4 Changes in version 1.49b:
9 causes Bison to produce a Generalized LR (GLR) parser, capable of handling
10 almost any context-free grammar, ambiguous or not. The new declarations
11 %dprec and %merge on grammar rules allow parse-time resolution of
12 ambiguities. Contributed by Paul Hilfinger.
15 When not in Yacc compatibility mode, when the output file was not
16 specified, runnning `bison foo/bar.y' created `foo/bar.c'. It
20 The undefined token was systematically mapped to 2 which prevented
21 the use of 2 from the user. This is no longer the case.
23 * Unknown token numbers
24 If yylex returned a code out of range, yyparse could die. This is
28 According to POSIX, the error token should be numbered as 256.
29 Bison extends this requirement by making it a preference: *if* the
30 user specified that one of her tokens is numbered 256, then error
31 will be mapped onto another number.
33 * Verbose error messages
34 They no longer report `..., expecting error or...' for state where
35 error recovery is possible.
38 Defaults to `$end' instead of `$'.
40 * Error recovery now conforms to documentation and to POSIX
41 When a Bison-generated parser encounters a syntax error, it now pops
42 the stack until it finds a state that allows shifting the error
43 token. Formerly, it popped the stack until it found a state that
44 allowed some non-error action other than a default reduction on the
45 error token. The new behavior has long been the documented behavior,
46 and has long been required by POSIX. For more details, please see
47 <http://mail.gnu.org/pipermail/bug-bison/2002-May/001452.html>.
50 Popped tokens and nonterminals are now reported.
53 Large grammars are now supported (large token numbers, large grammar
54 size (= sum of the LHS and RHS lengths), large LALR tables).
56 * Explicit initial rule
57 Bison used to play hacks with the initial rule, which the user does
58 not write. It is now explicit, and visible in the reports and
62 Before, Bison reported the useless rules, but, although not used,
63 included them in the parsers. They are now actually removed.
65 * Useless rules, useless nonterminals
66 They are now reported, as a warning, with their locations.
69 Rules that can never be reduced because of conflicts are now
72 * Incorrect `Token not used'
77 exp: '0' %prec useful;
79 where a token was used to set the precedence of the last rule,
80 bison reported both `useful' and `useless' as useless tokens.
82 * Revert the C++ namespace changes introduced in 1.31
83 as they caused too many portability hassles.
86 By an accident of design, the default computation of @$ was
87 performed after another default computation was performed: @$ = @1.
88 The latter is now removed: YYLLOC_DEFAULT is fully responsible of
89 the computation of @$.
92 The token end of file may be specified by the user, in which case,
93 the user symbol is used in the reports, the graphs, and the verbose
94 error messages instead of `$end', which remains being the defaults.
98 %token YYEOF 0 "end of file"
101 This old option, which has been broken for ages, is removed.
104 Croatian, thanks to Denis Lackovic.
106 * Incorrect token definitions
107 When fed with `%token 'a' "A"', Bison used to output `#define 'a' 65'.
109 * Token definitions as enums
110 Tokens are output both as the traditional #define's, and, provided
111 the compiler supports ANSI C or is a C++ compiler, as enums.
112 This helps debuggers producing symbols instead of values.
115 In addition to --verbose, bison supports --report=THINGS, which
116 produces additional information:
118 complete the core item sets with their closure
120 explicitly associate lookaheads to items
122 describe shift/reduce conflicts solving.
123 Bison used to systematically output this information on top of
124 the report. Solved conflicts are now attached to their states.
127 Previous versions don't complain when there is a type clash on
128 the default action if the rule has a mid-rule action, such as in:
136 Changes in version 1.35, 2002-03-25:
139 Some projects use Bison's C parser with C++ compilers, and define
140 YYSTYPE as a class. The recent adjustment of C parsers for data
141 alignment and 64 bit architectures made this impossible.
143 Because for the time being no real solution for C++ parser
144 generation exists, kludges were implemented in the parser to
145 maintain this use. In the future, when Bison has C++ parsers, this
146 kludge will be disabled.
148 This kludge also addresses some C++ problems when the stack was
152 Changes in version 1.34, 2002-03-12:
154 * File name clashes are detected
155 $ bison foo.y -d -o foo.x
156 fatal error: header and parser would both be named `foo.x'
158 * A missing `;' at the end of a rule triggers a warning
159 In accordance with POSIX, and in agreement with other
160 Yacc implementations, Bison will mandate this semicolon in the near
161 future. This eases the implementation of a Bison parser of Bison
162 grammars by making this grammar LALR(1) instead of LR(2). To
163 facilitate the transition, this release introduces a warning.
165 * Revert the C++ namespace changes introduced in 1.31, as they caused too
166 many portability hassles.
168 * DJGPP support added.
170 * Fix test suite portability problems.
172 Changes in version 1.33, 2002-02-07:
175 Groff could not be compiled for the definition of size_t was lacking
176 under some conditions.
181 Changes in version 1.32, 2002-01-23:
183 * Fix Yacc output file names
187 * Italian, Dutch translations
189 Changes in version 1.31, 2002-01-14:
193 * GNU Gettext and %expect
194 GNU Gettext asserts 10 s/r conflicts, but there are 7. Now that
195 Bison dies on incorrect %expectations, we fear there will be
196 too many bug reports for Gettext, so _for the time being_, %expect
197 does not trigger an error when the input file is named `plural.y'.
199 * Use of alloca in parsers
200 If YYSTACK_USE_ALLOCA is defined to 0, then the parsers will use
201 malloc exclusively. Since 1.29, but was not NEWS'ed.
203 alloca is used only when compiled with GCC, to avoid portability
206 * When the generated parser lacks debugging code, YYDEBUG is now 0
207 (as POSIX requires) instead of being undefined.
210 Bison has always permitted actions such as { $$ = $1 }: it adds the
211 ending semicolon. Now if in Yacc compatibility mode, the semicolon
212 is no longer output: one has to write { $$ = $1; }.
214 * Better C++ compliance
215 The output parsers try to respect C++ namespaces.
216 [This turned out to be a failed experiment, and it was reverted later.]
219 Fixed bugs when reporting useless nonterminals.
222 The parsers work properly on 64 bit hosts.
225 Some calls to strerror resulted in scrambled or missing error messages.
228 When the number of shift/reduce conflicts is correct, don't issue
231 * The verbose report includes the rule line numbers.
233 * Rule line numbers are fixed in traces.
235 * Swedish translation
238 Verbose parse error messages from the parsers are better looking.
239 Before: parse error: unexpected `'/'', expecting `"number"' or `'-'' or `'(''
240 Now: parse error: unexpected '/', expecting "number" or '-' or '('
242 * Fixed parser memory leaks.
243 When the generated parser was using malloc to extend its stacks, the
244 previous allocations were not freed.
246 * Fixed verbose output file.
247 Some newlines were missing.
248 Some conflicts in state descriptions were missing.
250 * Fixed conflict report.
251 Option -v was needed to get the result.
255 Mismatches are errors, not warnings.
257 * Fixed incorrect processing of some invalid input.
259 * Fixed CPP guards: 9foo.h uses BISON_9FOO_H instead of 9FOO_H.
261 * Fixed some typos in the documentation.
263 * %token MY_EOF 0 is supported.
264 Before, MY_EOF was silently renumbered as 257.
266 * doc/refcard.tex is updated.
268 * %output, %file-prefix, %name-prefix.
272 New, aliasing `--output-file'.
274 Changes in version 1.30, 2001-10-26:
276 * `--defines' and `--graph' have now an optionnal argument which is the
277 output file name. `-d' and `-g' do not change, they do not take any
280 * `%source_extension' and `%header_extension' are removed, failed
285 Changes in version 1.29, 2001-09-07:
287 * The output file does not define const, as this caused problems when used
288 with common autoconfiguration schemes. If you still use ancient compilers
289 that lack const, compile with the equivalent of the C compiler option
290 `-Dconst='. autoconf's AC_C_CONST macro provides one way to do this.
292 * Added `-g' and `--graph'.
294 * The Bison manual is now distributed under the terms of the GNU FDL.
296 * The input and the output files has automatically a similar extension.
298 * Russian translation added.
300 * NLS support updated; should hopefully be less troublesome.
302 * Added the old Bison reference card.
304 * Added `--locations' and `%locations'.
306 * Added `-S' and `--skeleton'.
308 * `%raw', `-r', `--raw' is disabled.
310 * Special characters are escaped when output. This solves the problems
311 of the #line lines with path names including backslashes.
314 `%yacc', `%fixed_output_files', `%defines', `%no_parser', `%verbose',
315 `%debug', `%source_extension' and `%header_extension'.
318 Automatic location tracking.
320 Changes in version 1.28, 1999-07-06:
322 * Should compile better now with K&R compilers.
326 * Fixed a problem with escaping the double quote character.
328 * There is now a FAQ.
330 Changes in version 1.27:
332 * The make rule which prevented bison.simple from being created on
333 some systems has been fixed.
335 Changes in version 1.26:
337 * Bison now uses automake.
339 * New mailing lists: <bug-bison@gnu.org> and <help-bison@gnu.org>.
341 * Token numbers now start at 257 as previously documented, not 258.
343 * Bison honors the TMPDIR environment variable.
345 * A couple of buffer overruns have been fixed.
347 * Problems when closing files should now be reported.
349 * Generated parsers should now work even on operating systems which do
350 not provide alloca().
352 Changes in version 1.25, 1995-10-16:
354 * Errors in the input grammar are not fatal; Bison keeps reading
355 the grammar file, and reports all the errors found in it.
357 * Tokens can now be specified as multiple-character strings: for
358 example, you could use "<=" for a token which looks like <=, instead
359 of chosing a name like LESSEQ.
361 * The %token_table declaration says to write a table of tokens (names
362 and numbers) into the parser file. The yylex function can use this
363 table to recognize multiple-character string tokens, or for other
366 * The %no_lines declaration says not to generate any #line preprocessor
367 directives in the parser file.
369 * The %raw declaration says to use internal Bison token numbers, not
370 Yacc-compatible token numbers, when token names are defined as macros.
372 * The --no-parser option produces the parser tables without including
373 the parser engine; a project can now use its own parser engine.
374 The actions go into a separate file called NAME.act, in the form of
375 a switch statement body.
377 Changes in version 1.23:
379 The user can define YYPARSE_PARAM as the name of an argument to be
380 passed into yyparse. The argument should have type void *. It should
381 actually point to an object. Grammar actions can access the variable
382 by casting it to the proper pointer type.
384 Line numbers in output file corrected.
386 Changes in version 1.22:
390 Changes in version 1.20:
392 Output file does not redefine const for C++.
400 Copyright (C) 2001, 2002 Free Software Foundation, Inc.
402 This file is part of GNU Autoconf.
404 GNU Autoconf is free software; you can redistribute it and/or modify
405 it under the terms of the GNU General Public License as published by
406 the Free Software Foundation; either version 2, or (at your option)
409 GNU Autoconf is distributed in the hope that it will be useful,
410 but WITHOUT ANY WARRANTY; without even the implied warranty of
411 MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the
412 GNU General Public License for more details.
414 You should have received a copy of the GNU General Public License
415 along with autoconf; see the file COPYING. If not, write to
416 the Free Software Foundation, Inc., 59 Temple Place - Suite 330,
417 Boston, MA 02111-1307, USA.