1 ------------------------------------------------------------------------------
3 -- GNAT COMPILER COMPONENTS --
9 -- Copyright (C) 1992-2005, Free Software Foundation, Inc. --
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 2, 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 COPYING. If not, write --
19 -- to the Free Software Foundation, 51 Franklin Street, Fifth Floor, --
20 -- Boston, MA 02110-1301, USA. --
22 -- GNAT was originally developed by the GNAT team at New York University. --
23 -- Extensive contributions were provided by Ada Core Technologies Inc. --
25 ------------------------------------------------------------------------------
27 -- Attribute handling is isolated in a separate package to ease the addition
28 -- of implementation defined attributes. Logically this processing belongs
29 -- in chapter 4. See Sem_Ch4 for a description of the relation of the
30 -- Analyze and Resolve routines for expression components.
32 -- This spec also documents all GNAT implementation defined pragmas
34 with Exp_Tss; use Exp_Tss;
35 with Snames; use Snames;
36 with Types; use Types;
40 type Attribute_Class_Array is array (Attribute_Id) of Boolean;
41 -- Type used to build attribute classification flag arrays
43 -----------------------------------------
44 -- Implementation Dependent Attributes --
45 -----------------------------------------
47 -- This section describes the implementation dependent attributes
48 -- provided in GNAT, as well as constructing an array of flags
49 -- indicating which attributes these are.
51 Attribute_Impl_Def : Attribute_Class_Array := Attribute_Class_Array'(
57 Attribute_Abort_Signal => True,
58 -- Standard'Abort_Signal (Standard is the only allowed prefix) provides
59 -- the entity for the special exception used to signal task abort or
60 -- asynchronous transfer of control. Normally this attribute should only
61 -- be used in the tasking runtime (it is highly peculiar, and completely
62 -- outside the normal semantics of Ada, for a user program to intercept
63 -- the abort exception).
69 Attribute_Address_Size => True,
70 -- Standard'Address_Size (Standard is the only allowed prefix) is
71 -- a static constant giving the number of bits in an Address. It
72 -- is used primarily for constructing the definition of Memory_Size
73 -- in package Standard, but may be freely used in user programs.
74 -- This is a static attribute.
80 Attribute_Asm_Input => True,
81 -- Used only in conjunction with the Asm and Asm_Volatile subprograms
82 -- in package Machine_Code to construct machine instructions. See
83 -- documentation in package Machine_Code in file s-maccod.ads.
89 Attribute_Asm_Output => True,
90 -- Used only in conjunction with the Asm and Asm_Volatile subprograms
91 -- in package Machine_Code to construct machine instructions. See
92 -- documentation in package Machine_Code in file s-maccod.ads.
98 Attribute_AST_Entry => True,
99 -- E'Ast_Entry, where E is a task entry, yields a value of the
100 -- predefined type System.DEC.AST_Handler, that enables the given
101 -- entry to be called when an AST occurs. If the name to which the
102 -- attribute applies has not been specified with the pragma AST_Entry,
103 -- the attribute returns the value No_Ast_Handler, and no AST occurs.
104 -- If the entry is for a task that is not callable (T'Callable False),
105 -- the exception program error is raised. If an AST occurs for an
106 -- entry of a task that is terminated, the program is erroneous.
108 -- The attribute AST_Entry is supported only in OpenVMS versions
109 -- of GNAT. It will be rejected as illegal in other GNAT versions.
115 Attribute_Bit => True,
116 -- Obj'Bit, where Obj is any object, yields the bit offset within the
117 -- storage unit (byte) that contains the first bit of storage allocated
118 -- for the object. The attribute value is of type Universal_Integer,
119 -- and is always a non-negative number not exceeding the value of
120 -- System.Storage_Unit.
122 -- For an object that is a variable or a constant allocated in a
123 -- register, the value is zero. (The use of this attribute does not
124 -- force the allocation of a variable to memory).
126 -- For an object that is a formal parameter, this attribute applies to
127 -- either the matching actual parameter or to a copy of the matching
130 -- For an access object the value is zero. Note that Obj.all'Bit is
131 -- subject to an Access_Check for the designated object. Similarly
132 -- for a record component X.C'Bit is subject to a discriminant check
133 -- and X(I).Bit and X(I1..I2)'Bit are subject to index checks.
135 -- This attribute is designed to be compatible with the DEC Ada
136 -- definition and implementation of the Bit attribute.
142 Attribute_Code_Address => True,
143 -- The reference subp'Code_Address, where subp is a subprogram entity,
144 -- gives the address of the first generated instruction for the sub-
145 -- program. This is often, but not always the same as the 'Address
146 -- value, which is the address to be used in a call. The differences
147 -- occur in the case of a nested procedure (where Address yields the
148 -- address of the trampoline code used to load the static link), and on
149 -- some systems which use procedure descriptors (in which case Address
150 -- yields the address of the descriptor).
152 -----------------------
153 -- Default_Bit_Order --
154 -----------------------
156 Attribute_Default_Bit_Order => True,
157 -- Standard'Default_Bit_Order (Standard is the only permissible prefix),
158 -- provides the value System.Default_Bit_Order as a Pos value (0 for
159 -- High_Order_First, 1 for Low_Order_First). This is used to construct
160 -- the definition of Default_Bit_Order in package System. This is a
167 Attribute_Elab_Body => True,
168 -- This attribute can only be applied to a program unit name. It returns
169 -- the entity for the corresponding elaboration procedure for elabor-
170 -- ating the body of the referenced unit. This is used in the main
171 -- generated elaboration procedure by the binder, and is not normally
172 -- used in any other context, but there may be specialized situations in
173 -- which it is useful to be able to call this elaboration procedure from
174 -- Ada code, e.g. if it is necessary to do selective reelaboration to
181 Attribute_Elab_Spec => True,
182 -- This attribute can only be applied to a program unit name. It
183 -- returns the entity for the corresponding elaboration procedure
184 -- for elaborating the spec of the referenced unit. This is used
185 -- in the main generated elaboration procedure by the binder, and
186 -- is not normally used in any other context, but there may be
187 -- specialized situations in which it is useful to be able to
188 -- call this elaboration procedure from Ada code, e.g. if it
189 -- is necessary to do selective reelaboration to fix some error.
195 Attribute_Elaborated => True,
196 -- Lunit'Elaborated, where Lunit is a library unit, yields a boolean
197 -- value indicating whether or not the body of the designated library
198 -- unit has been elaborated yet.
204 Attribute_Enum_Rep => True,
205 -- For every enumeration subtype S, S'Enum_Rep denotes a function
206 -- with the following specification:
208 -- function S'Enum_Rep (Arg : S'Base) return universal_integer;
210 -- The function returns the representation value for the given
211 -- enumeration value. This will be equal to the 'Pos value in the
212 -- absence of an enumeration representation clause. This is a static
213 -- attribute (i.e. the result is static if the argument is static).
219 Attribute_Fixed_Value => True,
220 -- For every fixed-point type S, S'Fixed_Value denotes a function
221 -- with the following specification:
223 -- function S'Fixed_Value (Arg : universal_integer) return S;
225 -- The value returned is the fixed-point value V such that
229 -- The effect is thus equivalent to first converting the argument to
230 -- the integer type used to represent S, and then doing an unchecked
231 -- conversion to the fixed-point type. This attribute is primarily
232 -- intended for use in implementation of the input-output functions for
233 -- fixed-point values.
235 -----------------------
236 -- Has_Discriminants --
237 -----------------------
239 Attribute_Has_Discriminants => True,
240 -- Gtyp'Has_Discriminants, where Gtyp is a generic formal type, yields
241 -- a Boolean value indicating whether or not the actual instantiation
242 -- type has discriminants.
248 Attribute_Img => True,
249 -- The 'Img function is defined for any prefix, P, that denotes an
250 -- object of scalar type T. P'Img is equivalent to T'Image (P). This
251 -- is convenient for debugging. For example:
253 -- Put_Line ("X = " & X'Img);
255 -- has the same meaning as the more verbose:
257 -- Put_Line ("X = " & Temperature_Type'Image (X));
259 -- where Temperature_Type is the subtype of the object X.
265 Attribute_Integer_Value => True,
266 -- For every integer type S, S'Integer_Value denotes a function
267 -- with the following specification:
269 -- function S'Integer_Value (Arg : universal_fixed) return S;
271 -- The value returned is the integer value V, such that
273 -- Arg = V * fixed-type'Small
275 -- The effect is thus equivalent to first doing an unchecked convert
276 -- from the fixed-point type to its corresponding implementation type,
277 -- and then converting the result to the target integer type. This
278 -- attribute is primarily intended for use in implementation of the
279 -- standard input-output functions for fixed-point values.
285 Attribute_Machine_Size => True,
286 -- This attribute is identical to the Object_Size attribute. It is
287 -- provided for compatibility with the DEC attribute of this name.
289 -----------------------
290 -- Maximum_Alignment --
291 -----------------------
293 Attribute_Maximum_Alignment => True,
294 -- Standard'Maximum_Alignment (Standard is the only permissible prefix)
295 -- provides the maximum useful alignment value for the target. This
296 -- is a static value that can be used to specify the alignment for an
297 -- object, guaranteeing that it is properly aligned in all cases. The
298 -- time this is useful is when an external object is imported and its
299 -- alignment requirements are unknown. This is a static attribute.
305 Attribute_Mechanism_Code => True,
306 -- function'Mechanism_Code yeilds an integer code for the mechanism
307 -- used for the result of function, and subprogram'Mechanism_Code (n)
308 -- yields the mechanism used for formal parameter number n (a static
309 -- integer value, 1 = first parameter). The code returned is:
311 -- 1 = by copy (value)
313 -- 3 = by descriptor (default descriptor type)
314 -- 4 = by descriptor (UBS unaligned bit string)
315 -- 5 = by descriptor (UBSB aligned bit string with arbitrary bounds)
316 -- 6 = by descriptor (UBA unaligned bit array)
317 -- 7 = by descriptor (S string, also scalar access type parameter)
318 -- 8 = by descriptor (SB string with arbitrary bounds)
319 -- 9 = by descriptor (A contiguous array)
320 -- 10 = by descriptor (NCA non-contiguous array)
326 Attribute_Null_Parameter => True,
327 -- A reference T'Null_Parameter denotes an (imaginary) object of type or
328 -- subtype T allocated at (machine) address zero. The attribute is
329 -- allowed only as the default expression of a formal parameter, or as
330 -- an actual expression of a subporgram call. In either case, the
331 -- subprogram must be imported.
333 -- The identity of the object is represented by the address zero in the
334 -- argument list, independent of the passing mechanism (explicit or
337 -- The reason that this capability is needed is that for a record or
338 -- other composite object passed by reference, there is no other way of
339 -- specifying that a zero address should be passed.
345 Attribute_Object_Size => True,
346 -- Type'Object_Size is the same as Type'Size for all types except
347 -- fixed-point types and discrete types. For fixed-point types and
348 -- discrete types, this attribute gives the size used for default
349 -- allocation of objects and components of the size. See section in
350 -- Einfo ("Handling of type'Size values") for further details.
352 -------------------------
353 -- Passed_By_Reference --
354 -------------------------
356 Attribute_Passed_By_Reference => True,
357 -- T'Passed_By_Reference for any subtype T returns a boolean value that
358 -- is true if the type is normally passed by reference and false if the
359 -- type is normally passed by copy in calls. For scalar types, the
360 -- result is always False and is static. For non-scalar types, the
361 -- result is non-static (since it is computed by Gigi).
367 Attribute_Range_Length => True,
368 -- T'Range_Length for any discrete type T yields the number of values
369 -- represented by the subtype (zero for a null range). The result is
370 -- static for static subtypes. Note that Range_Length applied to the
371 -- index subtype of a one dimensional array always gives the same result
372 -- as Range applied to the array itself. The result is of type universal
379 Attribute_Storage_Unit => True,
380 -- Standard'Storage_Unit (Standard is the only permissible prefix)
381 -- provides the value System.Storage_Unit, and is intended primarily
382 -- for constructing this definition in package System (see note above
383 -- in Default_Bit_Order description). The is a static attribute.
389 Attribute_Target_Name => True,
390 -- Standard'Target_Name yields the string identifying the target for the
391 -- compilation, taken from Sdefault.Target_Name.
397 Attribute_To_Address => True,
398 -- System'To_Address (Address is the only permissible prefix) is a
399 -- function that takes any integer value, and converts it into an
400 -- address value. The semantics is to first convert the integer value to
401 -- type Integer_Address according to normal conversion rules, and then
402 -- to convert this to an address using the same semantics as the
403 -- System.Storage_Elements.To_Address function. The important difference
404 -- is that this is a static attribute so it can be used in
405 -- initializations in preealborate packages.
411 Attribute_Type_Class => True,
412 -- T'Type_Class for any type or subtype T yields the value of the type
413 -- class for the full type of T. If T is a generic formal type, then the
414 -- value is the value for the corresponding actual subtype. The value of
415 -- this attribute is of type System.Aux_DEC.Type_Class, which has the
416 -- following definition:
418 -- type Type_Class is
419 -- (Type_Class_Enumeration,
420 -- Type_Class_Integer,
421 -- Type_Class_Fixed_Point,
422 -- Type_Class_Floating_Point,
424 -- Type_Class_Record,
425 -- Type_Class_Access,
427 -- Type_Class_Address);
429 -- Protected types yield the value Type_Class_Task, which thus applies
430 -- to all concurrent types. This attribute is designed to be compatible
431 -- with the DEC Ada attribute of the same name.
433 -- Note: if pragma Extend_System is used to merge the definitions of
434 -- Aux_DEC into System, then the type Type_Class can be referenced
435 -- as an entity within System, as can its enumeration literals.
441 Attribute_UET_Address => True,
442 -- Unit'UET_Address, where Unit is a program unit, yields the address
443 -- of the unit exception table for the specified unit. This is only
444 -- used in the internal implementation of exception handling. See the
445 -- implementation of unit Ada.Exceptions for details on its use.
447 ------------------------------
448 -- Universal_Literal_String --
449 ------------------------------
451 Attribute_Universal_Literal_String => True,
452 -- The prefix of 'Universal_Literal_String must be a named number. The
453 -- static result is the string consisting of the characters of the
454 -- number as defined in the original source. This allows the user
455 -- program to access the actual text of named numbers without
456 -- intermediate conversions and without the need to enclose the strings
457 -- in quotes (which would preclude their use as numbers). This is used
458 -- internally for the construction of values of the floating-point
459 -- attributes from the file ttypef.ads, but may also be used by user
462 -------------------------
463 -- Unrestricted_Access --
464 -------------------------
466 Attribute_Unrestricted_Access => True,
467 -- The Unrestricted_Access attribute is similar to Access except that
468 -- all accessibility and aliased view checks are omitted. This is very
469 -- much a user-beware attribute. Basically its status is very similar
470 -- to Address, for which it is a desirable replacement where the value
471 -- desired is an access type. In other words, its effect is identical
472 -- to first taking 'Address and then doing an unchecked conversion to
473 -- a desired access type. Note that in GNAT, but not necessarily in
474 -- other implementations, the use of static chains for inner level
475 -- subprograms means that Unrestricted_Access applied to a subprogram
476 -- yields a value that can be called as long as the subprogram is in
477 -- scope (normal Ada 95 accessibility rules restrict this usage).
483 Attribute_VADS_Size => True,
484 -- Typ'VADS_Size yields the Size value typically yielded by some Ada 83
485 -- compilers. The differences between VADS_Size and Size is that for
486 -- scalar types for which no Size has been specified, VADS_Size yields
487 -- the Object_Size rather than the Value_Size. For example, while
488 -- Natural'Size is typically 31, the value of Natural'VADS_Size is 32.
489 -- For all other types, Size and VADS_Size yield the same value.
495 Attribute_Value_Size => True,
496 -- Type'Value_Size is the number of bits required to represent value of
497 -- the given subtype. It is the same as Type'Size, but, unlike Size, may
498 -- be set for non-first subtypes. See section in Einfo ("Handling of
499 -- type'Size values") for further details.
505 Attribute_Word_Size => True,
506 -- Standard'Word_Size (Standard is the only permissible prefix)
507 -- provides the value System.Word_Size, and is intended primarily
508 -- for constructing this definition in package System (see note above
509 -- in Default_Bit_Order description). This is a static attribute.
517 procedure Analyze_Attribute (N : Node_Id);
518 -- Performs bottom up semantic analysis of an attribute. Note that the
519 -- parser has already checked that type returning attributes appear only
520 -- in appropriate contexts (i.e. in subtype marks, or as prefixes for
521 -- other attributes).
523 procedure Resolve_Attribute (N : Node_Id; Typ : Entity_Id);
524 -- Performs type resolution of attribute. If the attribute yields a
525 -- universal value, mark its type as that of the context. On the other
526 -- hand, if the context itself is universal (as in T'Val (T'Pos (X)), mark
527 -- the type as being the largest type of that class that can be used at
528 -- run-time. This is correct since either the value gets folded (in which
529 -- case it doesn't matter what type of the class we give if, since the
530 -- folding uses universal arithmetic anyway) or it doesn't get folded (in
531 -- which case it is going to be dealt with at runtime, and the largest type
534 function Stream_Attribute_Available
537 Partial_View : Entity_Id := Empty) return Boolean;
538 -- For a limited type Typ, return True iff the given attribute is
539 -- available. For Ada 05, availability is defined by 13.13.2(36/1). For Ada
540 -- 95, an attribute is considered to be available if it has been specified
541 -- using an attribute definition clause for the type, or for its full view,
542 -- or for an ancestor of either. Parameter Partial_View is used only
543 -- internally, when checking for an attribute definition clause that is not
544 -- visible (Ada 95 only).
OSDN Git Service
