4 Changes in version 1.49b:
7 The undefined token was systematically mapped to 2 which prevented
8 the use of 2 from the user. This is no longer the case.
11 If yylex returned a code out of range, yyparse could die. This is
15 According to POSIX, the error token should be numbered as 256.
16 Bison extends this requirement by making it a preference: *if* the
17 user specified that one of her tokens is numbered 256, then error
18 will be mapped onto another number.
21 Are now supported (large token numbers, large grammar size (= sum of
22 the LHS and RHS lengths), large LALR tables).
24 * The initial rule is explicit.
25 Bison used to play hacks with the initial rule, which the user does
26 not write. It is now explicit, and visible in the reports and
29 * Useless rules are actually removed.
30 Before, Bison reported the useless rules, but, although not used,
31 included them in the parsers.
33 * False `Token not used' report fixed.
38 exp: '0' %prec useful;
40 where a token was used to set the precedence of the last rule,
41 bison reported both `useful' and `useless' as useless tokens.
43 * Revert the C++ namespace changes introduced in 1.31, as they caused too
44 many portability hassles.
47 By an accident of design, the default computation of @$ was
48 performed after another default computation was performed: @$ = @1.
49 The latter is now removed: YYLLOC_DEFAULT is fully responsible of
50 the computation of @$.
53 The token end of file may be specified by the user, in which case,
54 the user symbol is used in the reports, the graphs, and the verbose
55 error messages instead of `$', which remains being the defaults.
59 %token YYEOF 0 "end of file"
62 This old option, which has been broken for ages, is removed.
65 Croatian, thanks to Denis Lackovic.
68 When fed with `%token 'a' "A"', Bison used to output `#define 'a' 65'.
71 Changes in version 1.35, 2002-03-25:
74 Some projects use Bison's C parser with C++ compilers, and define
75 YYSTYPE as a class. The recent adjustment of C parsers for data
76 alignment and 64 bit architectures made this impossible.
78 Because for the time being no real solution for C++ parser
79 generation exists, kludges were implemented in the parser to
80 maintain this use. In the future, when Bison has C++ parsers, this
81 kludge will be disabled.
83 This kludge also addresses some C++ problems when the stack was
87 Changes in version 1.34, 2002-03-12:
89 * File name clashes are detected
90 $ bison foo.y -d -o foo.x
91 fatal error: header and parser would both be named `foo.x'
93 * A missing `;' at the end of a rule triggers a warning
94 In accordance with POSIX, and in agreement with other
95 Yacc implementations, Bison will mandate this semicolon in the near
96 future. This eases the implementation of a Bison parser of Bison
97 grammars by making this grammar LALR(1) instead of LR(2). To
98 facilitate the transition, this release introduces a warning.
100 * Revert the C++ namespace changes introduced in 1.31, as they caused too
101 many portability hassles.
103 * DJGPP support added.
105 * Fix test suite portability problems.
107 Changes in version 1.33, 2002-02-07:
110 Groff could not be compiled for the definition of size_t was lacking
111 under some conditions.
116 Changes in version 1.32, 2002-01-23:
118 * Fix Yacc output file names
122 * Italian, Dutch translations
124 Changes in version 1.31, 2002-01-14:
128 * GNU Gettext and %expect
129 GNU Gettext asserts 10 s/r conflicts, but there are 7. Now that
130 Bison dies on incorrect %expectations, we fear there will be
131 too many bug reports for Gettext, so _for the time being_, %expect
132 does not trigger an error when the input file is named `plural.y'.
134 * Use of alloca in parsers
135 If YYSTACK_USE_ALLOCA is defined to 0, then the parsers will use
136 malloc exclusively. Since 1.29, but was not NEWS'ed.
138 alloca is used only when compiled with GCC, to avoid portability
141 * When the generated parser lacks debugging code, YYDEBUG is now 0
142 (as POSIX requires) instead of being undefined.
145 Bison has always permitted actions such as { $$ = $1 }: it adds the
146 ending semicolon. Now if in Yacc compatibility mode, the semicolon
147 is no longer output: one has to write { $$ = $1; }.
149 * Better C++ compliance
150 The output parsers try to respect C++ namespaces.
151 [This turned out to be a failed experiment, and it was reverted later.]
154 Fixed bugs when reporting useless nonterminals.
157 The parsers work properly on 64 bit hosts.
160 Some calls to strerror resulted in scrambled or missing error messages.
163 When the number of shift/reduce conflicts is correct, don't issue
166 * The verbose report includes the rule line numbers.
168 * Rule line numbers are fixed in traces.
170 * Swedish translation
173 Verbose parse error messages from the parsers are better looking.
174 Before: parse error: unexpected `'/'', expecting `"number"' or `'-'' or `'(''
175 Now: parse error: unexpected '/', expecting "number" or '-' or '('
177 * Fixed parser memory leaks.
178 When the generated parser was using malloc to extend its stacks, the
179 previous allocations were not freed.
181 * Fixed verbose output file.
182 Some newlines were missing.
183 Some conflicts in state descriptions were missing.
185 * Fixed conflict report.
186 Option -v was needed to get the result.
190 Mismatches are errors, not warnings.
192 * Fixed incorrect processing of some invalid input.
194 * Fixed CPP guards: 9foo.h uses BISON_9FOO_H instead of 9FOO_H.
196 * Fixed some typos in the documentation.
198 * %token MY_EOF 0 is supported.
199 Before, MY_EOF was silently renumbered as 257.
201 * doc/refcard.tex is updated.
203 * %output, %file-prefix, %name-prefix.
207 New, aliasing `--output-file'.
209 Changes in version 1.30, 2001-10-26:
211 * `--defines' and `--graph' have now an optionnal argument which is the
212 output file name. `-d' and `-g' do not change, they do not take any
215 * `%source_extension' and `%header_extension' are removed, failed
220 Changes in version 1.29, 2001-09-07:
222 * The output file does not define const, as this caused problems when used
223 with common autoconfiguration schemes. If you still use ancient compilers
224 that lack const, compile with the equivalent of the C compiler option
225 `-Dconst='. autoconf's AC_C_CONST macro provides one way to do this.
227 * Added `-g' and `--graph'.
229 * The Bison manual is now distributed under the terms of the GNU FDL.
231 * The input and the output files has automatically a similar extension.
233 * Russian translation added.
235 * NLS support updated; should hopefully be less troublesome.
237 * Added the old Bison reference card.
239 * Added `--locations' and `%locations'.
241 * Added `-S' and `--skeleton'.
243 * `%raw', `-r', `--raw' is disabled.
245 * Special characters are escaped when output. This solves the problems
246 of the #line lines with path names including backslashes.
249 `%yacc', `%fixed_output_files', `%defines', `%no_parser', `%verbose',
250 `%debug', `%source_extension' and `%header_extension'.
253 Automatic location tracking.
255 Changes in version 1.28, 1999-07-06:
257 * Should compile better now with K&R compilers.
261 * Fixed a problem with escaping the double quote character.
263 * There is now a FAQ.
265 Changes in version 1.27:
267 * The make rule which prevented bison.simple from being created on
268 some systems has been fixed.
270 Changes in version 1.26:
272 * Bison now uses automake.
274 * New mailing lists: <bug-bison@gnu.org> and <help-bison@gnu.org>.
276 * Token numbers now start at 257 as previously documented, not 258.
278 * Bison honors the TMPDIR environment variable.
280 * A couple of buffer overruns have been fixed.
282 * Problems when closing files should now be reported.
284 * Generated parsers should now work even on operating systems which do
285 not provide alloca().
287 Changes in version 1.25, 1995-10-16:
289 * Errors in the input grammar are not fatal; Bison keeps reading
290 the grammar file, and reports all the errors found in it.
292 * Tokens can now be specified as multiple-character strings: for
293 example, you could use "<=" for a token which looks like <=, instead
294 of chosing a name like LESSEQ.
296 * The %token_table declaration says to write a table of tokens (names
297 and numbers) into the parser file. The yylex function can use this
298 table to recognize multiple-character string tokens, or for other
301 * The %no_lines declaration says not to generate any #line preprocessor
302 directives in the parser file.
304 * The %raw declaration says to use internal Bison token numbers, not
305 Yacc-compatible token numbers, when token names are defined as macros.
307 * The --no-parser option produces the parser tables without including
308 the parser engine; a project can now use its own parser engine.
309 The actions go into a separate file called NAME.act, in the form of
310 a switch statement body.
312 Changes in version 1.23:
314 The user can define YYPARSE_PARAM as the name of an argument to be
315 passed into yyparse. The argument should have type void *. It should
316 actually point to an object. Grammar actions can access the variable
317 by casting it to the proper pointer type.
319 Line numbers in output file corrected.
321 Changes in version 1.22:
325 Changes in version 1.20:
327 Output file does not redefine const for C++.
335 Copyright (C) 2001, 2002 Free Software Foundation, Inc.
337 This file is part of GNU Autoconf.
339 GNU Autoconf is free software; you can redistribute it and/or modify
340 it under the terms of the GNU General Public License as published by
341 the Free Software Foundation; either version 2, or (at your option)
344 GNU Autoconf is distributed in the hope that it will be useful,
345 but WITHOUT ANY WARRANTY; without even the implied warranty of
346 MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the
347 GNU General Public License for more details.
349 You should have received a copy of the GNU General Public License
350 along with autoconf; see the file COPYING. If not, write to
351 the Free Software Foundation, Inc., 59 Temple Place - Suite 330,
352 Boston, MA 02111-1307, USA.