- with a name that is an explicit dereference, there is
2467 -- nothing to be done at this point.
2470 N_Attribute_Reference,
2471 N_And_Then,
2472 N_Indexed_Component,
2473 N_Or_Else,
2474 N_Range,
2475 N_Selected_Component,
2476 N_Slice)
2478 then
2481 -- For procedure or function calls, set the type of the name,
2482 -- and also the entity pointer for the prefix.
2486 then
2493 then
2498 -- For all other cases, just set the type of the Name
2500 else
2508 -- Move to next interpretation
2516 -- At this stage Found indicates whether or not an acceptable
2517 -- interpretation exists. If not, then we have an error, except that if
2518 -- the context is Any_Type as a result of some other error, then we
2519 -- suppress the error report.
2524 -- If type we are looking for is Void, then this is the procedure
2525 -- call case, and the error is simply that what we gave is not a
2526 -- procedure name (we think of procedure calls as expressions with
2527 -- types internally, but the user doesn't think of them this way).
2531 -- Special case message if function used as a procedure
2536 then
2537 Error_Msg_NE
2541 -- Otherwise give general message (not clear what cases this
2542 -- covers, but no harm in providing for them).
2544 else
2550 -- Otherwise we do have a subexpression with the wrong type
2552 -- Check for the case of an allocator which uses an access type
2553 -- instead of the designated type. This is a common error and we
2554 -- specialize the message, posting an error on the operand of the
2555 -- allocator, complaining that we expected the designated type of
2556 -- the allocator.
2562 then
2566 -- Check for view mismatch on Null in instances, for which the
2567 -- view-swapping mechanism has no identifier.
2573 then
2579 -- Check for an aggregate. Sometimes we can get bogus aggregates
2580 -- from misuse of parentheses, and we are about to complain about
2581 -- the aggregate without even looking inside it.
2583 -- Instead, if we have an aggregate of type Any_Composite, then
2584 -- analyze and resolve the component fields, and then only issue
2585 -- another message if we get no errors doing this (otherwise
2586 -- assume that the errors in the aggregate caused the problem).
2590 then
2591 -- Disable expansion in any case. If there is a type mismatch
2592 -- it may be fatal to try to expand the aggregate. The flag
2593 -- would otherwise be set to false when the error is posted.
2597 declare
2599 -- Check one aggregate, and set Found to True if we have a
2600 -- definite error in any of its elements
2603 -- Check one element of aggregate and set Found to True if
2604 -- we definitely have an error in the element.
2606 ----------------
2607 -- Check_Aggr --
2608 ----------------
2613 begin
2626 -- If this is a default-initialized component, then
2627 -- there is nothing to check. The box will be
2628 -- replaced by the appropriate call during late
2629 -- expansion.
2640 ----------------
2641 -- Check_Elmt --
2642 ----------------
2645 begin
2646 -- If we have a nested aggregate, go inside it (to
2647 -- attempt a naked analyze-resolve of the aggregate can
2648 -- cause undesirable cascaded errors). Do not resolve
2649 -- expression if it needs a type from context, as for
2650 -- integer * fixed expression.
2655 else
2660 then
2670 begin
2675 -- Looks like we have a type error, but check for special case
2676 -- of Address wanted, integer found, with the configuration pragma
2677 -- Allow_Integer_Address active. If we have this case, introduce
2678 -- an unchecked conversion to allow the integer expression to be
2679 -- treated as an Address. The reverse case of integer wanted,
2680 -- Address found, is treated in an analogous manner.
2689 -- That special Allow_Integer_Address check did not appply, so we
2690 -- have a real type error. If an error message was issued already,
2691 -- Found got reset to True, so if it's still False, issue standard
2692 -- Wrong_Type message.
2696 declare
2699 begin
2705 -- Protected operation: retrieve operation name
2709 else
2714 Error_Msg_NE
2720 declare
2724 begin
2731 Error_Msg_NE
2737 else
2741 else
2747 Resolution_Failed;
2751 -- Test if we have more than one interpretation for the context
2754 Resolution_Failed;
2758 -- Only one intepretation
2760 else
2761 -- In Ada 2005, if we have something like "X : T := 2 + 2;", where
2762 -- the "+" on T is abstract, and the operands are of universal type,
2763 -- the above code will have (incorrectly) resolved the "+" to the
2764 -- universal one in Standard. Therefore check for this case and give
2765 -- an error. We can't do this earlier, because it would cause legal
2766 -- cases to get errors (when some other type has an abstract "+").
2772 then
2777 then
2778 Error_Msg_NE
2787 -- Here we have an acceptable interpretation for the context
2789 -- Propagate type information and normalize tree for various
2790 -- predefined operations. If the context only imposes a class of
2791 -- types, rather than a specific type, propagate the actual type
2792 -- downward.
2798 Typ = Any_Discrete
2799 then
2802 -- Any_Fixed is legal in a real context only if a specific fixed-
2803 -- point type is imposed. If Norman Cohen can be confused by this,
2804 -- it deserves a separate message.
2808 then
2815 -- A user-defined operator is transformed into a function call at
2816 -- this point, so that further processing knows that operators are
2817 -- really operators (i.e. are predefined operators). User-defined
2818 -- operators that are intrinsic are just renamings of the predefined
2819 -- ones, and need not be turned into calls either, but if they rename
2820 -- a different operator, we must transform the node accordingly.
2821 -- Instantiations of Unchecked_Conversion are intrinsic but are
2822 -- treated as functions, even if given an operator designator.
2827 then
2833 and then
2835 N_Subprogram_Renaming_Declaration
2836 then
2839 -- If the node is rewritten, it will be fully resolved in
2840 -- Rewrite_Renamed_Operator.
2851 when N_Aggregate => Resolve_Aggregate (N, Ctx_Type);
2853 when N_Allocator => Resolve_Allocator (N, Ctx_Type);
2855 when N_Short_Circuit
2856 => Resolve_Short_Circuit (N, Ctx_Type);
2858 when N_Attribute_Reference
2859 => Resolve_Attribute (N, Ctx_Type);
2861 when N_Case_Expression
2862 => Resolve_Case_Expression (N, Ctx_Type);
2864 when N_Character_Literal
2865 => Resolve_Character_Literal (N, Ctx_Type);
2867 when N_Expanded_Name
2868 => Resolve_Entity_Name (N, Ctx_Type);
2870 when N_Explicit_Dereference
2871 => Resolve_Explicit_Dereference (N, Ctx_Type);
2873 when N_Expression_With_Actions
2874 => Resolve_Expression_With_Actions (N, Ctx_Type);
2876 when N_Extension_Aggregate
2877 => Resolve_Extension_Aggregate (N, Ctx_Type);
2879 when N_Function_Call
2880 => Resolve_Call (N, Ctx_Type);
2882 when N_Identifier
2883 => Resolve_Entity_Name (N, Ctx_Type);
2885 when N_If_Expression
2886 => Resolve_If_Expression (N, Ctx_Type);
2888 when N_Indexed_Component
2889 => Resolve_Indexed_Component (N, Ctx_Type);
2891 when N_Integer_Literal
2892 => Resolve_Integer_Literal (N, Ctx_Type);
2894 when N_Membership_Test
2895 => Resolve_Membership_Op (N, Ctx_Type);
2897 when N_Null => Resolve_Null (N, Ctx_Type);
2899 when N_Op_And | N_Op_Or | N_Op_Xor
2900 => Resolve_Logical_Op (N, Ctx_Type);
2902 when N_Op_Eq | N_Op_Ne
2903 => Resolve_Equality_Op (N, Ctx_Type);
2905 when N_Op_Lt | N_Op_Le | N_Op_Gt | N_Op_Ge
2906 => Resolve_Comparison_Op (N, Ctx_Type);
2908 when N_Op_Not => Resolve_Op_Not (N, Ctx_Type);
2910 when N_Op_Add | N_Op_Subtract | N_Op_Multiply |
2911 N_Op_Divide | N_Op_Mod | N_Op_Rem
2913 => Resolve_Arithmetic_Op (N, Ctx_Type);
2915 when N_Op_Concat => Resolve_Op_Concat (N, Ctx_Type);
2917 when N_Op_Expon => Resolve_Op_Expon (N, Ctx_Type);
2919 when N_Op_Plus | N_Op_Minus | N_Op_Abs
2920 => Resolve_Unary_Op (N, Ctx_Type);
2922 when N_Op_Shift => Resolve_Shift (N, Ctx_Type);
2924 when N_Procedure_Call_Statement
2925 => Resolve_Call (N, Ctx_Type);
2927 when N_Operator_Symbol
2928 => Resolve_Operator_Symbol (N, Ctx_Type);
2930 when N_Qualified_Expression
2931 => Resolve_Qualified_Expression (N, Ctx_Type);
2933 -- Why is the following null, needs a comment ???
2935 when N_Quantified_Expression
2936 => null;
2938 when N_Raise_Expression
2939 => Resolve_Raise_Expression (N, Ctx_Type);
2941 when N_Raise_xxx_Error
2942 => Set_Etype (N, Ctx_Type);
2944 when N_Range => Resolve_Range (N, Ctx_Type);
2946 when N_Real_Literal
2947 => Resolve_Real_Literal (N, Ctx_Type);
2949 when N_Reference => Resolve_Reference (N, Ctx_Type);
2951 when N_Selected_Component
2952 => Resolve_Selected_Component (N, Ctx_Type);
2954 when N_Slice => Resolve_Slice (N, Ctx_Type);
2956 when N_String_Literal
2957 => Resolve_String_Literal (N, Ctx_Type);
2959 when N_Type_Conversion
2960 => Resolve_Type_Conversion (N, Ctx_Type);
2962 when N_Unchecked_Expression =>
2963 Resolve_Unchecked_Expression (N, Ctx_Type);
2965 when N_Unchecked_Type_Conversion =>
2966 Resolve_Unchecked_Type_Conversion (N, Ctx_Type);
2967 end case;
2969 -- Ada 2012 (AI05-0149): Apply an (implicit) conversion to an
2970 -- expression of an anonymous access type that occurs in the context
2971 -- of a named general access type, except when the expression is that
2972 -- of a membership test. This ensures proper legality checking in
2973 -- terms of allowed conversions (expressions that would be illegal to
2974 -- convert implicitly are allowed in membership tests).
2976 if Ada_Version >= Ada_2012
2977 and then Ekind (Ctx_Type) = E_General_Access_Type
2978 and then Ekind (Etype (N)) = E_Anonymous_Access_Type
2979 and then Nkind (Parent (N)) not in N_Membership_Test
2980 then
2981 Rewrite (N, Convert_To (Ctx_Type, Relocate_Node (N)));
2982 Analyze_And_Resolve (N, Ctx_Type);
2983 end if;
2985 -- If the subexpression was replaced by a non-subexpression, then
2986 -- all we do is to expand it. The only legitimate case we know of
2987 -- is converting procedure call statement to entry call statements,
2988 -- but there may be others, so we are making this test general.
2990 if Nkind (N) not in N_Subexpr then
2991 Debug_A_Exit ("resolving ", N, " (done)");
2992 Expand (N);
2993 Ghost_Mode := Save_Ghost_Mode;
2994 return;
2995 end if;
2997 -- The expression is definitely NOT overloaded at this point, so
2998 -- we reset the Is_Overloaded flag to avoid any confusion when
2999 -- reanalyzing the node.
3001 Set_Is_Overloaded (N, False);
3003 -- Freeze expression type, entity if it is a name, and designated
3004 -- type if it is an allocator (RM 13.14(10,11,13)).
3006 -- Now that the resolution of the type of the node is complete, and
3007 -- we did not detect an error, we can expand this node. We skip the
3008 -- expand call if we are in a default expression, see section
3009 -- "Handling of Default Expressions" in Sem spec.
3011 Debug_A_Exit ("resolving ", N, " (done)");
3013 -- We unconditionally freeze the expression, even if we are in
3014 -- default expression mode (the Freeze_Expression routine tests this
3015 -- flag and only freezes static types if it is set).
3017 -- Ada 2012 (AI05-177): The declaration of an expression function
3018 -- does not cause freezing, but we never reach here in that case.
3019 -- Here we are resolving the corresponding expanded body, so we do
3020 -- need to perform normal freezing.
3022 Freeze_Expression (N);
3024 -- Now we can do the expansion
3026 Expand (N);
3027 end if;
3029 Ghost_Mode := Save_Ghost_Mode;
3030 end Resolve;
3032 -------------
3033 -- Resolve --
3034 -------------
3036 -- Version with check(s) suppressed
3038 procedure Resolve (N : Node_Id; Typ : Entity_Id; Suppress : Check_Id) is
3039 begin
3040 if Suppress = All_Checks then
3041 declare
3042 Sva : constant Suppress_Array := Scope_Suppress.Suppress;
3043 begin
3044 Scope_Suppress.Suppress := (others => True);
3045 Resolve (N, Typ);
3046 Scope_Suppress.Suppress := Sva;
3047 end;
3049 else
3050 declare
3051 Svg : constant Boolean := Scope_Suppress.Suppress (Suppress);
3052 begin
3053 Scope_Suppress.Suppress (Suppress) := True;
3054 Resolve (N, Typ);
3055 Scope_Suppress.Suppress (Suppress) := Svg;
3056 end;
3057 end if;
3058 end Resolve;
3060 -------------
3061 -- Resolve --
3062 -------------
3064 -- Version with implicit type
3066 procedure Resolve (N : Node_Id) is
3067 begin
3068 Resolve (N, Etype (N));
3069 end Resolve;
3071 ---------------------
3072 -- Resolve_Actuals --
3073 ---------------------
3075 procedure Resolve_Actuals (N : Node_Id; Nam : Entity_Id) is
3076 Loc : constant Source_Ptr := Sloc (N);
3077 A : Node_Id;
3078 A_Id : Entity_Id;
3079 A_Typ : Entity_Id;
3080 F : Entity_Id;
3081 F_Typ : Entity_Id;
3082 Prev : Node_Id := Empty;
3083 Orig_A : Node_Id;
3084 Real_F : Entity_Id;
3086 Real_Subp : Entity_Id;
3087 -- If the subprogram being called is an inherited operation for
3088 -- a formal derived type in an instance, Real_Subp is the subprogram
3089 -- that will be called. It may have different formal names than the
3090 -- operation of the formal in the generic, so after actual is resolved
3091 -- the name of the actual in a named association must carry the name
3092 -- of the actual of the subprogram being called.
3094 procedure Check_Aliased_Parameter;
3095 -- Check rules on aliased parameters and related accessibility rules
3096 -- in (RM 3.10.2 (10.2-10.4)).
3098 procedure Check_Argument_Order;
3099 -- Performs a check for the case where the actuals are all simple
3100 -- identifiers that correspond to the formal names, but in the wrong
3101 -- order, which is considered suspicious and cause for a warning.
3103 procedure Check_Prefixed_Call;
3104 -- If the original node is an overloaded call in prefix notation,
3106 -- Try_Object_Operation has already verified that there is a valid
3107 -- interpretation, but the form of the actual can only be determined
3108 -- once the primitive operation is identified.
3111 -- Emit an error concerning the illegal usage of an effectively volatile
3112 -- object in interfering context (SPARK RM 7.13(12)).
3115 -- If the actual is missing in a call, insert in the actuals list
3116 -- an instance of the default expression. The insertion is always
3117 -- a named association.
3119 procedure Property_Error
3123 -- Emit an error concerning variable Var with entity Var_Id that has
3124 -- enabled property Prop_Nam when it acts as an actual parameter in a
3125 -- call and the corresponding formal parameter is of mode IN.
3128 -- Check whether T1 and T2, or their full views, are derived from a
3129 -- common type. Used to enforce the restrictions on array conversions
3130 -- of AI95-00246.
3133 -- Predicate to determine whether an actual that is a concatenation
3134 -- will be evaluated statically and does not need a transient scope.
3135 -- This must be determined before the actual is resolved and expanded
3136 -- because if needed the transient scope must be introduced earlier.
3138 -----------------------------
3139 -- Check_Aliased_Parameter --
3140 -----------------------------
3145 begin
3153 else
3160 -- In a generic body assume the worst for generic formals:
3161 -- they can have a constrained partial view (AI05-041).
3167 then
3170 else
3175 else
3187 then
3195 then
3196 Error_Msg_N
3202 --------------------------
3203 -- Check_Argument_Order --
3204 --------------------------
3207 begin
3208 -- Nothing to do if no parameters, or original node is neither a
3209 -- function call nor a procedure call statement (happens in the
3210 -- operator-transformed-to-function call case), or the call does
3211 -- not come from source, or this warning is off.
3213 if not Warn_On_Parameter_Order
3217 then
3221 declare
3224 begin
3225 -- Nothing to do if only one parameter
3231 -- Here if at least two arguments
3233 declare
3239 -- Set True if an out of order case is found
3241 begin
3242 -- Collect identifier names of actuals, fail if any actual is
3243 -- not a simple identifier, and record max length of name.
3249 else
3255 -- If we got this far, all actuals are identifiers and the list
3256 -- of their names is stored in the Actuals array.
3261 -- If we ran out of formals, that's odd, probably an error
3262 -- which will be detected elsewhere, but abandon the search.
3268 -- If name matches and is in order OK
3273 else
3274 -- If no match, see if it is elsewhere in list and if so
3275 -- flag potential wrong order if type is compatible.
3279 and then
3281 then
3287 -- No match
3295 -- If Formals left over, also probably an error, skip warning
3301 -- Here we give the warning if something was out of order
3304 Error_Msg_N
3311 -------------------------
3312 -- Check_Prefixed_Call --
3313 -------------------------
3322 begin
3323 -- Check whether the call is a prefixed call, with or without
3324 -- additional actuals.
3327 or else
3333 then
3336 then
3337 -- Introduce dereference on object in prefix
3339 New_A :=
3347 then
3348 -- Introduce an implicit 'Access in prefix
3351 Error_Msg_NE
3367 ---------------------------------------
3368 -- Flag_Effectively_Volatile_Objects --
3369 ---------------------------------------
3373 -- Determine whether arbitrary node N denotes an effectively volatile
3374 -- object and if it does, emit an error.
3376 -----------------
3377 -- Flag_Object --
3378 -----------------
3383 begin
3384 -- Do not consider nested function calls because they have already
3385 -- been processed during their own resolution.
3397 then
3398 Error_Msg_N
3410 -- Start of processing for Flag_Effectively_Volatile_Objects
3412 begin
3416 --------------------
3417 -- Insert_Default --
3418 --------------------
3424 begin
3425 -- Missing argument in call, nothing to insert
3430 else
3431 -- Note that we do a full New_Copy_Tree, so that any associated
3432 -- Itypes are properly copied. This may not be needed any more,
3433 -- but it does no harm as a safety measure. Defaults of a generic
3434 -- formal may be out of bounds of the corresponding actual (see
3435 -- cc1311b) and an additional check may be required.
3437 Actval :=
3438 New_Copy_Tree
3443 -- Propagate dimension information, if any.
3449 then
3455 then
3458 then
3459 -- If default is a real literal, do not introduce a
3460 -- conversion whose effect may depend on the run-time
3461 -- size of universal real.
3465 else
3476 -- Resolve aggregates with their base type, to avoid scope
3477 -- anomalies: the subtype was first built in the subprogram
3478 -- declaration, and the current call may be nested.
3482 else
3486 else
3489 -- See note above concerning aggregates
3493 then
3496 -- Resolve entities with their own type, which may differ from
3497 -- the type of a reference in a generic context (the view
3498 -- swapping mechanism did not anticipate the re-analysis of
3499 -- default values in calls).
3504 else
3509 -- If default is a tag indeterminate function call, propagate tag
3510 -- to obtain proper dispatching.
3514 then
3519 -- If the default expression raises constraint error, then just
3520 -- silently replace it with an N_Raise_Constraint_Error node, since
3521 -- we already gave the warning on the subprogram spec. If node is
3522 -- already a Raise_Constraint_Error leave as is, to prevent loops in
3523 -- the warnings removal machinery.
3527 then
3535 Assoc :=
3540 -- Case of insertion is first named actual
3544 then
3551 else
3555 else
3559 -- Case of insertion is not first named actual
3561 else
3562 Set_Next_Named_Actual
3574 --------------------
3575 -- Property_Error --
3576 --------------------
3578 procedure Property_Error
3582 is
3583 begin
3585 Error_Msg_NE
3591 -------------------
3592 -- Same_Ancestor --
3593 -------------------
3599 begin
3602 then
3608 then
3615 --------------------------
3616 -- Static_Concatenation --
3617 --------------------------
3620 begin
3627 -- Concatenation is static when both operands are static and
3628 -- the concatenation operator is a predefined one.
3631 and then
3633 and then
3638 declare
3640 begin
3643 and then
3647 else
3653 -- Start of processing for Resolve_Actuals
3655 begin
3656 Check_Argument_Order;
3660 and then In_Instance
3663 then
3665 else
3670 Check_Prefixed_Call;
3684 -- If we have an error in any actual or formal, indicated by a type
3685 -- of Any_Type, then abandon resolution attempt, and set result type
3686 -- to Any_Type. Skip this if the actual is a Raise_Expression, whose
3687 -- type is imposed from context.
3691 then
3698 -- Case where actual is present
3700 -- If the actual is an entity, generate a reference to it now. We
3701 -- do this before the actual is resolved, because a formal of some
3702 -- protected subprogram, or a task discriminant, will be rewritten
3703 -- during expansion, and the source entity reference may be lost.
3708 then
3714 then
3721 -- RM 6.4.1(12): For an out parameter that is passed by
3722 -- copy, the formal parameter object is created, and:
3724 -- * For an access type, the formal parameter is initialized
3725 -- from the value of the actual, without checking that the
3726 -- value satisfies any constraint, any predicate, or any
3727 -- exclusion of the null value.
3729 -- * For a scalar type that has the Default_Value aspect
3730 -- specified, the formal parameter is initialized from the
3731 -- value of the actual, without checking that the value
3732 -- satisfies any constraint or any predicate.
3733 -- I do not understand why this case is included??? this is
3734 -- not a case where an OUT parameter is treated as IN OUT.
3736 -- * For a composite type with discriminants or that has
3737 -- implicit initial values for any subcomponents, the
3738 -- behavior is as for an in out parameter passed by copy.
3740 -- Hence for these cases we generate the read reference now
3741 -- (the write reference will be generated later by
3742 -- Note_Possible_Modification).
3745 and then
3747 or else
3749 and then
3751 or else
3754 or else Is_Partially_Initialized_Type
3756 then
3766 then
3767 -- If style checking mode on, check match of formal name
3775 -- If the formal is Out or In_Out, do not resolve and expand the
3776 -- conversion, because it is subsequently expanded into explicit
3777 -- temporaries and assignments. However, the object of the
3778 -- conversion can be resolved. An exception is the case of tagged
3779 -- type conversion with a class-wide actual. In that case we want
3780 -- the tag check to occur and no temporary will be needed (no
3781 -- representation change can occur) and the parameter is passed by
3782 -- reference, so we go ahead and resolve the type conversion.
3783 -- Another exception is the case of reference to component or
3784 -- subcomponent of a bit-packed array, in which case we want to
3785 -- defer expansion to the point the in and out assignments are
3786 -- performed.
3791 then
3794 then
3795 -- In a view conversion, the conversion must be legal in
3796 -- both directions, and thus both component types must be
3797 -- aliased, or neither (4.6 (8)).
3799 -- The extra rule in 4.6 (24.9.2) seems unduly restrictive:
3800 -- the privacy requirement should not apply to generic
3801 -- types, and should be checked in an instance. ARG query
3802 -- is in order ???
3806 then
3807 Error_Msg_N
3811 -- Comment here??? what set of cases???
3813 elsif
3815 then
3816 -- Check view conv between unrelated by ref array types
3820 then
3821 Error_Msg_N
3825 -- In Ada 2005 mode, check view conversion component
3826 -- type cannot be private, tagged, or volatile. Note
3827 -- that we only apply this to source conversions. The
3828 -- generated code can contain conversions which are
3829 -- not subject to this test, and we cannot extract the
3830 -- component type in such cases since it is not present.
3834 then
3835 declare
3837 Component_Type
3839 begin
3844 then
3845 Error_Msg_N
3856 -- Resolve expression if conversion is all OK
3861 then
3865 -- If the actual is a function call that returns a limited
3866 -- unconstrained object that needs finalization, create a
3867 -- transient scope for it, so that it can receive the proper
3868 -- finalization list.
3873 and then Expander_Active
3875 then
3879 -- A small optimization: if one of the actuals is a concatenation
3880 -- create a block around a procedure call to recover stack space.
3881 -- This alleviates stack usage when several procedure calls in
3882 -- the same statement list use concatenation. We do not perform
3883 -- this wrapping for code statements, where the argument is a
3884 -- static string, and we want to preserve warnings involving
3885 -- sequences of such statements.
3889 and then Expander_Active
3890 and then
3894 then
3898 else
3902 and then
3905 then
3906 Error_Msg_N
3919 -- (Ada 2005: AI-251): If the actual is an allocator whose
3920 -- directly designated type is a class-wide interface, we build
3921 -- an anonymous access type to use it as the type of the
3922 -- allocator. Later, when the subprogram call is expanded, if
3923 -- the interface has a secondary dispatch table the expander
3924 -- will add a type conversion to force the correct displacement
3925 -- of the pointer.
3928 declare
3934 begin
3937 then
3940 Set_Directly_Designated_Type
3945 -- Ada 2005, AI-162:If the actual is an allocator, the
3946 -- innermost enclosing statement is the master of the
3947 -- created object. This needs to be done with expansion
3948 -- enabled only, otherwise the transient scope will not
3949 -- be removed in the expansion of the wrapped construct.
3952 and then Expander_Active
3953 then
3963 -- (Ada 2005): The call may be to a primitive operation of a
3964 -- tagged synchronized type, declared outside of the type. In
3965 -- this case the controlling actual must be converted to its
3966 -- corresponding record type, which is the formal type. The
3967 -- actual may be a subtype, either because of a constraint or
3968 -- because it is a generic actual, so use base type to locate
3969 -- concurrent type.
3976 then
3977 -- If the actual is overloaded, look for an interpretation
3978 -- that has a synchronized type.
3983 else
3984 declare
3988 begin
3993 then
4003 declare
4006 begin
4009 else
4013 -- Tagged synchronized type (case 1): the actual is a
4014 -- concurrent type.
4018 then
4020 Unchecked_Convert_To
4024 -- Tagged synchronized type (case 2): the formal is a
4025 -- concurrent type.
4028 and then Present
4033 then
4036 -- Common case
4038 else
4043 -- Not a synchronized operation
4045 else
4053 -- An actual cannot be an untagged formal incomplete type
4058 then
4059 Error_Msg_N
4065 N_Procedure_Call_Statement)
4066 then
4067 -- In formal mode, check that actual parameters matching
4068 -- formals of tagged types are objects (or ancestor type
4069 -- conversions of objects), not general expressions.
4076 declare
4081 begin
4083 Check_SPARK_05_Restriction
4086 -- In formal mode, the only view conversions are those
4087 -- involving ancestor conversion of an extended type.
4089 elsif not
4095 then
4096 if Ekind_In
4098 then
4099 Check_SPARK_05_Restriction
4102 else
4103 Check_SPARK_05_Restriction
4107 else
4112 else
4116 -- In formal mode, the only view conversions are those
4117 -- involving ancestor conversion of an extended type.
4121 then
4122 Check_SPARK_05_Restriction
4128 -- has warnings suppressed, then we reset Never_Set_In_Source for
4129 -- the calling entity. The reason for this is to catch cases like
4130 -- GNAT.Spitbol.Patterns.Vstring_Var where the called subprogram
4131 -- uses trickery to modify an IN parameter.
4138 then
4142 -- Perform error checks for IN and IN OUT parameters
4146 -- Check unset reference. For scalar parameters, it is clearly
4147 -- wrong to pass an uninitialized value as either an IN or
4148 -- IN-OUT parameter. For composites, it is also clearly an
4149 -- error to pass a completely uninitialized value as an IN
4150 -- parameter, but the case of IN OUT is trickier. We prefer
4151 -- not to give a warning here. For example, suppose there is
4152 -- a routine that sets some component of a record to False.
4153 -- It is perfectly reasonable to make this IN-OUT and allow
4154 -- either initialized or uninitialized records to be passed
4155 -- in this case.
4157 -- For partially initialized composite values, we also avoid
4158 -- warnings, since it is quite likely that we are passing a
4159 -- partially initialized value and only the initialized fields
4160 -- will in fact be read in the subprogram.
4165 then
4169 -- In Ada 83 we cannot pass an OUT parameter as an IN or IN OUT
4170 -- actual to a nested call, since this constitutes a reading of
4171 -- the parameter, which is not allowed.
4176 then
4181 -- Case of OUT or IN OUT parameter
4185 -- For an Out parameter, check for useless assignment. Note
4186 -- that we can't set Last_Assignment this early, because we may
4187 -- kill current values in Resolve_Call, and that call would
4188 -- clobber the Last_Assignment field.
4190 -- Note: call Warn_On_Useless_Assignment before doing the check
4191 -- below for Is_OK_Variable_For_Out_Formal so that the setting
4192 -- of Referenced_As_LHS/Referenced_As_Out_Formal properly
4193 -- reflects the last assignment, not this one.
4200 then
4205 -- Validate the form of the actual. Note that the call to
4206 -- Is_OK_Variable_For_Out_Formal generates the required
4207 -- reference in this case.
4209 -- A call to an initialization procedure for an aggregate
4210 -- component may initialize a nested component of a constant
4211 -- designated object. In this context the object is variable.
4215 then
4219 and then
4222 then
4223 Error_Msg_N
4229 -- What's the following about???
4233 else
4234 Kill_All_Checks;
4243 -- Apply appropriate constraint/predicate checks for IN [OUT] case
4247 -- Apply predicate tests except in certain special cases. Note
4248 -- that it might be more consistent to apply these only when
4249 -- expansion is active (in Exp_Ch6.Expand_Actuals), as we do
4250 -- for the outbound predicate tests ???
4256 -- Apply required constraint checks
4258 -- Gigi looks at the check flag and uses the appropriate types.
4259 -- For now since one flag is used there is an optimization
4260 -- which might not be done in the IN OUT case since Gigi does
4261 -- not do any analysis. More thought required about this ???
4263 -- In fact is this comment obsolete??? doesn't the expander now
4264 -- generate all these tests anyway???
4277 then
4280 -- For view conversions of a discriminated object, apply
4281 -- check to object itself, the conversion alreay has the
4282 -- proper type.
4286 then
4293 then
4299 then
4302 else
4306 -- Ada 2005 (AI-231): Note that the controlling parameter case
4307 -- already existed in Ada 95, which is partially checked
4308 -- elsewhere (see Checks), and we don't want the warning
4309 -- message to differ.
4314 then
4316 Apply_Compile_Time_Constraint_Error
4322 Apply_Compile_Time_Constraint_Error
4331 -- Checks for OUT parameters and IN OUT parameters
4335 -- If there is a type conversion, to make sure the return value
4336 -- meets the constraints of the variable before the conversion.
4340 Apply_Scalar_Range_Check
4342 else
4343 Apply_Range_Check
4347 -- If no conversion apply scalar range checks and length checks
4348 -- base on the subtype of the actual (NOT that of the formal).
4350 else
4355 then
4357 else
4362 -- Note: we do not apply the predicate checks for the case of
4363 -- OUT and IN OUT parameters. They are instead applied in the
4364 -- Expand_Actuals routine in Exp_Ch6.
4367 -- An actual associated with an access parameter is implicitly
4368 -- converted to the anonymous access type of the formal and must
4369 -- satisfy the legality checks for access conversions.
4373 Error_Msg_N
4377 -- If the actual is an access selected component of a variable,
4378 -- the call may modify its designated object. It is reasonable
4379 -- to treat this as a potential modification of the enclosing
4380 -- record, to prevent spurious warnings that it should be
4381 -- declared as a constant, because intuitively programmers
4382 -- regard the designated subcomponent as part of the record.
4387 then
4392 -- Check bad case of atomic/volatile argument (RM C.6(12))
4396 then
4399 then
4400 Error_Msg_NE
4406 then
4407 Error_Msg_NE
4413 -- Check that subprograms don't have improper controlling
4414 -- arguments (RM 3.9.2 (9)).
4416 -- A primitive operation may have an access parameter of an
4417 -- incomplete tagged type, but a dispatching call is illegal
4418 -- if the type is still incomplete.
4424 declare
4426 begin
4430 then
4431 Error_Msg_NE
4442 -- Apply legality rule 3.9.2 (9/1)
4447 and then not In_Instance
4448 then
4453 Error_Msg_NE
4457 -- Apply the checks described in 3.10.2(27): if the context is a
4458 -- specific access-to-object, the actual cannot be class-wide.
4459 -- Use base type to exclude access_to_subprogram cases.
4466 and then
4471 -- Disable these checks for call to imported C++ subprograms
4473 and then not
4477 then
4478 Error_Msg_N
4483 Error_Msg_NE
4488 Check_Aliased_Parameter;
4492 -- If it is a named association, treat the selector_name as a
4493 -- proper identifier, and mark the corresponding entity.
4497 -- Ignore reference in SPARK mode, as it refers to an entity not
4498 -- in scope at the point of reference, so the reference should
4499 -- be ignored for computing effects of subprograms.
4501 and then not GNATprove_Mode
4502 then
4503 -- If subprogram is overridden, use name of formal that
4504 -- is being called.
4510 else
4524 -- The following checks are only relevant when SPARK_Mode is on as
4525 -- they are not standard Ada legality rule. Internally generated
4526 -- temporaries are ignored.
4530 -- An effectively volatile object may act as an actual when the
4531 -- corresponding formal is of a non-scalar effectively volatile
4532 -- type (SPARK RM 7.1.3(11)).
4536 then
4539 -- An effectively volatile object may act as an actual in a
4540 -- call to an instance of Unchecked_Conversion.
4541 -- (SPARK RM 7.1.3(11)).
4546 -- The actual denotes an object
4549 Error_Msg_N
4553 -- Otherwise the actual denotes an expression. Inspect the
4554 -- expression and flag each effectively volatile object with
4555 -- enabled property Async_Writers or Effective_Reads as illegal
4556 -- because it apprears within an interfering context. Note that
4557 -- this is usually done in Resolve_Entity_Name, but when the
4558 -- effectively volatile object appears as an actual in a call,
4559 -- the call must be resolved first.
4561 else
4565 -- Detect an external variable with an enabled property that
4566 -- does not match the mode of the corresponding formal in a
4567 -- procedure call. Functions are not considered because they
4568 -- cannot have effectively volatile formal parameters in the
4569 -- first place.
4576 then
4589 -- A formal parameter of a specific tagged type whose related
4590 -- subprogram is subject to pragma Extensions_Visible with value
4591 -- "False" cannot act as an actual in a subprogram with value
4592 -- "True" (SPARK RM 6.1.7(3)).
4596 Extensions_Visible_True
4597 then
4598 Error_Msg_N
4601 Error_Msg_NE
4605 -- The actual parameter of a Ghost subprogram whose formal is of
4606 -- mode IN OUT or OUT must be a Ghost variable (SPARK RM 6.9(12)).
4614 then
4615 Error_Msg_NE
4621 else
4628 -- Case where actual is not present
4630 else
4631 Insert_Default;
4642 -----------------------
4643 -- Resolve_Allocator --
4644 -----------------------
4656 procedure Check_Allocator_Discrim_Accessibility
4659 -- Check that accessibility level associated with an access discriminant
4660 -- initialized in an allocator by the expression Disc_Exp is not deeper
4661 -- than the level of the allocator type Alloc_Typ. An error message is
4662 -- issued if this condition is violated. Specialized checks are done for
4663 -- the cases of a constraint expression which is an access attribute or
4664 -- an access discriminant.
4667 -- If the allocator is an actual in a call, it is allowed to be class-
4668 -- wide when the context is not because it is a controlling actual.
4670 -------------------------------------------
4671 -- Check_Allocator_Discrim_Accessibility --
4672 -------------------------------------------
4674 procedure Check_Allocator_Discrim_Accessibility
4677 is
4678 begin
4681 then
4682 Error_Msg_N
4685 -- When the expression is an Access attribute the level of the prefix
4686 -- object must not be deeper than that of the allocator's type.
4690 Attribute_Access
4693 then
4694 Error_Msg_N
4696 Disc_Exp);
4698 -- When the expression is an access discriminant the check is against
4699 -- the level of the prefix object.
4705 then
4706 Error_Msg_N
4708 Disc_Exp);
4710 -- All other cases are legal
4712 else
4717 ----------------------------
4718 -- In_Dispatching_Context --
4719 ----------------------------
4724 begin
4730 -- Start of processing for Resolve_Allocator
4732 begin
4733 -- Replace general access with specific type
4743 -- For qualified expression, resolve the expression using the given
4744 -- subtype (nothing to do for type mark, subtype indication)
4749 and then not In_Dispatching_Context
4750 then
4751 Error_Msg_N
4759 -- Allocators generated by the build-in-place expansion mechanism
4760 -- are explicitly marked as coming from source but do not need to be
4761 -- checked for limited initialization. To exclude this case, ensure
4762 -- that the parent of the allocator is a source node.
4767 and then not In_Instance_Body
4768 then
4771 Error_Msg_N
4774 else
4775 Error_Msg_N
4783 -- A qualified expression requires an exact match of the type. Class-
4784 -- wide matching is not allowed.
4789 then
4793 -- Calls to build-in-place functions are not currently supported in
4794 -- allocators for access types associated with a simple storage pool.
4795 -- Supporting such allocators may require passing additional implicit
4796 -- parameters to build-in-place functions (or a significant revision
4797 -- of the current b-i-p implementation to unify the handling for
4798 -- multiple kinds of storage pools). ???
4802 then
4803 declare
4806 begin
4808 and then
4809 Present (Get_Rep_Pragma
4811 then
4812 Error_Msg_N
4819 -- A special accessibility check is needed for allocators that
4820 -- constrain access discriminants. The level of the type of the
4821 -- expression used to constrain an access discriminant cannot be
4822 -- deeper than the type of the allocator (in contrast to access
4823 -- parameters, where the level of the actual can be arbitrary).
4825 -- We can't use Valid_Conversion to perform this check because in
4826 -- general the type of the allocator is unrelated to the type of
4827 -- the access discriminant.
4831 then
4838 then
4841 -- Get the first component expression of the aggregate
4851 else
4855 else
4874 else
4879 else
4888 -- For a subtype mark or subtype indication, freeze the subtype
4890 else
4894 Error_Msg_N
4898 -- A special accessibility check is needed for allocators that
4899 -- constrain access discriminants. The level of the type of the
4900 -- expression used to constrain an access discriminant cannot be
4901 -- deeper than the type of the allocator (in contrast to access
4902 -- parameters, where the level of the actual can be arbitrary).
4903 -- We can't use Valid_Conversion to perform this check because
4904 -- in general the type of the allocator is unrelated to the type
4905 -- of the access discriminant.
4910 then
4920 else
4934 -- Ada 2005 (AI-344): A class-wide allocator requires an accessibility
4935 -- check that the level of the type of the created object is not deeper
4936 -- than the level of the allocator's access type, since extensions can
4937 -- now occur at deeper levels than their ancestor types. This is a
4938 -- static accessibility level check; a run-time check is also needed in
4939 -- the case of an initialized allocator with a class-wide argument (see
4940 -- Expand_Allocator_Expression).
4944 then
4945 declare
4948 begin
4953 else
4959 then
4962 Error_Msg_N
4971 -- Do not apply Ada 2005 accessibility checks on a class-wide
4972 -- allocator if the type given in the allocator is a formal
4973 -- type. A run-time check will be performed in the instance.
4983 -- Check for allocation from an empty storage pool
4988 -- If the context is an unchecked conversion, as may happen within an
4989 -- inlined subprogram, the allocator is being resolved with its own
4990 -- anonymous type. In that case, if the target type has a specific
4991 -- storage pool, it must be inherited explicitly by the allocator type.
4995 then
4996 Set_Associated_Storage_Pool
5004 -- Check that an allocator with task parts isn't for a nested access
5005 -- type when restriction No_Task_Hierarchy applies.
5009 then
5013 -- An illegal allocator may be rewritten as a raise Program_Error
5014 -- statement.
5018 -- An anonymous access discriminant is the definition of a
5019 -- coextension.
5023 N_Discriminant_Specification
5024 then
5025 declare
5029 begin
5032 -- Ada 2012 AI05-0052: If the designated type of the allocator
5033 -- is limited, then the allocator shall not be used to define
5034 -- the value of an access discriminant unless the discriminated
5035 -- type is immutably limited.
5040 then
5041 Error_Msg_N
5047 -- Avoid marking an allocator as a dynamic coextension if it is
5048 -- within a static construct.
5054 -- Cleanup for potential static coextensions
5056 else
5062 -- Report a simple error: if the designated object is a local task,
5063 -- its body has not been seen yet, and its activation will fail an
5064 -- elaboration check.
5071 then
5077 -- Ada 2012 (AI05-0111-3): Detect an attempt to allocate a task or a
5078 -- type with a task component on a subpool. This action must raise
5079 -- Program_Error at runtime.
5085 then
5097 ---------------------------
5098 -- Resolve_Arithmetic_Op --
5099 ---------------------------
5101 -- Used for resolving all arithmetic operators except exponentiation
5112 -- We do the resolution using the base type, because intermediate values
5113 -- in expressions always are of the base type, not a subtype of it.
5116 -- Returns True if N is in a context that expects "any real type"
5119 -- Return True iff given type is Integer or universal real/integer
5122 -- Choose type of integer literal in fixed-point operation to conform
5123 -- to available fixed-point type. T is the type of the other operand,
5124 -- which is needed to determine the expected type of N.
5127 -- Set operand type to T if universal
5129 -------------------------------
5130 -- Expected_Type_Is_Any_Real --
5131 -------------------------------
5134 begin
5135 -- N is the expression after "delta" in a fixed_point_definition;
5136 -- see RM-3.5.9(6):
5139 N_Decimal_Fixed_Point_Definition,
5141 -- N is one of the bounds in a real_range_specification;
5142 -- see RM-3.5.7(5):
5144 N_Real_Range_Specification,
5146 -- N is the expression of a delta_constraint;
5147 -- see RM-J.3(3):
5149 N_Delta_Constraint);
5152 -----------------------------
5153 -- Is_Integer_Or_Universal --
5154 -----------------------------
5161 begin
5167 else
5173 then
5184 ----------------------------
5185 -- Set_Mixed_Mode_Operand --
5186 ----------------------------
5192 begin
5195 -- A universal integer literal is resolved as standard integer
5196 -- except in the case of a fixed-point result, where we leave it
5197 -- as universal (to be handled by Exp_Fixd later on)
5201 else
5209 then
5210 -- A universal real can appear in a fixed-type context. We resolve
5211 -- the literal with that context, even though this might raise an
5212 -- exception prematurely (the other operand may be zero).
5219 then
5220 -- Integer arg in mixed-mode operation. Resolve with universal
5221 -- type, in case preference rule must be applied.
5227 then
5228 -- Not a mixed-mode operation, resolve with context
5234 -- N may itself be a mixed-mode operation, so use context type
5241 then
5242 -- Must be (fixed * fixed) operation, operand must have one
5243 -- compatible interpretation.
5250 then
5251 -- C * F(X) in a fixed context, where C is a real literal or a
5252 -- fixed-point expression. F must have either a fixed type
5253 -- interpretation or an integer interpretation, but not both.
5260 else
5267 else
5275 -- Reanalyze the literal with the fixed type of the context. If
5276 -- context is Universal_Fixed, we are within a conversion, leave
5277 -- the literal as a universal real because there is no usable
5278 -- fixed type, and the target of the conversion plays no role in
5279 -- the resolution.
5281 declare
5285 begin
5288 else
5294 then
5296 else
5304 else
5309 ----------------------
5310 -- Set_Operand_Type --
5311 ----------------------
5314 begin
5317 then
5322 -- Start of processing for Resolve_Arithmetic_Op
5324 begin
5329 then
5333 -- Special-case for mixed-mode universal expressions or fixed point type
5334 -- operation: each argument is resolved separately. The same treatment
5335 -- is required if one of the operands of a fixed point operation is
5336 -- universal real, since in this case we don't do a conversion to a
5337 -- specific fixed-point type (instead the expander handles the case).
5339 -- Set the type of the node to its universal interpretation because
5340 -- legality checks on an exponentiation operand need the context.
5345 then
5356 or else
5359 then
5364 -- If context is a fixed type and one operand is integer, the other
5365 -- is resolved with the type of the context.
5370 then
5377 then
5381 else
5386 -- Check the rule in RM05-4.5.5(19.1/2) disallowing universal_fixed
5387 -- multiplying operators from being used when the expected type is
5388 -- also universal_fixed. Note that B_Typ will be Universal_Fixed in
5389 -- some cases where the expected type is actually Any_Real;
5390 -- Expected_Type_Is_Any_Real takes care of that case.
5394 then
5398 N_Unchecked_Type_Conversion)
5399 then
5406 else
5409 and then not
5411 N_Unchecked_Type_Conversion)
5412 then
5413 Error_Msg_N
5418 -- The expected type is "any real type" in contexts like
5420 -- type T is delta <universal_fixed-expression> ...
5422 -- in which case we need to set the type to Universal_Real
5423 -- so that static expression evaluation will work properly.
5427 else
5437 then
5442 -- If no previous errors, this is only possible if one operand is
5443 -- overloaded and the context is universal. Resolve as such.
5448 else
5450 and then
5452 then
5456 -- If the context is Universal_Fixed and the operands are also
5457 -- universal fixed, this is an error, unless there is only one
5458 -- applicable fixed_point type (usually Duration).
5466 else
5471 else
5476 -- If one of the arguments was resolved to a non-universal type.
5477 -- label the result of the operation itself with the same type.
5478 -- Do the same for the universal argument, if any.
5490 -- In SPARK, a multiplication or division with operands of fixed point
5491 -- types must be qualified or explicitly converted to identify the
5492 -- result type.
5497 and then
5499 then
5500 Check_SPARK_05_Restriction
5504 -- Set overflow and division checking bit
5511 -- Give warning if explicit division by zero
5515 then
5521 or else
5524 then
5525 -- Specialize the warning message according to the operation.
5526 -- When SPARK_Mode is On, force a warning instead of an error
5527 -- in that case, as this likely corresponds to deactivated
5528 -- code. The following warnings are for the case
5533 -- For division, we have two cases, for float division
5534 -- of an unconstrained float type, on a machine where
5535 -- Machine_Overflows is false, we don't get an exception
5536 -- at run-time, but rather an infinity or Nan. The Nan
5537 -- case is pretty obscure, so just warn about infinities.
5541 and then not Machine_Overflows_On_Target
5542 then
5543 Error_Msg_N
5547 -- For all other cases, we get a Constraint_Error
5549 else
5550 Apply_Compile_Time_Constraint_Error
5557 Apply_Compile_Time_Constraint_Error
5563 Apply_Compile_Time_Constraint_Error
5568 -- Division by zero can only happen with division, rem,
5569 -- and mod operations.
5575 -- In GNATprove mode, we enable the division check so that
5576 -- GNATprove will issue a message if it cannot be proved.
5582 -- Otherwise just set the flag to check at run time
5584 else
5589 -- If Restriction No_Implicit_Conditionals is active, then it is
5590 -- violated if either operand can be negative for mod, or for rem
5591 -- if both operands can be negative.
5595 then
5596 declare
5603 -- Set if corresponding operand might be negative
5605 begin
5606 Determine_Range
5610 Determine_Range
5614 -- Check if we will be generating conditionals. There are two
5615 -- cases where that can happen, first for REM, the only case
5616 -- is largest negative integer mod -1, where the division can
5617 -- overflow, but we still have to give the right result. The
5618 -- front end generates a test for this annoying case. Here we
5619 -- just test if both operands can be negative (that's what the
5620 -- expander does, so we match its logic here).
5622 -- The second case is mod where either operand can be negative.
5623 -- In this case, the back end has to generate additional tests.
5626 or else
5628 then
5639 ------------------
5640 -- Resolve_Call --
5641 ------------------
5644 function Same_Or_Aliased_Subprograms
5647 -- Returns True if the subprogram entity S is the same as E or else
5648 -- S is an alias of E.
5650 ---------------------------------
5651 -- Same_Or_Aliased_Subprograms --
5652 ---------------------------------
5654 function Same_Or_Aliased_Subprograms
5657 is
5659 begin
5663 -- Local variables
5677 -- Start of processing for Resolve_Call
5679 begin
5680 -- The context imposes a unique interpretation with type Typ on a
5681 -- procedure or function call. Find the entity of the subprogram that
5682 -- yields the expected type, and propagate the corresponding formal
5683 -- constraints on the actuals. The caller has established that an
5684 -- interpretation exists, and emitted an error if not unique.
5686 -- First deal with the case of a call to an access-to-subprogram,
5687 -- dereference made explicit in Analyze_Call.
5693 else
5694 -- Find the interpretation whose type (a subprogram type) has a
5695 -- return type that is compatible with the context. Analysis of
5696 -- the node has established that one exists.
5715 -- If the prefix is not an entity, then resolve it
5721 -- For an indirect call, we always invalidate checks, since we do not
5722 -- know whether the subprogram is local or global. Yes we could do
5723 -- better here, e.g. by knowing that there are no local subprograms,
5724 -- but it does not seem worth the effort. Similarly, we kill all
5725 -- knowledge of current constant values.
5727 Kill_Current_Values;
5729 -- If this is a procedure call which is really an entry call, do
5730 -- the conversion of the procedure call to an entry call. Protected
5731 -- operations use the same circuitry because the name in the call
5732 -- can be an arbitrary expression with special resolution rules.
5737 then
5741 -- Kill checks and constant values, as above for indirect case
5742 -- Who knows what happens when another task is activated?
5744 Kill_Current_Values;
5747 -- Normal subprogram call with name established in Resolve
5753 -- Otherwise we must have the case of an overloaded call
5755 else
5758 -- Initialize Nam to prevent warning (we know it will be assigned
5759 -- in the loop below, but the compiler does not know that).
5779 then
5780 -- The prefix is a parameterless function call that returns an access
5781 -- to subprogram. If parameters are present in the current call, add
5782 -- add an explicit dereference. We use the base type here because
5783 -- within an instance these may be subtypes.
5785 -- The dereference is added either in Analyze_Call or here. Should
5786 -- be consolidated ???
5795 -- Check that a call to Current_Task does not occur in an entry body
5798 declare
5801 begin
5803 loop
5806 -- Exclude calls that occur within the default of a formal
5807 -- parameter of the entry, since those are evaluated outside
5808 -- of the body.
5815 then
5818 Error_Msg_NE
5832 -- Check that a procedure call does not occur in the context of the
5833 -- entry call statement of a conditional or timed entry call. Note that
5834 -- the case of a call to a subprogram renaming of an entry will also be
5835 -- rejected. The test for N not being an N_Entry_Call_Statement is
5836 -- defensive, covering the possibility that the processing of entry
5837 -- calls might reach this point due to later modifications of the code
5838 -- above.
5843 then
5847 -- Ada 2005 (AI-345): If a procedure_call_statement is used
5848 -- for a procedure_or_entry_call, the procedure_name or
5849 -- procedure_prefix of the procedure_call_statement shall denote
5850 -- an entry renamed by a procedure, or (a view of) a primitive
5851 -- subprogram of a limited interface whose first parameter is
5852 -- a controlling parameter.
5857 then
5858 Error_Msg_N
5863 -- If the SPARK_05 restriction is active, we are not allowed
5864 -- to have a call to a subprogram before we see its completion.
5869 -- Don't flag strange internal calls
5874 -- Only flag calls in extended main source
5879 -- Exclude enumeration literals from this processing
5882 then
5883 Check_SPARK_05_Restriction
5887 -- Check that this is not a call to a protected procedure or entry from
5888 -- within a protected function.
5892 -- Freeze the subprogram name if not in a spec-expression. Note that
5893 -- we freeze procedure calls as well as function calls. Procedure calls
5894 -- are not frozen according to the rules (RM 13.14(14)) because it is
5895 -- impossible to have a procedure call to a non-frozen procedure in
5896 -- pure Ada, but in the code that we generate in the expander, this
5897 -- rule needs extending because we can generate procedure calls that
5898 -- need freezing.
5900 -- In Ada 2012, expression functions may be called within pre/post
5901 -- conditions of subsequent functions or expression functions. Such
5902 -- calls do not freeze when they appear within generated bodies,
5903 -- (including the body of another expression function) which would
5904 -- place the freeze node in the wrong scope. An expression function
5905 -- is frozen in the usual fashion, by the appearance of a real body,
5906 -- or at the end of a declarative part.
5909 and then not In_Spec_Expression
5911 and then
5914 then
5918 -- For a predefined operator, the type of the result is the type imposed
5919 -- by context, except for a predefined operation on universal fixed.
5920 -- Otherwise The type of the call is the type returned by the subprogram
5921 -- being called.
5928 -- If the subprogram returns an array type, and the context requires the
5929 -- component type of that array type, the node is really an indexing of
5930 -- the parameterless call. Resolve as such. A pathological case occurs
5931 -- when the type of the component is an access to the array type. In
5932 -- this case the call is truly ambiguous.
5935 and then
5938 or else
5941 and then
5943 then
5944 declare
5949 begin
5952 then
5953 Error_Msg_N
5955 else
5958 -- The called entity may be an explicit dereference, in which
5959 -- case there is no entity to set.
5967 or else
5969 and then
5971 then
5974 -- Indexed call to a parameterless function
5976 Index_Node :=
5978 Prefix =>
5981 else
5982 -- An Ada 2005 prefixed call to a primitive operation
5983 -- whose first parameter is the prefix. This prefix was
5984 -- prepended to the parameter list, which is actually a
5985 -- list of indexes. Remove the prefix in order to build
5986 -- the proper indexed component.
5988 Index_Node :=
5990 Prefix =>
5993 Parameter_Associations =>
5994 New_List
5999 -- Preserve the parenthesis count of the node
6003 -- Since we are correcting a node classification error made
6004 -- by the parser, we call Replace rather than Rewrite.
6018 else
6022 -- In the case where the call is to an overloaded subprogram, Analyze
6023 -- calls Normalize_Actuals once per overloaded subprogram. Therefore in
6024 -- such a case Normalize_Actuals needs to be called once more to order
6025 -- the actuals correctly. Otherwise the call will have the ordering
6026 -- given by the last overloaded subprogram whether this is the correct
6027 -- one being called or not.
6034 -- In any case, call is fully resolved now. Reset Overload flag, to
6035 -- prevent subsequent overload resolution if node is analyzed again
6040 -- A Ghost entity must appear in a specific context
6046 -- If we are calling the current subprogram from immediately within its
6047 -- body, then that is the case where we can sometimes detect cases of
6048 -- infinite recursion statically. Do not try this in case restriction
6049 -- No_Recursion is in effect anyway, and do it only for source calls.
6054 -- Check violation of SPARK_05 restriction which does not permit
6055 -- a subprogram body to contain a call to the subprogram directly.
6059 then
6060 Check_SPARK_05_Restriction
6064 -- Issue warning for possible infinite recursion in the absence
6065 -- of the No_Recursion restriction.
6070 then
6071 -- Here we detected and flagged an infinite recursion, so we do
6072 -- not need to test the case below for further warnings. Also we
6073 -- are all done if we now have a raise SE node.
6079 -- If call is to immediately containing subprogram, then check for
6080 -- the case of a possible run-time detectable infinite recursion.
6082 else
6086 -- Although in general case, recursion is not statically
6087 -- checkable, the case of calling an immediately containing
6088 -- subprogram is easy to catch.
6092 -- If the recursive call is to a parameterless subprogram,
6093 -- then even if we can't statically detect infinite
6094 -- recursion, this is pretty suspicious, and we output a
6095 -- warning. Furthermore, we will try later to detect some
6096 -- cases here at run time by expanding checking code (see
6097 -- Detect_Infinite_Recursion in package Exp_Ch6).
6099 -- If the recursive call is within a handler, do not emit a
6100 -- warning, because this is a common idiom: loop until input
6101 -- is correct, catch illegal input in handler and restart.
6107 then
6108 -- For the case of a procedure call. We give the message
6109 -- only if the call is the first statement in a sequence
6110 -- of statements, or if all previous statements are
6111 -- simple assignments. This is simply a heuristic to
6112 -- decrease false positives, without losing too many good
6113 -- warnings. The idea is that these previous statements
6114 -- may affect global variables the procedure depends on.
6115 -- We also exclude raise statements, that may arise from
6116 -- constraint checks and are probably unrelated to the
6117 -- intended control flow.
6121 then
6122 declare
6124 begin
6128 N_Raise_Constraint_Error)
6129 then
6138 -- Do not give warning if we are in a conditional context
6140 declare
6142 begin
6148 then
6153 -- Here warning is to be issued
6169 -- Check obsolescent reference to Ada.Characters.Handling subprogram
6173 -- If subprogram name is a predefined operator, it was given in
6174 -- functional notation. Replace call node with operator node, so
6175 -- that actuals can be resolved appropriately.
6183 then
6189 -- Create a transient scope if the resulting type requires it
6191 -- There are several notable exceptions:
6193 -- a) In init procs, the transient scope overhead is not needed, and is
6194 -- even incorrect when the call is a nested initialization call for a
6195 -- component whose expansion may generate adjust calls. However, if the
6196 -- call is some other procedure call within an initialization procedure
6197 -- (for example a call to Create_Task in the init_proc of the task
6198 -- run-time record) a transient scope must be created around this call.
6200 -- b) Enumeration literal pseudo-calls need no transient scope
6202 -- c) Intrinsic subprograms (Unchecked_Conversion and source info
6203 -- functions) do not use the secondary stack even though the return
6204 -- type may be unconstrained.
6206 -- d) Calls to a build-in-place function, since such functions may
6207 -- allocate their result directly in a target object, and cases where
6208 -- the result does get allocated in the secondary stack are checked for
6209 -- within the specialized Exp_Ch6 procedures for expanding those
6210 -- build-in-place calls.
6212 -- e) If the subprogram is marked Inline_Always, then even if it returns
6213 -- an unconstrained type the call does not require use of the secondary
6214 -- stack. However, inlining will only take place if the body to inline
6215 -- is already present. It may not be available if e.g. the subprogram is
6216 -- declared in a child instance.
6218 -- If this is an initialization call for a type whose construction
6219 -- uses the secondary stack, and it is not a nested call to initialize
6220 -- a component, we do need to create a transient scope for it. We
6221 -- check for this by traversing the type in Check_Initialization_Call.
6227 then
6233 then
6236 elsif Expander_Active
6239 and then
6241 or else
6243 then
6246 -- If the call appears within the bounds of a loop, it will
6247 -- be rewritten and reanalyzed, nothing left to do here.
6254 and then not Within_Init_Proc
6255 then
6259 -- A protected function cannot be called within the definition of the
6260 -- enclosing protected type, unless it is part of a pre/postcondition
6261 -- on another protected operation.
6266 and then not In_Spec_Expression
6267 then
6268 Error_Msg_NE
6272 -- Propagate interpretation to actuals, and add default expressions
6273 -- where needed.
6278 -- Overloaded literals are rewritten as function calls, for purpose of
6279 -- resolution. After resolution, we can replace the call with the
6280 -- literal itself.
6286 -- Avoid validation, since it is a static function call
6292 -- If the subprogram is not global, then kill all saved values and
6293 -- checks. This is a bit conservative, since in many cases we could do
6294 -- better, but it is not worth the effort. Similarly, we kill constant
6295 -- values. However we do not need to do this for internal entities
6296 -- (unless they are inherited user-defined subprograms), since they
6297 -- are not in the business of molesting local values.
6299 -- If the flag Suppress_Value_Tracking_On_Calls is set, then we also
6300 -- kill all checks and values for calls to global subprograms. This
6301 -- takes care of the case where an access to a local subprogram is
6302 -- taken, and could be passed directly or indirectly and then called
6303 -- from almost any context.
6305 -- Note: we do not do this step till after resolving the actuals. That
6306 -- way we still take advantage of the current value information while
6307 -- scanning the actuals.
6309 -- We suppress killing values if we are processing the nodes associated
6310 -- with N_Freeze_Entity nodes. Otherwise the declaration of a tagged
6311 -- type kills all the values as part of analyzing the code that
6312 -- initializes the dispatch tables.
6316 or else Suppress_Value_Tracking_On_Call
6321 then
6322 Kill_Current_Values;
6325 -- If we are warning about unread OUT parameters, this is the place to
6326 -- set Last_Assignment for OUT and IN OUT parameters. We have to do this
6327 -- after the above call to Kill_Current_Values (since that call clears
6328 -- the Last_Assignment field of all local variables).
6333 then
6334 declare
6338 begin
6348 then
6358 -- If the subprogram is a primitive operation, check whether or not
6359 -- it is a correct dispatching call.
6363 then
6368 and then not In_Instance
6369 then
6373 -- If this is a dispatching call, generate the appropriate reference,
6374 -- for better source navigation in GPS.
6378 then
6381 -- Normal case, not a dispatching call: generate a call reference
6383 else
6391 -- Check for violation of restriction No_Specific_Termination_Handlers
6392 -- and warn on a potentially blocking call to Abort_Task.
6396 or else
6398 then
6405 -- A call to Ada.Real_Time.Timing_Events.Set_Handler to set a relative
6406 -- timing event violates restriction No_Relative_Delay (AI-0211). We
6407 -- need to check the second argument to determine whether it is an
6408 -- absolute or relative timing event.
6413 then
6417 -- Issue an error for a call to an eliminated subprogram. This routine
6418 -- will not perform the check if the call appears within a default
6419 -- expression.
6423 -- In formal mode, the primitive operations of a tagged type or type
6424 -- extension do not include functions that return the tagged type.
6430 then
6434 -- Implement rule in 12.5.1 (23.3/2): In an instance, if the actual is
6435 -- class-wide and the call dispatches on result in a context that does
6436 -- not provide a tag, the call raises Program_Error.
6439 and then In_Instance
6444 then
6445 -- Verify that none of the formals are controlling
6447 declare
6451 begin
6473 -- Check for calling a function with OUT or IN OUT parameter when the
6474 -- calling context (us right now) is not Ada 2012, so does not allow
6475 -- OUT or IN OUT parameters in function calls. Functions declared in
6476 -- a predefined unit are OK, as they may be called indirectly from a
6477 -- user-declared instantiation.
6483 then
6488 -- Check the dimensions of the actuals in the call. For function calls,
6489 -- propagate the dimensions from the returned type to N.
6493 -- All done, evaluate call and deal with elaboration issues
6498 -- In GNATprove mode, expansion is disabled, but we want to inline some
6499 -- subprograms to facilitate formal verification. Indirect calls through
6500 -- a subprogram type or within a generic cannot be inlined. Inlining is
6501 -- performed only for calls subject to SPARK_Mode on.
6503 if GNATprove_Mode
6506 and then not Inside_A_Generic
6507 then
6514 -- Nothing to do if the subprogram is not eligible for inlining in
6515 -- GNATprove mode.
6519 then
6522 -- Calls cannot be inlined inside assertions, as GNATprove treats
6523 -- assertions as logic expressions.
6526 Cannot_Inline
6529 -- Calls cannot be inlined inside default expressions
6532 Cannot_Inline
6535 -- Inlining should not be performed during pre-analysis
6539 -- With the one-pass inlining technique, a call cannot be
6540 -- inlined if the corresponding body has not been seen yet.
6543 Cannot_Inline
6546 -- Nothing to do if there is no body to inline, indicating that
6547 -- the subprogram is not suitable for inlining in GNATprove
6548 -- mode.
6553 -- Do not inline calls inside expression functions, as this
6554 -- would prevent interpreting them as logical formulas in
6555 -- GNATprove.
6558 and then
6560 then
6561 Cannot_Inline
6565 -- Calls cannot be inlined inside potentially unevaluated
6566 -- expressions, as this would create complex actions inside
6567 -- expressions, that are not handled by GNATprove.
6570 Cannot_Inline
6574 -- Otherwise, inline the call
6576 else
6586 -----------------------------
6587 -- Resolve_Case_Expression --
6588 -----------------------------
6596 begin
6603 -- When the expression is of a scalar subtype different from the
6604 -- result subtype, then insert a conversion to ensure the generation
6605 -- of a constraint check.
6615 -- Apply RM 4.5.7 (17/3): whether the expression is statically or
6616 -- dynamically tagged must be known statically.
6624 Error_Msg_N
6637 -------------------------------
6638 -- Resolve_Character_Literal --
6639 -------------------------------
6645 begin
6646 -- Verify that the character does belong to the type of the context
6651 -- Wide_Wide_Character literals must always be defined, since the set
6652 -- of wide wide character literals is complete, i.e. if a character
6653 -- literal is accepted by the parser, then it is OK for wide wide
6654 -- character (out of range character literals are rejected).
6659 -- Always accept character literal for type Any_Character, which
6660 -- occurs in error situations and in comparisons of literals, both
6661 -- of which should accept all literals.
6666 -- For Standard.Character or a type derived from it, check that the
6667 -- literal is in range.
6674 -- For Standard.Wide_Character or a type derived from it, check that the
6675 -- literal is in range.
6682 -- For Standard.Wide_Wide_Character or a type derived from it, we
6683 -- know the literal is in range, since the parser checked.
6688 -- If the entity is already set, this has already been resolved in a
6689 -- generic context, or comes from expansion. Nothing else to do.
6694 -- Otherwise we have a user defined character type, and we can use the
6695 -- standard visibility mechanisms to locate the referenced entity.
6697 else
6710 -- If we fall through, then the literal does not match any of the
6711 -- entries of the enumeration type. This isn't just a constraint error
6712 -- situation, it is an illegality (see RM 4.2).
6714 Error_Msg_NE
6718 ---------------------------
6719 -- Resolve_Comparison_Op --
6720 ---------------------------
6722 -- Context requires a boolean type, and plays no role in resolution.
6723 -- Processing identical to that for equality operators. The result type is
6724 -- the base type, which matters when pathological subtypes of booleans with
6725 -- limited ranges are used.
6732 begin
6733 -- If this is an intrinsic operation which is not predefined, use the
6734 -- types of its declared arguments to resolve the possibly overloaded
6735 -- operands. Otherwise the operands are unambiguous and specify the
6736 -- expected type.
6741 else
6752 -- Skip remaining processing if already set to Any_Type
6758 -- Deal with other error cases
6762 T = Any_Character
6763 then
6766 else
6774 -- Resolve the operands if types OK
6783 -- In SPARK, ordering operators <, <=, >, >= are not defined for Boolean
6784 -- types or array types except String.
6787 Check_SPARK_05_Restriction
6792 then
6793 Check_SPARK_05_Restriction
6797 -- Check comparison on unordered enumeration
6801 Error_Msg_NE
6806 -- Evaluate the relation (note we do this after the above check since
6807 -- this Eval call may change N to True/False.
6813 -----------------------------------------
6814 -- Resolve_Discrete_Subtype_Indication --
6815 -----------------------------------------
6817 procedure Resolve_Discrete_Subtype_Indication
6820 is
6824 begin
6832 else
6842 Error_Msg_NE
6845 -- Rewrite the constraint as a range of Typ
6846 -- to allow compilation to proceed further.
6858 else
6862 -- Additionally, we must check that the bounds are compatible
6863 -- with the given subtype, which might be different from the
6864 -- type of the context.
6868 -- ??? If the above check statically detects a Constraint_Error
6869 -- it replaces the offending bound(s) of the range R with a
6870 -- Constraint_Error node. When the itype which uses these bounds
6871 -- is frozen the resulting call to Duplicate_Subexpr generates
6872 -- a new temporary for the bounds.
6874 -- Unfortunately there are other itypes that are also made depend
6875 -- on these bounds, so when Duplicate_Subexpr is called they get
6876 -- a forward reference to the newly created temporaries and Gigi
6877 -- aborts on such forward references. This is probably sign of a
6878 -- more fundamental problem somewhere else in either the order of
6879 -- itype freezing or the way certain itypes are constructed.
6881 -- To get around this problem we call Remove_Side_Effects right
6882 -- away if either bounds of R are a Constraint_Error.
6884 declare
6888 begin
6904 -------------------------
6905 -- Resolve_Entity_Name --
6906 -------------------------
6908 -- Used to resolve identifiers and expanded names
6911 function Is_Assignment_Or_Object_Expression
6914 -- Determine whether node Context denotes an assignment statement or an
6915 -- object declaration whose expression is node Expr.
6917 ----------------------------------------
6918 -- Is_Assignment_Or_Object_Expression --
6919 ----------------------------------------
6921 function Is_Assignment_Or_Object_Expression
6924 is
6925 begin
6927 N_Object_Declaration)
6929 then
6932 -- Check whether a construct that yields a name is the expression of
6933 -- an assignment statement or an object declaration.
6936 N_Explicit_Dereference,
6937 N_Indexed_Component,
6938 N_Selected_Component,
6939 N_Slice)
6941 or else
6943 N_Unchecked_Type_Conversion)
6945 then
6946 return
6947 Is_Assignment_Or_Object_Expression
6951 -- Otherwise the context is not an assignment statement or an object
6952 -- declaration.
6954 else
6959 -- Local variables
6964 -- Start of processing for Resolve_Entity_Name
6966 begin
6967 -- If garbage from errors, set to Any_Type and return
6974 -- Replace named numbers by corresponding literals. Note that this is
6975 -- the one case where Resolve_Entity_Name must reset the Etype, since
6976 -- it is currently marked as universal.
6986 -- For enumeration literals, we need to make sure that a proper style
6987 -- check is done, since such literals are overloaded, and thus we did
6988 -- not do a style check during the first phase of analysis.
6994 -- Case of (sub)type name appearing in a context where an expression
6995 -- is expected. This is legal if occurrence is a current instance.
6996 -- See RM 8.6 (17/3).
7002 -- Any other use is an error
7004 else
7005 Error_Msg_N
7009 -- Check discriminant use if entity is discriminant in current scope,
7010 -- i.e. discriminant of record or concurrent type currently being
7011 -- analyzed. Uses in corresponding body are unrestricted.
7016 then
7019 -- A parameterless generic function cannot appear in a context that
7020 -- requires resolution.
7025 -- In Ada 83 an OUT parameter cannot be read
7030 or else Is_Assignment_Or_Object_Expression
7033 then
7038 -- In all other cases, just do the possible static evaluation
7040 else
7041 -- A deferred constant that appears in an expression must have a
7042 -- completion, unless it has been removed by in-place expansion of
7043 -- an aggregate. A constant that is a renaming does not need
7044 -- initialization.
7050 and then not In_Spec_Expression
7053 then
7057 then
7059 else
7060 Error_Msg_N
7070 -- When the entity appears in a parameter association, retrieve the
7071 -- related subprogram call.
7079 -- The following checks are only relevant when SPARK_Mode is on as
7080 -- they are not standard Ada legality rules.
7084 -- An effectively volatile object subject to enabled properties
7085 -- Async_Writers or Effective_Reads must appear in non-interfering
7086 -- context (SPARK RM 7.1.3(12)).
7093 then
7094 SPARK_Msg_N
7099 -- Check for possible elaboration issues with respect to reads of
7100 -- variables. The act of renaming the variable is not considered a
7101 -- read as it simply establishes an alias.
7105 then
7109 -- The variable may eventually become a constituent of a single
7110 -- protected/task type. Record the reference now and verify its
7111 -- legality when analyzing the contract of the variable
7112 -- (SPARK RM 9.3).
7119 -- A Ghost entity must appear in a specific context
7127 -------------------
7128 -- Resolve_Entry --
7129 -------------------
7141 -- If the bounds of the entry family being called depend on task
7142 -- discriminants, build a new index subtype where a discriminant is
7143 -- replaced with the value of the discriminant of the target task.
7144 -- The target task is the prefix of the entry name in the call.
7146 -----------------------
7147 -- Actual_Index_Type --
7148 -----------------------
7158 -- If the bound is given by a discriminant, replace with a reference
7159 -- to the discriminant of the same name in the target task. If the
7160 -- entry name is the target of a requeue statement and the entry is
7161 -- in the current protected object, the bound to be used is the
7162 -- discriminal of the object (see Apply_Range_Checks for details of
7163 -- the transformation).
7165 -----------------------------
7166 -- Actual_Discriminant_Ref --
7167 -----------------------------
7173 begin
7178 then
7184 then
7185 -- Note: here Bound denotes a discriminant of the corresponding
7186 -- record type tskV, whose discriminal is a formal of the
7187 -- init-proc tskVIP. What we want is the body discriminal,
7188 -- which is associated to the discriminant of the original
7189 -- concurrent type tsk.
7191 return New_Occurrence_Of
7194 else
7195 Ref :=
7205 -- Start of processing for Actual_Index_Type
7207 begin
7210 then
7213 else
7227 -- Start of processing for Resolve_Entry
7229 begin
7230 -- Find name of entry being called, and resolve prefix of name with its
7231 -- own type. The prefix can be overloaded, and the name and signature of
7232 -- the entry must be taken into account.
7236 -- Case of dealing with entry family within the current tasks
7240 else
7246 -- Entry call to an entry (or entry family) in the current task. This
7247 -- is legal even though the task will deadlock. Rewrite as call to
7248 -- current task.
7250 -- This can also be a call to an entry in an enclosing task. If this
7251 -- is a single task, we have to retrieve its name, because the scope
7252 -- of the entry is the task type, not the object. If the enclosing
7253 -- task is a task type, the identity of the task is given by its own
7254 -- self variable.
7256 -- Finally this can be a requeue on an entry of the same task or
7257 -- protected object.
7264 then
7265 -- S is an enclosing task or protected object. The concurrent
7266 -- declaration has been converted into a type declaration, and
7267 -- the object itself has an object declaration that follows
7268 -- the type in the same declarative part.
7280 -- Call to current task. Will be transformed into call to Self
7287 New_N :=
7290 Selector_Name =>
7297 then
7298 -- Use the entry name (which must be unique at this point) to find
7299 -- the prefix that returns the corresponding task/protected type.
7301 declare
7307 begin
7329 -- Up to this point the expression could have been the actual in a
7330 -- simple entry call, and be given by a named association.
7334 else
7340 ------------------------
7341 -- Resolve_Entry_Call --
7342 ------------------------
7354 begin
7355 -- We kill all checks here, because it does not seem worth the effort to
7356 -- do anything better, an entry call is a big operation.
7358 Kill_All_Checks;
7360 -- Processing of the name is similar for entry calls and protected
7361 -- operation calls. Once the entity is determined, we can complete
7362 -- the resolution of the actuals.
7364 -- The selector may be overloaded, in the case of a protected object
7365 -- with overloaded functions. The type of the context is used for
7366 -- resolution.
7371 then
7372 declare
7376 begin
7395 -- Simple entry call
7403 -- Call to member of entry family
7410 -- We cannot in general check the maximum depth of protected entry calls
7411 -- at compile time. But we can tell that any protected entry call at all
7412 -- violates a specified nesting depth of zero.
7418 -- Use context type to disambiguate a protected function that can be
7419 -- called without actuals and that returns an array type, and where the
7420 -- argument list may be an indexing of the returned value.
7425 and then
7431 and then
7432 Covers
7435 then
7436 declare
7439 begin
7440 Index_Node :=
7442 Prefix =>
7446 -- Since we are correcting a node classification error made by the
7447 -- parser, we call Replace rather than Rewrite.
7460 then
7462 -- Note the entity being called before rewriting the call, so that
7463 -- it appears used at this point.
7467 -- Rewrite as call to the precondition wrapper, adding the task
7468 -- object to the list of actuals. If the call is to a member of an
7469 -- entry family, include the index as well.
7471 declare
7475 begin
7484 New_Call :=
7486 Name =>
7491 -- Preanalyze and resolve new call. Current procedure is called
7492 -- from Resolve_Call, after which expansion will take place.
7499 -- The operation name may have been overloaded. Order the actuals
7500 -- according to the formals of the resolved entity, and set the return
7501 -- type to that of the operation.
7508 -- Reset the Is_Overloaded flag, since resolution is now completed
7510 -- Simple entry call
7515 -- Call to a member of an entry family
7525 -- Create a call reference to the entry
7533 -- Verify that a procedure call cannot masquerade as an entry
7534 -- call where an entry call is expected.
7539 then
7544 then
7549 then
7554 -- After resolution, entry calls and protected procedure calls are
7555 -- changed into entry calls, for expansion. The structure of the node
7556 -- does not change, so it can safely be done in place. Protected
7557 -- function calls must keep their structure because they are
7558 -- subexpressions.
7562 -- A protected operation that is not a function may modify the
7563 -- corresponding object, and cannot apply to a constant. If this
7564 -- is an internal call, the prefix is the type itself.
7570 then
7571 Error_Msg_N
7573 Entry_Name);
7587 -- Protected functions can return on the secondary stack, in which
7588 -- case we must trigger the transient scope mechanism.
7590 elsif Expander_Active
7592 then
7597 -------------------------
7598 -- Resolve_Equality_Op --
7599 -------------------------
7601 -- Both arguments must have the same type, and the boolean context does
7602 -- not participate in the resolution. The first pass verifies that the
7603 -- interpretation is not ambiguous, and the type of the left argument is
7604 -- correctly set, or is Any_Type in case of ambiguity. If both arguments
7605 -- are strings or aggregates, allocators, or Null, they are ambiguous even
7606 -- though they carry a single (universal) type. Diagnose this case here.
7614 -- The resolution rule for if expressions requires that each such must
7615 -- have a unique type. This means that if several dependent expressions
7616 -- are of a non-null anonymous access type, and the context does not
7617 -- impose an expected type (as can be the case in an equality operation)
7618 -- the expression must be rejected.
7621 -- Attempt to explain the nature of a redundant comparison with True. If
7622 -- the expression N is too complex, this routine issues a general error
7623 -- message.
7626 -- In the case of allocators and access attributes, the context must
7627 -- provide an indication of the specific access type to be used. If
7628 -- one operand is of such a "generic" access type, check whether there
7629 -- is a specific visible access type that has the same designated type.
7630 -- This is semantically dubious, and of no interest to any real code,
7631 -- but c48008a makes it all worthwhile.
7633 -------------------------
7634 -- Check_If_Expression --
7635 -------------------------
7641 begin
7648 then
7654 ------------------------
7655 -- Explain_Redundancy --
7656 ------------------------
7663 begin
7666 -- Strip the operand down to an entity
7668 loop
7671 else
7676 -- The construct denotes an entity
7681 -- Do not generate an error message when the comparison is done
7682 -- against the enumeration literal Standard.True.
7686 -- Build a customized error message
7713 -- The construct is too complex to disect, issue a general message
7715 else
7720 -----------------------------
7721 -- Find_Unique_Access_Type --
7722 -----------------------------
7729 begin
7731 E_Access_Attribute_Type)
7732 then
7736 E_Access_Attribute_Type)
7737 then
7739 else
7751 then
7764 -- Start of processing for Resolve_Equality_Op
7766 begin
7777 T = Any_Character
7778 then
7781 else
7790 then
7799 -- If expressions must have a single type, and if the context does
7800 -- not impose one the dependent expressions cannot be anonymous
7801 -- access types.
7803 -- Why no similar processing for case expressions???
7807 E_Anonymous_Access_Subprogram_Type)
7809 E_Anonymous_Access_Subprogram_Type)
7810 then
7818 -- In SPARK, equality operators = and /= for array types other than
7819 -- String are only defined when, for each index position, the
7820 -- operands have equal static bounds.
7824 -- Protect call to Matching_Static_Array_Bounds to avoid costly
7825 -- operation if not needed.
7833 then
7834 Check_SPARK_05_Restriction
7839 -- If the unique type is a class-wide type then it will be expanded
7840 -- into a dispatching call to the predefined primitive. Therefore we
7841 -- check here for potential violation of such restriction.
7847 if Warn_On_Redundant_Constructs
7852 then
7853 Error_Msg_N -- CODEFIX
7863 -- If this is an inequality, it may be the implicit inequality
7864 -- created for a user-defined operation, in which case the corres-
7865 -- ponding equality operation is not intrinsic, and the operation
7866 -- cannot be constant-folded. Else fold.
7871 or else Is_Intrinsic_Subprogram
7873 then
7879 then
7883 -- Ada 2005: If one operand is an anonymous access type, convert the
7884 -- other operand to it, to ensure that the underlying types match in
7885 -- the back-end. Same for access_to_subprogram, and the conversion
7886 -- verifies that the types are subtype conformant.
7888 -- We apply the same conversion in the case one of the operands is a
7889 -- private subtype of the type of the other.
7891 -- Why the Expander_Active test here ???
7893 if Expander_Active
7894 and then
7896 E_Anonymous_Access_Subprogram_Type)
7898 then
7918 ----------------------------------
7919 -- Resolve_Explicit_Dereference --
7920 ----------------------------------
7928 -- The candidate prefix type, if overloaded
7933 begin
7937 -- A useful optimization: check whether the dereference denotes an
7938 -- element of a container, and if so rewrite it as a call to the
7939 -- corresponding Element function.
7941 -- Disabled for now, on advice of ARG. A more restricted form of the
7942 -- predicate might be acceptable ???
7944 -- if Is_Container_Element (N) then
7945 -- return;
7946 -- end if;
7950 -- Use the context type to select the prefix that has the correct
7951 -- designated type. Keep the first match, which will be the inner-
7952 -- most.
7959 then
7964 -- Remove access types that do not match, but preserve access
7965 -- to subprogram interpretations, in case a further dereference
7966 -- is needed (see below).
7979 else
7980 -- If no interpretation covers the designated type of the prefix,
7981 -- this is the pathological case where not all implementations of
7982 -- the prefix allow the interpretation of the node as a call. Now
7983 -- that the expected type is known, Remove other interpretations
7984 -- from prefix, rewrite it as a call, and resolve again, so that
7985 -- the proper call node is generated.
7996 New_N :=
7998 Name =>
8009 -- If not overloaded, resolve P with its own type
8011 else
8015 -- If the prefix might be null, add an access check
8019 then
8023 -- If the designated type is a packed unconstrained array type, and the
8024 -- explicit dereference is not in the context of an attribute reference,
8025 -- then we must compute and set the actual subtype, since it is needed
8026 -- by Gigi. The reason we exclude the attribute case is that this is
8027 -- handled fine by Gigi, and in fact we use such attributes to build the
8028 -- actual subtype. We also exclude generated code (which builds actual
8029 -- subtypes directly if they are needed).
8036 then
8042 -- Note: No Eval processing is required for an explicit dereference,
8043 -- because such a name can never be static.
8047 -------------------------------------
8048 -- Resolve_Expression_With_Actions --
8049 -------------------------------------
8052 begin
8055 -- If N has no actions, and its expression has been constant folded,
8056 -- then rewrite N as just its expression. Note, we can't do this in
8057 -- the general case of Is_Empty_List (Actions (N)) as this would cause
8058 -- Expression (N) to be expanded again.
8062 then
8067 ----------------------------------
8068 -- Resolve_Generalized_Indexing --
8069 ----------------------------------
8077 begin
8078 -- In ASIS mode, propagate the information about the indexes back to
8079 -- to the original indexing node. The generalized indexing is either
8080 -- a function call, or a dereference of one. The actuals include the
8081 -- prefix of the original node, which is the container expression.
8090 loop
8099 -- If expression is to be reanalyzed, reset Generalized_Indexing
8100 -- to recreate call node, as is the case when the expression is
8101 -- part of an expression function.
8110 else
8116 ---------------------------
8117 -- Resolve_If_Expression --
8118 ---------------------------
8127 begin
8132 -- When the "then" expression is of a scalar subtype different from the
8133 -- result subtype, then insert a conversion to ensure the generation of
8134 -- a constraint check. The same is done for the else part below, again
8135 -- comparing subtypes rather than base types.
8139 then
8144 -- If ELSE expression present, just resolve using the determined type
8145 -- If type is universal, resolve to any member of the class.
8154 else
8164 -- Apply RM 4.5.7 (17/3): whether the expression is statically or
8165 -- dynamically tagged must be known statically.
8170 then
8176 -- If no ELSE expression is present, root type must be Standard.Boolean
8177 -- and we provide a Standard.True result converted to the appropriate
8178 -- Boolean type (in case it is a derived boolean type).
8181 Else_Expr :=
8186 else
8195 -------------------------------
8196 -- Resolve_Indexed_Component --
8197 -------------------------------
8205 begin
8213 -- Use the context type to select the prefix that yields the correct
8214 -- component type.
8216 declare
8223 begin
8230 and then
8231 Covers
8234 then
8243 else
8249 else
8260 else
8267 -- If prefix is access type, dereference to get real array type.
8268 -- Note: we do not apply an access check because the expander always
8269 -- introduces an explicit dereference, and the check will happen there.
8275 -- If name was overloaded, set component type correctly now
8276 -- If a misplaced call to an entry family (which has no index types)
8277 -- return. Error will be diagnosed from calling context.
8281 else
8288 -- The prefix may have resolved to a string literal, in which case its
8289 -- etype has a special representation. This is only possible currently
8290 -- if the prefix is a static concatenation, written in functional
8291 -- notation.
8296 else
8303 else
8314 -- Do not generate the warning on suspicious index if we are analyzing
8315 -- package Ada.Tags; otherwise we will report the warning with the
8316 -- Prims_Ptr field of the dispatch table.
8319 or else not
8321 Ada_Tags)
8322 then
8327 -- If the array type is atomic, and the component is not atomic, then
8328 -- this is worth a warning, since we have a situation where the access
8329 -- to the component may cause extra read/writes of the atomic array
8330 -- object, or partial word accesses, which could be unexpected.
8336 and then Has_Atomic_Components
8339 then
8340 Error_Msg_N
8342 Error_Msg_N
8347 -----------------------------
8348 -- Resolve_Integer_Literal --
8349 -----------------------------
8352 begin
8357 --------------------------------
8358 -- Resolve_Intrinsic_Operator --
8359 --------------------------------
8368 -- If the operand is a literal, it cannot be the expression in a
8369 -- conversion. Use a qualified expression instead.
8371 ---------------------
8372 -- Convert_Operand --
8373 ---------------------
8379 begin
8381 Res :=
8387 else
8394 -- Start of processing for Resolve_Intrinsic_Operator
8396 begin
8397 -- We must preserve the original entity in a generic setting, so that
8398 -- the legality of the operation can be verified in an instance.
8413 -- If the result or operand types are private, rewrite with unchecked
8414 -- conversions on the operands and the result, to expose the proper
8415 -- underlying numeric type.
8420 then
8425 else
8446 then
8447 -- Add explicit conversion where needed, and save interpretations in
8448 -- case operands are overloaded.
8455 else
8461 else
8471 else
8476 --------------------------------------
8477 -- Resolve_Intrinsic_Unary_Operator --
8478 --------------------------------------
8480 procedure Resolve_Intrinsic_Unary_Operator
8483 is
8488 begin
8507 else
8512 ------------------------
8513 -- Resolve_Logical_Op --
8514 ------------------------
8519 begin
8522 -- Predefined operations on scalar types yield the base type. On the
8523 -- other hand, logical operations on arrays yield the type of the
8524 -- arguments (and the context).
8528 else
8532 -- The following test is required because the operands of the operation
8533 -- may be literals, in which case the resulting type appears to be
8534 -- compatible with a signed integer type, when in fact it is compatible
8535 -- only with modular types. If the context itself is universal, the
8536 -- operation is illegal.
8544 Error_Msg_N
8552 then
8556 -- Replace AND by AND THEN, or OR by OR ELSE, if Short_Circuit_And_Or
8557 -- is active and the result type is standard Boolean (do not mess with
8558 -- ops that return a nonstandard Boolean type, because something strange
8559 -- is going on).
8561 -- Note: you might expect this replacement to be done during expansion,
8562 -- but that doesn't work, because when the pragma Short_Circuit_And_Or
8563 -- is used, no part of the right operand of an "and" or "or" operator
8564 -- should be executed if the left operand would short-circuit the
8565 -- evaluation of the corresponding "and then" or "or else". If we left
8566 -- the replacement to expansion time, then run-time checks associated
8567 -- with such operands would be evaluated unconditionally, due to being
8568 -- before the condition prior to the rewriting as short-circuit forms
8569 -- during expansion.
8571 if Short_Circuit_And_Or
8574 then
8575 -- Mark the corresponding putative SCO operator as truly a logical
8576 -- (and short-circuit) operator.
8589 -- Case of OR changed to OR ELSE
8591 else
8599 -- Return now, since analysis of the rewritten ops will take care of
8600 -- other reference bookkeeping and expression folding.
8615 -- In SPARK, logical operations AND, OR and XOR for arrays are defined
8616 -- only when both operands have same static lower and higher bounds. Of
8617 -- course the types have to match, so only check if operands are
8618 -- compatible and the node itself has no errors.
8622 then
8623 declare
8627 begin
8628 -- Protect call to Matching_Static_Array_Bounds to avoid costly
8629 -- operation if not needed.
8636 then
8637 Check_SPARK_05_Restriction
8644 ---------------------------
8645 -- Resolve_Membership_Op --
8646 ---------------------------
8648 -- The context can only be a boolean type, and does not determine the
8649 -- arguments. Arguments should be unambiguous, but the preference rule for
8650 -- universal types applies.
8660 -- Analysis has determined a unique type for the left operand. Use it to
8661 -- resolve the disjuncts.
8663 ----------------------------
8664 -- Resolve_Set_Membership --
8665 ----------------------------
8671 begin
8672 -- If the left operand is overloaded, find type compatible with not
8673 -- overloaded alternative of the right operand.
8682 else
8687 -- Unclear how to resolve expression if all alternatives are also
8688 -- overloaded.
8694 else
8703 -- Alternative is an expression, a range
8704 -- or a subtype mark.
8708 then
8715 -- Check for duplicates for discrete case
8718 declare
8727 begin
8728 -- Loop checking duplicates. This is quadratic, but giant sets
8729 -- are unlikely in this context so it's a reasonable choice.
8736 N_Character_Literal)
8738 then
8756 -- Start of processing for Resolve_Membership_Op
8758 begin
8764 Resolve_Set_Membership;
8768 and then
8770 or else
8773 then
8776 -- Ada 2005 (AI-251): Support the following case:
8778 -- type I is interface;
8779 -- type T is tagged ...
8781 -- function Test (O : I'Class) is
8782 -- begin
8783 -- return O in T'Class.
8784 -- end Test;
8786 -- In this case we have nothing else to do. The membership test will be
8787 -- done at run time.
8794 then
8796 else
8800 -- If mixed-mode operations are present and operands are all literal,
8801 -- the only interpretation involves Duration, which is probably not
8802 -- the intention of the programmer.
8817 then
8823 else
8828 -- Here after resolving membership operation
8835 ------------------
8836 -- Resolve_Null --
8837 ------------------
8842 begin
8843 -- Handle restriction against anonymous null access values This
8844 -- restriction can be turned off using -gnatdj.
8846 -- Ada 2005 (AI-231): Remove restriction
8849 and then not Debug_Flag_J
8852 then
8853 -- In the common case of a call which uses an explicitly null value
8854 -- for an access parameter, give specialized error message.
8857 Error_Msg_N
8860 -- Standard message for all other cases (are there any?)
8862 else
8863 Error_Msg_N
8868 -- Ada 2005 (AI-231): Generate the null-excluding check in case of
8869 -- assignment to a null-excluding object
8874 then
8876 Insert_Action
8881 else
8888 -- In a distributed context, null for a remote access to subprogram may
8889 -- need to be replaced with a special record aggregate. In this case,
8890 -- return after having done the transformation.
8895 then
8899 -- The null literal takes its type from the context
8904 -----------------------
8905 -- Resolve_Op_Concat --
8906 -----------------------
8910 -- We wish to avoid deep recursion, because concatenations are often
8911 -- deeply nested, as in A&B&...&Z. Therefore, we walk down the left
8912 -- operands nonrecursively until we find something that is not a simple
8913 -- concatenation (A in this case). We resolve that, and then walk back
8914 -- up the tree following Parent pointers, calling Resolve_Op_Concat_Rest
8915 -- to do the rest of the work at each level. The Parent pointers allow
8916 -- us to avoid recursion, and thus avoid running out of memory. See also
8917 -- Sem_Ch4.Analyze_Concatenation, where a similar approach is used.
8922 begin
8923 -- The following code is equivalent to:
8925 -- Resolve_Op_Concat_First (NN, Typ);
8926 -- Resolve_Op_Concat_Arg (N, ...);
8927 -- Resolve_Op_Concat_Rest (N, Typ);
8929 -- where the Resolve_Op_Concat_Arg call recurses back here if the left
8930 -- operand is a concatenation.
8932 -- Walk down left operands
8934 loop
8943 -- Now (given the above example) NN is A&B and Op1 is A
8945 -- First resolve Op1 ...
8949 -- ... then walk NN back up until we reach N (where we started), calling
8950 -- Resolve_Op_Concat_Rest along the way.
8952 loop
8959 Check_SPARK_05_Restriction
8964 ---------------------------
8965 -- Resolve_Op_Concat_Arg --
8966 ---------------------------
8968 procedure Resolve_Op_Concat_Arg
8973 is
8977 begin
8979 if Is_Comp
8983 then
8985 else
8989 -- If both Array & Array and Array & Component are visible, there is a
8990 -- potential ambiguity that must be reported.
8995 then
8999 -- If the operation is user-defined and not overloaded use its
9000 -- profile. The operation may be a renaming, in which case it has
9001 -- been rewritten, and we want the original profile.
9006 then
9008 Etype
9012 -- Otherwise an aggregate may match both the array type and the
9013 -- component type.
9015 else
9020 else
9024 then
9025 declare
9030 begin
9035 -- Special-case the error message when the overloading is
9036 -- caused by a function that yields an array and can be
9037 -- called without parameters.
9042 Error_Msg_NE
9044 Error_Msg_NE
9048 else
9056 or else
9058 then
9059 Error_Msg_N -- CODEFIX
9077 else
9082 else
9086 -- Concatenation is restricted in SPARK: each operand must be either a
9087 -- string literal, the name of a string constant, a static character or
9088 -- string expression, or another concatenation. Arg cannot be a
9089 -- concatenation here as callers of Resolve_Op_Concat_Arg call it
9090 -- separately on each final operand, past concatenation operations.
9094 Check_SPARK_05_Restriction
9102 then
9103 Check_SPARK_05_Restriction
9107 -- Do not issue error on an operand that is neither a character nor a
9108 -- string, as the error is issued in Resolve_Op_Concat.
9110 else
9117 -----------------------------
9118 -- Resolve_Op_Concat_First --
9119 -----------------------------
9126 begin
9127 -- The parser folds an enormous sequence of concatenations of string
9128 -- literals into "" & "...", where the Is_Folded_In_Parser flag is set
9129 -- in the right operand. If the expression resolves to a predefined "&"
9130 -- operator, all is well. Otherwise, the parser's folding is wrong, so
9131 -- we give an error. See P_Simple_Expression in Par.Ch4.
9136 then
9151 ----------------------------
9152 -- Resolve_Op_Concat_Rest --
9153 ----------------------------
9159 begin
9168 -- If this is not a static concatenation, but the result is a string
9169 -- type (and not an array of strings) ensure that static string operands
9170 -- have their subtypes properly constructed.
9174 then
9180 ----------------------
9181 -- Resolve_Op_Expon --
9182 ----------------------
9187 begin
9188 -- Catch attempts to do fixed-point exponentiation with universal
9189 -- operands, which is a case where the illegality is not caught during
9190 -- normal operator analysis. This is not done in preanalysis mode
9191 -- since the tree is not fully decorated during preanalysis.
9202 then
9212 then
9219 then
9223 -- We do the resolution using the base type, because intermediate values
9224 -- in expressions are always of the base type, not a subtype of it.
9229 -- For integer types, right argument must be in Natural range
9244 -- Evaluate the exponentiation operator for dimensioned type
9247 else
9251 -- Set overflow checking bit. Much cleverer code needed here eventually
9252 -- and perhaps the Resolve routines should be separated for the various
9253 -- arithmetic operations, since they will need different processing. ???
9262 --------------------
9263 -- Resolve_Op_Not --
9264 --------------------
9270 -- This function determines if the parent node is a boolean operator or
9271 -- operation (comparison op, membership test, or short circuit form) and
9272 -- the not in question is the left operand of this operation. Note that
9273 -- if the not is in parens, then false is returned.
9275 -----------------------
9276 -- Parent_Is_Boolean --
9277 -----------------------
9280 begin
9284 else
9286 when N_Op_And |
9287 N_Op_Eq |
9288 N_Op_Ge |
9289 N_Op_Gt |
9290 N_Op_Le |
9291 N_Op_Lt |
9292 N_Op_Ne |
9293 N_Op_Or |
9294 N_Op_Xor |
9295 N_In |
9296 N_Not_In |
9297 N_And_Then |
9298 N_Or_Else =>
9308 -- Start of processing for Resolve_Op_Not
9310 begin
9311 -- Predefined operations on scalar types yield the base type. On the
9312 -- other hand, logical operations on arrays yield the type of the
9313 -- arguments (and the context).
9317 else
9321 -- Straightforward case of incorrect arguments
9328 -- Special case of probable missing parens
9332 Error_Msg_N
9335 else
9336 Error_Msg_N
9343 -- OK resolution of NOT
9345 else
9346 -- Warn if non-boolean types involved. This is a case like not a < b
9347 -- where a and b are modular, where we will get (not a) < b and most
9348 -- likely not (a < b) was intended.
9350 if Warn_On_Questionable_Missing_Parens
9352 and then Parent_Is_Boolean
9353 then
9357 -- Warn on double negation if checking redundant constructs
9359 if Warn_On_Redundant_Constructs
9364 then
9368 -- Complete resolution and evaluation of NOT
9378 -----------------------------
9379 -- Resolve_Operator_Symbol --
9380 -----------------------------
9382 -- Nothing to be done, all resolved already
9388 begin
9392 ----------------------------------
9393 -- Resolve_Qualified_Expression --
9394 ----------------------------------
9402 begin
9405 -- Protect call to Matching_Static_Array_Bounds to avoid costly
9406 -- operation if not needed.
9413 then
9414 Check_SPARK_05_Restriction
9418 -- A qualified expression requires an exact match of the type, class-
9419 -- wide matching is not allowed. However, if the qualifying type is
9420 -- specific and the expression has a class-wide type, it may still be
9421 -- okay, since it can be the result of the expansion of a call to a
9422 -- dispatching function, so we also have to check class-wideness of the
9423 -- type of the expression's original node.
9426 or else
9430 then
9434 -- If the target type is unconstrained, then we reset the type of the
9435 -- result from the type of the expression. For other cases, the actual
9436 -- subtype of the expression is the target type.
9440 then
9447 -- If we still have a qualified expression after the static evaluation,
9448 -- then apply a scalar range check if needed. The reason that we do this
9449 -- after the Eval call is that otherwise, the application of the range
9450 -- check may convert an illegal static expression and result in warning
9451 -- rather than giving an error (e.g Integer'(Integer'Last + 1)).
9457 -- Finally, check whether a predicate applies to the target type. This
9458 -- comes from AI12-0100. As for type conversions, check the enclosing
9459 -- context to prevent an infinite expansion.
9465 or else
9467 then
9476 ------------------------------
9477 -- Resolve_Raise_Expression --
9478 ------------------------------
9481 begin
9485 else
9490 -------------------
9491 -- Resolve_Range --
9492 -------------------
9499 -- Returns True if N is of the form X'First .. X'Last where X is the
9500 -- same entity for both attributes.
9502 --------------------
9503 -- First_Last_Ref --
9504 --------------------
9510 begin
9515 then
9516 declare
9519 begin
9523 then
9532 -- Start of processing for Resolve_Range
9534 begin
9539 -- Check for inappropriate range on unordered enumeration type
9543 -- Exclude X'First .. X'Last if X is the same entity for both
9545 and then not First_Last_Ref
9546 then
9548 Error_Msg_NE
9555 -- We have to check the bounds for being within the base range as
9556 -- required for a non-static context. Normally this is automatic and
9557 -- done as part of evaluating expressions, but the N_Range node is an
9558 -- exception, since in GNAT we consider this node to be a subexpression,
9559 -- even though in Ada it is not. The circuit in Sem_Eval could check for
9560 -- this, but that would put the test on the main evaluation path for
9561 -- expressions.
9566 -- Check for an ambiguous range over character literals. This will
9567 -- happen with a membership test involving only literals.
9575 -- If bounds are static, constant-fold them, so size computations are
9576 -- identical between front-end and back-end. Do not perform this
9577 -- transformation while analyzing generic units, as type information
9578 -- would be lost when reanalyzing the constant node in the instance.
9591 --------------------------
9592 -- Resolve_Real_Literal --
9593 --------------------------
9598 begin
9599 -- Special processing for fixed-point literals to make sure that the
9600 -- value is an exact multiple of small where this is required. We skip
9601 -- this for the universal real case, and also for generic types.
9607 then
9608 declare
9615 begin
9616 -- Case of literal is not an exact multiple of the Small
9620 -- For a source program literal for a decimal fixed-point type,
9621 -- this is statically illegal (RM 4.9(36)).
9626 then
9630 -- Generate a warning if literal from source
9633 and then Warn_On_Bad_Fixed_Value
9634 then
9635 Error_Msg_N
9637 N);
9640 -- Replace literal by a value that is the exact representation
9641 -- of a value of the type, i.e. a multiple of the small value,
9642 -- by truncation, since Machine_Rounds is false for all GNAT
9643 -- fixed-point types (RM 4.9(38)).
9653 -- In all cases, set the corresponding integer field
9659 -- Now replace the actual type by the expected type as usual
9665 -----------------------
9666 -- Resolve_Reference --
9667 -----------------------
9672 begin
9673 -- Replace general access with specific type
9681 -- If we are taking the reference of a volatile entity, then treat it as
9682 -- a potential modification of this entity. This is too conservative,
9683 -- but necessary because remove side effects can cause transformations
9684 -- of normal assignments into reference sequences that otherwise fail to
9685 -- notice the modification.
9692 --------------------------------
9693 -- Resolve_Selected_Component --
9694 --------------------------------
9709 -- Check whether this is the initialization of a component within an
9710 -- init proc (by assignment or call to another init proc). If true,
9711 -- there is no need for a discriminant check.
9713 --------------------
9714 -- Init_Component --
9715 --------------------
9718 begin
9719 return Inside_Init_Proc
9725 -- Start of processing for Resolve_Selected_Component
9727 begin
9730 -- Use the context type to select the prefix that has a selector
9731 -- of the correct name and type.
9739 else
9743 -- Locate selected component. For a private prefix the selector
9744 -- can denote a discriminant.
9748 -- The visible components of a class-wide type are those of
9749 -- the root type.
9759 then
9766 else
9770 Error_Msg_N
9775 else
9778 -- There may be an implicit dereference. Retrieve
9779 -- designated record type.
9783 else
9789 -- Resolution chooses the new interpretation.
9790 -- Find the component with the right name.
9795 loop
9812 -- There must be a legal interpretation at this point
9819 else
9820 -- Resolve prefix with its type
9825 -- Generate cross-reference. We needed to wait until full overloading
9826 -- resolution was complete to do this, since otherwise we can't tell if
9827 -- we are an lvalue or not.
9831 else
9835 -- If prefix is an access type, the node will be transformed into an
9836 -- explicit dereference during expansion. The type of the node is the
9837 -- designated type of that of the prefix.
9842 else
9846 -- Set flag for expander if discriminant check required on a component
9847 -- appearing within a variant.
9853 and then
9856 and then not Init_Component
9857 then
9865 -- If the prefix is a record conversion, this may be a renamed
9866 -- discriminant whose bounds differ from those of the original
9867 -- one, so we must ensure that a range check is performed.
9872 then
9876 -- Note: No Eval processing is required, because the prefix is of a
9877 -- record type, or protected type, and neither can possibly be static.
9879 -- If the record type is atomic, and the component is non-atomic, then
9880 -- this is worth a warning, since we have a situation where the access
9881 -- to the component may cause extra read/writes of the atomic array
9882 -- object, or partial word accesses, both of which may be unexpected.
9888 then
9889 Error_Msg_N
9892 Error_Msg_N
9900 -------------------
9901 -- Resolve_Shift --
9902 -------------------
9909 begin
9910 -- We do the resolution using the base type, because intermediate values
9911 -- in expressions always are of the base type, not a subtype of it.
9924 ---------------------------
9925 -- Resolve_Short_Circuit --
9926 ---------------------------
9933 begin
9934 -- Ensure all actions associated with the left operand (e.g.
9935 -- finalization of transient controlled objects) are fully evaluated
9936 -- locally within an expression with actions. This is particularly
9937 -- helpful for coverage analysis. However this should not happen in
9938 -- generics or if Minimize_Expression_With_Actions is set.
9941 declare
9943 begin
9951 -- Set Comes_From_Source on L to preserve warnings for unset
9952 -- reference.
9961 -- Check for issuing warning for always False assert/check, this happens
9962 -- when assertions are turned off, in which case the pragma Assert/Check
9963 -- was transformed into:
9965 -- if False and then <condition> then ...
9967 -- and we detect this pattern
9969 if Warn_On_Assertion_Failure
9976 then
9977 declare
9980 begin
9981 -- Special handling of Asssert pragma
9985 then
9986 declare
9988 Original_Node
9989 (Expression
9992 begin
9993 -- Don't warn if original condition is explicit False,
9994 -- since obviously the failure is expected in this case.
9998 then
10001 -- Issue warning. We do not want the deletion of the
10002 -- IF/AND-THEN to take this message with it. We achieve this
10003 -- by making sure that the expanded code points to the Sloc
10004 -- of the expression, not the original pragma.
10006 else
10007 -- Note: Use Error_Msg_F here rather than Error_Msg_N.
10008 -- The source location of the expression is not usually
10009 -- the best choice here. For example, it gets located on
10010 -- the last AND keyword in a chain of boolean expressiond
10011 -- AND'ed together. It is best to put the message on the
10012 -- first character of the assertion, which is the effect
10013 -- of the First_Node call here.
10015 Error_Msg_F
10017 Expression
10022 -- Similar processing for Check pragma
10026 then
10027 -- Don't want to warn if original condition is explicit False
10029 declare
10031 Original_Node
10032 (Expression
10034 begin
10037 then
10040 -- Post warning
10042 else
10043 -- Again use Error_Msg_F rather than Error_Msg_N, see
10044 -- comment above for an explanation of why we do this.
10046 Error_Msg_F
10048 Expression
10056 -- Continue with processing of short circuit
10065 -------------------
10066 -- Resolve_Slice --
10067 -------------------
10076 begin
10079 -- Use the context type to select the prefix that yields the correct
10080 -- array type.
10082 declare
10089 begin
10097 then
10105 else
10110 else
10121 else
10131 -- If the prefix is an access to an unconstrained array, we must use
10132 -- the actual subtype of the object to perform the index checks. The
10133 -- object denoted by the prefix is implicit in the node, so we build
10134 -- an explicit representation for it in order to compute the actual
10135 -- subtype.
10140 declare
10144 begin
10155 then
10158 -- If the name is a selected component that depends on discriminants,
10159 -- build an actual subtype for it. This can happen only when the name
10160 -- itself is overloaded; otherwise the actual subtype is created when
10161 -- the selected component is analyzed.
10164 and then Full_Analysis
10166 then
10167 declare
10170 begin
10175 -- Maybe this should just be "else", instead of checking for the
10176 -- specific case of slice??? This is needed for the case where the
10177 -- prefix is an Image attribute, which gets expanded to a slice, and so
10178 -- has a constrained subtype which we want to use for the slice range
10179 -- check applied below (the range check won't get done if the
10180 -- unconstrained subtype of the 'Image is used).
10186 -- Obtain the type of the array index
10190 else
10194 -- If name was overloaded, set slice type correctly now
10198 -- Handle the generation of a range check that compares the array index
10199 -- against the discrete_range. The check is not applied to internally
10200 -- built nodes associated with the expansion of dispatch tables. Check
10201 -- that Ada.Tags has already been loaded to avoid extra dependencies on
10202 -- the unit.
10204 if Tagged_Type_Expansion
10210 then
10213 -- The discrete_range is specified by a subtype indication. Create a
10214 -- shallow copy and inherit the type, parent and source location from
10215 -- the discrete_range. This ensures that the range check is inserted
10216 -- relative to the slice and that the runtime exception points to the
10217 -- proper construct.
10226 -- The discrete_range is a regular range. Resolve the bounds and remove
10227 -- their side effects.
10229 else
10246 -- Check bad use of type with predicates
10248 declare
10251 begin
10254 then
10256 else
10261 Bad_Predicated_Subtype_Use
10266 -- Otherwise here is where we check suspicious indexes
10277 ----------------------------
10278 -- Resolve_String_Literal --
10279 ----------------------------
10290 begin
10291 -- For a string appearing in a concatenation, defer creation of the
10292 -- string_literal_subtype until the end of the resolution of the
10293 -- concatenation, because the literal may be constant-folded away. This
10294 -- is a useful optimization for long concatenation expressions.
10296 -- If the string is an aggregate built for a single character (which
10297 -- happens in a non-static context) or a is null string to which special
10298 -- checks may apply, we build the subtype. Wide strings must also get a
10299 -- string subtype if they come from a one character aggregate. Strings
10300 -- generated by attributes might be static, but it is often hard to
10301 -- determine whether the enclosing context is static, so we generate
10302 -- subtypes for them as well, thus losing some rarer optimizations ???
10303 -- Same for strings that come from a static conversion.
10305 Need_Check :=
10314 -- If the resolving type is itself a string literal subtype, we can just
10315 -- reuse it, since there is no point in creating another.
10321 and then not Need_Check
10323 N_Attribute_Reference,
10324 N_Qualified_Expression,
10325 N_Type_Conversion)
10326 then
10329 -- Do not generate a string literal subtype for the default expression
10330 -- of a formal parameter in GNATprove mode. This is because the string
10331 -- subtype is associated with the freezing actions of the subprogram,
10332 -- however freezing is disabled in GNATprove mode and as a result the
10333 -- subtype is unavailable.
10335 elsif GNATprove_Mode
10337 then
10340 -- Otherwise we must create a string literal subtype. Note that the
10341 -- whole idea of string literal subtypes is simply to avoid the need
10342 -- for building a full fledged array subtype for each literal.
10344 else
10350 or else Need_Check
10351 then
10358 then
10364 -- The validity of a null string has been checked in the call to
10365 -- Eval_String_Literal.
10370 -- Always accept string literal with component type Any_Character, which
10371 -- occurs in error situations and in comparisons of literals, both of
10372 -- which should accept all literals.
10377 -- If the type is bit-packed, then we always transform the string
10378 -- literal into a full fledged aggregate.
10383 -- Deal with cases of Wide_Wide_String, Wide_String, and String
10385 else
10386 -- For Standard.Wide_Wide_String, or any other type whose component
10387 -- type is Standard.Wide_Wide_Character, we know that all the
10388 -- characters in the string must be acceptable, since the parser
10389 -- accepted the characters as valid character literals.
10394 -- For the case of Standard.String, or any other type whose component
10395 -- type is Standard.Character, we must make sure that there are no
10396 -- wide characters in the string, i.e. that it is entirely composed
10397 -- of characters in range of type Character.
10399 -- If the string literal is the result of a static concatenation, the
10400 -- test has already been performed on the components, and need not be
10401 -- repeated.
10405 then
10409 -- If we are out of range, post error. This is one of the
10410 -- very few places that we place the flag in the middle of
10411 -- a token, right under the offending wide character. Not
10412 -- quite clear if this is right wrt wide character encoding
10413 -- sequences, but it's only an error message.
10415 Error_Msg
10422 -- For the case of Standard.Wide_String, or any other type whose
10423 -- component type is Standard.Wide_Character, we must make sure that
10424 -- there are no wide characters in the string, i.e. that it is
10425 -- entirely composed of characters in range of type Wide_Character.
10427 -- If the string literal is the result of a static concatenation,
10428 -- the test has already been performed on the components, and need
10429 -- not be repeated.
10433 then
10437 -- If we are out of range, post error. This is one of the
10438 -- very few places that we place the flag in the middle of
10439 -- a token, right under the offending wide character.
10441 -- This is not quite right, because characters in general
10442 -- will take more than one character position ???
10444 Error_Msg
10451 -- If the root type is not a standard character, then we will convert
10452 -- the string into an aggregate and will let the aggregate code do
10453 -- the checking. Standard Wide_Wide_Character is also OK here.
10455 else
10459 -- See if the component type of the array corresponding to the string
10460 -- has compile time known bounds. If yes we can directly check
10461 -- whether the evaluation of the string will raise constraint error.
10462 -- Otherwise we need to transform the string literal into the
10463 -- corresponding character aggregate and let the aggregate code do
10464 -- the checking.
10468 -- Check for the case of full range, where we are definitely OK
10474 -- Here the range is not the complete base type range, so check
10476 declare
10484 begin
10487 then
10493 then
10494 Apply_Compile_Time_Constraint_Error
10496 CE_Range_Check_Failed,
10507 -- If we got here we meed to transform the string literal into the
10508 -- equivalent qualified positional array aggregate. This is rather
10509 -- heavy artillery for this situation, but it is hard work to avoid.
10511 declare
10516 begin
10517 -- Build the character literals, we give them source locations that
10518 -- correspond to the string positions, which is a bit tricky given
10519 -- the possible presence of wide character escape sequences.
10533 -- Should we have a call to Skip_Wide here ???
10535 -- ??? else
10536 -- Skip_Wide (P);
10544 Expression =>
10551 -----------------------------
10552 -- Resolve_Type_Conversion --
10553 -----------------------------
10565 -- Set to False to suppress cases where we want to suppress the test
10566 -- for redundancy to avoid possible false positives on this warning.
10568 begin
10569 if not Conv_OK
10571 then
10575 -- If the Operand Etype is Universal_Fixed, then the conversion is
10576 -- never redundant. We need this check because by the time we have
10577 -- finished the rather complex transformation, the conversion looks
10578 -- redundant when it is not.
10583 -- If the operand is marked as Any_Fixed, then special processing is
10584 -- required. This is also a case where we suppress the test for a
10585 -- redundant conversion, since most certainly it is not redundant.
10590 -- Mixed-mode operation involving a literal. Context must be a fixed
10591 -- type which is applied to the literal subsequently.
10599 or else
10601 then
10602 -- Return if expression is ambiguous
10607 -- If nothing else, the available fixed type is Duration
10609 else
10613 -- Resolve the real operand with largest available precision
10617 else
10623 -- If the operand is a literal (it could be a non-static and
10624 -- illegal exponentiation) check whether the use of Duration
10625 -- is potentially inaccurate.
10630 then
10631 Error_Msg_N
10634 Error_Msg_N
10641 then
10644 else
10653 -- In SPARK, a type conversion between array types should be restricted
10654 -- to types which have matching static bounds.
10656 -- Protect call to Matching_Static_Array_Bounds to avoid costly
10657 -- operation if not needed.
10664 then
10665 Check_SPARK_05_Restriction
10669 -- In formal mode, the operand of an ancestor type conversion must be an
10670 -- object (not an expression).
10678 then
10684 -- Note: we do the Eval_Type_Conversion call before applying the
10685 -- required checks for a subtype conversion. This is important, since
10686 -- both are prepared under certain circumstances to change the type
10687 -- conversion to a constraint error node, but in the case of
10688 -- Eval_Type_Conversion this may reflect an illegality in the static
10689 -- case, and we would miss the illegality (getting only a warning
10690 -- message), if we applied the type conversion checks first.
10694 -- Even when evaluation is not possible, we may be able to simplify the
10695 -- conversion or its expression. This needs to be done before applying
10696 -- checks, since otherwise the checks may use the original expression
10697 -- and defeat the simplifications. This is specifically the case for
10698 -- elimination of the floating-point Truncation attribute in
10699 -- float-to-int conversions.
10703 -- If after evaluation we still have a type conversion, then we may need
10704 -- to apply checks required for a subtype conversion.
10706 -- Skip these type conversion checks if universal fixed operands
10707 -- operands involved, since range checks are handled separately for
10708 -- these cases (in the appropriate Expand routines in unit Exp_Fixd).
10714 then
10718 -- Issue warning for conversion of simple object to its own type. We
10719 -- have to test the original nodes, since they may have been rewritten
10720 -- by various optimizations.
10724 -- Here we test for a redundant conversion if the warning mode is
10725 -- active (and was not locally reset), and we have a type conversion
10726 -- from source not appearing in a generic instance.
10728 if Test_Redundant
10731 and then not In_Instance
10732 then
10736 -- If the node is part of a larger expression, the Target_Type
10737 -- may not be the original type of the node if the context is a
10738 -- condition. Recover original type to see if conversion is needed.
10742 then
10746 -- If we have an entity name, then give the warning if the entity
10747 -- is the right type, or if it is a loop parameter covered by the
10748 -- original type (that's needed because loop parameters have an
10749 -- odd subtype coming from the bounds).
10752 and then
10754 or else
10758 -- If not an entity, then type of expression must match
10761 then
10762 -- One more check, do not give warning if the analyzed conversion
10763 -- has an expression with non-static bounds, and the bounds of the
10764 -- target are static. This avoids junk warnings in cases where the
10765 -- conversion is necessary to establish staticness, for example in
10766 -- a case statement.
10770 then
10773 -- Finally, if this type conversion occurs in a context requiring
10774 -- a prefix, and the expression is a qualified expression then the
10775 -- type conversion is not redundant, since a qualified expression
10776 -- is not a prefix, whereas a type conversion is. For example, "X
10777 -- := T'(Funx(...)).Y;" is illegal because a selected component
10778 -- requires a prefix, but a type conversion makes it legal: "X :=
10779 -- T(T'(Funx(...))).Y;"
10781 -- In Ada 2012, a qualified expression is a name, so this idiom is
10782 -- no longer needed, but we still suppress the warning because it
10783 -- seems unfriendly for warnings to pop up when you switch to the
10784 -- newer language version.
10788 N_Indexed_Component,
10789 N_Selected_Component,
10790 N_Slice,
10791 N_Explicit_Dereference)
10792 then
10795 -- Never warn on conversion to Long_Long_Integer'Base since
10796 -- that is most likely an artifact of the extended overflow
10797 -- checking and comes from complex expanded code.
10802 -- Here we give the redundant conversion warning. If it is an
10803 -- entity, give the name of the entity in the message. If not,
10804 -- just mention the expression.
10806 -- Shoudn't we test Warn_On_Redundant_Constructs here ???
10808 else
10811 Error_Msg_NE -- CODEFIX
10814 else
10815 Error_Msg_NE
10823 -- Ada 2005 (AI-251): Handle class-wide interface type conversions.
10824 -- No need to perform any interface conversion if the type of the
10825 -- expression coincides with the target type.
10828 and then Expander_Active
10830 then
10831 declare
10835 begin
10836 -- If the type of the operand is a limited view, use nonlimited
10837 -- view when available. If it is a class-wide type, recover the
10838 -- class-wide type of the nonlimited view.
10842 then
10858 -- Conversion from interface type
10862 -- Ada 2005 (AI-217): Handle entities from limited views
10866 Error_Msg_NE -- CODEFIX
10869 Error_Msg_N
10871 N);
10874 and then not
10877 then
10879 Error_Msg_NE -- CODEFIX
10882 Error_Msg_N
10884 N);
10886 else
10890 -- Conversion to interface type
10894 -- Handle subtypes
10901 or else Interface_Present_In_Ancestor
10904 then
10906 else
10909 Error_Msg_N
10917 -- Ada 2012: if target type has predicates, the result requires a
10918 -- predicate check. If the context is a call to another predicate
10919 -- check we must prevent infinite recursion.
10925 or else
10927 then
10930 else
10935 -- If at this stage we have a real to integer conversion, make sure
10936 -- that the Do_Range_Check flag is set, because such conversions in
10937 -- general need a range check. We only need this if expansion is off
10938 -- or we are in GNATProve mode.
10944 then
10948 -- Generating C code a type conversion of an access to constrained
10949 -- array type to access to unconstrained array type involves building
10950 -- a fat pointer which in general cannot be generated on the fly. We
10951 -- remove side effects in order to store the result of the conversion
10952 -- into a temporary.
10954 if Generate_C_Code
10961 then
10966 ----------------------
10967 -- Resolve_Unary_Op --
10968 ----------------------
10977 begin
10980 Check_SPARK_05_Restriction
10984 -- Deal with intrinsic unary operators
10990 then
10995 -- Deal with universal cases
10998 or else
11000 then
11007 -- Generate warning for expressions like abs (x mod 2)
11009 if Warn_On_Redundant_Constructs
11011 then
11015 Error_Msg_N -- CODEFIX
11020 -- Deal with reference generation
11027 -- Set overflow checking bit. Much cleverer code needed here eventually
11028 -- and perhaps the Resolve routines should be separated for the various
11029 -- arithmetic operations, since they will need different processing ???
11037 -- Generate warning for expressions like -5 mod 3 for integers. No need
11038 -- to worry in the floating-point case, since parens do not affect the
11039 -- result so there is no point in giving in a warning.
11041 declare
11050 begin
11051 if Warn_On_Questionable_Missing_Parens
11055 then
11058 -- We are looking for cases where the right operand is not
11059 -- parenthesized, and is a binary operator, multiply, divide, or
11060 -- mod. These are the cases where the grouping can affect results.
11064 then
11065 -- For mod, we always give the warning, since the value is
11066 -- affected by the parenthesization (e.g. (-5) mod 315 /=
11067 -- -(5 mod 315)). But for the other cases, the only concern is
11068 -- overflow, e.g. for the case of 8 big signed (-(2 * 64)
11069 -- overflows, but (-2) * 64 does not). So we try to give the
11070 -- message only when overflow is possible.
11074 then
11079 else
11085 else
11089 -- Note that the test below is deliberately excluding the
11090 -- largest negative number, since that is a potentially
11091 -- troublesome case (e.g. -2 * x, where the result is the
11092 -- largest negative integer has an overflow with 2 * x).
11099 -- For the multiplication case, the only case we have to worry
11100 -- about is when (-a)*b is exactly the largest negative number
11101 -- so that -(a*b) can cause overflow. This can only happen if
11102 -- a is a power of 2, and more generally if any operand is a
11103 -- constant that is not a power of 2, then the parentheses
11104 -- cannot affect whether overflow occurs. We only bother to
11105 -- test the left most operand
11107 -- Loop looking at left operands for one that has known value
11114 -- Operand value of 0 or 1 skips warning
11119 -- Otherwise check power of 2, if power of 2, warn, if
11120 -- anything else, skip warning.
11122 else
11126 else
11135 -- Keep looking at left operands
11140 -- For rem or "/" we can only have a problematic situation
11141 -- if the divisor has a value of minus one or one. Otherwise
11142 -- overflow is impossible (divisor > 1) or we have a case of
11143 -- division by zero in any case.
11148 then
11152 -- If we fall through warning should be issued
11154 -- Shouldn't we test Warn_On_Questionable_Missing_Parens ???
11156 Error_Msg_N
11163 ----------------------------------
11164 -- Resolve_Unchecked_Expression --
11165 ----------------------------------
11167 procedure Resolve_Unchecked_Expression
11170 is
11171 begin
11176 ---------------------------------------
11177 -- Resolve_Unchecked_Type_Conversion --
11178 ---------------------------------------
11180 procedure Resolve_Unchecked_Type_Conversion
11183 is
11189 begin
11190 -- Resolve operand using its own type
11194 -- In an inlined context, the unchecked conversion may be applied
11195 -- to a literal, in which case its type is the type of the context.
11196 -- (In other contexts conversions cannot apply to literals).
11198 if In_Inlined_Body
11202 then
11210 ------------------------------
11211 -- Rewrite_Operator_As_Call --
11212 ------------------------------
11219 begin
11226 New_N :=
11237 ------------------------------
11238 -- Rewrite_Renamed_Operator --
11239 ------------------------------
11241 procedure Rewrite_Renamed_Operator
11245 is
11250 begin
11251 -- Do not perform this transformation within a pre/postcondition,
11252 -- because the expression will be re-analyzed, and the transformation
11253 -- might affect the visibility of the operator, e.g. in an instance.
11254 -- Note that fully analyzed and expanded pre/postconditions appear as
11255 -- pragma Check equivalents.
11261 -- Rewrite the operator node using the real operator, not its renaming.
11262 -- Exclude user-defined intrinsic operations of the same name, which are
11263 -- treated separately and rewritten as calls.
11272 -- Indicate that both the original entity and its renaming are
11273 -- referenced at this point.
11284 -- If the context type is private, add the appropriate conversions so
11285 -- that the operator is applied to the full view. This is done in the
11286 -- routines that resolve intrinsic operators.
11290 when N_Op_Add | N_Op_Subtract | N_Op_Multiply | N_Op_Divide |
11291 N_Op_Expon | N_Op_Mod | N_Op_Rem =>
11304 -- Operator renames a user-defined operator of the same name. Use the
11305 -- original operator in the node, which is the one Gigi knows about.
11312 -----------------------
11313 -- Set_Slice_Subtype --
11314 -----------------------
11316 -- Build an implicit subtype declaration to represent the type delivered by
11317 -- the slice. This is an abbreviated version of an array subtype. We define
11318 -- an index subtype for the slice, using either the subtype name or the
11319 -- discrete range of the slice. To be consistent with index usage elsewhere
11320 -- we create a list header to hold the single index. This list is not
11321 -- otherwise attached to the syntax tree.
11332 begin
11338 else
11339 -- We force the evaluation of a range. This is definitely needed in
11340 -- the renamed case, and seems safer to do unconditionally. Note in
11341 -- any case that since we will create and insert an Itype referring
11342 -- to this range, we must make sure any side effect removal actions
11343 -- are inserted before the Itype definition.
11349 -- If the discrete range is given by a subtype indication, the
11350 -- type of the slice is the base of the subtype mark.
11353 declare
11355 begin
11364 -- Take a new copy of Drange (where bounds have been rewritten to
11365 -- reference side-effect-free names). Using a separate tree ensures
11366 -- that further expansion (e.g. while rewriting a slice assignment
11367 -- into a FOR loop) does not attempt to remove side effects on the
11368 -- bounds again (which would cause the bounds in the index subtype
11369 -- definition to refer to temporaries before they are defined) (the
11370 -- reason is that some names are considered side effect free here
11371 -- for the subtype, but not in the context of a loop iteration
11372 -- scheme).
11393 -- The Etype of the existing Slice node is reset to this slice subtype.
11394 -- Its bounds are obtained from its first index.
11398 -- For packed slice subtypes, freeze immediately (except in the case of
11399 -- being in a "spec expression" where we never freeze when we first see
11400 -- the expression).
11405 -- For all other cases insert an itype reference in the slice's actions
11406 -- so that the itype is frozen at the proper place in the tree (i.e. at
11407 -- the point where actions for the slice are analyzed). Note that this
11408 -- is different from freezing the itype immediately, which might be
11409 -- premature (e.g. if the slice is within a transient scope). This needs
11410 -- to be done only if expansion is enabled.
11417 --------------------------------
11418 -- Set_String_Literal_Subtype --
11419 --------------------------------
11427 begin
11439 -- The low bound is set from the low bound of the corresponding index
11440 -- type. Note that we do not store the high bound in the string literal
11441 -- subtype, but it can be deduced if necessary from the length and the
11442 -- low bound.
11447 -- If the lower bound is not static we create a range for the string
11448 -- literal, using the index type and the known length of the literal.
11449 -- The index type is not necessarily Positive, so the upper bound is
11450 -- computed as T'Val (T'Pos (Low_Bound) + L - 1).
11452 else
11453 declare
11459 Prefix =>
11463 Left_Opnd =>
11466 Prefix =>
11468 Expressions =>
11470 Right_Opnd =>
11479 begin
11481 Set_String_Literal_Low_Bound
11484 else
11485 -- If the index type is an enumeration type, build bounds
11486 -- expression with attributes.
11488 Set_String_Literal_Low_Bound
11492 Prefix =>
11499 -- Build bona fide subtype for the string, and wrap it in an
11500 -- unchecked conversion, because the backend expects the
11501 -- String_Literal_Subtype to have a static lower bound.
11503 Index_Subtype :=
11510 -- In the context, the Index_Type may already have a constraint,
11511 -- so use common base type on string subtype. The base type may
11512 -- be used when generating attributes of the string, for example
11513 -- in the context of a slice assignment.
11538 ------------------------------
11539 -- Simplify_Type_Conversion --
11540 ------------------------------
11543 begin
11545 declare
11550 begin
11551 -- Special processing if the conversion is the expression of a
11552 -- Rounding or Truncation attribute reference. In this case we
11553 -- replace:
11555 -- ityp (ftyp'Rounding (x)) or ityp (ftyp'Truncation (x))
11557 -- by
11559 -- ityp (x)
11561 -- with the Float_Truncate flag set to False or True respectively,
11562 -- which is more efficient.
11565 and then
11571 Name_Truncation)
11572 then
11573 declare
11576 begin
11586 -----------------------------
11587 -- Unique_Fixed_Point_Type --
11588 -----------------------------
11597 -- Give error messages for true ambiguity. Messages are posted on node
11598 -- N, and entities T1, T2 are the possible interpretations.
11600 -----------------------
11601 -- Fixed_Point_Error --
11602 -----------------------
11605 begin
11611 -- Start of processing for Unique_Fixed_Point_Type
11613 begin
11614 -- The operations on Duration are visible, so Duration is always a
11615 -- possible interpretation.
11619 -- Look for fixed-point types in enclosing scopes
11628 then
11630 Fixed_Point_Error;
11632 else
11643 -- Look for visible fixed type declarations in the context
11654 then
11656 Fixed_Point_Error;
11658 else
11671 Error_Msg_NE
11673 else
11674 Error_Msg_NE
11681 ----------------------
11682 -- Valid_Conversion --
11683 ----------------------
11685 function Valid_Conversion
11690 is
11695 function Conversion_Check
11698 -- Little routine to post Msg if Valid is False, returns Valid value
11701 -- If Report_Errs, then calls Errout.Error_Msg_N with its arguments
11703 procedure Conversion_Error_NE
11707 -- If Report_Errs, then calls Errout.Error_Msg_NE with its arguments
11709 function Valid_Tagged_Conversion
11712 -- Specifically test for validity of tagged conversions
11715 -- Check index and component conformance, and accessibility levels if
11716 -- the component types are anonymous access types (Ada 2005).
11718 ----------------------
11719 -- Conversion_Check --
11720 ----------------------
11722 function Conversion_Check
11725 is
11726 begin
11727 if not Valid
11729 -- A generic unit has already been analyzed and we have verified
11730 -- that a particular conversion is OK in that context. Since the
11731 -- instance is reanalyzed without relying on the relationships
11732 -- established during the analysis of the generic, it is possible
11733 -- to end up with inconsistent views of private types. Do not emit
11734 -- the error message in such cases. The rest of the machinery in
11735 -- Valid_Conversion still ensures the proper compatibility of
11736 -- target and operand types.
11738 and then not In_Instance
11739 then
11746 ------------------------
11747 -- Conversion_Error_N --
11748 ------------------------
11751 begin
11757 -------------------------
11758 -- Conversion_Error_NE --
11759 -------------------------
11761 procedure Conversion_Error_NE
11765 is
11766 begin
11772 ----------------------------
11773 -- Valid_Array_Conversion --
11774 ----------------------------
11777 is
11792 begin
11793 -- Error if wrong number of dimensions
11795 if
11797 then
11798 Conversion_Error_N
11802 -- Number of dimensions matches
11804 else
11805 -- Loop through indexes of the two arrays
11813 -- Error if index types are incompatible
11819 then
11820 Conversion_Error_N
11822 Operand);
11830 -- If component types have same base type, all set
11835 -- Here if base types of components are not the same. The only
11836 -- time this is allowed is if we have anonymous access types.
11838 -- The conversion of arrays of anonymous access types can lead
11839 -- to dangling pointers. AI-392 formalizes the accessibility
11840 -- checks that must be applied to such conversions to prevent
11841 -- out-of-scope references.
11843 elsif Ekind_In
11845 E_Anonymous_Access_Subprogram_Type)
11847 and then
11849 then
11852 then
11855 Conversion_Error_N
11865 -- Conversion not allowed because of accessibility levels
11867 else
11868 Conversion_Error_N
11874 else
11878 -- All other cases where component base types do not match
11880 else
11881 Conversion_Error_N
11883 Operand);
11887 -- Check that component subtypes statically match. For numeric
11888 -- types this means that both must be either constrained or
11889 -- unconstrained. For enumeration types the bounds must match.
11890 -- All of this is checked in Subtypes_Statically_Match.
11892 if not Subtypes_Statically_Match
11894 then
11895 Conversion_Error_N
11904 -----------------------------
11905 -- Valid_Tagged_Conversion --
11906 -----------------------------
11908 function Valid_Tagged_Conversion
11911 is
11912 begin
11913 -- Upward conversions are allowed (RM 4.6(22))
11917 then
11920 -- Downward conversion are allowed if the operand is class-wide
11921 -- (RM 4.6(23)).
11925 then
11930 then
11931 return
11935 -- Ada 2005 (AI-251): The conversion to/from interface types is
11936 -- always valid. The types involved may be class-wide (sub)types.
11940 then
11943 -- If the operand is a class-wide type obtained through a limited_
11944 -- with clause, and the context includes the nonlimited view, use
11945 -- it to determine whether the conversion is legal.
11951 then
11956 then
11959 else
11960 Conversion_Error_NE
11967 -- Start of processing for Valid_Conversion
11969 begin
11973 declare
11981 begin
11982 -- Remove procedure calls, which syntactically cannot appear in
11983 -- this context, but which cannot be removed by type checking,
11984 -- because the context does not impose a type.
11986 -- The node may be labelled overloaded, but still contain only one
11987 -- interpretation because others were discarded earlier. If this
11988 -- is the case, retain the single interpretation if legal.
11996 then
11997 -- More than one candidate interpretation is available
12005 -- When compiling for a system where Address is of a visible
12006 -- integer type, spurious ambiguities can be produced when
12007 -- arithmetic operations have a literal operand and return
12008 -- System.Address or a descendant of it. These ambiguities
12009 -- are usually resolved by the context, but for conversions
12010 -- there is no context type and the removal of the spurious
12011 -- operations must be done explicitly here.
12013 if not Address_Is_Private
12015 then
12040 Conversion_Error_N
12043 -- If the interpretation involves a standard operator, use
12044 -- the location of the type, which may be user-defined.
12048 else
12052 Conversion_Error_N -- CODEFIX
12057 else
12061 Conversion_Error_N -- CODEFIX
12073 -- Deal with conversion of integer type to address if the pragma
12074 -- Allow_Integer_Address is in effect. We convert the conversion to
12075 -- an unchecked conversion in this case and we are all done.
12083 -- If we are within a child unit, check whether the type of the
12084 -- expression has an ancestor in a parent unit, in which case it
12085 -- belongs to its derivation class even if the ancestor is private.
12086 -- See RM 7.3.1 (5.2/3).
12090 -- Numeric types
12094 -- A universal fixed expression can be converted to any numeric type
12099 -- Also no need to check when in an instance or inlined body, because
12100 -- the legality has been established when the template was analyzed.
12101 -- Furthermore, numeric conversions may occur where only a private
12102 -- view of the operand type is visible at the instantiation point.
12103 -- This results in a spurious error if we check that the operand type
12104 -- is a numeric type.
12106 -- Note: in a previous version of this unit, the following tests were
12107 -- applied only for generated code (Comes_From_Source set to False),
12108 -- but in fact the test is required for source code as well, since
12109 -- this situation can arise in source code.
12114 -- Otherwise we need the conversion check
12116 else
12117 return Conversion_Check
12119 or else
12125 -- Array types
12131 then
12132 Conversion_Error_N
12136 else
12140 -- Ada 2005 (AI-251): Internally generated conversions of access to
12141 -- interface types added to force the displacement of the pointer to
12142 -- reference the corresponding dispatch table.
12147 then
12150 -- Ada 2005 (AI-251): Anonymous access types where target references an
12151 -- interface type.
12155 E_Anonymous_Access_Type)
12157 then
12158 -- Check the static accessibility rule of 4.6(17). Note that the
12159 -- check is not enforced when within an instance body, since the
12160 -- RM requires such cases to be caught at run time.
12162 -- If the operand is a rewriting of an allocator no check is needed
12163 -- because there are no accessibility issues.
12171 then
12172 -- In an instance, this is a run-time check, but one we know
12173 -- will fail, so generate an appropriate warning. The raise
12174 -- will be generated by Expand_N_Type_Conversion.
12178 Conversion_Error_N
12180 Operand);
12183 else
12184 Conversion_Error_N
12186 Operand);
12190 -- Special accessibility checks are needed in the case of access
12191 -- discriminants declared for a limited type.
12195 then
12196 -- When the operand is a selected access discriminant the check
12197 -- needs to be made against the level of the object denoted by
12198 -- the prefix of the selected name (Object_Access_Level handles
12199 -- checking the prefix of the operand for this case).
12204 then
12205 -- In an instance, this is a run-time check, but one we know
12206 -- will fail, so generate an appropriate warning. The raise
12207 -- will be generated by Expand_N_Type_Conversion.
12211 Conversion_Error_N
12216 -- Real error if not in instance body
12218 else
12219 Conversion_Error_N
12226 -- The case of a reference to an access discriminant from
12227 -- within a limited type declaration (which will appear as
12228 -- a discriminal) is always illegal because the level of the
12229 -- discriminant is considered to be deeper than any (nameable)
12230 -- access type.
12234 and then
12237 then
12238 Conversion_Error_N
12240 Operand);
12248 -- General and anonymous access types
12251 E_Anonymous_Access_Type)
12252 and then
12253 Conversion_Check
12255 and then not
12257 E_Access_Protected_Subprogram_Type),
12259 then
12262 then
12263 Conversion_Error_N
12268 -- Check the static accessibility rule of 4.6(17). Note that the
12269 -- check is not enforced when within an instance body, since the RM
12270 -- requires such cases to be caught at run time.
12275 N_Object_Declaration
12276 then
12277 -- Ada 2012 (AI05-0149): Perform legality checking on implicit
12278 -- conversions from an anonymous access type to a named general
12279 -- access type. Such conversions are not allowed in the case of
12280 -- access parameters and stand-alone objects of an anonymous
12281 -- access type. The implicit conversion case is recognized by
12282 -- testing that Comes_From_Source is False and that it's been
12283 -- rewritten. The Comes_From_Source test isn't sufficient because
12284 -- nodes in inlined calls to predefined library routines can have
12285 -- Comes_From_Source set to False. (Is there a better way to test
12286 -- for implicit conversions???)
12293 then
12296 -- Implicit conversions aren't allowed for objects of an
12297 -- anonymous access type, since such objects have nonstatic
12298 -- levels in Ada 2012.
12301 N_Object_Declaration
12302 then
12303 Conversion_Error_N
12308 -- Implicit conversions aren't allowed for anonymous access
12309 -- parameters. The "not Is_Local_Anonymous_Access_Type" test
12310 -- is done to exclude anonymous access results.
12314 N_Function_Specification,
12315 N_Procedure_Specification)
12316 then
12317 Conversion_Error_N
12322 -- This is a case where there's an enclosing object whose
12323 -- to which the "statically deeper than" relationship does
12324 -- not apply (such as an access discriminant selected from
12325 -- a dereference of an access parameter).
12329 then
12330 Conversion_Error_N
12335 -- In other cases, the level of the operand's type must be
12336 -- statically less deep than that of the target type, else
12337 -- implicit conversion is disallowed (by RM12-8.6(27.1/3)).
12341 then
12342 Conversion_Error_N
12351 then
12352 -- In an instance, this is a run-time check, but one we know
12353 -- will fail, so generate an appropriate warning. The raise
12354 -- will be generated by Expand_N_Type_Conversion.
12358 Conversion_Error_N
12360 Operand);
12363 -- If not in an instance body, this is a real error
12365 else
12366 -- Avoid generation of spurious error message
12369 Conversion_Error_N
12371 Operand);
12377 -- Special accessibility checks are needed in the case of access
12378 -- discriminants declared for a limited type.
12382 then
12383 -- When the operand is a selected access discriminant the check
12384 -- needs to be made against the level of the object denoted by
12385 -- the prefix of the selected name (Object_Access_Level handles
12386 -- checking the prefix of the operand for this case).
12391 then
12392 -- In an instance, this is a run-time check, but one we know
12393 -- will fail, so generate an appropriate warning. The raise
12394 -- will be generated by Expand_N_Type_Conversion.
12398 Conversion_Error_N
12403 -- If not in an instance body, this is a real error
12405 else
12406 Conversion_Error_N
12413 -- The case of a reference to an access discriminant from
12414 -- within a limited type declaration (which will appear as
12415 -- a discriminal) is always illegal because the level of the
12416 -- discriminant is considered to be deeper than any (nameable)
12417 -- access type.
12420 and then
12423 then
12424 Conversion_Error_N
12426 Operand);
12432 -- In the presence of limited_with clauses we have to use nonlimited
12433 -- views, if available.
12437 -- Helper function to handle limited views
12439 --------------------------
12440 -- Full_Designated_Type --
12441 --------------------------
12446 begin
12447 -- Handle the limited view of a type
12451 then
12453 else
12458 -- Local Declarations
12466 -- Start of processing for Check_Limited
12468 begin
12472 else
12474 Conversion_Error_NE
12479 -- Ada 2005 AI-384: legality rule is symmetric in both
12480 -- designated types. The conversion is legal (with possible
12481 -- constraint check) if either designated type is
12482 -- unconstrained.
12485 or else
12487 and then
12490 then
12491 -- Special case, if Value_Size has been used to make the
12492 -- sizes different, the conversion is not allowed even
12493 -- though the subtypes statically match.
12498 then
12499 Conversion_Error_NE
12502 Conversion_Error_NE
12507 -- Normal case where conversion is allowed
12509 else
12513 else
12514 Error_Msg_NE
12522 -- Access to subprogram types. If the operand is an access parameter,
12523 -- the type has a deeper accessibility that any master, and cannot be
12524 -- assigned. We must make an exception if the conversion is part of an
12525 -- assignment and the target is the return object of an extended return
12526 -- statement, because in that case the accessibility check takes place
12527 -- after the return.
12531 -- Note: this test of Opnd_Type is there to prevent entering this
12532 -- branch in the case of a remote access to subprogram type, which
12533 -- is internally represented as an E_Record_Type.
12536 then
12540 and then
12544 then
12545 Conversion_Error_N
12547 Operand);
12548 Conversion_Error_N
12551 Operand);
12553 Error_Msg_NE
12558 -- Check that the designated types are subtype conformant
12564 -- Check the static accessibility rule of 4.6(20)
12568 then
12569 Conversion_Error_N
12571 Operand);
12573 -- Check that if the operand type is declared in a generic body,
12574 -- then the target type must be declared within that same body
12575 -- (enforces last sentence of 4.6(20)).
12578 declare
12584 begin
12591 Conversion_Error_N
12600 -- Remote access to subprogram types
12604 then
12605 -- It is valid to convert from one RAS type to another provided
12606 -- that their specification statically match.
12608 -- Note: at this point, remote access to subprogram types have been
12609 -- expanded to their E_Record_Type representation, and we need to
12610 -- go back to the original access type definition using the
12611 -- Corresponding_Remote_Type attribute in order to check that the
12612 -- designated profiles match.
12617 Check_Subtype_Conformant
12620 Old_Id =>
12622 Err_Loc =>
12623 N);
12626 -- If it was legal in the generic, it's legal in the instance
12631 -- If both are tagged types, check legality of view conversions
12634 and then
12636 then
12639 -- Types derived from the same root type are convertible
12644 -- In an instance or an inlined body, there may be inconsistent views of
12645 -- the same type, or of types derived from a common root.
12648 and then
12651 then
12654 -- Special check for common access type error case
12658 then
12660 Conversion_Error_NE -- CODEFIX
12664 -- Here we have a real conversion error
12666 else
12667 Conversion_Error_NE