1 ------------------------------------------------------------------------------
3 -- GNAT COMPILER COMPONENTS --
9 -- Copyright (C) 1992-2003 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, 59 Temple Place - Suite 330, Boston, --
20 -- MA 02111-1307, 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 with Atree; use Atree;
28 with Casing; use Casing;
29 with Csets; use Csets;
30 with Debug; use Debug;
31 with Elists; use Elists;
32 with Errout; use Errout;
33 with Fname; use Fname;
35 with Namet; use Namet;
36 with Nlists; use Nlists;
37 with Nmake; use Nmake;
39 with Output; use Output;
40 with Scans; use Scans;
42 with Sinput; use Sinput;
43 with Sinput.L; use Sinput.L;
44 with Sinfo; use Sinfo;
45 with Snames; use Snames;
48 with Tbuild; use Tbuild;
50 function Par (Configuration_Pragmas : Boolean) return List_Id is
52 Num_Library_Units : Natural := 0;
53 -- Count number of units parsed (relevant only in syntax check only mode,
54 -- since in semantics check mode only a single unit is permitted anyway)
56 Save_Config_Switches : Config_Switches_Type;
57 -- Variable used to save values of config switches while we parse the
58 -- new unit, to be restored on exit for proper recursive behavior.
60 Loop_Block_Count : Nat := 0;
61 -- Counter used for constructing loop/block names (see the routine
62 -- Par.Ch5.Get_Loop_Block_Name)
68 -- When an error is encountered, a call is made to one of the Error_Msg
69 -- routines to record the error. If the syntax scan is not derailed by the
70 -- error (e.g. a complaint that logical operators are inconsistent in an
71 -- EXPRESSION), then control returns from the Error_Msg call, and the
72 -- parse continues unimpeded.
74 -- If on the other hand, the Error_Msg represents a situation from which
75 -- the parser cannot recover locally, the exception Error_Resync is raised
76 -- immediately after the call to Error_Msg. Handlers for Error_Resync
77 -- are located at strategic points to resynchronize the parse. For example,
78 -- when an error occurs in a statement, the handler skips to the next
79 -- semicolon and continues the scan from there.
81 -- Each parsing procedure contains a note with the heading "Error recovery"
82 -- which shows if it can propagate the Error_Resync exception. In order
83 -- not to propagate the exception, a procedure must either contain its own
84 -- handler for this exception, or it must not call any other routines which
85 -- propagate the exception.
87 -- Note: the arrangement of Error_Resync handlers is such that it should
88 -- never be possible to transfer control through a procedure which made
89 -- an entry in the scope stack, invalidating the contents of the stack.
91 Error_Resync : exception;
92 -- Exception raised on error that is not handled locally, see above.
94 Last_Resync_Point : Source_Ptr;
95 -- The resynchronization routines in Par.Sync run a risk of getting
96 -- stuck in an infinite loop if they do not skip a token, and the caller
97 -- keeps repeating the same resync call. On the other hand, if they skip
98 -- a token unconditionally, some recovery opportunities are missed. The
99 -- variable Last_Resync_Point records the token location previously set
100 -- by a Resync call, and if a subsequent Resync call occurs at the same
101 -- location, then the Resync routine does guarantee to skip a token.
103 --------------------------------------------
104 -- Handling Semicolon Used in Place of IS --
105 --------------------------------------------
107 -- The following global variables are used in handling the error situation
108 -- of using a semicolon in place of IS in a subprogram declaration as in:
110 -- procedure X (Y : Integer);
116 -- The two contexts in which this can appear are at the outer level, and
117 -- within a declarative region. At the outer level, we know something is
118 -- wrong as soon as we see the Q (or begin, if there are no declarations),
119 -- and we can immediately decide that the semicolon should have been IS.
121 -- The situation in a declarative region is more complex. The declaration
122 -- of Q could belong to the outer region, and we do not know that we have
123 -- an error until we hit the begin. It is still not clear at this point
124 -- from a syntactic point of view that something is wrong, because the
125 -- begin could belong to the enclosing subprogram or package. However, we
126 -- can incorporate a bit of semantic knowledge and note that the body of
127 -- X is missing, so we definitely DO have an error. We diagnose this error
128 -- as semicolon in place of IS on the subprogram line.
130 -- There are two styles for this diagnostic. If the begin immediately
131 -- follows the semicolon, then we can place a flag (IS expected) right
132 -- on the semicolon. Otherwise we do not detect the error until we hit
133 -- the begin which refers back to the line with the semicolon.
135 -- To control the process in the second case, the following global
136 -- variables are set to indicate that we have a subprogram declaration
137 -- whose body is required and has not yet been found. The prefix SIS
138 -- stands for "Subprogram IS" handling.
140 SIS_Entry_Active : Boolean;
141 -- Set True to indicate that an entry is active (i.e. that a subprogram
142 -- declaration has been encountered, and no body for this subprogram has
143 -- been encountered). The remaining fields are valid only if this is True.
146 -- Subprogram designator
148 SIS_Sloc : Source_Ptr;
149 -- Source location of FUNCTION/PROCEDURE keyword
151 SIS_Ecol : Column_Number;
152 -- Column number of FUNCTION/PROCEDURE keyword
154 SIS_Semicolon_Sloc : Source_Ptr;
155 -- Source location of semicolon at end of subprogram declaration
157 SIS_Declaration_Node : Node_Id;
158 -- Pointer to tree node for subprogram declaration
160 SIS_Missing_Semicolon_Message : Error_Msg_Id;
161 -- Used to save message ID of missing semicolon message (which will be
162 -- modified to missing IS if necessary). Set to No_Error_Msg in the
163 -- normal (non-error) case.
165 -- Five things can happen to an active SIS entry
167 -- 1. If a BEGIN is encountered with an SIS entry active, then we have
168 -- exactly the situation in which we know the body of the subprogram is
169 -- missing. After posting an error message, we change the spec to a body,
170 -- rechaining the declarations that intervened between the spec and BEGIN.
172 -- 2. Another subprogram declaration or body is encountered. In this
173 -- case the entry gets overwritten with the information for the new
174 -- subprogram declaration. We don't catch some nested cases this way,
175 -- but it doesn't seem worth the effort.
177 -- 3. A nested declarative region (e.g. package declaration or package
178 -- body) is encountered. The SIS active indication is reset at the start
179 -- of such a nested region. Again, like case 2, this causes us to miss
180 -- some nested cases, but it doesn't seen worth the effort to stack and
181 -- unstack the SIS information. Maybe we will reconsider this if we ever
182 -- get a complaint about a missed case :-)
184 -- 4. We encounter a valid pragma INTERFACE or IMPORT that effectively
185 -- supplies the missing body. In this case we reset the entry.
187 -- 5. We encounter the end of the declarative region without encoutering
188 -- a BEGIN first. In this situation we simply reset the entry. We know
189 -- that there is a missing body, but it seems more reasonable to let the
190 -- later semantic checking discover this.
192 --------------------------------------------
193 -- Handling IS Used in Place of Semicolon --
194 --------------------------------------------
196 -- This is a somewhat trickier situation, and we can't catch it in all
197 -- cases, but we do our best to detect common situations resulting from
198 -- a "cut and paste" operation which forgets to change the IS to semicolon.
199 -- Consider the following example:
214 -- The trouble is that the section of text from PROCEDURE B through END;
215 -- consitutes a valid procedure body, and the danger is that we find out
216 -- far too late that something is wrong (indeed most compilers will behave
217 -- uncomfortably on the above example).
219 -- We have two approaches to helping to control this situation. First we
220 -- make every attempt to avoid swallowing the last END; if we can be
221 -- sure that some error will result from doing so. In particular, we won't
222 -- accept the END; unless it is exactly correct (in particular it must not
223 -- have incorrect name tokens), and we won't accept it if it is immediately
224 -- followed by end of file, WITH or SEPARATE (all tokens that unmistakeably
225 -- signal the start of a compilation unit, and which therefore allow us to
226 -- reserve the END; for the outer level.) For more details on this aspect
227 -- of the handling, see package Par.Endh.
229 -- If we can avoid eating up the END; then the result in the absense of
230 -- any additional steps would be to post a missing END referring back to
231 -- the subprogram with the bogus IS. Similarly, if the enclosing package
232 -- has no BEGIN, then the result is a missing BEGIN message, which again
233 -- refers back to the subprogram header.
235 -- Such an error message is not too bad (it's already a big improvement
236 -- over what many parsers do), but it's not ideal, because the declarations
237 -- following the IS have been absorbed into the wrong scope. In the above
238 -- case, this could result for example in a bogus complaint that the body
239 -- of D was missing from the package.
241 -- To catch at least some of these cases, we take the following additional
242 -- steps. First, a subprogram body is marked as having a suspicious IS if
243 -- the declaration line is followed by a line which starts with a symbol
244 -- that can start a declaration in the same column, or to the left of the
245 -- column in which the FUNCTION or PROCEDURE starts (normal style is to
246 -- indent any declarations which really belong a subprogram). If such a
247 -- subprogram encounters a missing BEGIN or missing END, then we decide
248 -- that the IS should have been a semicolon, and the subprogram body node
249 -- is marked (by setting the Bad_Is_Detected flag true. Note that we do
250 -- not do this for library level procedures, only for nested procedures,
251 -- since for library level procedures, we must have a body.
253 -- The processing for a declarative part checks to see if the last
254 -- declaration scanned is marked in this way, and if it is, the tree
255 -- is modified to reflect the IS being interpreted as a semicolon.
257 ---------------------------------------------------
258 -- Parser Type Definitions and Control Variables --
259 ---------------------------------------------------
261 -- The following variable and associated type declaration are used by the
262 -- expression parsing routines to return more detailed information about
263 -- the categorization of a parsed expression.
265 type Expr_Form_Type is (
266 EF_Simple_Name, -- Simple name, i.e. possibly qualified identifier
267 EF_Name, -- Simple expression which could also be a name
268 EF_Simple, -- Simple expression which is not call or name
269 EF_Range_Attr, -- Range attribute reference
270 EF_Non_Simple); -- Expression that is not a simple expression
272 Expr_Form : Expr_Form_Type;
274 -- The following type is used for calls to P_Subprogram, P_Package, P_Task,
275 -- P_Protected to indicate which of several possibilities is acceptable.
277 type Pf_Rec is record
278 Spcn : Boolean; -- True if specification OK
279 Decl : Boolean; -- True if declaration OK
280 Gins : Boolean; -- True if generic instantiation OK
281 Pbod : Boolean; -- True if proper body OK
282 Rnam : Boolean; -- True if renaming declaration OK
283 Stub : Boolean; -- True if body stub OK
284 Fil1 : Boolean; -- Filler to fill to 8 bits
285 Fil2 : Boolean; -- Filler to fill to 8 bits
287 pragma Pack (Pf_Rec);
289 function T return Boolean renames True;
290 function F return Boolean renames False;
292 Pf_Decl_Gins_Pbod_Rnam_Stub : constant Pf_Rec :=
293 Pf_Rec'(F, T, T, T, T, T, F, F);
294 Pf_Decl : constant Pf_Rec :=
295 Pf_Rec'(F, T, F, F, F, F, F, F);
296 Pf_Decl_Gins_Pbod_Rnam : constant Pf_Rec :=
297 Pf_Rec'(F, T, T, T, T, F, F, F);
298 Pf_Decl_Pbod : constant Pf_Rec :=
299 Pf_Rec'(F, T, F, T, F, F, F, F);
300 Pf_Pbod : constant Pf_Rec :=
301 Pf_Rec'(F, F, F, T, F, F, F, F);
302 Pf_Spcn : constant Pf_Rec :=
303 Pf_Rec'(T, F, F, F, F, F, F, F);
304 -- The above are the only allowed values of Pf_Rec arguments
306 type SS_Rec is record
307 Eftm : Boolean; -- ELSIF can terminate sequence
308 Eltm : Boolean; -- ELSE can terminate sequence
309 Extm : Boolean; -- EXCEPTION can terminate sequence
310 Ortm : Boolean; -- OR can terminate sequence
311 Sreq : Boolean; -- at least one statement required
312 Tatm : Boolean; -- THEN ABORT can terminate sequence
313 Whtm : Boolean; -- WHEN can terminate sequence
314 Unco : Boolean; -- Unconditional terminate after one statement
316 pragma Pack (SS_Rec);
318 SS_Eftm_Eltm_Sreq : constant SS_Rec := SS_Rec'(T, T, F, F, T, F, F, F);
319 SS_Eltm_Ortm_Tatm : constant SS_Rec := SS_Rec'(F, T, F, T, F, T, F, F);
320 SS_Extm_Sreq : constant SS_Rec := SS_Rec'(F, F, T, F, T, F, F, F);
321 SS_None : constant SS_Rec := SS_Rec'(F, F, F, F, F, F, F, F);
322 SS_Ortm_Sreq : constant SS_Rec := SS_Rec'(F, F, F, T, T, F, F, F);
323 SS_Sreq : constant SS_Rec := SS_Rec'(F, F, F, F, T, F, F, F);
324 SS_Sreq_Whtm : constant SS_Rec := SS_Rec'(F, F, F, F, T, F, T, F);
325 SS_Whtm : constant SS_Rec := SS_Rec'(F, F, F, F, F, F, T, F);
326 SS_Unco : constant SS_Rec := SS_Rec'(F, F, F, F, F, F, F, T);
328 Label_List : Elist_Id;
329 -- List of label nodes for labels appearing in the current compilation.
330 -- Used by Par.Labl to construct the corresponding implicit declarations.
336 -- The scope table, also referred to as the scope stack, is used to
337 -- record the current scope context. It is organized as a stack, with
338 -- inner nested entries corresponding to higher entries on the stack.
339 -- An entry is made when the parser encounters the opening of a nested
340 -- construct (such as a record, task, package etc.), and then package
341 -- Par.Endh uses this stack to deal with END lines (including properly
342 -- dealing with END nesting errors).
345 -- Type of end entry required for this scope. The last two entries are
346 -- used only in the subprogram body case to mark the case of a suspicious
347 -- IS, or a bad IS (i.e. suspicions confirmed by missing BEGIN or END).
348 -- See separate section on dealing with IS used in place of semicolon.
349 -- Note that for many purposes E_Name, E_Suspicious_Is and E_Bad_Is are
350 -- treated the same (E_Suspicious_Is and E_Bad_Is are simply special cases
351 -- of E_Name). They are placed at the end of the enumeration so that a
352 -- test for >= E_Name catches all three cases efficiently.
354 (E_Dummy, -- dummy entry at outer level
358 E_Record, -- END RECORD;
359 E_Select, -- END SELECT;
360 E_Name, -- END [name];
361 E_Suspicious_Is, -- END [name]; (case of suspicious IS)
362 E_Bad_Is); -- END [name]; (case of bad IS)
364 -- The following describes a single entry in the scope table
366 type Scope_Table_Entry is record
368 -- Type of end entry, as per above description
371 -- A flag indicating whether the label, if present, is required to
372 -- appear on the end line. It is referenced only in the case of
373 -- Etyp = E_Name or E_Suspicious_Is where the name may or may not be
374 -- required (yes for labeled block, no in other cases). Note that for
375 -- all cases except begin, the question of whether a label is required
376 -- can be determined from the other fields (for loop, it is required if
377 -- it is present, and for the other constructs it is never required or
380 Ecol : Column_Number;
381 -- Contains the absolute column number (with tabs expanded) of the
382 -- the expected column of the end assuming normal Ada indentation
383 -- usage. If the RM_Column_Check mode is set, this value is used for
384 -- generating error messages about indentation. Otherwise it is used
385 -- only to control heuristic error recovery actions.
388 -- This field is used only for the LOOP and BEGIN cases, and is the
389 -- Node_Id value of the label name. For all cases except child units,
390 -- this value is an entity whose Chars field contains the name pointer
391 -- that identifies the label uniquely. For the child unit case the Labl
392 -- field references an N_Defining_Program_Unit_Name node for the name.
393 -- For cases other than LOOP or BEGIN, the Label field is set to Error,
394 -- indicating that it is an error to have a label on the end line.
395 -- (this is really a misuse of Error since there is no Error ???)
398 -- Points to the list of declarations (i.e. the declarative part)
399 -- associated with this construct. It is set only in the END [name]
400 -- cases, and is set to No_List for all other cases which do not have a
401 -- declarative unit associated with them. This is used for determining
402 -- the proper location for implicit label declarations.
405 -- Empty except in the case of entries for IF and CASE statements,
406 -- in which case it contains the N_If_Statement or N_Case_Statement
407 -- node. This is used for setting the End_Span field.
410 -- Source location of the opening token of the construct. This is
411 -- used to refer back to this line in error messages (such as missing
412 -- or incorrect end lines). The Sloc field is not used, and is not set,
413 -- if a label is present (the Labl field provides the text name of the
414 -- label in this case, which is fine for error messages).
417 -- S_Is is relevant only if Etyp is set to E_Suspicious_Is or
418 -- E_Bad_Is. It records the location of the IS that is considered
422 -- A boolean flag that is set true if the opening entry is the dubious
423 -- result of some prior error, e.g. a record entry where the record
424 -- keyword was missing. It is used to suppress the issuing of a
425 -- corresponding junk complaint about the end line (we do not want
426 -- to complain about a missing end record when there was no record).
429 -- The following declares the scope table itself. The Last field is the
430 -- stack pointer, so that Scope.Table (Scope.Last) is the top entry. The
431 -- oldest entry, at Scope_Stack (0), is a dummy entry with Etyp set to
432 -- E_Dummy, and the other fields undefined. This dummy entry ensures that
433 -- Scope_Stack (Scope_Stack_Ptr).Etyp can always be tested, and that the
434 -- scope stack pointer is always in range.
436 package Scope is new Table.Table (
437 Table_Component_Type => Scope_Table_Entry,
438 Table_Index_Type => Int,
439 Table_Low_Bound => 0,
441 Table_Increment => 100,
442 Table_Name => "Scope");
444 ---------------------------------
445 -- Parsing Routines by Chapter --
446 ---------------------------------
448 -- Uncommented declarations in this section simply parse the construct
449 -- corresponding to their name, and return an ID value for the Node or
450 -- List that is created.
453 function P_Identifier return Node_Id;
454 function P_Pragma return Node_Id;
456 function P_Pragmas_Opt return List_Id;
457 -- This function scans for a sequence of pragmas in other than a
458 -- declaration sequence or statement sequence context. All pragmas
459 -- can appear except pragmas Assert and Debug, which are only allowed
460 -- in a declaration or statement sequence context.
462 procedure P_Pragmas_Misplaced;
463 -- Skips misplaced pragmas with a complaint
465 procedure P_Pragmas_Opt (List : List_Id);
466 -- Parses optional pragmas and appends them to the List
470 Missing_Begin_Msg : Error_Msg_Id;
471 -- This variable is set by a call to P_Declarative_Part. Normally it
472 -- is set to No_Error_Msg, indicating that no special processing is
473 -- required by the caller. The special case arises when a statement
474 -- is found in the sequence of declarations. In this case the Id of
475 -- the message issued ("declaration expected") is preserved in this
476 -- variable, then the caller can change it to an appropriate missing
477 -- begin message if indeed the BEGIN is missing.
479 function P_Access_Definition return Node_Id;
480 function P_Access_Type_Definition return Node_Id;
481 function P_Array_Type_Definition return Node_Id;
482 function P_Basic_Declarative_Items return List_Id;
483 function P_Constraint_Opt return Node_Id;
484 function P_Declarative_Part return List_Id;
485 function P_Defining_Identifier return Node_Id;
486 function P_Discrete_Choice_List return List_Id;
487 function P_Discrete_Range return Node_Id;
488 function P_Discrete_Subtype_Definition return Node_Id;
489 function P_Known_Discriminant_Part_Opt return List_Id;
490 function P_Signed_Integer_Type_Definition return Node_Id;
491 function P_Range return Node_Id;
492 function P_Range_Or_Subtype_Mark return Node_Id;
493 function P_Range_Constraint return Node_Id;
494 function P_Record_Definition return Node_Id;
495 function P_Subtype_Indication return Node_Id;
496 function P_Subtype_Mark return Node_Id;
497 function P_Subtype_Mark_Resync return Node_Id;
498 function P_Unknown_Discriminant_Part_Opt return Boolean;
500 procedure P_Component_Items (Decls : List_Id);
501 -- Scan out one or more component items and append them to the
502 -- given list. Only scans out more than one declaration in the
503 -- case where the source has a single declaration with multiple
504 -- defining identifiers.
506 function Init_Expr_Opt (P : Boolean := False) return Node_Id;
507 -- If an initialization expression is present (:= expression), then
508 -- it is scanned out and returned, otherwise Empty is returned if no
509 -- initialization expression is present. This procedure also handles
510 -- certain common error cases cleanly. The parameter P indicates if
511 -- a right paren can follow the expression (default = no right paren
514 procedure Skip_Declaration (S : List_Id);
515 -- Used when scanning statements to skip past a mispaced declaration
516 -- The declaration is scanned out and appended to the given list.
517 -- Token is known to be a declaration token (in Token_Class_Declk)
518 -- on entry, so there definition is a declaration to be scanned.
520 function P_Subtype_Indication (Subtype_Mark : Node_Id) return Node_Id;
521 -- This version of P_Subtype_Indication is called when the caller has
522 -- already scanned out the subtype mark which is passed as a parameter.
524 function P_Subtype_Mark_Attribute (Type_Node : Node_Id) return Node_Id;
525 -- Parse a subtype mark attribute. The caller has already parsed the
526 -- subtype mark, which is passed in as the argument, and has checked
527 -- that the current token is apostrophe.
532 function P_Aggregate return Node_Id;
533 function P_Expression return Node_Id;
534 function P_Expression_No_Right_Paren return Node_Id;
535 function P_Expression_Or_Range_Attribute return Node_Id;
536 function P_Function_Name return Node_Id;
537 function P_Name return Node_Id;
538 function P_Qualified_Simple_Name return Node_Id;
539 function P_Qualified_Simple_Name_Resync return Node_Id;
540 function P_Simple_Expression return Node_Id;
541 function P_Simple_Expression_Or_Range_Attribute return Node_Id;
543 function P_Qualified_Expression
544 (Subtype_Mark : Node_Id)
546 -- This routine scans out a qualified expression when the caller has
547 -- already scanned out the name and apostrophe of the construct.
553 function P_Statement_Name (Name_Node : Node_Id) return Node_Id;
554 -- Given a node representing a name (which is a call), converts it
555 -- to the syntactically corresponding procedure call statement.
557 function P_Sequence_Of_Statements (SS_Flags : SS_Rec) return List_Id;
558 -- The argument indicates the acceptable termination tokens.
559 -- See body in Par.Ch5 for details of the use of this parameter.
561 procedure Parse_Decls_Begin_End (Parent : Node_Id);
562 -- Parses declarations and handled statement sequence, setting
563 -- fields of Parent node appropriately.
568 function P_Designator return Node_Id;
569 function P_Defining_Program_Unit_Name return Node_Id;
570 function P_Formal_Part return List_Id;
571 function P_Parameter_Profile return List_Id;
572 function P_Return_Statement return Node_Id;
573 function P_Subprogram_Specification return Node_Id;
575 procedure P_Mode (Node : Node_Id);
576 -- Sets In_Present and/or Out_Present flags in Node scanning past
577 -- IN, OUT or IN OUT tokens in the source.
579 function P_Subprogram (Pf_Flags : Pf_Rec) return Node_Id;
580 -- Scans out any construct starting with either of the keywords
581 -- PROCEDURE or FUNCTION. The parameter indicates which possible
582 -- possible kinds of construct (body, spec, instantiation etc.)
583 -- are permissible in the current context.
588 function P_Package (Pf_Flags : Pf_Rec) return Node_Id;
589 -- Scans out any construct starting with the keyword PACKAGE. The
590 -- parameter indicates which possible kinds of construct (body, spec,
591 -- instantiation etc.) are permissible in the current context.
595 function P_Use_Clause return Node_Id;
599 function P_Abort_Statement return Node_Id;
600 function P_Abortable_Part return Node_Id;
601 function P_Accept_Statement return Node_Id;
602 function P_Delay_Statement return Node_Id;
603 function P_Entry_Body return Node_Id;
604 function P_Protected return Node_Id;
605 function P_Requeue_Statement return Node_Id;
606 function P_Select_Statement return Node_Id;
607 function P_Task return Node_Id;
608 function P_Terminate_Alternative return Node_Id;
612 function P_Compilation_Unit return Node_Id;
613 -- Note: this function scans a single compilation unit, and
614 -- checks that an end of file follows this unit, diagnosing
615 -- any unexpected input as an error, and then skipping it, so
616 -- that Token is set to Tok_EOF on return. An exception is in
617 -- syntax-only mode, where multiple compilation units are
618 -- permitted. In this case, P_Compilation_Unit does not check
619 -- for end of file and there may be more compilation units to
620 -- scan. The caller can uniquely detect this situation by the
621 -- fact that Token is not set to Tok_EOF on return.
625 function P_Handled_Sequence_Of_Statements return Node_Id;
626 function P_Raise_Statement return Node_Id;
628 function Parse_Exception_Handlers return List_Id;
629 -- Parses the partial construct EXCEPTION followed by a list of
630 -- exception handlers which appears in a number of productions,
631 -- and returns the list of exception handlers.
636 function P_Generic return Node_Id;
637 function P_Generic_Actual_Part_Opt return List_Id;
641 function P_Representation_Clause return Node_Id;
643 function P_Code_Statement (Subtype_Mark : Node_Id) return Node_Id;
644 -- Function to parse a code statement. The caller has scanned out
645 -- the name to be used as the subtype mark (but has not checked that
646 -- it is suitable for use as a subtype mark, i.e. is either an
647 -- identifier or a selected component). The current token is an
648 -- apostrophe and the following token is either a left paren or
649 -- RANGE (the latter being an error to be caught by P_Code_Statement.
652 -- Note: the parsing for annexe J features (i.e. obsolescent features)
653 -- is found in the logical section where these features would be if
654 -- they were not obsolescent. In particular:
656 -- Delta constraint is parsed by P_Delta_Constraint (3.5.9)
657 -- At clause is parsed by P_At_Clause (13.1)
658 -- Mod clause is parsed by P_Mod_Clause (13.5.1)
664 -- Routines for handling end lines, including scope recovery
668 function Check_End return Boolean;
669 -- Called when an end sequence is required. In the absence of an error
670 -- situation, Token contains Tok_End on entry, but in a missing end
671 -- case, this may not be the case. Pop_End_Context is used to determine
672 -- the appropriate action to be taken. The returned result is True if
673 -- an End sequence was encountered and False if no End sequence was
674 -- present. This occurs if the END keyword encountered was determined
675 -- to be improper and deleted (i.e. Pop_End_Context set End_Action to
676 -- Skip_And_Reject). Note that the END sequence includes a semicolon,
677 -- except in the case of END RECORD, where a semicolon follows the END
678 -- RECORD, but is not part of the record type definition itself.
681 -- Skip past an end sequence. On entry Token contains Tok_End, and we
682 -- we know that the end sequence is syntactically incorrect, and that
683 -- an appropriate error message has already been posted. The mission
684 -- is simply to position the scan pointer to be the best guess of the
685 -- position after the end sequence. We do not issue any additional
686 -- error messages while carrying this out.
688 procedure End_Statements (Parent : Node_Id := Empty);
689 -- Called when an end is required or expected to terminate a sequence
690 -- of statements. The caller has already made an appropriate entry in
691 -- the Scope.Table to describe the expected form of the end. This can
692 -- only be used in cases where the only appropriate terminator is end.
693 -- If Parent is non-empty, then if a correct END line is encountered,
694 -- the End_Label field of Parent is set appropriately.
698 ------------------------------------
699 -- Resynchronization After Errors --
700 ------------------------------------
702 -- These procedures are used to resynchronize after errors. Following an
703 -- error which is not immediately locally recoverable, the exception
704 -- Error_Resync is raised. The handler for Error_Resync typically calls
705 -- one of these recovery procedures to resynchronize the source position
706 -- to a point from which parsing can be restarted.
708 -- Note: these procedures output an information message that tokens are
709 -- being skipped, but this message is output only if the option for
710 -- Multiple_Errors_Per_Line is set in Options.
714 procedure Resync_Choice;
715 -- Used if an error occurs scanning a choice. The scan pointer is
716 -- advanced to the next vertical bar, arrow, or semicolon, whichever
717 -- comes first. We also quit if we encounter an end of file.
719 procedure Resync_Expression;
720 -- Used if an error is detected during the parsing of an expression.
721 -- It skips past tokens until either a token which cannot be part of
722 -- an expression is encountered (an expression terminator), or if a
723 -- comma or right parenthesis or vertical bar is encountered at the
724 -- current parenthesis level (a parenthesis level counter is maintained
725 -- to carry out this test).
727 procedure Resync_Past_Semicolon;
728 -- Used if an error occurs while scanning a sequence of declarations.
729 -- The scan pointer is positioned past the next semicolon and the scan
730 -- resumes. The scan is also resumed on encountering a token which
731 -- starts a declaration (but we make sure to skip at least one token
732 -- in this case, to avoid getting stuck in a loop).
734 procedure Resync_To_Semicolon;
735 -- Similar to Resync_Past_Semicolon, except that the scan pointer is
736 -- left pointing to the semicolon rather than past it.
738 procedure Resync_Past_Semicolon_Or_To_Loop_Or_Then;
739 -- Used if an error occurs while scanning a sequence of statements.
740 -- The scan pointer is positioned past the next semicolon, or to the
741 -- next occurrence of either then or loop, and the scan resumes.
743 procedure Resync_To_When;
744 -- Used when an error occurs scanning an entry index specification.
745 -- The scan pointer is positioned to the next WHEN (or to IS or
746 -- semicolon if either of these appear before WHEN, indicating
747 -- another error has occurred).
749 procedure Resync_Semicolon_List;
750 -- Used if an error occurs while scanning a parenthesized list of items
751 -- separated by semicolons. The scan pointer is advanced to the next
752 -- semicolon or right parenthesis at the outer parenthesis level, or
753 -- to the next is or RETURN keyword occurence, whichever comes first.
755 procedure Resync_Cunit;
756 -- Synchronize to next token which could be the start of a compilation
757 -- unit, or to the end of file token.
761 -------------------------
762 -- Token Scan Routines --
763 -------------------------
765 -- Routines to check for expected tokens
769 -- Procedures with names of the form T_xxx, where Tok_xxx is a token
770 -- name, check that the current token matches the required token, and
771 -- if so, scan past it. If not, an error is issued indicating that
772 -- the required token is not present (xxx expected). In most cases, the
773 -- scan pointer is not moved in the not-found case, but there are some
774 -- exceptions to this, see for example T_Id, where the scan pointer is
775 -- moved across a literal appearing where an identifier is expected.
783 procedure T_Colon_Equal;
787 procedure T_Greater_Greater;
788 procedure T_Identifier;
791 procedure T_Left_Paren;
800 procedure T_Right_Paren;
801 procedure T_Semicolon;
808 -- Procedures have names of the form TF_xxx, where Tok_xxx is a token
809 -- name check that the current token matches the required token, and
810 -- if so, scan past it. If not, an error message is issued indicating
811 -- that the required token is not present (xxx expected).
813 -- If the missing token is at the end of the line, then control returns
814 -- immediately after posting the message. If there are remaining tokens
815 -- on the current line, a search is conducted to see if the token
816 -- appears later on the current line, as follows:
818 -- A call to Scan_Save is issued and a forward search for the token
819 -- is carried out. If the token is found on the current line before a
820 -- semicolon, then it is scanned out and the scan continues from that
821 -- point. If not the scan is restored to the point where it was missing.
827 procedure TF_Semicolon;
833 ----------------------
834 -- Utility Routines --
835 ----------------------
839 function Bad_Spelling_Of (T : Token_Type) return Boolean;
840 -- This function is called in an error situation. It checks if the
841 -- current token is an identifier whose name is a plausible bad
842 -- spelling of the given keyword token, and if so, issues an error
843 -- message, sets Token from T, and returns True. Otherwise Token is
844 -- unchanged, and False is returned.
846 procedure Check_Bad_Layout;
847 -- Check for bad indentation in RM checking mode. Used for statements
848 -- and declarations. Checks if current token is at start of line and
849 -- is exdented from the current expected end column, and if so an
850 -- error message is generated.
852 procedure Check_Misspelling_Of (T : Token_Type);
853 pragma Inline (Check_Misspelling_Of);
854 -- This is similar to the function above, except that it does not
855 -- return a result. It is typically used in a situation where any
856 -- identifier is an error, and it makes sense to simply convert it
857 -- to the given token if it is a plausible misspelling of it.
859 procedure Check_95_Keyword (Token_95, Next : Token_Type);
860 -- This routine checks if the token after the current one matches the
861 -- Next argument. If so, the scan is backed up to the current token
862 -- and Token_Type is changed to Token_95 after issuing an appropriate
863 -- error message ("(Ada 83) keyword xx cannot be used"). If not,
864 -- the scan is backed up with Token_Type unchanged. This routine
865 -- is used to deal with an attempt to use a 95 keyword in Ada 83
866 -- mode. The caller has typically checked that the current token,
867 -- an identifier, matches one of the 95 keywords.
869 procedure Check_Simple_Expression (E : Node_Id);
870 -- Given an expression E, that has just been scanned, so that Expr_Form
871 -- is still set, outputs an error if E is a non-simple expression. E is
872 -- not modified by this call.
874 procedure Check_Simple_Expression_In_Ada_83 (E : Node_Id);
875 -- Like Check_Simple_Expression, except that the error message is only
876 -- given when operating in Ada 83 mode, and includes "in Ada 83".
878 function Check_Subtype_Mark (Mark : Node_Id) return Node_Id;
879 -- Called to check that a node representing a name (or call) is
880 -- suitable for a subtype mark, i.e, that it is an identifier or
881 -- a selected component. If so, or if it is already Error, then
882 -- it is returned unchanged. Otherwise an error message is issued
883 -- and Error is returned.
885 function Comma_Present return Boolean;
886 -- Used in comma delimited lists to determine if a comma is present, or
887 -- can reasonably be assumed to have been present (an error message is
888 -- generated in the latter case). If True is returned, the scan has been
889 -- positioned past the comma. If False is returned, the scan position
890 -- is unchanged. Note that all comma-delimited lists are terminated by
891 -- a right paren, so the only legitimate tokens when Comma_Present is
892 -- called are right paren and comma. If some other token is found, then
893 -- Comma_Present has the job of deciding whether it is better to pretend
894 -- a comma was present, post a message for a missing comma and return
895 -- True, or return False and let the caller diagnose the missing right
898 procedure Discard_Junk_Node (N : Node_Id);
899 procedure Discard_Junk_List (L : List_Id);
900 pragma Inline (Discard_Junk_Node);
901 pragma Inline (Discard_Junk_List);
902 -- These procedures do nothing at all, their effect is simply to discard
903 -- the argument. A typical use is to skip by some junk that is not
904 -- expected in the current context.
906 procedure Ignore (T : Token_Type);
907 -- If current token matches T, then give an error message and skip
908 -- past it, otherwise the call has no effect at all. T may be any
909 -- reserved word token, or comma, left or right paren, or semicolon.
911 function Is_Reserved_Identifier return Boolean;
912 -- Test if current token is a reserved identifier. This test is based
913 -- on the token being a keyword and being spelled in typical identifier
914 -- style (i.e. starting with an upper case letter).
916 procedure Merge_Identifier (Prev : Node_Id; Nxt : Token_Type);
917 -- Called when the previous token is an identifier (whose Token_Node
918 -- value is given by Prev) to check if current token is an identifier
919 -- that can be merged with the previous one adding an underscore. The
920 -- merge is only attempted if the following token matches Nxt. If all
921 -- conditions are met, an error message is issued, and the merge is
922 -- carried out, modifying the Chars field of Prev.
924 procedure No_Constraint;
925 -- Called in a place where no constraint is allowed, but one might
926 -- appear due to a common error (e.g. after the type mark in a procedure
927 -- parameter. If a constraint is present, an error message is posted,
928 -- and the constraint is scanned and discarded.
930 function No_Right_Paren (Expr : Node_Id) return Node_Id;
931 -- Function to check for no right paren at end of expression, returns
932 -- its argument if no right paren, else flags paren and returns Error.
934 procedure Push_Scope_Stack;
935 pragma Inline (Push_Scope_Stack);
936 -- Push a new entry onto the scope stack. Scope.Last (the stack pointer)
937 -- is incremented. The Junk field is preinitialized to False. The caller
938 -- is expected to fill in all remaining entries of the new new top stack
939 -- entry at Scope.Table (Scope.Last).
941 procedure Pop_Scope_Stack;
942 -- Pop an entry off the top of the scope stack. Scope_Last (the scope
943 -- table stack pointer) is decremented by one. It is a fatal error to
944 -- try to pop off the dummy entry at the bottom of the stack (i.e.
945 -- Scope.Last must be non-zero at the time of call).
947 function Separate_Present return Boolean;
948 -- Determines if the current token is either Tok_Separate, or an
949 -- identifier that is a possible misspelling of "separate" followed
950 -- by a semicolon. True is returned if so, otherwise False.
952 procedure Signal_Bad_Attribute;
953 -- The current token is an identifier that is supposed to be an
954 -- attribute identifier but is not. This routine posts appropriate
955 -- error messages, including a check for a near misspelling.
957 function Token_Is_At_Start_Of_Line return Boolean;
958 pragma Inline (Token_Is_At_Start_Of_Line);
959 -- Determines if the current token is the first token on the line
961 function Token_Is_At_End_Of_Line return Boolean;
962 -- Determines if the current token is the last token on the line
966 ---------------------------------------
967 -- Specialized Syntax Check Routines --
968 ---------------------------------------
970 function Prag (Pragma_Node : Node_Id; Semi : Source_Ptr) return Node_Id;
971 -- This function is passed a tree for a pragma that has been scanned out.
972 -- The pragma is syntactically well formed according to the general syntax
973 -- for pragmas and the pragma identifier is for one of the recognized
974 -- pragmas. It performs specific syntactic checks for specific pragmas.
975 -- The result is the input node if it is OK, or Error otherwise. The
976 -- reason that this is separated out is to facilitate the addition
977 -- of implementation defined pragmas. The second parameter records the
978 -- location of the semicolon following the pragma (this is needed for
979 -- correct processing of the List and Page pragmas). The returned value
980 -- is a copy of Pragma_Node, or Error if an error is found. Note that
981 -- at the point where Prag is called, the right paren ending the pragma
982 -- has been scanned out, and except in the case of pragma Style_Checks,
983 -- so has the following semicolon. For Style_Checks, the caller delays
984 -- the scanning of the semicolon so that it will be scanned using the
985 -- settings from the Style_Checks pragma preceding it.
987 -------------------------
988 -- Subsidiary Routines --
989 -------------------------
992 -- This procedure creates implicit label declarations for all label that
993 -- are declared in the current unit. Note that this could conceptually
994 -- be done at the point where the labels are declared, but it is tricky
995 -- to do it then, since the tree is not hooked up at the point where the
996 -- label is declared (e.g. a sequence of statements is not yet attached
997 -- to its containing scope at the point a label in the sequence is found)
1000 -- This procedure loads all subsidiary units that are required by this
1001 -- unit, including with'ed units, specs for bodies, and parents for child
1002 -- units. It does not load bodies for inlined procedures and generics,
1003 -- since we don't know till semantic analysis is complete what is needed.
1009 -- The package bodies can see all routines defined in all other subpackages
1029 package body Ch2 is separate;
1030 package body Ch3 is separate;
1031 package body Ch4 is separate;
1032 package body Ch5 is separate;
1033 package body Ch6 is separate;
1034 package body Ch7 is separate;
1035 package body Ch8 is separate;
1036 package body Ch9 is separate;
1037 package body Ch10 is separate;
1038 package body Ch11 is separate;
1039 package body Ch12 is separate;
1040 package body Ch13 is separate;
1042 package body Endh is separate;
1043 package body Tchk is separate;
1044 package body Sync is separate;
1045 package body Util is separate;
1047 function Prag (Pragma_Node : Node_Id; Semi : Source_Ptr) return Node_Id
1050 procedure Labl is separate;
1051 procedure Load is separate;
1057 -- This function is the parse routine called at the outer level. It parses
1058 -- the current compilation unit and adds implicit label declarations.
1061 -- Deal with configuration pragmas case first
1063 if Configuration_Pragmas then
1065 Ecount : constant Int := Serious_Errors_Detected;
1066 Pragmas : constant List_Id := Empty_List;
1071 if Token = Tok_EOF then
1074 elsif Token /= Tok_Pragma then
1075 Error_Msg_SC ("only pragmas allowed in configuration file");
1081 if Serious_Errors_Detected > Ecount then
1085 if Chars (P_Node) > Last_Configuration_Pragma_Name
1086 and then Chars (P_Node) /= Name_Source_Reference
1089 ("only configuration pragmas allowed " &
1090 "in configuration file");
1094 Append (P_Node, Pragmas);
1099 -- Normal case of compilation unit
1102 Save_Opt_Config_Switches (Save_Config_Switches);
1104 -- Special processing for language defined units. For this purpose
1105 -- we do NOT consider the renamings in annex J as predefined. That
1106 -- allows users to compile their own versions of these files, and
1107 -- in particular, in the VMS implementation, the DEC versions can
1108 -- be substituted for the standard Ada 95 versions.
1110 if Is_Predefined_File_Name
1111 (Fname => File_Name (Current_Source_File),
1112 Renamings_Included => False)
1114 Set_Opt_Config_Switches
1115 (Is_Internal_File_Name (File_Name (Current_Source_File)));
1117 -- If this is the main unit, disallow compilation unless the -gnatg
1118 -- (GNAT mode) switch is set (from a user point of view, the rule is
1119 -- that language defined units cannot be recompiled).
1121 -- However, an exception is s-rpc, and its children. We test this
1122 -- by looking at the character after the minus, the rule is that
1123 -- System.RPC and its children are the only children in System
1124 -- whose second level name can start with the letter r.
1126 Get_Name_String (File_Name (Current_Source_File));
1128 if (Name_Len < 3 or else Name_Buffer (1 .. 3) /= "s-r")
1129 and then Current_Source_Unit = Main_Unit
1130 and then not GNAT_Mode
1131 and then Operating_Mode = Generate_Code
1133 Error_Msg_SC ("language defined units may not be recompiled");
1137 -- The following loop runs more than once only in syntax check mode
1138 -- where we allow multiple compilation units in the same file.
1141 Set_Opt_Config_Switches
1142 (Is_Internal_File_Name (File_Name (Current_Source_File)));
1144 -- Initialize scope table and other parser control variables
1146 Compiler_State := Parsing;
1148 Scope.Increment_Last;
1149 Scope.Table (0).Etyp := E_Dummy;
1150 SIS_Entry_Active := False;
1151 Last_Resync_Point := No_Location;
1153 Label_List := New_Elmt_List;
1154 Discard_Node (P_Compilation_Unit);
1156 -- If we are not at an end of file, then this means that we are
1157 -- in syntax scan mode, and we can have another compilation unit,
1158 -- otherwise we will exit from the loop.
1160 exit when Token = Tok_EOF;
1161 Restore_Opt_Config_Switches (Save_Config_Switches);
1164 -- Now that we have completely parsed the source file, we can
1165 -- complete the source file table entry.
1167 Complete_Source_File_Entry;
1169 -- An internal error check, the scope stack should now be empty
1171 pragma Assert (Scope.Last = 0);
1173 -- Remaining steps are to create implicit label declarations and to
1174 -- load required subsidiary sources. These steps are required only
1175 -- if we are doing semantic checking.
1177 if Operating_Mode /= Check_Syntax or else Debug_Flag_F then
1182 -- Restore settings of switches saved on entry
1184 Restore_Opt_Config_Switches (Save_Config_Switches);
1185 Set_Comes_From_Source_Default (False);
OSDN Git Service
