]>
Commit | Line | Data |
---|---|---|
1 | Bison News | |
2 | ---------- | |
3 | ||
4 | Changes in version 1.49b: | |
5 | ||
6 | * Undefined token | |
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. | |
9 | ||
10 | * Unknown token numbers | |
11 | If yylex returned a code out of range, yyparse could die. This is | |
12 | no longer the case. | |
13 | ||
14 | * Error token | |
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. | |
19 | ||
20 | * Large grammars | |
21 | Large grammars are now supported (large token numbers, large grammar | |
22 | size (= sum of the LHS and RHS lengths), large LALR tables). | |
23 | ||
24 | * Explicit initial rule | |
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 | |
27 | graphs as rule 0. | |
28 | ||
29 | * Useless rules | |
30 | Before, Bison reported the useless rules, but, although not used, | |
31 | included them in the parsers. They are now actually removed. | |
32 | ||
33 | * Incorrect `Token not used' | |
34 | On a grammar such as | |
35 | ||
36 | %token useless useful | |
37 | %% | |
38 | exp: '0' %prec useful; | |
39 | ||
40 | where a token was used to set the precedence of the last rule, | |
41 | bison reported both `useful' and `useless' as useless tokens. | |
42 | ||
43 | * Revert the C++ namespace changes introduced in 1.31 | |
44 | as they caused too many portability hassles. | |
45 | ||
46 | * Default locations | |
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 @$. | |
51 | ||
52 | * Token end-of-file | |
53 | The token end of file may be specified by the user, in which case, | |
54 | the user symbol is used in the reports, the graphs, and the verbose | |
55 | error messages instead of `$', which remains being the defaults. | |
56 | For instance | |
57 | %token YYEOF 0 | |
58 | or | |
59 | %token YYEOF 0 "end of file" | |
60 | ||
61 | * Semantic parser | |
62 | This old option, which has been broken for ages, is removed. | |
63 | ||
64 | * New translations | |
65 | Croatian, thanks to Denis Lackovic. | |
66 | ||
67 | * Incorrect token definitions | |
68 | When fed with `%token 'a' "A"', Bison used to output `#define 'a' 65'. | |
69 | ||
70 | * Token definitions as enums | |
71 | Tokens are output both as the traditional #define's, and, provided | |
72 | the compiler supports ANSI C or is a C++ compiler, as enums. | |
73 | This helps debuggers producing symbols instead of values. | |
74 | ||
75 | \f | |
76 | Changes in version 1.35, 2002-03-25: | |
77 | ||
78 | * C Skeleton | |
79 | Some projects use Bison's C parser with C++ compilers, and define | |
80 | YYSTYPE as a class. The recent adjustment of C parsers for data | |
81 | alignment and 64 bit architectures made this impossible. | |
82 | ||
83 | Because for the time being no real solution for C++ parser | |
84 | generation exists, kludges were implemented in the parser to | |
85 | maintain this use. In the future, when Bison has C++ parsers, this | |
86 | kludge will be disabled. | |
87 | ||
88 | This kludge also addresses some C++ problems when the stack was | |
89 | extended. | |
90 | ||
91 | \f | |
92 | Changes in version 1.34, 2002-03-12: | |
93 | ||
94 | * File name clashes are detected | |
95 | $ bison foo.y -d -o foo.x | |
96 | fatal error: header and parser would both be named `foo.x' | |
97 | ||
98 | * A missing `;' at the end of a rule triggers a warning | |
99 | In accordance with POSIX, and in agreement with other | |
100 | Yacc implementations, Bison will mandate this semicolon in the near | |
101 | future. This eases the implementation of a Bison parser of Bison | |
102 | grammars by making this grammar LALR(1) instead of LR(2). To | |
103 | facilitate the transition, this release introduces a warning. | |
104 | ||
105 | * Revert the C++ namespace changes introduced in 1.31, as they caused too | |
106 | many portability hassles. | |
107 | ||
108 | * DJGPP support added. | |
109 | ||
110 | * Fix test suite portability problems. | |
111 | \f | |
112 | Changes in version 1.33, 2002-02-07: | |
113 | ||
114 | * Fix C++ issues | |
115 | Groff could not be compiled for the definition of size_t was lacking | |
116 | under some conditions. | |
117 | ||
118 | * Catch invalid @n | |
119 | As is done with $n. | |
120 | \f | |
121 | Changes in version 1.32, 2002-01-23: | |
122 | ||
123 | * Fix Yacc output file names | |
124 | ||
125 | * Portability fixes | |
126 | ||
127 | * Italian, Dutch translations | |
128 | \f | |
129 | Changes in version 1.31, 2002-01-14: | |
130 | ||
131 | * Many Bug Fixes | |
132 | ||
133 | * GNU Gettext and %expect | |
134 | GNU Gettext asserts 10 s/r conflicts, but there are 7. Now that | |
135 | Bison dies on incorrect %expectations, we fear there will be | |
136 | too many bug reports for Gettext, so _for the time being_, %expect | |
137 | does not trigger an error when the input file is named `plural.y'. | |
138 | ||
139 | * Use of alloca in parsers | |
140 | If YYSTACK_USE_ALLOCA is defined to 0, then the parsers will use | |
141 | malloc exclusively. Since 1.29, but was not NEWS'ed. | |
142 | ||
143 | alloca is used only when compiled with GCC, to avoid portability | |
144 | problems as on AIX. | |
145 | ||
146 | * When the generated parser lacks debugging code, YYDEBUG is now 0 | |
147 | (as POSIX requires) instead of being undefined. | |
148 | ||
149 | * User Actions | |
150 | Bison has always permitted actions such as { $$ = $1 }: it adds the | |
151 | ending semicolon. Now if in Yacc compatibility mode, the semicolon | |
152 | is no longer output: one has to write { $$ = $1; }. | |
153 | ||
154 | * Better C++ compliance | |
155 | The output parsers try to respect C++ namespaces. | |
156 | [This turned out to be a failed experiment, and it was reverted later.] | |
157 | ||
158 | * Reduced Grammars | |
159 | Fixed bugs when reporting useless nonterminals. | |
160 | ||
161 | * 64 bit hosts | |
162 | The parsers work properly on 64 bit hosts. | |
163 | ||
164 | * Error messages | |
165 | Some calls to strerror resulted in scrambled or missing error messages. | |
166 | ||
167 | * %expect | |
168 | When the number of shift/reduce conflicts is correct, don't issue | |
169 | any warning. | |
170 | ||
171 | * The verbose report includes the rule line numbers. | |
172 | ||
173 | * Rule line numbers are fixed in traces. | |
174 | ||
175 | * Swedish translation | |
176 | ||
177 | * Parse errors | |
178 | Verbose parse error messages from the parsers are better looking. | |
179 | Before: parse error: unexpected `'/'', expecting `"number"' or `'-'' or `'('' | |
180 | Now: parse error: unexpected '/', expecting "number" or '-' or '(' | |
181 | ||
182 | * Fixed parser memory leaks. | |
183 | When the generated parser was using malloc to extend its stacks, the | |
184 | previous allocations were not freed. | |
185 | ||
186 | * Fixed verbose output file. | |
187 | Some newlines were missing. | |
188 | Some conflicts in state descriptions were missing. | |
189 | ||
190 | * Fixed conflict report. | |
191 | Option -v was needed to get the result. | |
192 | ||
193 | * %expect | |
194 | Was not used. | |
195 | Mismatches are errors, not warnings. | |
196 | ||
197 | * Fixed incorrect processing of some invalid input. | |
198 | ||
199 | * Fixed CPP guards: 9foo.h uses BISON_9FOO_H instead of 9FOO_H. | |
200 | ||
201 | * Fixed some typos in the documentation. | |
202 | ||
203 | * %token MY_EOF 0 is supported. | |
204 | Before, MY_EOF was silently renumbered as 257. | |
205 | ||
206 | * doc/refcard.tex is updated. | |
207 | ||
208 | * %output, %file-prefix, %name-prefix. | |
209 | New. | |
210 | ||
211 | * --output | |
212 | New, aliasing `--output-file'. | |
213 | \f | |
214 | Changes in version 1.30, 2001-10-26: | |
215 | ||
216 | * `--defines' and `--graph' have now an optionnal argument which is the | |
217 | output file name. `-d' and `-g' do not change, they do not take any | |
218 | argument. | |
219 | ||
220 | * `%source_extension' and `%header_extension' are removed, failed | |
221 | experiment. | |
222 | ||
223 | * Portability fixes. | |
224 | \f | |
225 | Changes in version 1.29, 2001-09-07: | |
226 | ||
227 | * The output file does not define const, as this caused problems when used | |
228 | with common autoconfiguration schemes. If you still use ancient compilers | |
229 | that lack const, compile with the equivalent of the C compiler option | |
230 | `-Dconst='. autoconf's AC_C_CONST macro provides one way to do this. | |
231 | ||
232 | * Added `-g' and `--graph'. | |
233 | ||
234 | * The Bison manual is now distributed under the terms of the GNU FDL. | |
235 | ||
236 | * The input and the output files has automatically a similar extension. | |
237 | ||
238 | * Russian translation added. | |
239 | ||
240 | * NLS support updated; should hopefully be less troublesome. | |
241 | ||
242 | * Added the old Bison reference card. | |
243 | ||
244 | * Added `--locations' and `%locations'. | |
245 | ||
246 | * Added `-S' and `--skeleton'. | |
247 | ||
248 | * `%raw', `-r', `--raw' is disabled. | |
249 | ||
250 | * Special characters are escaped when output. This solves the problems | |
251 | of the #line lines with path names including backslashes. | |
252 | ||
253 | * New directives. | |
254 | `%yacc', `%fixed_output_files', `%defines', `%no_parser', `%verbose', | |
255 | `%debug', `%source_extension' and `%header_extension'. | |
256 | ||
257 | * @$ | |
258 | Automatic location tracking. | |
259 | \f | |
260 | Changes in version 1.28, 1999-07-06: | |
261 | ||
262 | * Should compile better now with K&R compilers. | |
263 | ||
264 | * Added NLS. | |
265 | ||
266 | * Fixed a problem with escaping the double quote character. | |
267 | ||
268 | * There is now a FAQ. | |
269 | \f | |
270 | Changes in version 1.27: | |
271 | ||
272 | * The make rule which prevented bison.simple from being created on | |
273 | some systems has been fixed. | |
274 | \f | |
275 | Changes in version 1.26: | |
276 | ||
277 | * Bison now uses automake. | |
278 | ||
279 | * New mailing lists: <bug-bison@gnu.org> and <help-bison@gnu.org>. | |
280 | ||
281 | * Token numbers now start at 257 as previously documented, not 258. | |
282 | ||
283 | * Bison honors the TMPDIR environment variable. | |
284 | ||
285 | * A couple of buffer overruns have been fixed. | |
286 | ||
287 | * Problems when closing files should now be reported. | |
288 | ||
289 | * Generated parsers should now work even on operating systems which do | |
290 | not provide alloca(). | |
291 | \f | |
292 | Changes in version 1.25, 1995-10-16: | |
293 | ||
294 | * Errors in the input grammar are not fatal; Bison keeps reading | |
295 | the grammar file, and reports all the errors found in it. | |
296 | ||
297 | * Tokens can now be specified as multiple-character strings: for | |
298 | example, you could use "<=" for a token which looks like <=, instead | |
299 | of chosing a name like LESSEQ. | |
300 | ||
301 | * The %token_table declaration says to write a table of tokens (names | |
302 | and numbers) into the parser file. The yylex function can use this | |
303 | table to recognize multiple-character string tokens, or for other | |
304 | purposes. | |
305 | ||
306 | * The %no_lines declaration says not to generate any #line preprocessor | |
307 | directives in the parser file. | |
308 | ||
309 | * The %raw declaration says to use internal Bison token numbers, not | |
310 | Yacc-compatible token numbers, when token names are defined as macros. | |
311 | ||
312 | * The --no-parser option produces the parser tables without including | |
313 | the parser engine; a project can now use its own parser engine. | |
314 | The actions go into a separate file called NAME.act, in the form of | |
315 | a switch statement body. | |
316 | \f | |
317 | Changes in version 1.23: | |
318 | ||
319 | The user can define YYPARSE_PARAM as the name of an argument to be | |
320 | passed into yyparse. The argument should have type void *. It should | |
321 | actually point to an object. Grammar actions can access the variable | |
322 | by casting it to the proper pointer type. | |
323 | ||
324 | Line numbers in output file corrected. | |
325 | \f | |
326 | Changes in version 1.22: | |
327 | ||
328 | --help option added. | |
329 | \f | |
330 | Changes in version 1.20: | |
331 | ||
332 | Output file does not redefine const for C++. | |
333 | ||
334 | Local Variables: | |
335 | mode: outline | |
336 | End: | |
337 | ||
338 | ----- | |
339 | ||
340 | Copyright (C) 2001, 2002 Free Software Foundation, Inc. | |
341 | ||
342 | This file is part of GNU Autoconf. | |
343 | ||
344 | GNU Autoconf is free software; you can redistribute it and/or modify | |
345 | it under the terms of the GNU General Public License as published by | |
346 | the Free Software Foundation; either version 2, or (at your option) | |
347 | any later version. | |
348 | ||
349 | GNU Autoconf is distributed in the hope that it will be useful, | |
350 | but WITHOUT ANY WARRANTY; without even the implied warranty of | |
351 | MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the | |
352 | GNU General Public License for more details. | |
353 | ||
354 | You should have received a copy of the GNU General Public License | |
355 | along with autoconf; see the file COPYING. If not, write to | |
356 | the Free Software Foundation, Inc., 59 Temple Place - Suite 330, | |
357 | Boston, MA 02111-1307, USA. |