4 Changes in version 1.49a:
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 Are now supported (large token numbers, large grammar size (= sum of
16 the LHS and RHS lengths).
18 * The initial rule is explicit.
19 Bison used to play hacks with the initial rule, which the user does
20 not write. It is now explicit, and visible in the reports and
23 * Useless rules are actually removed.
24 Before, Bison reported the useless rules, but, although not used,
25 included them in the parsers.
27 * False `Token not used' report fixed.
32 exp: '0' %prec useful;
34 where a token was used to set the precedence of the last rule,
35 bison reported both `useful' and `useless' as useless tokens.
37 * Revert the C++ namespace changes introduced in 1.31, as they caused too
38 many portability hassles.
41 Bison no longer dumps core when there are too many items, it just
45 The token end of file may be specified by the user, in which case,
46 the user symbol is used in the reports, the graphs, and the verbose
47 error messages instead of `$', which remains being the defaults.
51 %token YYEOF 0 "end of file"
53 Changes in version 1.31, 2002-01-14:
57 * GNU Gettext and %expect
58 GNU Gettext asserts 10 s/r conflicts, but there are 7. Now that
59 Bison dies on incorrect %expectations, we fear there will be
60 too many bug reports for Gettext, so _for the time being_, %expect
61 does not trigger an error when the input file is named `plural.y'.
63 * Use of alloca in parsers
64 If YYSTACK_USE_ALLOCA is defined to 0, then the parsers will use
65 malloc exclusively. Since 1.29, but was not NEWS'ed.
67 alloca is used only when compiled with GCC, to avoid portability
70 * When the generated parser lacks debugging code, YYDEBUG is now 0
71 (as POSIX requires) instead of being undefined.
74 Bison has always permitted actions such as { $$ = $1 }: it adds the
75 ending semicolon. Now if in Yacc compatibility mode, the semicolon
76 is no longer output: one has to write { $$ = $1; }.
78 * Better C++ compliance
79 The output parsers try to respect C++ namespaces.
82 Fixed bugs when reporting useless nonterminals.
85 The parsers work properly on 64 bit hosts.
88 Some calls to strerror resulted in scrambled or missing error messages.
91 When the number of shift/reduce conflicts is correct, don't issue
94 * The verbose report includes the rule line numbers.
96 * Rule line numbers are fixed in traces.
101 Verbose parse error messages from the parsers are better looking.
102 Before: parse error: unexpected `'/'', expecting `"number"' or `'-'' or `'(''
103 Now: parse error: unexpected '/', expecting "number" or '-' or '('
105 * Fixed parser memory leaks.
106 When the generated parser was using malloc to extend its stacks, the
107 previous allocations were not freed.
109 * Fixed verbose output file.
110 Some newlines were missing.
111 Some conflicts in state descriptions were missing.
113 * Fixed conflict report.
114 Option -v was needed to get the result.
118 Mismatches are errors, not warnings.
120 * Fixed incorrect processing of some invalid input.
122 * Fixed CPP guards: 9foo.h uses BISON_9FOO_H instead of 9FOO_H.
124 * Fixed some typos in the documentation.
126 * %token MY_EOF 0 is supported.
127 Before, MY_EOF was silently renumbered as 257.
129 * doc/refcard.tex is updated.
131 * %output, %file-prefix, %name-prefix.
135 New, aliasing `--output-file'.
137 Changes in version 1.30:
139 * `--defines' and `--graph' have now an optionnal argument which is the
140 output file name. `-d' and `-g' do not change, they do not take any
143 * `%source_extension' and `%header_extension' are removed, failed
148 Changes in version 1.29, 2001-09-07:
150 * The output file does not define const, as this caused problems when used
151 with common autoconfiguration schemes. If you still use ancient compilers
152 that lack const, compile with the equivalent of the C compiler option
153 `-Dconst='. autoconf's AC_C_CONST macro provides one way to do this.
155 * Added `-g' and `--graph'.
157 * The Bison manual is now distributed under the terms of the GNU FDL.
159 * The input and the output files has automatically a similar extension.
161 * Russian translation added.
163 * NLS support updated; should hopefully be less troublesome.
165 * Added the old Bison reference card.
167 * Added `--locations' and `%locations'.
169 * Added `-S' and `--skeleton'.
171 * `%raw', `-r', `--raw' is disabled.
173 * Special characters are escaped when output. This solves the problems
174 of the #line lines with path names including backslashes.
177 `%yacc', `%fixed_output_files', `%defines', `%no_parser', `%verbose',
178 `%debug', `%source_extension' and `%header_extension'.
181 Automatic location tracking.
183 Changes in version 1.28, 1999-07-06:
185 * Should compile better now with K&R compilers.
189 * Fixed a problem with escaping the double quote character.
191 * There is now a FAQ.
193 Changes in version 1.27:
195 * The make rule which prevented bison.simple from being created on
196 some systems has been fixed.
198 Changes in version 1.26:
200 * Bison now uses automake.
202 * New mailing lists: <bug-bison@gnu.org> and <help-bison@gnu.org>.
204 * Token numbers now start at 257 as previously documented, not 258.
206 * Bison honors the TMPDIR environment variable.
208 * A couple of buffer overruns have been fixed.
210 * Problems when closing files should now be reported.
212 * Generated parsers should now work even on operating systems which do
213 not provide alloca().
215 Changes in version 1.25, 1995-10-16:
217 * Errors in the input grammar are not fatal; Bison keeps reading
218 the grammar file, and reports all the errors found in it.
220 * Tokens can now be specified as multiple-character strings: for
221 example, you could use "<=" for a token which looks like <=, instead
222 of chosing a name like LESSEQ.
224 * The %token_table declaration says to write a table of tokens (names
225 and numbers) into the parser file. The yylex function can use this
226 table to recognize multiple-character string tokens, or for other
229 * The %no_lines declaration says not to generate any #line preprocessor
230 directives in the parser file.
232 * The %raw declaration says to use internal Bison token numbers, not
233 Yacc-compatible token numbers, when token names are defined as macros.
235 * The --no-parser option produces the parser tables without including
236 the parser engine; a project can now use its own parser engine.
237 The actions go into a separate file called NAME.act, in the form of
238 a switch statement body.
240 Changes in version 1.23:
242 The user can define YYPARSE_PARAM as the name of an argument to be
243 passed into yyparse. The argument should have type void *. It should
244 actually point to an object. Grammar actions can access the variable
245 by casting it to the proper pointer type.
247 Line numbers in output file corrected.
249 Changes in version 1.22:
253 Changes in version 1.20:
255 Output file does not redefine const for C++.