1------------------------------------------------------------------------------ 2-- -- 3-- GNAT COMPILER COMPONENTS -- 4-- -- 5-- P A R -- 6-- -- 7-- B o d y -- 8-- -- 9-- Copyright (C) 1992-2013, 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------------------------------------------------------------------------------ 25 26with Aspects; use Aspects; 27with Atree; use Atree; 28with Casing; use Casing; 29with Debug; use Debug; 30with Elists; use Elists; 31with Errout; use Errout; 32with Fname; use Fname; 33with Lib; use Lib; 34with Namet; use Namet; 35with Namet.Sp; use Namet.Sp; 36with Nlists; use Nlists; 37with Nmake; use Nmake; 38with Opt; use Opt; 39with Output; use Output; 40with Par_SCO; use Par_SCO; 41with Restrict; use Restrict; 42with Scans; use Scans; 43with Scn; use Scn; 44with Sem_Util; use Sem_Util; 45with Sinput; use Sinput; 46with Sinput.L; use Sinput.L; 47with Sinfo; use Sinfo; 48with Snames; use Snames; 49with Style; 50with Stylesw; use Stylesw; 51with Table; 52with Tbuild; use Tbuild; 53 54--------- 55-- Par -- 56--------- 57 58function Par (Configuration_Pragmas : Boolean) return List_Id is 59 60 Num_Library_Units : Natural := 0; 61 -- Count number of units parsed (relevant only in syntax check only mode, 62 -- since in semantics check mode only a single unit is permitted anyway). 63 64 Save_Config_Switches : Config_Switches_Type; 65 -- Variable used to save values of config switches while we parse the 66 -- new unit, to be restored on exit for proper recursive behavior. 67 68 Loop_Block_Count : Nat := 0; 69 -- Counter used for constructing loop/block names (see the routine 70 -- Par.Ch5.Get_Loop_Block_Name). 71 72 Inside_Record_Definition : Boolean := False; 73 -- Flag set True within a record definition. Used to control warning 74 -- for redefinition of standard entities (not issued for field names). 75 76 -------------------- 77 -- Error Recovery -- 78 -------------------- 79 80 -- When an error is encountered, a call is made to one of the Error_Msg 81 -- routines to record the error. If the syntax scan is not derailed by the 82 -- error (e.g. a complaint that logical operators are inconsistent in an 83 -- EXPRESSION), then control returns from the Error_Msg call, and the 84 -- parse continues unimpeded. 85 86 -- If on the other hand, the Error_Msg represents a situation from which 87 -- the parser cannot recover locally, the exception Error_Resync is raised 88 -- immediately after the call to Error_Msg. Handlers for Error_Resync 89 -- are located at strategic points to resynchronize the parse. For example, 90 -- when an error occurs in a statement, the handler skips to the next 91 -- semicolon and continues the scan from there. 92 93 -- Each parsing procedure contains a note with the heading "Error recovery" 94 -- which shows if it can propagate the Error_Resync exception. In order 95 -- not to propagate the exception, a procedure must either contain its own 96 -- handler for this exception, or it must not call any other routines which 97 -- propagate the exception. 98 99 -- Note: the arrangement of Error_Resync handlers is such that it should 100 -- never be possible to transfer control through a procedure which made 101 -- an entry in the scope stack, invalidating the contents of the stack. 102 103 Error_Resync : exception; 104 -- Exception raised on error that is not handled locally, see above 105 106 Last_Resync_Point : Source_Ptr; 107 -- The resynchronization routines in Par.Sync run a risk of getting 108 -- stuck in an infinite loop if they do not skip a token, and the caller 109 -- keeps repeating the same resync call. On the other hand, if they skip 110 -- a token unconditionally, some recovery opportunities are missed. The 111 -- variable Last_Resync_Point records the token location previously set 112 -- by a Resync call, and if a subsequent Resync call occurs at the same 113 -- location, then the Resync routine does guarantee to skip a token. 114 115 -------------------------------------------- 116 -- Handling Semicolon Used in Place of IS -- 117 -------------------------------------------- 118 119 -- The following global variables are used in handling the error situation 120 -- of using a semicolon in place of IS in a subprogram declaration as in: 121 122 -- procedure X (Y : Integer); 123 -- Q : Integer; 124 -- begin 125 -- ... 126 -- end; 127 128 -- The two contexts in which this can appear are at the outer level, and 129 -- within a declarative region. At the outer level, we know something is 130 -- wrong as soon as we see the Q (or begin, if there are no declarations), 131 -- and we can immediately decide that the semicolon should have been IS. 132 133 -- The situation in a declarative region is more complex. The declaration 134 -- of Q could belong to the outer region, and we do not know that we have 135 -- an error until we hit the begin. It is still not clear at this point 136 -- from a syntactic point of view that something is wrong, because the 137 -- begin could belong to the enclosing subprogram or package. However, we 138 -- can incorporate a bit of semantic knowledge and note that the body of 139 -- X is missing, so we definitely DO have an error. We diagnose this error 140 -- as semicolon in place of IS on the subprogram line. 141 142 -- There are two styles for this diagnostic. If the begin immediately 143 -- follows the semicolon, then we can place a flag (IS expected) right 144 -- on the semicolon. Otherwise we do not detect the error until we hit 145 -- the begin which refers back to the line with the semicolon. 146 147 -- To control the process in the second case, the following global 148 -- variables are set to indicate that we have a subprogram declaration 149 -- whose body is required and has not yet been found. The prefix SIS 150 -- stands for "Subprogram IS" handling. 151 152 SIS_Entry_Active : Boolean := False; 153 -- Set True to indicate that an entry is active (i.e. that a subprogram 154 -- declaration has been encountered, and no body for this subprogram has 155 -- been encountered). The remaining fields are valid only if this is True. 156 157 SIS_Labl : Node_Id; 158 -- Subprogram designator 159 160 SIS_Sloc : Source_Ptr; 161 -- Source location of FUNCTION/PROCEDURE keyword 162 163 SIS_Ecol : Column_Number; 164 -- Column number of FUNCTION/PROCEDURE keyword 165 166 SIS_Semicolon_Sloc : Source_Ptr; 167 -- Source location of semicolon at end of subprogram declaration 168 169 SIS_Declaration_Node : Node_Id; 170 -- Pointer to tree node for subprogram declaration 171 172 SIS_Missing_Semicolon_Message : Error_Msg_Id; 173 -- Used to save message ID of missing semicolon message (which will be 174 -- modified to missing IS if necessary). Set to No_Error_Msg in the 175 -- normal (non-error) case. 176 177 -- Five things can happen to an active SIS entry 178 179 -- 1. If a BEGIN is encountered with an SIS entry active, then we have 180 -- exactly the situation in which we know the body of the subprogram is 181 -- missing. After posting an error message, we change the spec to a body, 182 -- rechaining the declarations that intervened between the spec and BEGIN. 183 184 -- 2. Another subprogram declaration or body is encountered. In this 185 -- case the entry gets overwritten with the information for the new 186 -- subprogram declaration. We don't catch some nested cases this way, 187 -- but it doesn't seem worth the effort. 188 189 -- 3. A nested declarative region (e.g. package declaration or package 190 -- body) is encountered. The SIS active indication is reset at the start 191 -- of such a nested region. Again, like case 2, this causes us to miss 192 -- some nested cases, but it doesn't seen worth the effort to stack and 193 -- unstack the SIS information. Maybe we will reconsider this if we ever 194 -- get a complaint about a missed case. 195 196 -- 4. We encounter a valid pragma INTERFACE or IMPORT that effectively 197 -- supplies the missing body. In this case we reset the entry. 198 199 -- 5. We encounter the end of the declarative region without encountering 200 -- a BEGIN first. In this situation we simply reset the entry. We know 201 -- that there is a missing body, but it seems more reasonable to let the 202 -- later semantic checking discover this. 203 204 ---------------------------------------------------- 205 -- Handling of Reserved Words Used as Identifiers -- 206 ---------------------------------------------------- 207 208 -- Note: throughout the parser, the terms reserved word and keyword are 209 -- used interchangeably to refer to the same set of reserved keywords 210 -- (including until, protected, etc). 211 212 -- If a reserved word is used in place of an identifier, the parser where 213 -- possible tries to recover gracefully. In particular, if the keyword is 214 -- clearly spelled using identifier casing, e.g. Until in a source program 215 -- using mixed case identifiers and lower case keywords, then the keyword 216 -- is treated as an identifier if it appears in a place where an identifier 217 -- is required. 218 219 -- The situation is more complex if the keyword is spelled with normal 220 -- keyword casing. In this case, the parser is more reluctant to consider 221 -- it to be intended as an identifier, unless it has some further 222 -- confirmation. 223 224 -- In the case of an identifier appearing in the identifier list of a 225 -- declaration, the appearance of a comma or colon right after the keyword 226 -- on the same line is taken as confirmation. For an enumeration literal, 227 -- a comma or right paren right after the identifier is also treated as 228 -- adequate confirmation. 229 230 -- The following type is used in calls to Is_Reserved_Identifier and 231 -- also to P_Defining_Identifier and P_Identifier. The default for all 232 -- these functions is that reserved words in reserved word case are not 233 -- considered to be reserved identifiers. The Id_Check value indicates 234 -- tokens, which if they appear immediately after the identifier, are 235 -- taken as confirming that the use of an identifier was expected 236 237 type Id_Check is 238 (None, 239 -- Default, no special token test 240 241 C_Comma_Right_Paren, 242 -- Consider as identifier if followed by comma or right paren 243 244 C_Comma_Colon, 245 -- Consider as identifier if followed by comma or colon 246 247 C_Do, 248 -- Consider as identifier if followed by DO 249 250 C_Dot, 251 -- Consider as identifier if followed by period 252 253 C_Greater_Greater, 254 -- Consider as identifier if followed by >> 255 256 C_In, 257 -- Consider as identifier if followed by IN 258 259 C_Is, 260 -- Consider as identifier if followed by IS 261 262 C_Left_Paren_Semicolon, 263 -- Consider as identifier if followed by left paren or semicolon 264 265 C_Use, 266 -- Consider as identifier if followed by USE 267 268 C_Vertical_Bar_Arrow); 269 -- Consider as identifier if followed by | or => 270 271 -------------------------------------------- 272 -- Handling IS Used in Place of Semicolon -- 273 -------------------------------------------- 274 275 -- This is a somewhat trickier situation, and we can't catch it in all 276 -- cases, but we do our best to detect common situations resulting from 277 -- a "cut and paste" operation which forgets to change the IS to semicolon. 278 -- Consider the following example: 279 280 -- package body X is 281 -- procedure A; 282 -- procedure B is 283 -- procedure C; 284 -- ... 285 -- procedure D is 286 -- begin 287 -- ... 288 -- end; 289 -- begin 290 -- ... 291 -- end; 292 293 -- The trouble is that the section of text from PROCEDURE B through END; 294 -- constitutes a valid procedure body, and the danger is that we find out 295 -- far too late that something is wrong (indeed most compilers will behave 296 -- uncomfortably on the above example). 297 298 -- We have two approaches to helping to control this situation. First we 299 -- make every attempt to avoid swallowing the last END; if we can be sure 300 -- that some error will result from doing so. In particular, we won't 301 -- accept the END; unless it is exactly correct (in particular it must not 302 -- have incorrect name tokens), and we won't accept it if it is immediately 303 -- followed by end of file, WITH or SEPARATE (all tokens that unmistakeably 304 -- signal the start of a compilation unit, and which therefore allow us to 305 -- reserve the END; for the outer level.) For more details on this aspect 306 -- of the handling, see package Par.Endh. 307 308 -- If we can avoid eating up the END; then the result in the absence of 309 -- any additional steps would be to post a missing END referring back to 310 -- the subprogram with the bogus IS. Similarly, if the enclosing package 311 -- has no BEGIN, then the result is a missing BEGIN message, which again 312 -- refers back to the subprogram header. 313 314 -- Such an error message is not too bad (it's already a big improvement 315 -- over what many parsers do), but it's not ideal, because the declarations 316 -- following the IS have been absorbed into the wrong scope. In the above 317 -- case, this could result for example in a bogus complaint that the body 318 -- of D was missing from the package. 319 320 -- To catch at least some of these cases, we take the following additional 321 -- steps. First, a subprogram body is marked as having a suspicious IS if 322 -- the declaration line is followed by a line which starts with a symbol 323 -- that can start a declaration in the same column, or to the left of the 324 -- column in which the FUNCTION or PROCEDURE starts (normal style is to 325 -- indent any declarations which really belong a subprogram). If such a 326 -- subprogram encounters a missing BEGIN or missing END, then we decide 327 -- that the IS should have been a semicolon, and the subprogram body node 328 -- is marked (by setting the Bad_Is_Detected flag true. Note that we do 329 -- not do this for library level procedures, only for nested procedures, 330 -- since for library level procedures, we must have a body. 331 332 -- The processing for a declarative part checks to see if the last 333 -- declaration scanned is marked in this way, and if it is, the tree 334 -- is modified to reflect the IS being interpreted as a semicolon. 335 336 --------------------------------------------------- 337 -- Parser Type Definitions and Control Variables -- 338 --------------------------------------------------- 339 340 -- The following variable and associated type declaration are used by the 341 -- expression parsing routines to return more detailed information about 342 -- the categorization of a parsed expression. 343 344 type Expr_Form_Type is ( 345 EF_Simple_Name, -- Simple name, i.e. possibly qualified identifier 346 EF_Name, -- Simple expression which could also be a name 347 EF_Simple, -- Simple expression which is not call or name 348 EF_Range_Attr, -- Range attribute reference 349 EF_Non_Simple); -- Expression that is not a simple expression 350 351 Expr_Form : Expr_Form_Type; 352 353 -- The following type is used for calls to P_Subprogram, P_Package, P_Task, 354 -- P_Protected to indicate which of several possibilities is acceptable. 355 356 type Pf_Rec is record 357 Spcn : Boolean; -- True if specification OK 358 Decl : Boolean; -- True if declaration OK 359 Gins : Boolean; -- True if generic instantiation OK 360 Pbod : Boolean; -- True if proper body OK 361 Rnam : Boolean; -- True if renaming declaration OK 362 Stub : Boolean; -- True if body stub OK 363 Pexp : Boolean; -- True if parameterized expression OK 364 Fil2 : Boolean; -- Filler to fill to 8 bits 365 end record; 366 pragma Pack (Pf_Rec); 367 368 function T return Boolean renames True; 369 function F return Boolean renames False; 370 371 Pf_Decl_Gins_Pbod_Rnam_Stub_Pexp : constant Pf_Rec := 372 Pf_Rec'(F, T, T, T, T, T, T, F); 373 Pf_Decl_Pexp : constant Pf_Rec := 374 Pf_Rec'(F, T, F, F, F, F, T, F); 375 Pf_Decl_Gins_Pbod_Rnam_Pexp : constant Pf_Rec := 376 Pf_Rec'(F, T, T, T, T, F, T, F); 377 Pf_Decl_Pbod_Pexp : constant Pf_Rec := 378 Pf_Rec'(F, T, F, T, F, F, T, F); 379 Pf_Pbod_Pexp : constant Pf_Rec := 380 Pf_Rec'(F, F, F, T, F, F, T, F); 381 Pf_Spcn : constant Pf_Rec := 382 Pf_Rec'(T, F, F, F, F, F, F, F); 383 -- The above are the only allowed values of Pf_Rec arguments 384 385 type SS_Rec is record 386 Eftm : Boolean; -- ELSIF can terminate sequence 387 Eltm : Boolean; -- ELSE can terminate sequence 388 Extm : Boolean; -- EXCEPTION can terminate sequence 389 Ortm : Boolean; -- OR can terminate sequence 390 Sreq : Boolean; -- at least one statement required 391 Tatm : Boolean; -- THEN ABORT can terminate sequence 392 Whtm : Boolean; -- WHEN can terminate sequence 393 Unco : Boolean; -- Unconditional terminate after one statement 394 end record; 395 pragma Pack (SS_Rec); 396 397 SS_Eftm_Eltm_Sreq : constant SS_Rec := SS_Rec'(T, T, F, F, T, F, F, F); 398 SS_Eltm_Ortm_Tatm : constant SS_Rec := SS_Rec'(F, T, F, T, F, T, F, F); 399 SS_Extm_Sreq : constant SS_Rec := SS_Rec'(F, F, T, F, T, F, F, F); 400 SS_None : constant SS_Rec := SS_Rec'(F, F, F, F, F, F, F, F); 401 SS_Ortm_Sreq : constant SS_Rec := SS_Rec'(F, F, F, T, T, F, F, F); 402 SS_Sreq : constant SS_Rec := SS_Rec'(F, F, F, F, T, F, F, F); 403 SS_Sreq_Whtm : constant SS_Rec := SS_Rec'(F, F, F, F, T, F, T, F); 404 SS_Whtm : constant SS_Rec := SS_Rec'(F, F, F, F, F, F, T, F); 405 SS_Unco : constant SS_Rec := SS_Rec'(F, F, F, F, F, F, F, T); 406 407 Goto_List : Elist_Id; 408 -- List of goto nodes appearing in the current compilation. Used to 409 -- recognize natural loops and convert them into bona fide loops for 410 -- optimization purposes. 411 412 Label_List : Elist_Id; 413 -- List of label nodes for labels appearing in the current compilation. 414 -- Used by Par.Labl to construct the corresponding implicit declarations. 415 416 ----------------- 417 -- Scope Table -- 418 ----------------- 419 420 -- The scope table, also referred to as the scope stack, is used to record 421 -- the current scope context. It is organized as a stack, with inner nested 422 -- entries corresponding to higher entries on the stack. An entry is made 423 -- when the parser encounters the opening of a nested construct (such as a 424 -- record, task, package etc.), and then package Par.Endh uses this stack 425 -- to deal with END lines (including properly dealing with END nesting 426 -- errors). 427 428 type SS_End_Type is 429 -- Type of end entry required for this scope. The last two entries are 430 -- used only in the subprogram body case to mark the case of a suspicious 431 -- IS, or a bad IS (i.e. suspicions confirmed by missing BEGIN or END). 432 -- See separate section on dealing with IS used in place of semicolon. 433 -- Note that for many purposes E_Name, E_Suspicious_Is and E_Bad_Is are 434 -- treated the same (E_Suspicious_Is and E_Bad_Is are simply special cases 435 -- of E_Name). They are placed at the end of the enumeration so that a 436 -- test for >= E_Name catches all three cases efficiently. 437 438 (E_Dummy, -- dummy entry at outer level 439 E_Case, -- END CASE; 440 E_If, -- END IF; 441 E_Loop, -- END LOOP; 442 E_Record, -- END RECORD; 443 E_Return, -- END RETURN; 444 E_Select, -- END SELECT; 445 E_Name, -- END [name]; 446 E_Suspicious_Is, -- END [name]; (case of suspicious IS) 447 E_Bad_Is); -- END [name]; (case of bad IS) 448 449 -- The following describes a single entry in the scope table 450 451 type Scope_Table_Entry is record 452 Etyp : SS_End_Type; 453 -- Type of end entry, as per above description 454 455 Lreq : Boolean; 456 -- A flag indicating whether the label, if present, is required to 457 -- appear on the end line. It is referenced only in the case of Etyp is 458 -- equal to E_Name or E_Suspicious_Is where the name may or may not be 459 -- required (yes for labeled block, no in other cases). Note that for 460 -- all cases except begin, the question of whether a label is required 461 -- can be determined from the other fields (for loop, it is required if 462 -- it is present, and for the other constructs it is never required or 463 -- allowed). 464 465 Ecol : Column_Number; 466 -- Contains the absolute column number (with tabs expanded) of the 467 -- expected column of the end assuming normal Ada indentation usage. If 468 -- the RM_Column_Check mode is set, this value is used for generating 469 -- error messages about indentation. Otherwise it is used only to 470 -- control heuristic error recovery actions. This value is zero origin. 471 472 Labl : Node_Id; 473 -- This field is used to provide the name of the construct being parsed 474 -- and indirectly its kind. For loops and blocks, the field contains the 475 -- source name or the generated one. For package specifications, bodies, 476 -- subprogram specifications and bodies the field holds the correponding 477 -- program unit name. For task declarations and bodies, protected types 478 -- and bodies, and accept statements the field hold the name of the type 479 -- or operation. For if-statements, case-statements, and selects, the 480 -- field is initialized to Error. 481 482 -- Note: this is a bit of an odd (mis)use of Error, since there is no 483 -- Error, but we use this value as a place holder to indicate that it 484 -- is an error to have a label on the end line. 485 486 -- Whenever the field is a name, it is attached to the parent node of 487 -- the construct being parsed. Thus the parent node indicates the kind 488 -- of construct whose parse tree is being built. This is used in error 489 -- recovery. 490 491 Decl : List_Id; 492 -- Points to the list of declarations (i.e. the declarative part) 493 -- associated with this construct. It is set only in the END [name] 494 -- cases, and is set to No_List for all other cases which do not have a 495 -- declarative unit associated with them. This is used for determining 496 -- the proper location for implicit label declarations. 497 498 Node : Node_Id; 499 -- Empty except in the case of entries for IF and CASE statements, in 500 -- which case it contains the N_If_Statement or N_Case_Statement node. 501 -- This is used for setting the End_Span field. 502 503 Sloc : Source_Ptr; 504 -- Source location of the opening token of the construct. This is used 505 -- to refer back to this line in error messages (such as missing or 506 -- incorrect end lines). The Sloc field is not used, and is not set, if 507 -- a label is present (the Labl field provides the text name of the 508 -- label in this case, which is fine for error messages). 509 510 S_Is : Source_Ptr; 511 -- S_Is is relevant only if Etyp is set to E_Suspicious_Is or E_Bad_Is. 512 -- It records the location of the IS that is considered to be 513 -- suspicious. 514 515 Junk : Boolean; 516 -- A boolean flag that is set true if the opening entry is the dubious 517 -- result of some prior error, e.g. a record entry where the record 518 -- keyword was missing. It is used to suppress the issuing of a 519 -- corresponding junk complaint about the end line (we do not want 520 -- to complain about a missing end record when there was no record). 521 end record; 522 523 -- The following declares the scope table itself. The Last field is the 524 -- stack pointer, so that Scope.Table (Scope.Last) is the top entry. The 525 -- oldest entry, at Scope_Stack (0), is a dummy entry with Etyp set to 526 -- E_Dummy, and the other fields undefined. This dummy entry ensures that 527 -- Scope_Stack (Scope_Stack_Ptr).Etyp can always be tested, and that the 528 -- scope stack pointer is always in range. 529 530 package Scope is new Table.Table ( 531 Table_Component_Type => Scope_Table_Entry, 532 Table_Index_Type => Int, 533 Table_Low_Bound => 0, 534 Table_Initial => 50, 535 Table_Increment => 100, 536 Table_Name => "Scope"); 537 538 ------------------------------------------ 539 -- Table for Handling Suspicious Labels -- 540 ------------------------------------------ 541 542 -- This is a special data structure which is used to deal very spefifically 543 -- with the following error case 544 545 -- label; 546 -- loop 547 -- ... 548 -- end loop label; 549 550 -- Similar cases apply to FOR, WHILE, DECLARE, or BEGIN 551 552 -- In each case the opening line looks like a procedure call because of 553 -- the semicolon. And the end line looks illegal because of an unexpected 554 -- label. If we did nothing special, we would just diagnose the label on 555 -- the end as unexpected. But that does not help point to the real error 556 -- which is that the semicolon after label should be a colon. 557 558 -- To deal with this, we build an entry in the Suspicious_Labels table 559 -- whenever we encounter an identifier followed by a semicolon, followed 560 -- by one of LOOP, FOR, WHILE, DECLARE, BEGIN. Then this entry is used to 561 -- issue the right message when we hit the END that confirms that this was 562 -- a bad label. 563 564 type Suspicious_Label_Entry is record 565 Proc_Call : Node_Id; 566 -- Node for the procedure call statement built for the label; construct 567 568 Semicolon_Loc : Source_Ptr; 569 -- Location of the possibly wrong semicolon 570 571 Start_Token : Source_Ptr; 572 -- Source location of the LOOP, FOR, WHILE, DECLARE, BEGIN token 573 end record; 574 575 package Suspicious_Labels is new Table.Table ( 576 Table_Component_Type => Suspicious_Label_Entry, 577 Table_Index_Type => Int, 578 Table_Low_Bound => 1, 579 Table_Initial => 50, 580 Table_Increment => 100, 581 Table_Name => "Suspicious_Labels"); 582 583 -- Now when we are about to issue a message complaining about an END label 584 -- that should not be there because it appears to end a construct that has 585 -- no label, we first search the suspicious labels table entry, using the 586 -- source location stored in the scope table as a key. If we find a match, 587 -- then we check that the label on the end matches the name in the call, 588 -- and if so, we issue a message saying the semicolon should be a colon. 589 590 -- Quite a bit of work, but really helpful in the case where it helps, and 591 -- the need for this is based on actual experience with tracking down this 592 -- kind of error (the eye often easily mistakes semicolon for colon). 593 594 -- Note: we actually have enough information to patch up the tree, but 595 -- this may not be worth the effort. Also we could deal with the same 596 -- situation for EXIT with a label, but for now don't bother with that. 597 598 --------------------------------- 599 -- Parsing Routines by Chapter -- 600 --------------------------------- 601 602 -- Uncommented declarations in this section simply parse the construct 603 -- corresponding to their name, and return an ID value for the Node or 604 -- List that is created. 605 606 ------------- 607 -- Par.Ch2 -- 608 ------------- 609 610 package Ch2 is 611 function P_Pragma (Skipping : Boolean := False) return Node_Id; 612 -- Scan out a pragma. If Skipping is True, then the caller is skipping 613 -- the pragma in the context of illegal placement (this is used to avoid 614 -- some junk cascaded messages). 615 616 function P_Identifier (C : Id_Check := None) return Node_Id; 617 -- Scans out an identifier. The parameter C determines the treatment 618 -- of reserved identifiers. See declaration of Id_Check for details. 619 620 function P_Pragmas_Opt return List_Id; 621 -- This function scans for a sequence of pragmas in other than a 622 -- declaration sequence or statement sequence context. All pragmas 623 -- can appear except pragmas Assert and Debug, which are only allowed 624 -- in a declaration or statement sequence context. 625 626 procedure P_Pragmas_Misplaced; 627 -- Skips misplaced pragmas with a complaint 628 629 procedure P_Pragmas_Opt (List : List_Id); 630 -- Parses optional pragmas and appends them to the List 631 end Ch2; 632 633 ------------- 634 -- Par.Ch3 -- 635 ------------- 636 637 package Ch3 is 638 Missing_Begin_Msg : Error_Msg_Id; 639 -- This variable is set by a call to P_Declarative_Part. Normally it 640 -- is set to No_Error_Msg, indicating that no special processing is 641 -- required by the caller. The special case arises when a statement 642 -- is found in the sequence of declarations. In this case the Id of 643 -- the message issued ("declaration expected") is preserved in this 644 -- variable, then the caller can change it to an appropriate missing 645 -- begin message if indeed the BEGIN is missing. 646 647 function P_Array_Type_Definition return Node_Id; 648 function P_Basic_Declarative_Items return List_Id; 649 function P_Constraint_Opt return Node_Id; 650 function P_Declarative_Part return List_Id; 651 function P_Discrete_Choice_List return List_Id; 652 function P_Discrete_Range return Node_Id; 653 function P_Discrete_Subtype_Definition return Node_Id; 654 function P_Known_Discriminant_Part_Opt return List_Id; 655 function P_Signed_Integer_Type_Definition return Node_Id; 656 function P_Range return Node_Id; 657 function P_Range_Constraint return Node_Id; 658 function P_Record_Definition return Node_Id; 659 function P_Subtype_Mark return Node_Id; 660 function P_Subtype_Mark_Resync return Node_Id; 661 function P_Unknown_Discriminant_Part_Opt return Boolean; 662 663 function P_Access_Definition 664 (Null_Exclusion_Present : Boolean) return Node_Id; 665 -- Ada 2005 (AI-231/AI-254): The caller parses the null-exclusion part 666 -- and indicates if it was present 667 668 function P_Access_Type_Definition 669 (Header_Already_Parsed : Boolean := False) return Node_Id; 670 -- Ada 2005 (AI-254): The formal is used to indicate if the caller has 671 -- parsed the null_exclusion part. In this case the caller has also 672 -- removed the ACCESS token 673 674 procedure P_Component_Items (Decls : List_Id); 675 -- Scan out one or more component items and append them to the given 676 -- list. Only scans out more than one declaration in the case where the 677 -- source has a single declaration with multiple defining identifiers. 678 679 function P_Defining_Identifier (C : Id_Check := None) return Node_Id; 680 -- Scan out a defining identifier. The parameter C controls the 681 -- treatment of errors in case a reserved word is scanned. See the 682 -- declaration of this type for details. 683 684 function P_Interface_Type_Definition 685 (Abstract_Present : Boolean) return Node_Id; 686 -- Ada 2005 (AI-251): Parse the interface type definition part. Abstract 687 -- Present indicates if the reserved word "abstract" has been previously 688 -- found. It is used to report an error message because interface types 689 -- are by definition abstract tagged. We generate a record_definition 690 -- node if the list of interfaces is empty; otherwise we generate a 691 -- derived_type_definition node (the first interface in this list is the 692 -- ancestor interface). 693 694 function P_Null_Exclusion 695 (Allow_Anonymous_In_95 : Boolean := False) return Boolean; 696 -- Ada 2005 (AI-231): Parse the null-excluding part. A True result 697 -- indicates that the null-excluding part was present. 698 -- 699 -- Allow_Anonymous_In_95 is True if we are in a context that allows 700 -- anonymous access types in Ada 95, in which case "not null" is legal 701 -- if it precedes "access". 702 703 function P_Subtype_Indication 704 (Not_Null_Present : Boolean := False) return Node_Id; 705 -- Ada 2005 (AI-231): The flag Not_Null_Present indicates that the 706 -- null-excluding part has been scanned out and it was present. 707 708 function P_Range_Or_Subtype_Mark 709 (Allow_Simple_Expression : Boolean := False) return Node_Id; 710 -- Scans out a range or subtype mark, and also permits a general simple 711 -- expression if Allow_Simple_Expression is set to True. 712 713 function Init_Expr_Opt (P : Boolean := False) return Node_Id; 714 -- If an initialization expression is present (:= expression), then 715 -- it is scanned out and returned, otherwise Empty is returned if no 716 -- initialization expression is present. This procedure also handles 717 -- certain common error cases cleanly. The parameter P indicates if 718 -- a right paren can follow the expression (default = no right paren 719 -- allowed). 720 721 procedure Skip_Declaration (S : List_Id); 722 -- Used when scanning statements to skip past a misplaced declaration 723 -- The declaration is scanned out and appended to the given list. 724 -- Token is known to be a declaration token (in Token_Class_Declk) 725 -- on entry, so there definition is a declaration to be scanned. 726 727 function P_Subtype_Indication 728 (Subtype_Mark : Node_Id; 729 Not_Null_Present : Boolean := False) return Node_Id; 730 -- This version of P_Subtype_Indication is called when the caller has 731 -- already scanned out the subtype mark which is passed as a parameter. 732 -- Ada 2005 (AI-231): The flag Not_Null_Present indicates that the 733 -- null-excluding part has been scanned out and it was present. 734 735 function P_Subtype_Mark_Attribute (Type_Node : Node_Id) return Node_Id; 736 -- Parse a subtype mark attribute. The caller has already parsed the 737 -- subtype mark, which is passed in as the argument, and has checked 738 -- that the current token is apostrophe. 739 end Ch3; 740 741 ------------- 742 -- Par.Ch4 -- 743 ------------- 744 745 package Ch4 is 746 function P_Aggregate return Node_Id; 747 function P_Expression return Node_Id; 748 function P_Expression_Or_Range_Attribute return Node_Id; 749 function P_Function_Name return Node_Id; 750 function P_Name return Node_Id; 751 function P_Qualified_Simple_Name return Node_Id; 752 function P_Qualified_Simple_Name_Resync return Node_Id; 753 function P_Simple_Expression return Node_Id; 754 function P_Simple_Expression_Or_Range_Attribute return Node_Id; 755 756 function P_Case_Expression return Node_Id; 757 -- Scans out a case expression. Called with Token pointing to the CASE 758 -- keyword, and returns pointing to the terminating right parent, 759 -- semicolon, or comma, but does not consume this terminating token. 760 761 function P_Expression_If_OK return Node_Id; 762 -- Scans out an expression allowing an unparenthesized case expression, 763 -- if expression, or quantified expression to appear without enclosing 764 -- parentheses. However, if such an expression is not preceded by a left 765 -- paren, and followed by a right paren, an error message will be output 766 -- noting that parenthesization is required. 767 768 function P_Expression_No_Right_Paren return Node_Id; 769 -- Scans out an expression in contexts where the expression cannot be 770 -- terminated by a right paren (gives better error recovery if an errant 771 -- right paren is found after the expression). 772 773 function P_Expression_Or_Range_Attribute_If_OK return Node_Id; 774 -- Scans out an expression or range attribute where a conditional 775 -- expression is permitted to appear without surrounding parentheses. 776 -- However, if such an expression is not preceded by a left paren, and 777 -- followed by a right paren, an error message will be output noting 778 -- that parenthesization is required. 779 780 function P_If_Expression return Node_Id; 781 -- Scans out an if expression. Called with Token pointing to the 782 -- IF keyword, and returns pointing to the terminating right paren, 783 -- semicolon or comma, but does not consume this terminating token. 784 785 function P_Qualified_Expression (Subtype_Mark : Node_Id) return Node_Id; 786 -- This routine scans out a qualified expression when the caller has 787 -- already scanned out the name and apostrophe of the construct. 788 789 function P_Quantified_Expression return Node_Id; 790 -- This routine scans out a quantified expression when the caller has 791 -- already scanned out the keyword "for" of the construct. 792 end Ch4; 793 794 ------------- 795 -- Par.Ch5 -- 796 ------------- 797 798 package Ch5 is 799 function P_Condition return Node_Id; 800 -- Scan out and return a condition. Note that an error is given if 801 -- the condition is followed by a right parenthesis. 802 803 function P_Condition (Cond : Node_Id) return Node_Id; 804 -- Similar to the above, but the caller has already scanned out the 805 -- conditional expression and passes it as an argument. This form of 806 -- the call does not check for a following right parenthesis. 807 808 function P_Loop_Parameter_Specification return Node_Id; 809 -- Used in loop constructs and quantified expressions. 810 811 function P_Sequence_Of_Statements (SS_Flags : SS_Rec) return List_Id; 812 -- The argument indicates the acceptable termination tokens. 813 -- See body in Par.Ch5 for details of the use of this parameter. 814 815 procedure Parse_Decls_Begin_End (Parent : Node_Id); 816 -- Parses declarations and handled statement sequence, setting 817 -- fields of Parent node appropriately. 818 end Ch5; 819 820 ------------- 821 -- Par.Ch6 -- 822 ------------- 823 824 package Ch6 is 825 function P_Designator return Node_Id; 826 function P_Defining_Program_Unit_Name return Node_Id; 827 function P_Formal_Part return List_Id; 828 function P_Parameter_Profile return List_Id; 829 function P_Return_Statement return Node_Id; 830 function P_Subprogram_Specification return Node_Id; 831 832 procedure P_Mode (Node : Node_Id); 833 -- Sets In_Present and/or Out_Present flags in Node scanning past IN, 834 -- OUT or IN OUT tokens in the source. 835 836 function P_Subprogram (Pf_Flags : Pf_Rec) return Node_Id; 837 -- Scans out any construct starting with either of the keywords 838 -- PROCEDURE or FUNCTION. The parameter indicates which possible 839 -- possible kinds of construct (body, spec, instantiation etc.) 840 -- are permissible in the current context. 841 end Ch6; 842 843 ------------- 844 -- Par.Ch7 -- 845 ------------- 846 847 package Ch7 is 848 function P_Package (Pf_Flags : Pf_Rec) return Node_Id; 849 -- Scans out any construct starting with the keyword PACKAGE. The 850 -- parameter indicates which possible kinds of construct (body, spec, 851 -- instantiation etc.) are permissible in the current context. 852 end Ch7; 853 854 ------------- 855 -- Par.Ch8 -- 856 ------------- 857 858 package Ch8 is 859 function P_Use_Clause return Node_Id; 860 end Ch8; 861 862 ------------- 863 -- Par.Ch9 -- 864 ------------- 865 866 package Ch9 is 867 function P_Abort_Statement return Node_Id; 868 function P_Abortable_Part return Node_Id; 869 function P_Accept_Statement return Node_Id; 870 function P_Delay_Statement return Node_Id; 871 function P_Entry_Body return Node_Id; 872 function P_Protected return Node_Id; 873 function P_Requeue_Statement return Node_Id; 874 function P_Select_Statement return Node_Id; 875 function P_Task return Node_Id; 876 function P_Terminate_Alternative return Node_Id; 877 end Ch9; 878 879 -------------- 880 -- Par.Ch10 -- 881 -------------- 882 883 package Ch10 is 884 function P_Compilation_Unit return Node_Id; 885 -- Note: this function scans a single compilation unit, and checks that 886 -- an end of file follows this unit, diagnosing any unexpected input as 887 -- an error, and then skipping it, so that Token is set to Tok_EOF on 888 -- return. An exception is in syntax-only mode, where multiple 889 -- compilation units are permitted. In this case, P_Compilation_Unit 890 -- does not check for end of file and there may be more compilation 891 -- units to scan. The caller can uniquely detect this situation by the 892 -- fact that Token is not set to Tok_EOF on return. 893 -- 894 -- What about multiple unit/file capability that now exists??? 895 -- 896 -- The Ignore parameter is normally set False. It is set True in the 897 -- multiple unit per file mode if we are skipping past a unit that we 898 -- are not interested in. 899 end Ch10; 900 901 -------------- 902 -- Par.Ch11 -- 903 -------------- 904 905 package Ch11 is 906 function P_Handled_Sequence_Of_Statements return Node_Id; 907 function P_Raise_Expression return Node_Id; 908 function P_Raise_Statement return Node_Id; 909 910 function Parse_Exception_Handlers return List_Id; 911 -- Parses the partial construct EXCEPTION followed by a list of 912 -- exception handlers which appears in a number of productions, and 913 -- returns the list of exception handlers. 914 end Ch11; 915 916 -------------- 917 -- Par.Ch12 -- 918 -------------- 919 920 package Ch12 is 921 function P_Generic return Node_Id; 922 function P_Generic_Actual_Part_Opt return List_Id; 923 end Ch12; 924 925 -------------- 926 -- Par.Ch13 -- 927 -------------- 928 929 package Ch13 is 930 function P_Representation_Clause return Node_Id; 931 932 function Aspect_Specifications_Present 933 (Strict : Boolean := Ada_Version < Ada_2012) return Boolean; 934 -- This function tests whether the next keyword is WITH followed by 935 -- something that looks reasonably like an aspect specification. If so, 936 -- True is returned. Otherwise False is returned. In either case control 937 -- returns with the token pointer unchanged (i.e. pointing to the WITH 938 -- token in the case where True is returned). This function takes care 939 -- of generating appropriate messages if aspect specifications appear 940 -- in versions of Ada prior to Ada 2012. The parameter strict can be 941 -- set to True, to be rather strict about considering something to be 942 -- an aspect specification. If Strict is False, then the circuitry is 943 -- rather more generous in considering something ill-formed to be an 944 -- attempt at an aspect specification. The default is more strict for 945 -- Ada versions before Ada 2012 (where aspect specifications are not 946 -- permitted). Note: this routine never checks the terminator token 947 -- for aspects so it does not matter whether the aspect specifications 948 -- are terminated by semicolon or some other character. 949 950 function Get_Aspect_Specifications 951 (Semicolon : Boolean := True) return List_Id; 952 -- Parse a list of aspects but do not attach them to a declaration node. 953 -- Subsidiary to the following procedure. Used when parsing a subprogram 954 -- specification that may be a declaration or a body. 955 956 procedure P_Aspect_Specifications 957 (Decl : Node_Id; 958 Semicolon : Boolean := True); 959 -- This procedure scans out a series of aspect spefications. If argument 960 -- Semicolon is True, a terminating semicolon is also scanned. If this 961 -- argument is False, the scan pointer is left pointing past the aspects 962 -- and the caller must check for a proper terminator. 963 -- 964 -- P_Aspect_Specifications is called with the current token pointing to 965 -- either a WITH keyword starting an aspect specification, or an 966 -- instance of the terminator token. In the former case, the aspect 967 -- specifications are scanned out including the terminator token if it 968 -- it is a semicolon, and the Has_Aspect_Specifications flag is set in 969 -- the given declaration node. A list of aspects is built and stored for 970 -- this declaration node using a call to Set_Aspect_Specifications. If 971 -- no WITH keyword is present, then this call has no effect other than 972 -- scanning out the terminator if it is a semicolon. 973 974 -- If Decl is Error on entry, any scanned aspect specifications are 975 -- ignored and a message is output saying aspect specifications not 976 -- permitted here. If Decl is Empty, then scanned aspect specifications 977 -- are also ignored, but no error message is given (this is used when 978 -- the caller has already taken care of the error message). 979 980 function P_Code_Statement (Subtype_Mark : Node_Id) return Node_Id; 981 -- Function to parse a code statement. The caller has scanned out 982 -- the name to be used as the subtype mark (but has not checked that 983 -- it is suitable for use as a subtype mark, i.e. is either an 984 -- identifier or a selected component). The current token is an 985 -- apostrophe and the following token is either a left paren or 986 -- RANGE (the latter being an error to be caught by P_Code_Statement. 987 end Ch13; 988 989 -- Note: the parsing for annexe J features (i.e. obsolescent features) 990 -- is found in the logical section where these features would be if 991 -- they were not obsolescent. In particular: 992 993 -- Delta constraint is parsed by P_Delta_Constraint (3.5.9) 994 -- At clause is parsed by P_At_Clause (13.1) 995 -- Mod clause is parsed by P_Mod_Clause (13.5.1) 996 997 -------------- 998 -- Par.Endh -- 999 -------------- 1000 1001 -- Routines for handling end lines, including scope recovery 1002 1003 package Endh is 1004 function Check_End 1005 (Decl : Node_Id := Empty; 1006 Is_Loc : Source_Ptr := No_Location) return Boolean; 1007 -- Called when an end sequence is required. In the absence of an error 1008 -- situation, Token contains Tok_End on entry, but in a missing end 1009 -- case, this may not be the case. Pop_End_Context is used to determine 1010 -- the appropriate action to be taken. The returned result is True if 1011 -- an End sequence was encountered and False if no End sequence was 1012 -- present. This occurs if the END keyword encountered was determined 1013 -- to be improper and deleted (i.e. Pop_End_Context set End_Action to 1014 -- Skip_And_Reject). Note that the END sequence includes a semicolon, 1015 -- except in the case of END RECORD, where a semicolon follows the END 1016 -- RECORD, but is not part of the record type definition itself. 1017 -- 1018 -- If Decl is non-empty, then aspect specifications are permitted 1019 -- following the end, and Decl is the declaration node with which 1020 -- these aspect specifications are to be associated. If Decl is empty, 1021 -- then aspect specifications are not permitted and will generate an 1022 -- error message. 1023 -- 1024 -- Is_Loc is set to other than the default only for the case of a 1025 -- package declaration. It points to the IS keyword of the declaration, 1026 -- and is used to specialize the error messages for misplaced aspect 1027 -- specifications in this case. Note that Decl is always Empty if Is_Loc 1028 -- is set. 1029 1030 procedure End_Skip; 1031 -- Skip past an end sequence. On entry Token contains Tok_End, and we 1032 -- we know that the end sequence is syntactically incorrect, and that 1033 -- an appropriate error message has already been posted. The mission 1034 -- is simply to position the scan pointer to be the best guess of the 1035 -- position after the end sequence. We do not issue any additional 1036 -- error messages while carrying this out. 1037 1038 procedure End_Statements 1039 (Parent : Node_Id := Empty; 1040 Decl : Node_Id := Empty; 1041 Is_Sloc : Source_Ptr := No_Location); 1042 -- Called when an end is required or expected to terminate a sequence 1043 -- of statements. The caller has already made an appropriate entry in 1044 -- the Scope.Table to describe the expected form of the end. This can 1045 -- only be used in cases where the only appropriate terminator is end. 1046 -- If Parent is non-empty, then if a correct END line is encountered, 1047 -- the End_Label field of Parent is set appropriately. 1048 -- 1049 -- If Decl is non-null, then it is a declaration node, and aspect 1050 -- specifications are permitted after the end statement. These aspect 1051 -- specifications, if present, are stored in this declaration node. 1052 -- If Decl is null, then aspect specifications are not permitted after 1053 -- the end statement. 1054 -- 1055 -- In the case where Decl is null, Is_Sloc determines the handling. If 1056 -- it is set to No_Location, then aspect specifications are ignored and 1057 -- an error message is given. Is_Sloc is used in the package declaration 1058 -- case to point to the IS, and is used to specialize the error emssages 1059 -- issued in this case. 1060 end Endh; 1061 1062 -------------- 1063 -- Par.Sync -- 1064 -------------- 1065 1066 -- These procedures are used to resynchronize after errors. Following an 1067 -- error which is not immediately locally recoverable, the exception 1068 -- Error_Resync is raised. The handler for Error_Resync typically calls 1069 -- one of these recovery procedures to resynchronize the source position 1070 -- to a point from which parsing can be restarted. 1071 1072 -- Note: these procedures output an information message that tokens are 1073 -- being skipped, but this message is output only if the option for 1074 -- Multiple_Errors_Per_Line is set in Options. 1075 1076 package Sync is 1077 procedure Resync_Choice; 1078 -- Used if an error occurs scanning a choice. The scan pointer is 1079 -- advanced to the next vertical bar, arrow, or semicolon, whichever 1080 -- comes first. We also quit if we encounter an end of file. 1081 1082 procedure Resync_Cunit; 1083 -- Synchronize to next token which could be the start of a compilation 1084 -- unit, or to the end of file token. 1085 1086 procedure Resync_Expression; 1087 -- Used if an error is detected during the parsing of an expression. 1088 -- It skips past tokens until either a token which cannot be part of 1089 -- an expression is encountered (an expression terminator), or if a 1090 -- comma or right parenthesis or vertical bar is encountered at the 1091 -- current parenthesis level (a parenthesis level counter is maintained 1092 -- to carry out this test). 1093 1094 procedure Resync_Past_Malformed_Aspect; 1095 -- Used when parsing aspect specifications to skip a malformed aspect. 1096 -- The scan pointer is positioned next to a comma, a semicolon or "is" 1097 -- when the aspect applies to a body. 1098 1099 procedure Resync_Past_Semicolon; 1100 -- Used if an error occurs while scanning a sequence of declarations. 1101 -- The scan pointer is positioned past the next semicolon and the scan 1102 -- resumes. The scan is also resumed on encountering a token which 1103 -- starts a declaration (but we make sure to skip at least one token 1104 -- in this case, to avoid getting stuck in a loop). 1105 1106 procedure Resync_Past_Semicolon_Or_To_Loop_Or_Then; 1107 -- Used if an error occurs while scanning a sequence of statements. The 1108 -- scan pointer is positioned past the next semicolon, or to the next 1109 -- occurrence of either then or loop, and the scan resumes. 1110 1111 procedure Resync_Semicolon_List; 1112 -- Used if an error occurs while scanning a parenthesized list of items 1113 -- separated by semicolons. The scan pointer is advanced to the next 1114 -- semicolon or right parenthesis at the outer parenthesis level, or 1115 -- to the next is or RETURN keyword occurrence, whichever comes first. 1116 1117 procedure Resync_To_Semicolon; 1118 -- Similar to Resync_Past_Semicolon, except that the scan pointer is 1119 -- left pointing to the semicolon rather than past it. 1120 1121 procedure Resync_To_When; 1122 -- Used when an error occurs scanning an entry index specification. The 1123 -- scan pointer is positioned to the next WHEN (or to IS or semicolon if 1124 -- either of these appear before WHEN, indicating another error has 1125 -- occurred). 1126 end Sync; 1127 1128 -------------- 1129 -- Par.Tchk -- 1130 -------------- 1131 1132 -- Routines to check for expected tokens 1133 1134 package Tchk is 1135 1136 -- Procedures with names of the form T_xxx, where Tok_xxx is a token 1137 -- name, check that the current token matches the required token, and 1138 -- if so, scan past it. If not, an error is issued indicating that 1139 -- the required token is not present (xxx expected). In most cases, the 1140 -- scan pointer is not moved in the not-found case, but there are some 1141 -- exceptions to this, see for example T_Id, where the scan pointer is 1142 -- moved across a literal appearing where an identifier is expected. 1143 1144 procedure T_Abort; 1145 procedure T_Arrow; 1146 procedure T_At; 1147 procedure T_Body; 1148 procedure T_Box; 1149 procedure T_Colon; 1150 procedure T_Colon_Equal; 1151 procedure T_Comma; 1152 procedure T_Dot_Dot; 1153 procedure T_For; 1154 procedure T_Greater_Greater; 1155 procedure T_Identifier; 1156 procedure T_In; 1157 procedure T_Is; 1158 procedure T_Left_Paren; 1159 procedure T_Loop; 1160 procedure T_Mod; 1161 procedure T_New; 1162 procedure T_Of; 1163 procedure T_Or; 1164 procedure T_Private; 1165 procedure T_Range; 1166 procedure T_Record; 1167 procedure T_Right_Paren; 1168 procedure T_Semicolon; 1169 procedure T_Then; 1170 procedure T_Type; 1171 procedure T_Use; 1172 procedure T_When; 1173 procedure T_With; 1174 1175 -- Procedures having names of the form TF_xxx, where Tok_xxx is a token 1176 -- name check that the current token matches the required token, and 1177 -- if so, scan past it. If not, an error message is issued indicating 1178 -- that the required token is not present (xxx expected). 1179 1180 -- If the missing token is at the end of the line, then control returns 1181 -- immediately after posting the message. If there are remaining tokens 1182 -- on the current line, a search is conducted to see if the token 1183 -- appears later on the current line, as follows: 1184 1185 -- A call to Scan_Save is issued and a forward search for the token 1186 -- is carried out. If the token is found on the current line before a 1187 -- semicolon, then it is scanned out and the scan continues from that 1188 -- point. If not the scan is restored to the point where it was missing. 1189 1190 procedure TF_Arrow; 1191 procedure TF_Is; 1192 procedure TF_Loop; 1193 procedure TF_Return; 1194 procedure TF_Semicolon; 1195 procedure TF_Then; 1196 procedure TF_Use; 1197 1198 -- Procedures with names of the form U_xxx, where Tok_xxx is a token 1199 -- name, are just like the corresponding T_xxx procedures except that 1200 -- an error message, if given, is unconditional. 1201 1202 procedure U_Left_Paren; 1203 procedure U_Right_Paren; 1204 end Tchk; 1205 1206 -------------- 1207 -- Par.Util -- 1208 -------------- 1209 1210 package Util is 1211 function Bad_Spelling_Of (T : Token_Type) return Boolean; 1212 -- This function is called in an error situation. It checks if the 1213 -- current token is an identifier whose name is a plausible bad 1214 -- spelling of the given keyword token, and if so, issues an error 1215 -- message, sets Token from T, and returns True. Otherwise Token is 1216 -- unchanged, and False is returned. 1217 1218 procedure Check_Bad_Layout; 1219 -- Check for bad indentation in RM checking mode. Used for statements 1220 -- and declarations. Checks if current token is at start of line and 1221 -- is exdented from the current expected end column, and if so an 1222 -- error message is generated. 1223 1224 procedure Check_Misspelling_Of (T : Token_Type); 1225 pragma Inline (Check_Misspelling_Of); 1226 -- This is similar to the function above, except that it does not 1227 -- return a result. It is typically used in a situation where any 1228 -- identifier is an error, and it makes sense to simply convert it 1229 -- to the given token if it is a plausible misspelling of it. 1230 1231 procedure Check_95_Keyword (Token_95, Next : Token_Type); 1232 -- This routine checks if the token after the current one matches the 1233 -- Next argument. If so, the scan is backed up to the current token 1234 -- and Token_Type is changed to Token_95 after issuing an appropriate 1235 -- error message ("(Ada 83) keyword xx cannot be used"). If not, 1236 -- the scan is backed up with Token_Type unchanged. This routine 1237 -- is used to deal with an attempt to use a 95 keyword in Ada 83 1238 -- mode. The caller has typically checked that the current token, 1239 -- an identifier, matches one of the 95 keywords. 1240 1241 procedure Check_Future_Keyword; 1242 -- Emit a warning if the current token is a valid identifier in the 1243 -- language version in use, but is a reserved word in a later language 1244 -- version (unless the language version in use is Ada 83). 1245 1246 procedure Check_Simple_Expression (E : Node_Id); 1247 -- Given an expression E, that has just been scanned, so that Expr_Form 1248 -- is still set, outputs an error if E is a non-simple expression. E is 1249 -- not modified by this call. 1250 1251 procedure Check_Simple_Expression_In_Ada_83 (E : Node_Id); 1252 -- Like Check_Simple_Expression, except that the error message is only 1253 -- given when operating in Ada 83 mode, and includes "in Ada 83". 1254 1255 function Check_Subtype_Mark (Mark : Node_Id) return Node_Id; 1256 -- Called to check that a node representing a name (or call) is 1257 -- suitable for a subtype mark, i.e, that it is an identifier or 1258 -- a selected component. If so, or if it is already Error, then 1259 -- it is returned unchanged. Otherwise an error message is issued 1260 -- and Error is returned. 1261 1262 function Comma_Present return Boolean; 1263 -- Used in comma delimited lists to determine if a comma is present, or 1264 -- can reasonably be assumed to have been present (an error message is 1265 -- generated in the latter case). If True is returned, the scan has been 1266 -- positioned past the comma. If False is returned, the scan position 1267 -- is unchanged. Note that all comma-delimited lists are terminated by 1268 -- a right paren, so the only legitimate tokens when Comma_Present is 1269 -- called are right paren and comma. If some other token is found, then 1270 -- Comma_Present has the job of deciding whether it is better to pretend 1271 -- a comma was present, post a message for a missing comma and return 1272 -- True, or return False and let the caller diagnose the missing right 1273 -- parenthesis. 1274 1275 procedure Discard_Junk_Node (N : Node_Id); 1276 procedure Discard_Junk_List (L : List_Id); 1277 pragma Inline (Discard_Junk_Node); 1278 pragma Inline (Discard_Junk_List); 1279 -- These procedures do nothing at all, their effect is simply to discard 1280 -- the argument. A typical use is to skip by some junk that is not 1281 -- expected in the current context. 1282 1283 procedure Ignore (T : Token_Type); 1284 -- If current token matches T, then give an error message and skip 1285 -- past it, otherwise the call has no effect at all. T may be any 1286 -- reserved word token, or comma, left or right paren, or semicolon. 1287 1288 function Is_Reserved_Identifier (C : Id_Check := None) return Boolean; 1289 -- Test if current token is a reserved identifier. This test is based 1290 -- on the token being a keyword and being spelled in typical identifier 1291 -- style (i.e. starting with an upper case letter). The parameter C 1292 -- determines the special treatment if a reserved word is encountered 1293 -- that has the normal casing of a reserved word. 1294 1295 procedure Merge_Identifier (Prev : Node_Id; Nxt : Token_Type); 1296 -- Called when the previous token is an identifier (whose Token_Node 1297 -- value is given by Prev) to check if current token is an identifier 1298 -- that can be merged with the previous one adding an underscore. The 1299 -- merge is only attempted if the following token matches Nxt. If all 1300 -- conditions are met, an error message is issued, and the merge is 1301 -- carried out, modifying the Chars field of Prev. 1302 1303 function Next_Token_Is (Tok : Token_Type) return Boolean; 1304 -- Looks at token after current one and returns True if the token type 1305 -- matches Tok. The scan is unconditionally restored on return. 1306 1307 procedure No_Constraint; 1308 -- Called in a place where no constraint is allowed, but one might 1309 -- appear due to a common error (e.g. after the type mark in a procedure 1310 -- parameter. If a constraint is present, an error message is posted, 1311 -- and the constraint is scanned and discarded. 1312 1313 procedure Push_Scope_Stack; 1314 pragma Inline (Push_Scope_Stack); 1315 -- Push a new entry onto the scope stack. Scope.Last (the stack pointer) 1316 -- is incremented. The Junk field is preinitialized to False. The caller 1317 -- is expected to fill in all remaining entries of the new top stack 1318 -- entry at Scope.Table (Scope.Last). 1319 1320 procedure Pop_Scope_Stack; 1321 -- Pop an entry off the top of the scope stack. Scope_Last (the scope 1322 -- table stack pointer) is decremented by one. It is a fatal error to 1323 -- try to pop off the dummy entry at the bottom of the stack (i.e. 1324 -- Scope.Last must be non-zero at the time of call). 1325 1326 function Separate_Present return Boolean; 1327 -- Determines if the current token is either Tok_Separate, or an 1328 -- identifier that is a possible misspelling of "separate" followed 1329 -- by a semicolon. True is returned if so, otherwise False. 1330 1331 procedure Signal_Bad_Attribute; 1332 -- The current token is an identifier that is supposed to be an 1333 -- attribute identifier but is not. This routine posts appropriate 1334 -- error messages, including a check for a near misspelling. 1335 1336 function Token_Is_At_Start_Of_Line return Boolean; 1337 pragma Inline (Token_Is_At_Start_Of_Line); 1338 -- Determines if the current token is the first token on the line 1339 1340 function Token_Is_At_End_Of_Line return Boolean; 1341 -- Determines if the current token is the last token on the line 1342 1343 procedure Warn_If_Standard_Redefinition (N : Node_Id); 1344 -- Issues a warning if Warn_On_Standard_Redefinition is set True, and 1345 -- the Node N (which is a Defining_Identifier node with the Chars field 1346 -- set) is a renaming of an entity in package Standard. 1347 1348 end Util; 1349 1350 -------------- 1351 -- Par.Prag -- 1352 -------------- 1353 1354 -- The processing for pragmas is split off from chapter 2 1355 1356 function Prag (Pragma_Node : Node_Id; Semi : Source_Ptr) return Node_Id; 1357 -- This function is passed a tree for a pragma that has been scanned out. 1358 -- The pragma is syntactically well formed according to the general syntax 1359 -- for pragmas and the pragma identifier is for one of the recognized 1360 -- pragmas. It performs specific syntactic checks for specific pragmas. 1361 -- The result is the input node if it is OK, or Error otherwise. The 1362 -- reason that this is separated out is to facilitate the addition 1363 -- of implementation defined pragmas. The second parameter records the 1364 -- location of the semicolon following the pragma (this is needed for 1365 -- correct processing of the List and Page pragmas). The returned value 1366 -- is a copy of Pragma_Node, or Error if an error is found. Note that 1367 -- at the point where Prag is called, the right paren ending the pragma 1368 -- has been scanned out, and except in the case of pragma Style_Checks, 1369 -- so has the following semicolon. For Style_Checks, the caller delays 1370 -- the scanning of the semicolon so that it will be scanned using the 1371 -- settings from the Style_Checks pragma preceding it. 1372 1373 -------------- 1374 -- Par.Labl -- 1375 -------------- 1376 1377 procedure Labl; 1378 -- This procedure creates implicit label declarations for all labels that 1379 -- are declared in the current unit. Note that this could conceptually be 1380 -- done at the point where the labels are declared, but it is tricky to do 1381 -- it then, since the tree is not hooked up at the point where the label is 1382 -- declared (e.g. a sequence of statements is not yet attached to its 1383 -- containing scope at the point a label in the sequence is found). 1384 1385 -------------- 1386 -- Par.Load -- 1387 -------------- 1388 1389 procedure Load; 1390 -- This procedure loads all subsidiary units that are required by this 1391 -- unit, including with'ed units, specs for bodies, and parents for child 1392 -- units. It does not load bodies for inlined procedures and generics, 1393 -- since we don't know till semantic analysis is complete what is needed. 1394 1395 ----------- 1396 -- Stubs -- 1397 ----------- 1398 1399 -- The package bodies can see all routines defined in all other subpackages 1400 1401 use Ch2; 1402 use Ch3; 1403 use Ch4; 1404 use Ch5; 1405 use Ch6; 1406 use Ch7; 1407 use Ch8; 1408 use Ch9; 1409 use Ch10; 1410 use Ch11; 1411 use Ch12; 1412 use Ch13; 1413 1414 use Endh; 1415 use Tchk; 1416 use Sync; 1417 use Util; 1418 1419 package body Ch2 is separate; 1420 package body Ch3 is separate; 1421 package body Ch4 is separate; 1422 package body Ch5 is separate; 1423 package body Ch6 is separate; 1424 package body Ch7 is separate; 1425 package body Ch8 is separate; 1426 package body Ch9 is separate; 1427 package body Ch10 is separate; 1428 package body Ch11 is separate; 1429 package body Ch12 is separate; 1430 package body Ch13 is separate; 1431 1432 package body Endh is separate; 1433 package body Tchk is separate; 1434 package body Sync is separate; 1435 package body Util is separate; 1436 1437 function Prag (Pragma_Node : Node_Id; Semi : Source_Ptr) return Node_Id 1438 is separate; 1439 1440 procedure Labl is separate; 1441 procedure Load is separate; 1442 1443-- Start of processing for Par 1444 1445begin 1446 Compiler_State := Parsing; 1447 1448 -- Deal with configuration pragmas case first 1449 1450 if Configuration_Pragmas then 1451 declare 1452 Pragmas : constant List_Id := Empty_List; 1453 P_Node : Node_Id; 1454 1455 begin 1456 loop 1457 if Token = Tok_EOF then 1458 Compiler_State := Analyzing; 1459 return Pragmas; 1460 1461 elsif Token /= Tok_Pragma then 1462 Error_Msg_SC ("only pragmas allowed in configuration file"); 1463 Compiler_State := Analyzing; 1464 return Error_List; 1465 1466 else 1467 P_Node := P_Pragma; 1468 1469 if Nkind (P_Node) = N_Pragma then 1470 1471 -- Give error if bad pragma 1472 1473 if not Is_Configuration_Pragma_Name (Pragma_Name (P_Node)) 1474 and then Pragma_Name (P_Node) /= Name_Source_Reference 1475 then 1476 if Is_Pragma_Name (Pragma_Name (P_Node)) then 1477 Error_Msg_N 1478 ("only configuration pragmas allowed " & 1479 "in configuration file", P_Node); 1480 else 1481 Error_Msg_N 1482 ("unrecognized pragma in configuration file", 1483 P_Node); 1484 end if; 1485 1486 -- Pragma is OK config pragma, so collect it 1487 1488 else 1489 Append (P_Node, Pragmas); 1490 end if; 1491 end if; 1492 end if; 1493 end loop; 1494 end; 1495 1496 -- Normal case of compilation unit 1497 1498 else 1499 Save_Opt_Config_Switches (Save_Config_Switches); 1500 1501 -- The following loop runs more than once in syntax check mode 1502 -- where we allow multiple compilation units in the same file 1503 -- and in Multiple_Unit_Per_file mode where we skip units till 1504 -- we get to the unit we want. 1505 1506 for Ucount in Pos loop 1507 Set_Opt_Config_Switches 1508 (Is_Internal_File_Name (File_Name (Current_Source_File)), 1509 Current_Source_Unit = Main_Unit); 1510 1511 -- Initialize scope table and other parser control variables 1512 1513 Compiler_State := Parsing; 1514 Scope.Init; 1515 Scope.Increment_Last; 1516 Scope.Table (0).Etyp := E_Dummy; 1517 SIS_Entry_Active := False; 1518 Last_Resync_Point := No_Location; 1519 1520 Goto_List := New_Elmt_List; 1521 Label_List := New_Elmt_List; 1522 1523 -- If in multiple unit per file mode, skip past ignored unit 1524 1525 if Ucount < Multiple_Unit_Index then 1526 1527 -- We skip in syntax check only mode, since we don't want to do 1528 -- anything more than skip past the unit and ignore it. This means 1529 -- we skip processing like setting up a unit table entry. 1530 1531 declare 1532 Save_Operating_Mode : constant Operating_Mode_Type := 1533 Operating_Mode; 1534 1535 Save_Style_Check : constant Boolean := Style_Check; 1536 1537 begin 1538 Operating_Mode := Check_Syntax; 1539 Style_Check := False; 1540 Discard_Node (P_Compilation_Unit); 1541 Operating_Mode := Save_Operating_Mode; 1542 Style_Check := Save_Style_Check; 1543 1544 -- If we are at an end of file, and not yet at the right unit, 1545 -- then we have a fatal error. The unit is missing. 1546 1547 if Token = Tok_EOF then 1548 Error_Msg_SC ("file has too few compilation units"); 1549 raise Unrecoverable_Error; 1550 end if; 1551 end; 1552 1553 -- Here if we are not skipping a file in multiple unit per file mode. 1554 -- Parse the unit that we are interested in. Note that in check 1555 -- syntax mode we are interested in all units in the file. 1556 1557 else 1558 declare 1559 Comp_Unit_Node : constant Node_Id := P_Compilation_Unit; 1560 1561 begin 1562 -- If parsing was successful and we are not in check syntax 1563 -- mode, check that language-defined units are compiled in GNAT 1564 -- mode. For this purpose we do NOT consider renamings in annex 1565 -- J as predefined. That allows users to compile their own 1566 -- versions of these files, and in particular, in the VMS 1567 -- implementation, the DEC versions can be substituted for the 1568 -- standard Ada 95 versions. Another exception is System.RPC 1569 -- and its children. This allows a user to supply their own 1570 -- communication layer. 1571 1572 if Comp_Unit_Node /= Error 1573 and then Operating_Mode = Generate_Code 1574 and then Current_Source_Unit = Main_Unit 1575 and then not GNAT_Mode 1576 then 1577 declare 1578 Uname : constant String := 1579 Get_Name_String 1580 (Unit_Name (Current_Source_Unit)); 1581 Name : String (1 .. Uname'Length - 2); 1582 1583 begin 1584 -- Because Unit_Name includes "%s"/"%b", we need to strip 1585 -- the last two characters to get the real unit name. 1586 1587 Name := Uname (Uname'First .. Uname'Last - 2); 1588 1589 if Name = "ada" or else 1590 Name = "interfaces" or else 1591 Name = "system" 1592 then 1593 Error_Msg 1594 ("language-defined units cannot be recompiled", 1595 Sloc (Unit (Comp_Unit_Node))); 1596 1597 elsif Name'Length > 4 1598 and then 1599 Name (Name'First .. Name'First + 3) = "ada." 1600 then 1601 Error_Msg 1602 ("user-defined descendents of package Ada " & 1603 "are not allowed", 1604 Sloc (Unit (Comp_Unit_Node))); 1605 1606 elsif Name'Length > 11 1607 and then 1608 Name (Name'First .. Name'First + 10) = "interfaces." 1609 then 1610 Error_Msg 1611 ("user-defined descendents of package Interfaces " & 1612 "are not allowed", 1613 Sloc (Unit (Comp_Unit_Node))); 1614 1615 elsif Name'Length > 7 1616 and then Name (Name'First .. Name'First + 6) = "system." 1617 and then Name /= "system.rpc" 1618 and then 1619 (Name'Length < 11 1620 or else Name (Name'First .. Name'First + 10) /= 1621 "system.rpc.") 1622 then 1623 Error_Msg 1624 ("user-defined descendents of package System " & 1625 "are not allowed", 1626 Sloc (Unit (Comp_Unit_Node))); 1627 end if; 1628 end; 1629 end if; 1630 end; 1631 1632 -- All done if at end of file 1633 1634 exit when Token = Tok_EOF; 1635 1636 -- If we are not at an end of file, it means we are in syntax 1637 -- check only mode, and we keep the loop going to parse all 1638 -- remaining units in the file. 1639 1640 end if; 1641 1642 Restore_Opt_Config_Switches (Save_Config_Switches); 1643 end loop; 1644 1645 -- Now that we have completely parsed the source file, we can complete 1646 -- the source file table entry. 1647 1648 Complete_Source_File_Entry; 1649 1650 -- An internal error check, the scope stack should now be empty 1651 1652 pragma Assert (Scope.Last = 0); 1653 1654 -- Here we make the SCO table entries for the main unit 1655 1656 if Generate_SCO then 1657 SCO_Record (Main_Unit); 1658 end if; 1659 1660 -- Remaining steps are to create implicit label declarations and to load 1661 -- required subsidiary sources. These steps are required only if we are 1662 -- doing semantic checking. 1663 1664 if Operating_Mode /= Check_Syntax or else Debug_Flag_F then 1665 Par.Labl; 1666 Par.Load; 1667 end if; 1668 1669 -- Restore settings of switches saved on entry 1670 1671 Restore_Opt_Config_Switches (Save_Config_Switches); 1672 Set_Comes_From_Source_Default (False); 1673 Compiler_State := Analyzing; 1674 return Empty_List; 1675 end if; 1676end Par; 1677