projects / bison.git / blame
| Commit | Line | Data |
|---|---|---|
| 8b3df748 AD |
1 | 2002-04-07 Akim Demaille <akim@epita.fr> |
| 2 | ||
| 3 | * src/lalr.c, src/LR0.c, src/closure.c, src/gram.c, src/reduce.c: | |
| 4 | Properly escape the symbols' TAG when outputting them. | |
| 5 | ||
| e601aa1d AD |
6 | 2002-04-07 Akim Demaille <akim@epita.fr> |
| 7 | ||
| 8 | * src/lalr.h (LA): Is a bitsetv, not bitset*. | |
| 9 | ||
| 10 | ||
| b0299a2e AD |
11 | 2002-04-07 Akim Demaille <akim@epita.fr> |
| 12 | ||
| 13 | * src/lalr.h, src/lalr.c (LAruleno): Replace with... | |
| 14 | (LArule): this, which is an array to rule_t*. | |
| 15 | * src/print.c, src/conflicts.c: Adjust. | |
| 16 | ||
| 17 | ||
| d7e1f00c AD |
18 | 2002-04-07 Akim Demaille <akim@epita.fr> |
| 19 | ||
| 20 | * src/gram.h (rule_t): Rename `number' as `user_number'. | |
| 21 | `number' is a new member. | |
| 22 | Adjust dependencies. | |
| 23 | * src/reduce.c (reduce_grammar_tables): Renumber rule_t.number. | |
| 24 | ||
| cc9305dd AD |
25 | 2002-04-07 Akim Demaille <akim@epita.fr> |
| 26 | ||
| 27 | As a result of the previous patch, it is no longer needed | |
| 28 | to reorder ritem itself. | |
| 29 | ||
| 30 | * src/reduce.c (reduce_grammar_tables): Don't sort RITEM. | |
| 31 | ||
| b0940840 AD |
32 | 2002-04-07 Akim Demaille <akim@epita.fr> |
| 33 | ||
| 34 | Be sure never to walk through RITEMS, but use only data related to | |
| 35 | the rules themselves. RITEMS should be banished. | |
| 36 | ||
| 37 | * src/output.c (output_token_translations): Rename as... | |
| 38 | (prepare_tokens): this. | |
| 39 | In addition to `translate', prepare the muscles `tname' and | |
| 40 | `toknum', which were handled by... | |
| 41 | (output_rule_data): this. | |
| 42 | Remove, and move the remainder of its outputs into... | |
| 43 | (prepare_rules): this new routines, which also merges content from | |
| 44 | (output_gram): this. | |
| 45 | (prepare_rules): Be sure never to walk through RITEMS. | |
| 46 | (output_stos): Rename as... | |
| 47 | (prepare_stos): this. | |
| 48 | (output): Always invoke prepare_states, after all, just don't use it | |
| 49 | in the output if you don't need it. | |
| 50 | ||
| 643a5994 AD |
51 | 2002-04-07 Akim Demaille <akim@epita.fr> |
| 52 | ||
| 53 | * src/LR0.c (new_state): Display `nstates' as the name of the | |
| 54 | newly created state. | |
| 55 | Adjust to initialize first_state and last_state if needed. | |
| 56 | Be sure to distinguish the initial from the final state. | |
| 57 | (new_states): Create the itemset of the initial state, and use | |
| 58 | new_state. | |
| 59 | * src/closure.c (closure): Now that the initial state has its | |
| 60 | items properly set, there is no need for a special case when | |
| 61 | creating `ruleset'. | |
| 62 | ||
| 63 | As a result, now the rule 0, reducing to $axiom, is visible in the | |
| 64 | outputs. Adjust the test suite. | |
| 65 | ||
| 66 | * tests/conflicts.at (Solved SR Conflicts) | |
| 67 | (Unresolved SR Conflicts): Adjust. | |
| 68 | * tests/regression.at (Web2c Report, Rule Line Numbers): Idem. | |
| 69 | * tests/conflicts.at (S/R in initial): New. | |
| 70 | ||
| b4c4ccc2 AD |
71 | 2002-04-07 Akim Demaille <akim@epita.fr> |
| 72 | ||
| 73 | * src/LR0.c (allocate_itemsets): Don't loop over ritem: loop over | |
| 74 | the RHS of the rules. | |
| 75 | * src/output.c (output_gram): Likewise. | |
| 76 | ||
| bba97eb2 AD |
77 | 2002-04-07 Akim Demaille <akim@epita.fr> |
| 78 | ||
| 79 | * src/gram.h (rule_t): `lhs' is now a pointer to the symbol's | |
| 80 | bucket. | |
| 81 | Adjust all dependencies. | |
| 82 | * src/reduce.c (nonterminals_reduce): Don't forget to renumber the | |
| 83 | `number' of the buckets too. | |
| 84 | * src/gram.h: Include `symtab.h'. | |
| 85 | (associativity): Move to... | |
| 86 | * src/symtab.h: here. | |
| 87 | No longer include `gram.h'. | |
| 88 | ||
| 89 | ||
| c3b407f4 AD |
90 | 2002-04-07 Akim Demaille <akim@epita.fr> |
| 91 | ||
| 92 | * src/gram.h, src/gram.c (rules_rhs_length): New. | |
| 93 | (ritem_longest_rhs): Use it. | |
| 94 | * src/gram.h (rule_t): `number' is a new member. | |
| 95 | * src/reader.c (packgram): Set it. | |
| 96 | * src/reduce.c (reduce_grammar_tables): Move the useless rules at | |
| 97 | the end of `rules', and count them out of `nrules'. | |
| 98 | (reduce_output, dump_grammar): Adjust. | |
| 99 | * src/print.c (print_grammar): It is no longer needed to check for | |
| 100 | the usefulness of a rule, as useless rules are beyond `nrules + 1'. | |
| 101 | * tests/reduce.at (Reduced Automaton): New test. | |
| 102 | ||
| 11652ab3 AD |
103 | 2002-04-07 Akim Demaille <akim@epita.fr> |
| 104 | ||
| 105 | * src/reduce.c (inaccessable_symbols): Fix a buglet: because of a | |
| 106 | lacking `+ 1' to nrules, Bison reported as useless a token if it | |
| 107 | was used solely to set the precedence of the last rule... | |
| 108 | ||
| 26b23c1a AD |
109 | 2002-04-07 Akim Demaille <akim@epita.fr> |
| 110 | ||
| 111 | * data/bison.c++, data/bison.simple: Don't output the current file | |
| 112 | name in #line, to avoid useless diffs between two identical | |
| 113 | outputs under different names. | |
| 114 | ||
| 18bcecb0 AD |
115 | 2002-04-07 Akim Demaille <akim@epita.fr> |
| 116 | ||
| 117 | * src/closure.c, src/print.c, src/reader.c, src/reduce.c: | |
| 118 | Normalize loops to using `< nrules + 1', not `<= nrules'. | |
| 119 | ||
| fa770c86 AD |
120 | 2002-04-07 Akim Demaille <akim@epita.fr> |
| 121 | ||
| 122 | * TODO: Update. | |
| 123 | ||
| d9b739c3 AD |
124 | 2002-04-07 Akim Demaille <akim@epita.fr> |
| 125 | ||
| 126 | * src/output.c, src/reader.c, src/symtab.c, src/symtab.h: Rename | |
| 127 | bucket.value as bucket.number. | |
| 128 | ||
| 99013900 AD |
129 | 2002-04-07 Akim Demaille <akim@epita.fr> |
| 130 | ||
| 131 | * src/closure.c, src/derives.c, src/gram.h, src/lalr.c, | |
| 132 | * src/nullable.c, src/output.c, src/print.c, src/print_graph.c, | |
| 133 | * src/reader.c, src/reduce.c: Let rule_t.rhs point directly to the | |
| 134 | RHS, instead of being an index in RITEMS. | |
| 135 | ||
| e966383b PE |
136 | 2002-04-04 Paul Eggert <eggert@twinsun.com> |
| 137 | ||
| 138 | * doc/bison.texinfo: Update copyright date. | |
| 139 | (Rpcalc Lexer, Symbols, Token Decl): Don't assume ASCII. | |
| 140 | (Symbols): Warn about running Bison in one character set, | |
| 141 | but compiling and/or running in an incompatible one. | |
| 142 | Warn about character code 256, too. | |
| 143 | ||
| 144 | 2002-04-03 Paul Eggert <eggert@twinsun.com> | |
| 145 | ||
| 146 | * src/bison.data (YYSTACK_ALLOC): Depend on whether | |
| 147 | YYERROR_VERBOSE is nonzero, not whether it is defined. | |
| 148 | ||
| 149 | Merge changes from bison-1_29-branch. | |
| c307773e | 150 | |
| 8d6c48b9 PE |
151 | 2002-03-20 Paul Eggert <eggert@twinsun.com> |
| 152 | ||
| 153 | Merge fixes from Debian bison_1.34-1.diff. | |
| 154 | ||
| 155 | * configure.in (AC_PREREQ): 2.53. | |
| 156 | ||
| e53c6322 AD |
157 | 2002-03-20 Akim Demaille <akim@epita.fr> |
| 158 | ||
| 159 | * src/conflicts.c (log_resolution): Argument `resolution' is const. | |
| 160 | ||
| 9ffbeca7 PE |
161 | 2002-03-19 Paul Eggert <eggert@twinsun.com> |
| 162 | ||
| 21db0b2a PE |
163 | * src/bison.simple (YYCOPY): New macro. |
| 164 | (YYSTACK_RELOCATE): Use it. | |
| 165 | Remove Type arg; no longer needed. All callers changed. | |
| 166 | (yymemcpy): Remove; no longer needed. | |
| 167 | ||
| 9ffbeca7 PE |
168 | * Makefile.am (AUTOMAKE_OPTIONS): 1.6. |
| 169 | * doc/Makefile.am (AUTOMAKE_OPTIONS): Remove. | |
| 170 | ||
| 642cb8f8 AD |
171 | 2002-03-19 Akim Demaille <akim@epita.fr> |
| 172 | ||
| 173 | Test and fix the #line outputs. | |
| 174 | ||
| 175 | * tests/atlocal.at (GCC): New. | |
| 176 | * tests/synclines.at (AT_TEST_SYNCLINE): New macro. | |
| 177 | (Prologue synch line, ,%union synch line, Postprologue synch line) | |
| 178 | (Action synch line, Epilogue synch line): New tests. | |
| 179 | * src/reader.c (parse_union_decl): Define the muscle stype_line. | |
| 180 | * data/bison.simple, data/bison.c++: Use it. | |
| 181 | ||
| 3c31a486 AD |
182 | 2002-03-19 Akim Demaille <akim@epita.fr> |
| 183 | ||
| 184 | * tests/regression.at (%nonassoc and eof, Unresolved SR Conflicts) | |
| 185 | (Solved SR Conflicts, %expect not enough, %expect right) | |
| 186 | (%expect too much): Move to... | |
| 187 | * tests/conflicts.at: this new file. | |
| 188 | ||
| 0d8bed56 AD |
189 | 2002-03-19 Akim Demaille <akim@epita.fr> |
| 190 | ||
| 191 | * data/m4sugar/m4sugar.m4: Update from CVS Autoconf. | |
| 192 | * data/bison.simple, data/bison.c++: Handle the `#define' part, so | |
| 193 | that we can move to enums for instance. | |
| 194 | * src/output.c (token_definitions_output): Output a list of | |
| 195 | `token-name, token-number' instead of the #define. | |
| 196 | (output_skeleton): Name this list `b4_tokens', not `b4_tokendefs'. | |
| 197 | ||
| 9208d17f AD |
198 | 2002-03-14 Akim Demaille <akim@epita.fr> |
| 199 | ||
| 200 | Use Gettext 0.11.1. | |
| 201 | ||
| af27eacb RA |
202 | 2002-03-09 Robert Anisko <robert@lrde.epita.fr> |
| 203 | ||
| 204 | * data/bison.c++: Make the user able to add members to the generated | |
| 205 | parser by subclassing. | |
| 206 | ||
| 9101a310 RA |
207 | 2002-03-05 Robert Anisko <robert@lrde.epita.fr> |
| 208 | ||
| 209 | * src/reader.c (read_additionnal_code): `c' should be an integer, not | |
| 210 | a character. | |
| 211 | Reported by Nicolas Tisserand and Nicolas Burrus. | |
| 212 | ||
| fff9bf0b RA |
213 | 2002-03-04 Robert Anisko <robert@lrde.epita.fr> |
| 214 | ||
| 215 | * src/reader.c: Warn about lacking semi-colons, do not complain. | |
| 216 | ||
| 64dba31e RA |
217 | 2002-03-04 Robert Anisko <robert@lrde.epita.fr> |
| 218 | ||
| 219 | * data/bison.c++: Remove a debug line. | |
| 220 | ||
| 374f5a14 RA |
221 | 2002-03-04 Robert Anisko <robert@lrde.epita.fr> |
| 222 | ||
| 223 | * data/bison.c++: Unmerge value as yylval and value as yyval. Unmerge | |
| 224 | location as yylloc and location as yyloc. Use YYLLOC_DEFAULT, and | |
| 225 | provide a default implementation. | |
| 226 | ||
| bfcf1f3a AD |
227 | 2002-03-04 Akim Demaille <akim@epita.fr> |
| 228 | ||
| 229 | * tests/input.at (Invalid $n, Invalid @n): Add the ending `;'. | |
| 230 | * tests/output.at (AT_CHECK_OUTPUT): Likewise. | |
| 231 | * tests/headers.at (AT_TEST_CPP_GUARD_H): Ditto. | |
| 232 | * tests/semantic.at (Parsing Guards): Similarly. | |
| 233 | * src/reader.at (readgram): Complain if the last rule is not ended | |
| 234 | with a semi-colon. | |
| 235 | ||
| 65ccf9fc AD |
236 | 2002-03-04 Akim Demaille <akim@epita.fr> |
| 237 | ||
| 238 | * src/warshall.h, src/warshall.c (bitmatrix_print): Move to... | |
| 239 | * src/closure.c: here. | |
| 240 | (set_firsts): Use bitsetv_reflexive_transitive_closure instead of | |
| 241 | RTC. | |
| 242 | * src/warshall.h, src/warshall.c: Remove. | |
| 243 | * tests/sets.at (Broken Closure): Adjust. | |
| 244 | ||
| d0039cbc AD |
245 | 2002-03-04 Akim Demaille <akim@epita.fr> |
| 246 | ||
| 247 | * src/output.c (output_skeleton): tempdir is const. | |
| 248 | bytes_read is unused. | |
| 249 | ||
| 345cea78 AD |
250 | 2002-03-04 Akim Demaille <akim@epita.fr> |
| 251 | ||
| 252 | * lib/bbitset.h, lib/bitset.c, lib/bitset.h, lib/bitsetv.c, | |
| 253 | * lib/bitsetv.h, lib/ebitset.c, lib/lbitset.c, lib/sbitset.c: | |
| 254 | Update. | |
| 255 | From Michael Hayes. | |
| 256 | ||
| 564801f7 AD |
257 | 2002-03-04 Akim Demaille <akim@epita.fr> |
| 258 | ||
| 259 | * src/closure.c (closure): `r' is unused. | |
| 260 | ||
| e5352bc7 AD |
261 | 2002-03-04 Akim Demaille <akim@epita.fr> |
| 262 | ||
| 263 | * tests/sets.at (Broken Closure): Add the ending `;'. | |
| 264 | * src/reader.at (readgram): Complain if a rule is not ended with a | |
| 265 | semi-colon. | |
| 266 | ||
| 914feea9 AD |
267 | 2002-03-04 Akim Demaille <akim@epita.fr> |
| 268 | ||
| 269 | * src/conflicts.c (set_conflicts): Use bitset_disjoint_p. | |
| 270 | (count_sr_conflicts): Use bitset_count. | |
| 271 | * src/reduce.c (inaccessable_symbols): Ditto. | |
| 272 | (bits_size): Remove. | |
| 273 | * src/warshall.h, src/warshall.c: Convert to bitsetv. | |
| 274 | ||
| f0250de6 AD |
275 | 2002-03-04 Akim Demaille <akim@epita.fr> |
| 276 | ||
| 277 | * src/closure.c, src/conflicts.c, src/lalr.c, src/print.c, | |
| 278 | * src/reduce.c: Remove the `bitset_zero's following the | |
| 279 | `bitset_create's, as now it is performed by the latter. | |
| 280 | ||
| ef017502 AD |
281 | 2002-03-04 Akim Demaille <akim@epita.fr> |
| 282 | ||
| 283 | * lib/bitset.c, lib/bitset.h, lib/bitsetv.c, lib/bitsetv.h, | |
| 284 | * lib/ebitset.c, lib/ebitset.h, lib/lbitset.c, lib/lbitset.h, | |
| 285 | * lib/sbitset.c, lib/sbitset.h, lib/bbitset.h: Update from the | |
| 286 | latest sources from Michael. | |
| 287 | ||
| 76514394 AD |
288 | 2002-03-04 Akim Demaille <akim@epita.fr> |
| 289 | ||
| 290 | * src/output.c (output): Don't free the grammar. | |
| 291 | * src/reader.c (grammar_free): New. | |
| 292 | * src/main.c (main): Call it and don't free symtab here. | |
| 293 | ||
| 55024580 AD |
294 | 2002-03-04 Akim Demaille <akim@epita.fr> |
| 295 | ||
| 296 | * src/lex.c (parse_percent_token): Be sure to 0-end token_buffer | |
| 297 | before returning. | |
| 298 | Reported by Benoit Perrot. | |
| 299 | ||
| f9abaa2c AD |
300 | 2002-03-04 Akim Demaille <akim@epita.fr> |
| 301 | ||
| 302 | Use bitset operations when possible, not loops over bits. | |
| 303 | ||
| 304 | * src/conflicts.c (set_conflicts, count_sr_conflicts): Use | |
| 305 | bitset_or. | |
| 306 | * src/print.c (print_reductions): Use bitset_and, bitset_andn. | |
| 307 | * src/reduce.c (useless_nonterminals): Formatting changes. | |
| 308 | * src/warshall.c (TC): Use bitset_or. | |
| 309 | ||
| 0e721e75 AD |
310 | 2002-03-04 Akim Demaille <akim@epita.fr> |
| 311 | ||
| 312 | * src/lalr.h, src/lalr.c (tokensetsize): Remove, unused. | |
| 313 | * src/system.h (BITS_PER_WORD, WORDSIZE, SETBIT, RESETBIT, BITISSET): | |
| 314 | Ditto. | |
| 315 | ||
| 0fb1ffb1 AD |
316 | 2002-03-04 Akim Demaille <akim@epita.fr> |
| 317 | ||
| 318 | * src/lalr.c (F): Now a bitset*. | |
| 319 | Adjust all dependencies. | |
| 320 | ||
| b86796bf AD |
321 | 2002-03-04 Akim Demaille <akim@epita.fr> |
| 322 | ||
| 323 | * src/conflicts.c (shiftset, lookaheadset): Now bitset. | |
| 324 | Adjust all dependencies. | |
| 325 | ||
| 602bbf31 AD |
326 | 2002-03-04 Akim Demaille <akim@epita.fr> |
| 327 | ||
| 328 | * src/L0.c, src/LR0.h (nstates): Be size_t. | |
| 329 | Adjust comparisons (signed vs unsigned). | |
| 330 | * src/conflics.c, src/lalr.c, src/lalr.h, src/output.c (LA): Now a | |
| 331 | bitset*. | |
| 332 | Adjust all dependencies. | |
| 333 | ||
| d8a0245c AD |
334 | 2002-03-04 Akim Demaille <akim@epita.fr> |
| 335 | ||
| 336 | * src/closure.c (firsts): Now, also a bitset. | |
| 337 | Adjust all dependencies. | |
| 338 | (varsetsize): Remove, now unused. | |
| 339 | * src/warshall.h, src/warshall.c: Now work on arrays of bitsets. | |
| 340 | ||
| 34ba9743 AD |
341 | 2002-03-04 Akim Demaille <akim@epita.fr> |
| 342 | ||
| 343 | * src/print.c: Convert to use bitset.h, not hand coded iterations | |
| 344 | over ints. | |
| 345 | ||
| ed86e78c AD |
346 | 2002-03-04 Akim Demaille <akim@epita.fr> |
| 347 | ||
| 348 | * src/reduce.c: Convert to use bitset.h, not hand coded BSet. | |
| 349 | ||
| dfdb1797 AD |
350 | 2002-03-04 Akim Demaille <akim@epita.fr> |
| 351 | ||
| 352 | * src/closure.c (ruleset): Be a bitset. | |
| 353 | (rulesetsize): Remove. | |
| 354 | ||
| 7086e707 AD |
355 | 2002-03-04 Akim Demaille <akim@epita.fr> |
| 356 | ||
| 357 | * lib/bitset-int.h, lib/bitset.c, lib/bitset.h, lib/bitsetv.c, | |
| 358 | * lib/bitsetv.h, lib/ebitset.c, lib/ebitset.h, lib/lbitset.c, | |
| 359 | * lib/lbitset.h, lib/sbitset.c, lib/sbitset.h: New. | |
| 360 | * src/closure.c (fderives): Be an array of bitsets. | |
| 361 | ||
| 98254360 RA |
362 | 2002-02-28 Robert Anisko <robert@lrde.epita.fr> |
| 363 | ||
| 364 | * data/bison.c++: Merge the two generated headers. Insert a copyright | |
| 365 | notice in each output file. | |
| 366 | ||
| a75c057f AD |
367 | 2002-02-28 Akim Demaille <akim@epita.fr> |
| 368 | ||
| 369 | * data/bison.c++: Copy the prologue of bison.simple to fetch | |
| 370 | useful M4 definitions, such as b4_header_guard. | |
| 371 | ||
| 06b00abc AD |
372 | 2002-02-25 Akim Demaille <akim@epita.fr> |
| 373 | ||
| 374 | * src/getargs.c (version): Give the name of the authors, and use a | |
| a75c057f AD |
375 | translator friendly scheme for the bgr |
| 376 | copyright notice. | |
| 06b00abc | 377 | |
| 70e7d534 AD |
378 | 2002-02-25 Akim Demaille <akim@epita.fr> |
| 379 | ||
| 380 | * src/output.c (header_output): Remove, now handled completely via | |
| 381 | M4. | |
| 382 | ||
| abe017f6 AD |
383 | 2002-02-25 Akim Demaille <akim@epita.fr> |
| 384 | ||
| 385 | * m4/m4.m4: New, from CVS Autoconf. | |
| 386 | * configure.in: Invoke it. | |
| 387 | * src/output.c (output_skeleton): Use its result instead of the | |
| 388 | hard coded name. | |
| 389 | ||
| 381fb12e AD |
390 | 2002-02-25 Akim Demaille <akim@epita.fr> |
| 391 | ||
| 392 | * lib/tempname.c, lib/mkstemp.c, m4/mkstemp.m4: New, stolen from | |
| 393 | Fileutils 4.1.5. | |
| 394 | * configure.in: Invoke UTILS_FUNC_MKSTEMP. | |
| 395 | * src/output.c (output_skeleton): Use mkstemp to create a real | |
| 396 | temporary file. | |
| 397 | Move the filling of `skeleton' and its muscle to... | |
| 398 | (prepare): here. | |
| 399 | (output): Move the definition of the prologue muscle to... | |
| 400 | (prepare): here. | |
| 401 | * src/system.h (DEFAULT_TMPDIR): New. | |
| 402 | ||
| 6f38107f PE |
403 | 2002-02-14 Paul Eggert <eggert@twinsun.com> |
| 404 | ||
| 405 | Remove the support for C++ namespace cleanliness; it was | |
| 406 | causing more problems than it was curing, since it didn't work | |
| 407 | properly on some nonstandard C++ compilers. This can wait | |
| 408 | for a proper C++ parser. | |
| 409 | ||
| 410 | * NEWS: Document this. | |
| 411 | * doc/bison.texinfo (Bison Parser, Debugging): Remove special mention | |
| 412 | of C++, as it's treated like C now. | |
| 413 | * src/bison.simple (YYSTD): Remove. | |
| 414 | (YYSIZE_T, YYFPRINTF, YYPARSE_PARAM_ARG, YYPARSE_PARAM_DECL): | |
| 415 | Treat C++ just like Standard C instead of trying to support | |
| 416 | namespace cleanliness. | |
| 417 | ||
| 80cce3da AD |
418 | 2002-02-14 Akim Demaille <akim@epita.fr> |
| 419 | ||
| 420 | * tests/regression.at (else): Adjust to Andreas' change. | |
| 421 | ||
| 842e8679 AD |
422 | 2002-02-14 Akim Demaille <akim@epita.fr> |
| 423 | ||
| 424 | * lib/Makefile.am (EXTRA_DIST): Ship strnlen.c. | |
| 425 | ||
| 4bda3f10 AD |
426 | 2002-02-13 Andreas Schwab <schwab@suse.de> |
| 427 | ||
| 428 | * src/output.c (output_rule_data): Don't output NULL, it might | |
| 429 | not be defined yet. | |
| 430 | ||
| 4162fa07 | 431 | 2002-02-11 Robert Anisko <robert@lrde.epita.fr> |
| b418ecd8 | 432 | |
| 4162fa07 RA |
433 | * data/bison.c++ (YYDEBUG, YYERROR_VERBOSE): After the prologue. |
| 434 | (Copyright notice): Update. | |
| b418ecd8 | 435 | |
| bd16a5dc AD |
436 | 2002-02-11 Akim Demaille <akim@epita.fr> |
| 437 | ||
| 438 | * tests/regression.at (%nonassoc and eof): Don't include | |
| 439 | nonportable headers. | |
| 440 | ||
| 8d69a1a3 RA |
441 | 2002-02-08 Robert Anisko <robert@lrde.epita.fr> |
| 442 | ||
| 443 | * data/bison.c++: Correct error recovery. Make the user able to | |
| 444 | initialize the starting location. | |
| 445 | ||
| 9b2d0677 AD |
446 | 2002-02-07 Akim Demaille <akim@epita.fr> |
| 447 | ||
| 448 | * tests/input.at: New. | |
| 449 | ||
| 69e2658b RA |
450 | 2002-02-07 Robert Anisko <robert@lrde.epita.fr> |
| 451 | ||
| 452 | * data/bison.c++: Replace some direct m4 expansions by constants. Be | |
| 9b2d0677 | 453 | more consistent when naming methods and variables. Put preprocessor |
| 69e2658b RA |
454 | directives around tables only needed for debugging. |
| 455 | ||
| 4aacc3a7 RA |
456 | 2002-02-07 Robert Anisko <robert@lrde.epita.fr> |
| 457 | ||
| 458 | * data/bison.c++ (yy::b4_name::print_): New method, replaces yyprint in | |
| 459 | C++ parsers. | |
| 460 | (yy::b4_name::parse): Use print_. | |
| 461 | ||
| 762a801e RA |
462 | 2002-02-07 Robert Anisko <robert@lrde.epita.fr> |
| 463 | ||
| 464 | * data/bison.c++ (yy::b4_name::parse): Error recovery is back. | |
| 465 | ||
| 4bb2bc3f RA |
466 | 2002-02-07 Robert Anisko <robert@lrde.epita.fr> |
| 467 | ||
| 468 | * data/bison.c++ (yy::b4_name::error_): New method, replaces yyerror in | |
| 469 | C++ parsers. | |
| 470 | (yy::b4_name::parse): Build verbose error messages, and use error_. | |
| 471 | ||
| 6b45a3ca RA |
472 | 2002-02-06 Robert Anisko <robert@lrde.epita.fr> |
| 473 | ||
| 474 | * data/bison.c++: Fix m4 quoting in comments. | |
| 475 | ||
| 50997c6e RA |
476 | 2002-02-06 Robert Anisko <robert@lrde.epita.fr> |
| 477 | ||
| 478 | * data/bison.c++: Adjust the parser code. Fix some muscles that were | |
| 479 | not expanded by m4. | |
| 480 | ||
| 3f3eed27 AD |
481 | 2002-02-05 Akim Demaille <akim@epita.fr> |
| 482 | ||
| 483 | * data/bison.c++: Adjust to the M4 back end. | |
| 484 | More is certainly needed. | |
| 485 | ||
| be2a1a68 AD |
486 | 2002-02-05 Akim Demaille <akim@epita.fr> |
| 487 | ||
| 488 | Give a try to M4 as a back end. | |
| 489 | ||
| 490 | * lib/readpipe.c: New, from wdiff. | |
| 491 | * src/Makefile.am (DEFS): Define PKGDATADIR, not BISON_SIMPLE and | |
| 492 | BISON_HAIRY. | |
| 493 | * src/system.h (BISON_HAIRY, BISON_SIMPLE): Remove the DOS and VMS | |
| 494 | specific values. Now it is m4 that performs the lookup. | |
| 495 | * src/parse-skel.y: Remove. | |
| 496 | * src/muscle_tab.c, src/muscle_tab.h (muscles_m4_output): New. | |
| 497 | * src/output.c (actions_output, guards_output) | |
| 498 | (token_definitions_output): No longer keeps track of the output | |
| 499 | line number, hence remove the second argument. | |
| 500 | (guards_output): Check against the guard member of a rule, not the | |
| 501 | action member. | |
| 502 | Adjust callers. | |
| 503 | (output_skeleton): Don't look for the skeleton location, let m4 do | |
| 504 | that. | |
| 505 | Create `/tmp/muscles.m4'. This is temporary, a proper temporary | |
| 506 | file will be used. | |
| 507 | Invoke `m4' on m4sugar.m4, muscles.m4, and the skeleton. | |
| 508 | (prepare): Given that for the time being changesyntax is not | |
| 509 | usable in M4, rename the muscles using `-' to `_'. | |
| 510 | Define `defines_flag', `output_parser_name' and `output_header_name'. | |
| 511 | * src/output.h (actions_output, guards_output) | |
| 512 | (token_definitions_output): Adjust prototypes. | |
| 513 | * src/scan-skel.l: Instead of scanning the skeletons, it now | |
| 514 | processes the output of m4: `__oline__' and `#output'. | |
| 515 | * data/bison.simple: Adjust to be used by M4(sugar). | |
| 516 | * tests/Makefile.am: Use check_SCRIPTS to make sure `bison' is up | |
| 517 | to date. | |
| 518 | * tests/bison.in: Use the secrete envvar `BISON_PKGDATADIR' | |
| 519 | instead of the dead `BISON_SIMPLE' and `BISON_HAIRY'. | |
| 520 | * data/m4sugar/m4sugar.m4, data/m4sugar/version.m4: New, | |
| 521 | shamelessly stolen from CVS Autoconf. | |
| 522 | ||
| beda758b AD |
523 | 2002-02-05 Akim Demaille <akim@epita.fr> |
| 524 | ||
| 525 | * lib/hash.c, lib/hash.h: Replace with Fileutils 4.1's version. | |
| 526 | * configure.in: Check for the declarations of free and malloc. | |
| 527 | * src/muscle_tab.c: Adjust. | |
| 528 | ||
| 5ece6d43 AD |
529 | 2002-02-05 Akim Demaille <akim@epita.fr> |
| 530 | ||
| 531 | * src/muscle_tab.c (muscle_init): Don't default to NULL muscle | |
| 532 | which have no values. | |
| 533 | ||
| 5bb18f9a AD |
534 | 2002-02-05 Akim Demaille <akim@epita.fr> |
| 535 | ||
| 536 | * src/bison.simple, src/bison.hairy, src/bison.c++: Move to... | |
| 537 | * data/: here. | |
| 538 | ||
| 894dd62e PE |
539 | 2002-01-29 Paul Eggert <eggert@twinsun.com> |
| 540 | ||
| 541 | * src/bison.simple (YYSIZE_T): Do not define merely because | |
| 542 | YYSTACK_USE_ALLOCA is nonzero or alloca or _ALLOCA_H are defined. | |
| 543 | On some platforms, <alloca.h> does not declare YYSTD (size_t). | |
| 544 | ||
| 82841af7 AD |
545 | 2002-01-27 Akim Demaille <akim@epita.fr> |
| 546 | ||
| 547 | Fix `%nonassoc and eof'. | |
| 548 | ||
| 549 | * src/state.c (errs_dup): Aaaah! The failure was due to bytes | |
| 550 | which were not properly copied! Replace | |
| 551 | memcpy (res->errs, src->errs, src->nerrs); | |
| 552 | with | |
| 553 | memcpy (res->errs, src->errs, src->nerrs * sizeof (src->errs[0])); | |
| 554 | !!! | |
| 555 | * tests/regression.at (%nonassoc and eof): Adjust to newest | |
| 556 | Autotest: `.' is not in the PATH. | |
| 557 | ||
| 318b76e9 AD |
558 | 2002-01-27 Akim Demaille <akim@epita.fr> |
| 559 | ||
| 560 | * tests/sets.at (AT_EXTRACT_SETS): New. | |
| 561 | (Nullable): Use it. | |
| 562 | (Firsts): New. | |
| 563 | ||
| 30d2f3d5 AD |
564 | 2002-01-26 Akim Demaille <akim@epita.fr> |
| 565 | ||
| 566 | * tests/actions.at, tests/calc.at, tests/headers.at, | |
| 567 | * tests/torture.at: Adjust to the newest Autotest which no longer | |
| 568 | forces `.' in the PATH. | |
| 569 | ||
| 30f8c395 AD |
570 | 2002-01-25 Akim Demaille <akim@epita.fr> |
| 571 | ||
| 572 | * tests/regression.at (%nonassoc and eof): New. | |
| 573 | Suggested by Robert Anisko. | |
| 574 | ||
| 29ae55f1 AD |
575 | 2002-01-24 Akim Demaille <akim@epita.fr> |
| 576 | ||
| 577 | Bison dumps core when trying to complain about broken input files. | |
| 578 | Reported by Cris van Pelt. | |
| 579 | ||
| 580 | * src/lex.c (parse_percent_token): Be sure to set token_buffer. | |
| 581 | * tests/regression.at (Invalid input: 1, Invalid input: 2): Merge | |
| 582 | into... | |
| 583 | (Invalid inputs): Strengthen: exercise parse_percent_token. | |
| 584 | ||
| 2b548aa6 RA |
585 | 2002-01-24 Robert Anisko <robert.anisko@epita.fr> |
| 586 | ||
| 587 | * src/Makefile.am: Add bison.c++. | |
| 588 | * src/bison.c++: New skeleton. | |
| 589 | ||
| bb0146c2 AD |
590 | 2002-01-21 Paolo Bonzini <bonzini@gnu.org> |
| 591 | ||
| 592 | * po/it.po: New. | |
| 593 | ||
| bec30531 AD |
594 | 2002-01-21 Kees Zeelenberg <kzlg@users.sourceforge.net> |
| 595 | ||
| 596 | * src/files.c (skeleton_find) [MSDOS]: Fix cp definition. | |
| 597 | ||
| fc6edc45 MA |
598 | 2002-01-20 Marc Autret <marc@gnu.org> |
| 599 | ||
| 600 | * src/files.c (compute_output_file_names): Fix | |
| 601 | ||
| 5e5d5415 MA |
602 | 2002-01-20 Marc Autret <marc@gnu.org> |
| 603 | ||
| 604 | * tests/output.at: New test. | |
| 605 | * src/files.c (compute_base_names): Don't map extensions when | |
| 606 | the YACC flag is set, use defaults. | |
| 607 | Reported by Evgeny Stambulchik. | |
| 608 | ||
| 44ea3fbd MA |
609 | 2002-01-20 Marc Autret <marc@gnu.org> |
| 610 | ||
| bb0146c2 | 611 | * src/system.h: Need to define __attribute__ away for non-GCC |
| 44ea3fbd MA |
612 | compilers as well (i.e. the vendor C compiler). |
| 613 | Suggested by Albert Chin-A-Young. | |
| 614 | ||
| 338963d1 TVH |
615 | 2002-01-11 Tim Van Holder <tim.van.holder@pandora.be> |
| 616 | ||
| 617 | * lib/hash.h, lib/hash.c: Renamed __P to PARAMS and used the | |
| 618 | canonical definition. | |
| 619 | * src/system.h: Use the canonical definition for PARAMS (avoids | |
| 620 | a conflict with the macro from lib/hash.h). | |
| 621 | ||
| c57b2479 AD |
622 | 2002-01-11 Akim Demaille <akim@epita.fr> |
| 623 | ||
| 624 | * configure.in: Use AC_FUNC_STRNLEN. | |
| d9e9746c | 625 | Fixes the failures observed on AIX 4.3 by H.Merijn Brand. |
| c57b2479 | 626 | |
| b85810ae AD |
627 | 2002-01-09 Akim Demaille <akim@epita.fr> |
| 628 | ||
| 629 | * src/files.c, src/files.h (output_infix): New. | |
| 630 | (tab_extension): Remove. | |
| 631 | (compute_base_names): Compute the former, drop the latter. | |
| 632 | * src/output.c (prepare): Insert the muscles `output-infix', and | |
| 633 | `output-suffix'. | |
| 634 | * src/parse-skel.y (string, string.1): New. | |
| 635 | (section.header): Use it. | |
| 636 | (section.yacc): Remove. | |
| 637 | (prefix): Remove too. | |
| 638 | * src/scan-skel.l: Adjust. | |
| 639 | * src/bison.simple, src/bison.hairy: Adjust. | |
| 640 | ||
| cae60122 AD |
641 | 2002-01-09 Akim Demaille <akim@epita.fr> |
| 642 | ||
| 643 | * configure.in (WERROR_CFLAGS): Compute it. | |
| 644 | * src/Makefile.am (CFLAGS): Pass it. | |
| 645 | * tests/atlocal.in (CFLAGS): Idem. | |
| 646 | * src/files.c: Fix a few warnings. | |
| 647 | (get_extension_index): Remove, unused. | |
| 648 | ||
| ae404801 AD |
649 | 2002-01-08 Akim Demaille <akim@epita.fr> |
| 650 | ||
| 651 | * src/getargs.c (AS_FILE_NAME): New. | |
| 652 | (getargs): Use it to convert DOSish file names. | |
| 653 | * src/files.c (base_name): Rename as full_base_name to avoid | |
| 654 | clashes with `base_name ()'. | |
| 655 | (filename_split): New. | |
| 656 | (compute_base_names): N-th rewrite, using filename_split. | |
| 657 | ||
| 22312b71 AD |
658 | 2002-01-08 Akim Demaille <akim@epita.fr> |
| 659 | ||
| 660 | * lib/basename.c, lib/dirname.h, lib/dirname.c, lib/memrchr.c: | |
| 661 | New, stolen from the Fileutils 4.1. | |
| 662 | * lib/Makefile.am (libbison_a_SOURCES): Adjust. | |
| 663 | * configure.in: Check for the presence of memrchr, and of its | |
| 664 | prototype. | |
| 665 | ||
| a67cef01 TVH |
666 | 2002-01-07 Tim Van Holder <tim.van.holder@pandora.be> |
| 667 | ||
| 668 | * lib/hash.h (__P): Added definition for this macro. | |
| 669 | * src/Makefile.am: Add parse-skel.c and scan-skel.c to | |
| 670 | BUILT_SOURCES, to ensure they are generated first. | |
| 671 | * src/parse-skel.y: Use YYERROR_VERBOSE instead of | |
| 672 | %error-verbose to allow bootstrapping with bison 1.30x. | |
| 673 | ||
| 2b25d624 AD |
674 | 2002-01-06 Akim Demaille <akim@epita.fr> |
| 675 | ||
| 676 | * src/reader.c (parse_braces): Don't fetch the next char, the | |
| 677 | convention is to fetch on entry. | |
| 678 | * tests/torture.at (GNU Cim Grammar): Reintroduce their weird | |
| 679 | 'switch' without a following semicolon. | |
| 680 | * tests/regression.at (braces parsing): New. | |
| 681 | ||
| 3460813b AD |
682 | 2002-01-06 Akim Demaille <akim@epita.fr> |
| 683 | ||
| 684 | Bison is dead wrong in its RR conflict reports. | |
| 685 | ||
| 686 | * tests/torture.at (GNU Cim Grammar): New. | |
| 687 | * src/conflicts.c (count_rr_conflicts): Fix. | |
| 688 | ||
| 73784c64 AD |
689 | 2002-01-06 Akim Demaille <akim@epita.fr> |
| 690 | ||
| 691 | Creating package.m4 from configure.ac causes too many problems. | |
| 692 | ||
| 693 | * tests/Makefile.am (package.m4): Create it by hand, | |
| 694 | AC_CONFIG_TESTDIR no longer does in the most recent CVS Autoconf. | |
| 695 | ||
| 25d81090 AD |
696 | 2002-01-06 Akim Demaille <akim@epita.fr> |
| 697 | ||
| 698 | * src/Makefile.am (bison_SOURCES): Add parse-skel.h and | |
| 699 | skeleton.h. | |
| 700 | ||
| a9b8959e PE |
701 | 2002-01-04 Paul Eggert <eggert@twinsun.com> |
| 702 | ||
| 703 | * doc/bison.texinfo (Debugging): | |
| 704 | Remove YYSTDERR; it's no longer defined or used. | |
| 705 | Also, s/cstdio.h/cstdio/. | |
| 706 | ||
| 25d81090 AD |
707 | 2002-01-03 Akim Demaille <akim@epita.fr> |
| 708 | ||
| 709 | * tests/bison.in, tests/atlocal.in: Adjust to CVS Autoconf. | |
| 710 | ||
| 1109455c AD |
711 | 2002-01-03 Akim Demaille <akim@epita.fr> |
| 712 | ||
| 713 | * src/parse-skel.y (process_skeleton): Don't bind the parser's | |
| 714 | tracing code to --trace, wait for a better --trace option, with | |
| 715 | args. | |
| 716 | ||
| 7ea5e977 AD |
717 | 2002-01-03 Akim Demaille <akim@epita.fr> |
| 718 | ||
| 719 | * src/bison.simple (YYSTDERR): Remove, replace `stderr'. | |
| 720 | The ISO C++ standard is extremely clear about it: stderr is | |
| 721 | considered a macro, not a regular symbol (see table 94 `Header | |
| 722 | <cstdio> synopsis', [lib.c.files] 27.8.2 C Library files). | |
| 723 | Therefore std:: does not apply to it. It still does with fprintf. | |
| 724 | Also, s/cstdio.h/cstdio/. | |
| 725 | ||
| fab5b110 AD |
726 | 2002-01-03 Akim Demaille <akim@epita.fr> |
| 727 | ||
| 728 | * lib/quotearg.c: Use `#include "..."' instead of `#include <...>' | |
| 729 | for non system headers. | |
| 730 | ||
| aed7fd9b AD |
731 | 2002-01-02 Akim Demaille <akim@epita.fr> |
| 732 | ||
| 733 | Equip the skeleton chain with location tracking, runtime trace, | |
| 734 | pure parser and scanner. | |
| 735 | ||
| 736 | * src/parse-skel.y: Request a pure parser, locations, and prefix | |
| 737 | renaming. | |
| 738 | (%union): Having several members with the same type does not help | |
| 739 | type mismatches, simplify. | |
| 740 | (YYPRINT, yyprint): New. | |
| 741 | (yyerror): ``Rename'' (there is a #define yyerror skel_error) as... | |
| 742 | (skel_error): this. | |
| 743 | Handle locations. | |
| 744 | * src/scan-skel.l: Adjust to these changes. | |
| 745 | * src/skeleton.h (LOCATION_RESET, LOCATION_LINES, LOCATION_STEP) | |
| 746 | (LOCATION_PRINT, skel_control_t): New. | |
| 747 | ||
| 24fad99e AD |
748 | 2001-12-30 Akim Demaille <akim@epita.fr> |
| 749 | ||
| 750 | * src/parse-skel.y: Get rid of the shift/reduce conflict: | |
| 751 | replace `gb' with BLANKS. | |
| 752 | * src/scan-skel.l: Adjust. | |
| 753 | ||
| a4b36db4 AD |
754 | 2001-12-30 Akim Demaille <akim@epita.fr> |
| 755 | ||
| 756 | * src/system.h: We don't need nor want bcopy. | |
| 757 | Throw away MS-DOS crap: we don't need getpid. | |
| 758 | * configure.in: We don't need strndup. It was even causing | |
| 759 | problems: because Flex includes the headers *before* us, | |
| 760 | _GNU_SOURCE is not defined by config.h, and therefore strndup was | |
| 761 | not visible. | |
| 762 | * lib/xstrndup.c: New. | |
| 763 | * src/scan-skel.l: Use it. | |
| 764 | Be sure to initialize yylval.muscle member when scanning a MUSCLE. | |
| 765 | * src/parse-skel.y: Use %directives instead of #defines. | |
| 766 | ||
| 1239777d AD |
767 | 2001-12-30 Akim Demaille <akim@epita.fr> |
| 768 | ||
| 769 | * src/skeleton.h: New. | |
| 770 | * src/output.c (output_parser, output_master_parser): Remove, dead | |
| 771 | code. | |
| 772 | * src/output.h (get_lines_number, actions_output, guards_output) | |
| 773 | (token_definitions_output): Prototype them. | |
| 774 | * src/parse-skel.y: Add the license notice. | |
| 775 | Include output.h and skeleton.h. | |
| 776 | (process_skeleton): Returns void, and takes a single parameter. | |
| 777 | * src/scan-skel.l: Add the license notice. | |
| 778 | Include skeleton.h. | |
| 779 | Don't use %option yylineno: it seems that then Flex imagines | |
| 780 | REJECT has been used, and therefore it won't reallocate its | |
| 781 | buffers (which makes no other sense to me than a bug). It results | |
| 782 | in warnings for `unused: yy_flex_realloc'. | |
| 783 | ||
| 9b3add5b RA |
784 | 2001-12-30 Robert Anisko <robert.anisko@epita.fr> |
| 785 | ||
| 786 | * src/muscle_tab.h (MUSCLE_INSERT_INT, MUSCLE_INSERT_STRING) | |
| 787 | (MUSCLE_INSERT_PREFIX): ...to there. | |
| 788 | * src/output.c (MUSCLE_INSERT_INT, MUSCLE_INSERT_STRING) | |
| 789 | (MUSCLE_INSERT_PREFIX): Move from here... | |
| 790 | ||
| 791 | * src/bison.hairy: Add a section directive. Put braces around muscle | |
| 792 | names. This parser skeleton is still broken, but Bison should not | |
| 793 | choke on a bad muscle 'syntax'. | |
| 794 | * src/bison.simple: Add a section directive. Put braces around muscle | |
| 795 | names. | |
| 796 | ||
| 797 | * src/files.h (strsuffix, stringappend): Add declarations. | |
| 798 | (tab_extension): Add declaration. | |
| 799 | (short_base_name): Add declaration. | |
| 800 | ||
| 801 | * src/files.c (strsuffix, stringappend): No longer static. These | |
| 802 | functions are used in the skeleton parser. | |
| 803 | (tab_extension): New. | |
| 804 | (compute_base_names): Use the computations done in this function | |
| fab5b110 | 805 | to guess if the generated parsers should have '.tab' in their |
| 9b3add5b RA |
806 | names. |
| 807 | (short_base_name): No longer static. | |
| 808 | ||
| 809 | * src/output.c (output_skeleton): New. | |
| 810 | (output): Disable call to output_master_parser, and give a try to | |
| 811 | a new skeleton handling system. | |
| 812 | (guards_output, actions_output): No longer static. | |
| 813 | (token_definitions_output, get_lines_number): No longer static. | |
| 814 | ||
| 815 | * configure.in: Use AM_PROG_LEX and AC_PROG_YACC. | |
| 816 | ||
| fab5b110 | 817 | * src/Makefile.am (bison_SOURCES): Add scan-skel.l and |
| 9b3add5b RA |
818 | parse-skel.y. |
| 819 | ||
| 820 | * src/parse-skel.y: New file. | |
| 821 | * src/scan-skel.l: New file. | |
| 822 | ||
| b5b61c61 AD |
823 | 2001-12-29 Akim Demaille <akim@epita.fr> |
| 824 | ||
| 825 | %name-prefix is broken. | |
| 826 | ||
| 827 | * src/files.c (spec_name_prefix): Initialize to NULL, not to "yy". | |
| 828 | Adjust all dependencies. | |
| 829 | * tests/headers.at (export YYLTYPE): Strengthen this test: use | |
| 830 | %name-prefix. | |
| 831 | ||
| 832 | Renaming yylval but not yylloc is not consistent. Now we do. | |
| 833 | ||
| 834 | * src/bison.simple: Prefix yylloc if used. | |
| 835 | * doc/bison.texinfo (Decl Summary): Document that. | |
| 836 | ||
| 8c9a50be AD |
837 | 2001-12-29 Akim Demaille <akim@epita.fr> |
| 838 | ||
| 839 | * doc/bison.texinfo: Promote `%long-directive' over | |
| 840 | `%long_directive'. | |
| 841 | Remove all references to fixed-output-files, yacc is enough. | |
| 842 | ||
| d99361e6 AD |
843 | 2001-12-29 Akim Demaille <akim@epita.fr> |
| 844 | ||
| 845 | * src/bison.simple: Define YYDEBUG and YYERROR_VERBOSE *after* the | |
| 846 | user prologue. These are defaults. | |
| 847 | * tests/actions.at (Mid-rule actions): Make sure the user can | |
| 848 | define YYDEBUG and YYERROR_VERBOSE. | |
| 849 | ||
| b9cecb91 AD |
850 | 2001-12-29 Akim Demaille <akim@epita.fr> |
| 851 | ||
| 852 | * src/output.c (header_output): Don't forget to export YYLTYPE and | |
| 853 | yylloc. | |
| 854 | * tests/headers.at (export YYLTYPE): New, make sure it does. | |
| 855 | * tests/regression.at (%union and --defines, Invalid CPP headers): | |
| 856 | Move to... | |
| 857 | * tests/headers.at: here. | |
| 858 | ||
| aea13e97 AD |
859 | 2001-12-29 Akim Demaille <akim@epita.fr> |
| 860 | ||
| 861 | * src/gram.h (rule_s): Member `assoc' is of type `associativity'. | |
| 862 | ||
| 931394cb AD |
863 | 2001-12-29 Akim Demaille <akim@epita.fr> |
| 864 | ||
| 865 | * tests/actions.at (Mid-rule actions): Output on a single line | |
| 866 | instead of several. | |
| 867 | ||
| 704a47c4 AD |
868 | 2001-12-29 Akim Demaille <akim@epita.fr> |
| 869 | ||
| 870 | * doc/bison.texinfo: Formatting changes. | |
| 871 | ||
| 091e20bb AD |
872 | 2001-12-29 Akim Demaille <akim@epita.fr> |
| 873 | ||
| 874 | Don't store the token defs in a muscle, just be ready to output it | |
| 875 | on command. Now possible via `symbols'. Fixes a memory leak. | |
| 876 | ||
| 877 | * src/output.c (token_definitions_output): New. | |
| 878 | (output_parser, header_output): Use it. | |
| 879 | * src/reader.c (symbols_save): Remove. | |
| 880 | ||
| cce71710 AD |
881 | 2001-12-29 Akim Demaille <akim@epita.fr> |
| 882 | ||
| 883 | * src/bison.simple: Do not provide a default for YYSTYPE and | |
| 884 | YYLTYPE before the user's prologue. Otherwise it's hardly... a | |
| 885 | default. | |
| 886 | ||
| 82c035a8 AD |
887 | 2001-12-29 Akim Demaille <akim@epita.fr> |
| 888 | ||
| 889 | Mid-rule actions are simply... ignored! | |
| 890 | ||
| 891 | * src/reader.c (readgram): Be sure to attach mid-rule actions to | |
| 892 | the empty-rule associated to the dummy symbol, not to the host | |
| 893 | rule. | |
| 894 | * tests/actions.at (Mid-rule actions): New. | |
| 895 | ||
| 8419d367 AD |
896 | 2001-12-29 Akim Demaille <akim@epita.fr> |
| 897 | ||
| 898 | Memory leak. | |
| 899 | ||
| 900 | * src/reader.c (reader): Free grammar. | |
| 901 | ||
| 375d5806 AD |
902 | 2001-12-29 Akim Demaille <akim@epita.fr> |
| 903 | ||
| 904 | Memory leak. | |
| 905 | ||
| 906 | * src/LR0.c (new_itemsets): Don't allocate `shift_symbol' here, | |
| 907 | since it allocates it for each state, although only one is needed. | |
| 908 | (allocate_storage): Do it here. | |
| 909 | ||
| f51cb8ff AD |
910 | 2001-12-29 Akim Demaille <akim@epita.fr> |
| 911 | ||
| 912 | * src/options.h, src/options.c (create_long_option_table): Rename | |
| 913 | as... | |
| 914 | (long_option_table_new): this, with a clearer prototype. | |
| 915 | (percent_table): Remove, unused, | |
| 916 | * src/getargs.c (getargs): Adjust. | |
| 917 | ||
| 29e88316 AD |
918 | 2001-12-29 Akim Demaille <akim@epita.fr> |
| 919 | ||
| 920 | * src/LR0.c, src/conflicts.c, src/lalr.c, src/lalr.h, src/output.c | |
| 921 | * src/print.c, src/print_graph.c, src/state.h: Rename state_table | |
| 922 | as states. | |
| 923 | ||
| b9f71f19 AD |
924 | 2001-12-29 Akim Demaille <akim@epita.fr> |
| 925 | ||
| 926 | * src/lalr.c (build_relations): Rename `states' as `states1'. | |
| 927 | Sorry, I don't understand exactly what it is, no better name... | |
| 928 | ||
| 1a2b5d37 AD |
929 | 2001-12-29 Akim Demaille <akim@epita.fr> |
| 930 | ||
| 931 | * src/closure.c, src/conflicts.c, src/derives.c, src/gram.c | |
| 932 | * src/gram.h, src/lalr.c, src/nullable.c, src/output.c, src/print.c | |
| 933 | * src/print_graph.c, src/reader.c, src/reduce.c: Rename rule_table | |
| 934 | as rules. | |
| 935 | ||
| 1cca533e AD |
936 | 2001-12-29 Akim Demaille <akim@epita.fr> |
| 937 | ||
| 938 | * src/gram.c (rprec, rprecsym, rassoc): Remove, unused since long | |
| 939 | ago. | |
| 940 | ||
| c03ae966 AD |
941 | 2001-12-29 Akim Demaille <akim@epita.fr> |
| 942 | ||
| 943 | * src/reader.c, src/reader.h (user_toknums): Remove. | |
| 944 | Adjust all users to use symbols[i]->user_token_number. | |
| 945 | ||
| 5a670b1e AD |
946 | 2001-12-29 Akim Demaille <akim@epita.fr> |
| 947 | ||
| 948 | * src/gram.c, src/gram.h (sprec, sassoc): Remove. | |
| 949 | Adjust all users to use symbols[i]->prec or ->assoc. | |
| 950 | ||
| ad949da9 AD |
951 | 2001-12-29 Akim Demaille <akim@epita.fr> |
| 952 | ||
| 953 | * src/reader.c, src/reader.h (tags): Remove. | |
| 954 | Adjust all users to use symbols[i]->tag. | |
| 955 | ||
| 0e78e603 AD |
956 | 2001-12-29 Akim Demaille <akim@epita.fr> |
| 957 | ||
| 958 | * src/gram.h, src/gram.c (symbols): New, similar to state_table | |
| 959 | and rule_table. | |
| 960 | * src/reader.c (packsymbols): Fill this table. | |
| 961 | Drop sprec. | |
| 962 | * src/conflicts.c (resolve_sr_conflict): Adjust. | |
| 963 | * src/reduce.c (reduce_grammar): Adjust: just sort symbols, a | |
| 964 | single table. | |
| 965 | Use symbols[i]->tag instead of tags[i]. | |
| 966 | ||
| 213e640e AD |
967 | 2001-12-29 Akim Demaille <akim@epita.fr> |
| 968 | ||
| 969 | * tests/calc.at (_AT_DATA_CALC_Y): Also use %union. | |
| 970 | In addition, put a comment in there, to replace... | |
| 971 | * tests/regression.at (%union and C comments): Remove. | |
| 972 | ||
| e7b8bef1 AD |
973 | 2001-12-29 Akim Demaille <akim@epita.fr> |
| 974 | ||
| 975 | * tests/regression.at (Web2c Actions): Blindly move the actual | |
| 976 | output as expected output. The contents *seem* right to me, but I | |
| 977 | can't pretend reading perfectly parser tables... Nonetheless, all | |
| 978 | the other tests pass correctly, the table look OK, even though the | |
| 979 | presence of `$axiom' is to be noted: AFAICS it is useless (but | |
| 980 | harmless). | |
| 981 | ||
| b68e7744 AD |
982 | 2001-12-29 Akim Demaille <akim@epita.fr> |
| 983 | ||
| 984 | * src/reader.c (readgram): Don't add the rule 0 if there were no | |
| 985 | rules read. In other words, add it _after_ having performed | |
| 986 | grammar sanity checks. | |
| 987 | Fixes the `tests/regression.at (Invalid input: 1)' Failure. | |
| 988 | ||
| 78d5bae9 AD |
989 | 2001-12-29 Akim Demaille <akim@epita.fr> |
| 990 | ||
| 991 | * tests/regression.at (Web2c Report): Catch up: the rule 0 is now | |
| 992 | visible, and some states have now a different number. | |
| 993 | ||
| ff442794 AD |
994 | 2001-12-29 Akim Demaille <akim@epita.fr> |
| 995 | ||
| 996 | * src/reader.c (readgram): Bind the initial rule's lineno to that | |
| 997 | of the first rule. | |
| 998 | * tests/regression.at (Rule Line Numbers, Unresolved SR Conflicts): | |
| 999 | (Solved SR Conflicts): Adjust rule 0's line number. | |
| 1000 | ||
| 610ab194 AD |
1001 | 2001-12-29 Akim Demaille <akim@epita.fr> |
| 1002 | ||
| 1003 | Fix the `GAWK Grammar' failure. | |
| 1004 | ||
| 1005 | * src/LR0.c (final_state): Initialize to -1 so that we do compute | |
| 1006 | the reductions of the first state which was mistakenly confused | |
| 1007 | with the final state because precisely final_state was initialized | |
| 1008 | to 0. | |
| 1009 | * tests/sets.at (Nullable): Adjust: state 0 does have lookaheads, | |
| 1010 | now noticed by Bison. | |
| 1011 | * tests/regression.at (Rule Line Numbers): Adjust: state 0 does | |
| 1012 | have a reduction on $default. | |
| 1013 | ||
| 29d29c8f AD |
1014 | 2001-12-29 Akim Demaille <akim@epita.fr> |
| 1015 | ||
| 1016 | * src/gram.c (ritem_print): Be sure to subtract 1 when displaying | |
| 1017 | rule line numbers. | |
| 1018 | * src/closure.c (print_closure): Likewise. | |
| 1019 | * src/derives.c (print_derives): Likewise. | |
| 1020 | * tests/sets.at (Nullable): Adjust: the rule numbers are correct | |
| 1021 | now. | |
| 1022 | ||
| 7c6b64d0 AD |
1023 | 2001-12-29 Akim Demaille <akim@epita.fr> |
| 1024 | ||
| 1025 | * src/lalr.c (lookaheads_print): New. | |
| 1026 | (lalr): Call it when --trace-flag. | |
| 1027 | * tests/sets.at (Nullable): Adjust: when tracing, the lookaheads | |
| 1028 | are dumped. | |
| 1029 | ||
| 3d4daee3 AD |
1030 | 2001-12-29 Akim Demaille <akim@epita.fr> |
| 1031 | ||
| 1032 | * src/derives.c (print_derives): Be sure to use `>= 0', not `> 0', | |
| 1033 | when walking through ritem, even via rule->rhs. | |
| 1034 | * src/reduce.c (dump_grammar, useful_production, reduce_output) | |
| 1035 | (useful_production, useless_nonterminals): Likewise. | |
| 1036 | (reduce_grammar_tables): Likewise, plus update nritems. | |
| 1037 | * src/nullable.c (set_nullable): Likewise. | |
| 1038 | * src/lalr.c (build_relations): Likewise. | |
| 1039 | * tests/sets.at (Nullable): Adjust. | |
| 1040 | Fortunately, now, the $axiom is no longer nullable. | |
| 1041 | ||
| 9e7f6bbd AD |
1042 | 2001-12-29 Akim Demaille <akim@epita.fr> |
| 1043 | ||
| 1044 | * src/LR0.c (generate_states): Use nritems, not nitems, nor using | |
| 1045 | the 0-sentinel. | |
| 1046 | * src/gram.c (ritem_longest_rhs): Likewise. | |
| 1047 | * src/reduce.c (nonterminals_reduce): Likewise. | |
| 1048 | * src/print_graph.c (print_graph): Likewise. | |
| 1049 | * src/output.c (output_rule_data): Likewise. | |
| 1050 | * src/nullable.c (set_nullable): Likewise. | |
| 1051 | ||
| 255ef638 AD |
1052 | 2001-12-29 Akim Demaille <akim@epita.fr> |
| 1053 | ||
| 1054 | * src/output.c: Comment changes. | |
| 1055 | ||
| 0d8a7363 AD |
1056 | 2001-12-27 Paul Eggert <eggert@twinsun.com> |
| 1057 | ||
| 1058 | * src/bison.simple (YYSTACK_ALLOC, YYSIZE_T): Remove special | |
| 1059 | cases for non-GNU systems like AIX, HP-UX, SGI, Sun, and | |
| 1060 | Sparc, as they were causing more porting problems than the | |
| 1061 | (minor) performance improvement was worth. | |
| 1062 | ||
| 1063 | Also, catch up with 1.31's YYSTD. | |
| 1064 | ||
| 3db472b9 AD |
1065 | 2001-12-27 Akim Demaille <akim@epita.fr> |
| 1066 | ||
| 1067 | * src/output.c (output_gram): Rely on nritems, not the | |
| 1068 | 0-sentinel. See below. | |
| 1069 | Use -1 as separator, not 0. | |
| 1070 | * src/bison.simple (yyparse): Subtract 1 to the rule numbers. | |
| 1071 | Rely on -1 as separator in yyrhs, instead of 0. | |
| 1072 | * tests/calc.at (AT_CHECK_CALC): Now, the parsers no longer issue | |
| 1073 | twice `Now at end of input', therefore there are two lines less to | |
| 1074 | expect. | |
| 1075 | ||
| b365aa05 AD |
1076 | 2001-12-27 Akim Demaille <akim@epita.fr> |
| 1077 | ||
| 1078 | * tests/regression.at (Unresolved SR Conflicts): | |
| 1079 | (Solved SR Conflicts, Rule Line Numbers): Adjust to the changes | |
| 1080 | below. | |
| 1081 | ||
| 30171f79 AD |
1082 | 2001-12-27 Akim Demaille <akim@epita.fr> |
| 1083 | ||
| 1084 | * src/LR0.c (new_state): Recognize the final state by the fact it | |
| 1085 | is reached by eoftoken. | |
| 1086 | (insert_start_shifting_state, insert_eof_shifting_state) | |
| 1087 | (insert_accepting_state, augment_automaton): Remove, since now | |
| 1088 | these states are automatically computed from the initial state. | |
| 1089 | (generate_states): Adjust. | |
| 1090 | * src/print.c: When reporting a rule number to the user, substract | |
| 1091 | 1, so that the axiom rule is rule 0, and the first user rule is 1. | |
| 1092 | * src/reduce.c: Likewise. | |
| 1093 | * src/print_graph.c (print_core): For the time being, just as for | |
| 1094 | the report, depend upon --trace-flags to dump the full set of | |
| 1095 | items. | |
| 1096 | * src/reader.c (readgram): Once the grammar read, insert the rule | |
| 1097 | 0: `$axiom: START-SYMBOL $'. | |
| 1098 | * tests/set.at: Adjust: rule 0 is now displayed, and since the | |
| 1099 | number of the states has changed (the final state is no longer | |
| 1100 | necessarily the last), catch up. | |
| 1101 | ||
| 75142d45 AD |
1102 | 2001-12-27 Akim Demaille <akim@epita.fr> |
| 1103 | ||
| 1104 | Try to make the use of the eoftoken valid. Given that its value | |
| 1105 | is 0 which was also used as a sentinel in ritem, (i) make sure >= 0 | |
| 1106 | is used instead of > 0 where appropriate, (ii), depend upon nritems | |
| 1107 | instead of the 0-sentinel. | |
| 1108 | ||
| 1109 | * src/gram.h, src/gram.c (nritems): New. | |
| 1110 | Expected to be duplication of nitems, but for the time being... | |
| 1111 | * src/reader.c (packgram): Assert nritems and nitems are equal. | |
| 1112 | * src/LR0.c (allocate_itemsets, new_itemsets): Adjust. | |
| 1113 | * src/closure.c (print_closure, print_fderives): Likewise. | |
| 1114 | * src/gram.c (ritem_print): Likewise. | |
| 1115 | * src/print.c (print_core, print_grammar): Likewise. | |
| 1116 | * src/print_graph.c: Likewise. | |
| 1117 | ||
| b7c49edf AD |
1118 | 2001-12-27 Akim Demaille <akim@epita.fr> |
| 1119 | ||
| 1120 | * src/main.c (main): If there are complains after grammar | |
| 1121 | reductions, then output the report anyway if requested, then die. | |
| 1122 | * src/symtab.c (bucket_new): Initialize `value' to -1, not 0. | |
| 1123 | * src/reader.c (eoftoken): New. | |
| 1124 | (parse_token_decl): If the token being defined has value `0', it | |
| 1125 | is the eoftoken. | |
| 1126 | (packsymbols): No longer hack `tags' to insert `$' by hand. | |
| 1127 | Be sure to preserve the value of the eoftoken. | |
| 1128 | (reader): Make sure eoftoken is defined. | |
| 1129 | Initialize nsyms to 0: now eoftoken is created just like the others. | |
| 1130 | * src/print.c (print_grammar): Don't special case the eof token. | |
| 1131 | * src/regression.at: Adjust: `$' has value 0, not -1, which was a | |
| 1132 | lie anyway, albeit pleasant. | |
| 1133 | * tests/calc.at: Exercise error messages with eoftoken. | |
| 1134 | Change the grammar so that empty input is invalid. | |
| 1135 | Adjust expectations. | |
| 1136 | When yyungeting, be sure to use a valid yylloc: use last_yylloc. | |
| 1137 | ||
| ec2da99f AD |
1138 | 2001-12-27 Akim Demaille <akim@epita.fr> |
| 1139 | ||
| 1140 | * configure.in: Check the protos of strchr ans strspn. | |
| 1141 | Replace strchr if needed. | |
| 1142 | * src/system.h: Provide the protos of strchr, strspn and memchr if | |
| 1143 | missing. | |
| 1144 | * lib/strchr.c: New. | |
| 1145 | * src/reader.c (symbols_save): Use strchr. | |
| 1146 | ||
| 8adfa272 AD |
1147 | 2001-12-27 Akim Demaille <akim@epita.fr> |
| 1148 | ||
| 1149 | * src/print.c, src/print_graph.c (escape): New. | |
| 1150 | Use it to quote the TAGS outputs. | |
| 1151 | * src/print_graph.c (print_state): Now errors are in red, and | |
| 1152 | reductions in green. | |
| 1153 | Prefer high to wide: output the state number on a line of its own. | |
| 1154 | ||
| 80dac38c AD |
1155 | 2001-12-27 Akim Demaille <akim@epita.fr> |
| 1156 | ||
| 1157 | * src/state.h, src/state.c (reductions_new): New. | |
| 1158 | * src/LR0.c (set_state_table): Let all the states have a | |
| 1159 | `reductions', even if reduced to 0. | |
| 1160 | (save_reductions): Adjust. | |
| 1161 | * src/lalr.c (initialize_LA, initialize_lookaheads): Adjust. | |
| 1162 | * src/print.c (print_reductions, print_actions): Adjust. | |
| 1163 | * src/output.c (action_row): Adjust. | |
| 1164 | ||
| 2cec70b9 AD |
1165 | 2001-12-27 Akim Demaille <akim@epita.fr> |
| 1166 | ||
| 1167 | * src/state.h, src/state.c (errs_new, errs_dup): New. | |
| 1168 | * src/LR0.c (set_state_table): Let all the states have an errs, | |
| 1169 | even if reduced to 0. | |
| 1170 | * src/print.c (print_errs, print_reductions): Adjust. | |
| 1171 | * src/output.c (output_actions, action_row): Adjust. | |
| 1172 | * src/conflicts.c (resolve_sr_conflict): Adjust. | |
| 1173 | ||
| 13ca549a AD |
1174 | 2001-12-27 Akim Demaille <akim@epita.fr> |
| 1175 | ||
| 1176 | * src/lalr.c (set_goto_map, initialize_F): Use SHIFT_SYMBOL. | |
| 1177 | ||
| 5092aba5 AD |
1178 | 2001-12-27 Akim Demaille <akim@epita.fr> |
| 1179 | ||
| 1180 | * src/conflicts.c, src/conflicts.h (print_reductions): Move to... | |
| 1181 | * src/print.c: here. | |
| 1182 | (lookaheadset, shiftset): New, used as additional storage by | |
| 1183 | print_reductions. | |
| 1184 | (print_results): Adjust. | |
| 1185 | (print_shifts, print_gotos, print_errs): New, extracted from... | |
| 1186 | (print_actions): here. | |
| 1187 | * src/print_graph.c (print_actions): Remove dead code. | |
| 1188 | ||
| 11e2beca AD |
1189 | 2001-12-27 Akim Demaille <akim@epita.fr> |
| 1190 | ||
| 1191 | * src/reader.c (copy_dollar, copy_at): Better checking of `n' in | |
| 1192 | `$n' and `@n'. | |
| 1193 | ||
| dac3c910 AD |
1194 | 2001-12-27 Akim Demaille <akim@epita.fr> |
| 1195 | ||
| 1196 | * src/lalr.c (add_lookback_edge): Use state_t instead of ints. | |
| 1197 | (build_relations): Adjust. | |
| 1198 | ||
| d0b0fefa AD |
1199 | 2001-12-27 Akim Demaille <akim@epita.fr> |
| 1200 | ||
| 1201 | * src/lalr.c (set_goto_map): Remove a wrong but benign loop | |
| 1202 | duplication. | |
| 1203 | ||
| adc8c848 AD |
1204 | 2001-12-27 Akim Demaille <akim@epita.fr> |
| 1205 | ||
| 1206 | * src/reader.c (packgram): Catch nitems overflows. | |
| 1207 | ||
| 14d293ac AD |
1208 | 2001-12-27 Akim Demaille <akim@epita.fr> |
| 1209 | ||
| 1210 | * src/files.c, src/files.h (guard_obstack): Remove. | |
| 1211 | * src/output.c (output): Adjust. | |
| 1212 | * src/reader.c (parse_braces): New, factoring... | |
| 1213 | (copy_action, copy_guard): these two which are renamed as... | |
| 1214 | (parse_action, parse_guard): these. | |
| 1215 | As a voluntary consequence, using braces around guards is now | |
| 1216 | mandatory. | |
| 1217 | ||
| f499b062 AD |
1218 | 2001-12-27 Akim Demaille <akim@epita.fr> |
| 1219 | ||
| 1220 | * src/gram.h (rule_t): `guard' and `guard_line' are new members. | |
| 1221 | * src/reader.c (symbol_list): `guard' and `guard_line' are new | |
| 1222 | members. | |
| 1223 | (symbol_list_new): Adjust. | |
| 1224 | (copy_action): action_line is the first line, not the last. | |
| 1225 | (copy_guard): Just as for actions, store the `action' only, not | |
| 1226 | the switch/case/break flesh. | |
| 1227 | Don't parse the user action that might follow the guard, let... | |
| 1228 | (readgram): do it, i.e., now, there can be an action after a | |
| 1229 | guard. | |
| 1230 | In other words the guard is just explicitly optional. | |
| 1231 | (packgram): Adjust. | |
| 1232 | * src/output.c (guards_output): New. | |
| 1233 | (output_parser): Call it when needed. | |
| 1234 | (output): Also free the guard and attrs obstacks. | |
| 1235 | * src/files.c, src/files.h (obstack_save): Remove. | |
| 1236 | (output_files): Remove. | |
| 1237 | As a result, if one needs the former `.act' file, using an | |
| 1238 | appropriate skeleton which requires actions and guards is now | |
| 1239 | required. | |
| 1240 | * src/main.c (main): Adjust. | |
| 1241 | * tests/semantic.at: New. | |
| 1242 | * tests/regression.at: Use `input.y' as input file name. | |
| 1243 | Avoid 8+3 problems by requiring input.c when the test needs the | |
| 1244 | parser. | |
| 1245 | ||
| d945f5cd AD |
1246 | 2001-12-27 Akim Demaille <akim@epita.fr> |
| 1247 | ||
| 1248 | * src/reader.c (symbol_list_new): Be sure to initialize all the | |
| 1249 | fields. | |
| 1250 | ||
| d200e455 AD |
1251 | 2001-12-27 Akim Demaille <akim@epita.fr> |
| 1252 | ||
| 1253 | All the hacks using a final pseudo state are now useless. | |
| 1254 | ||
| 1255 | * src/LR0.c (set_state_table): state_table holds exactly nstates. | |
| 1256 | * src/lalr.c (nLA): New. | |
| 1257 | (initialize_LA, compute_lookaheads, initialize_lookaheads): Use it | |
| 1258 | instead of lookaheadsp from the pseudo state (nstate + 1). | |
| 1259 | ||
| f9507c28 AD |
1260 | 2001-12-27 Akim Demaille <akim@epita.fr> |
| 1261 | ||
| 1262 | * src/output.c (action_row, token_actions): Use a state_t instead | |
| 1263 | of a integer, and nlookaheads instead of the following state's | |
| 1264 | lookaheadsp. | |
| 1265 | ||
| 065fbd27 AD |
1266 | 2001-12-27 Akim Demaille <akim@epita.fr> |
| 1267 | ||
| 1268 | * src/conflicts.c (log_resolution, flush_shift) | |
| 1269 | (resolve_sr_conflict, set_conflicts, solve_conflicts) | |
| 1270 | (count_sr_conflicts, count_rr_conflicts, conflicts_output) | |
| 1271 | (conflicts_print, print_reductions): Use a state_t instead of an | |
| 1272 | integer when referring to a state. | |
| 1273 | As much as possible, depend upon nlookaheads, instead of the | |
| 1274 | `lookaheadsp' member of the following state (since lookaheads of | |
| 1275 | successive states are successive, the difference between state n + 1 | |
| 1276 | and n served as the number of lookaheads for state n). | |
| 1277 | * src/lalr.c (add_lookback_edge): Likewise. | |
| 1278 | * src/print.c (print_core, print_actions, print_state) | |
| 1279 | (print_results): Likewise. | |
| 1280 | * src/print_graph.c (print_core, print_actions, print_state) | |
| 1281 | (print_graph): Likewise. | |
| 1282 | * src/conflicts.h: Adjust. | |
| 1283 | ||
| 1b177bd7 AD |
1284 | 2001-12-27 Akim Demaille <akim@epita.fr> |
| 1285 | ||
| 1286 | * src/bison.hairy: Formatting/comment changes. | |
| 1287 | ANSIfy. | |
| 1288 | Remove `register' indications. | |
| 1289 | Add plenty of `static'. | |
| 1290 | ||
| 7742ddeb AD |
1291 | 2001-12-27 Akim Demaille <akim@epita.fr> |
| 1292 | ||
| 1293 | * src/output.c (prepare): Drop the muscle `ntbase' which | |
| 1294 | duplicates ntokens. | |
| 1295 | * src/bison.simple: Formatting/comment changes. | |
| 1296 | Use YYNTOKENS only, which is documented, but not YYNTBASE, which | |
| 1297 | is an undocumented synonym. | |
| 1298 | ||
| 1fa14068 AD |
1299 | 2001-12-22 Akim Demaille <akim@epita.fr> |
| 1300 | ||
| 1301 | * src/output.c (output_table_data): Change the prototype to use | |
| 1302 | `int' for array ranges: some invocations do pass an int, not a | |
| 1303 | short. | |
| 1304 | Reported by Wayne Green. | |
| 1305 | ||
| b9752825 AD |
1306 | 2001-12-22 Akim Demaille <akim@epita.fr> |
| 1307 | ||
| 1308 | Some actions of web2c.y are improperly triggered. | |
| 1309 | Reported by Mike Castle. | |
| 1310 | ||
| 1311 | * src/lalr.c (traverse): s/F (i)[k] = F (j)[k]/F (j)[k] = F (i)[k]/. | |
| 1312 | * tests/regression.at (Web2c): Rename as... | |
| 1313 | (Web2c Report): this. | |
| 1314 | (Web2c Actions): New. | |
| 1315 | ||
| 776209d6 AD |
1316 | 2001-12-22 Akim Demaille <akim@epita.fr> |
| 1317 | ||
| 1318 | Reductions in web2c.y are improperly reported. | |
| 1319 | Reported by Mike Castle. | |
| 1320 | ||
| 1321 | * src/conflicts.c (print_reductions): Fix. | |
| 1322 | * tests/regression.at (Web2c): New. | |
| 1323 | ||
| 275fc3ad AD |
1324 | 2001-12-18 Akim Demaille <akim@epita.fr> |
| 1325 | ||
| 1326 | Some host fail on `assert (!"foo")', which expands to | |
| 1327 | ((!"foo") ? (void)0 : __assert("!"foo."", __FILE__, __LINE__)) | |
| 1328 | Reported by Nelson Beebee. | |
| 1329 | ||
| 1330 | * src/output.c, src/vcg.c: Replace `assert (!"it succeeded")' with | |
| 1331 | `#define it_succeeded 0' and `assert (it_succeeded)'. | |
| 1332 | ||
| 897668ee MA |
1333 | 2001-12-17 Marc Autret <autret_m@epita.fr> |
| 1334 | ||
| 1335 | * src/bison.simple: Don't hard code the skeleton line and filename. | |
| 1336 | * src/output.c (output_parser): Rename 'line' as 'output_line'. | |
| 1337 | New line counter 'skeleton_line' (skeleton-line muscle). | |
| 1338 | ||
| ab3399e0 PE |
1339 | 2001-12-17 Paul Eggert <eggert@twinsun.com> |
| 1340 | ||
| 1341 | * NEWS, doc/bison.texinfo, doc/bison.1, doc/bison.rnh: Document that | |
| 1342 | YYDEBUG must be defined to a nonzero value. | |
| 1343 | ||
| 1344 | * src/bison.simple (yytname): Do not assume that the user defines | |
| 1345 | YYDEBUG to a properly parenthesized expression. | |
| 1346 | ||
| 3877f72b AD |
1347 | 2001-12-17 Akim Demaille <akim@epita.fr> |
| 1348 | ||
| 1349 | * src/state.h (state_t): Rename lookaheads as lookaheadsp. | |
| 1350 | nlookaheads is a new member. | |
| 1351 | Adjust all users. | |
| 1352 | * src/lalr.h (nlookaheads): Remove this orphan declaration. | |
| 1353 | * src/lalr.c (initialize_lookaheads): Set nlookaheads for each | |
| 1354 | state. | |
| 776209d6 | 1355 | |
| 331dbc1b AD |
1356 | 2001-12-17 Akim Demaille <akim@epita.fr> |
| 1357 | ||
| 1358 | * src/files.h, src/files.c (open_files, close_files): Remove. | |
| 1359 | * src/main.c (main): Don't open/close files, nor invoke lex_free, | |
| 1360 | let... | |
| 1361 | * src/reader.c (reader): Do it. | |
| 776209d6 | 1362 | |
| be750e4c AD |
1363 | 2001-12-17 Akim Demaille <akim@epita.fr> |
| 1364 | ||
| 1365 | * src/conflicts.c (print_reductions): Formatting changes. | |
| 776209d6 | 1366 | |
| 709ae8c6 AD |
1367 | 2001-12-17 Akim Demaille <akim@epita.fr> |
| 1368 | ||
| 1369 | * src/conflicts.c (flush_shift): Also adjust lookaheadset. | |
| 1370 | (flush_reduce): New. | |
| 1371 | (resolve_sr_conflict): Adjust. | |
| 776209d6 | 1372 | |
| f87685c3 AD |
1373 | 2001-12-17 Akim Demaille <akim@epita.fr> |
| 1374 | ||
| 1375 | * src/output.c (output_obstack): Be static and rename as... | |
| 1376 | (format_obstack): this, to avoid any confusion with files.c's | |
| 1377 | output_obstack. | |
| 1378 | * src/reader.h (muscle_obstack): Move to... | |
| 1379 | * src/output.h: here, since it's defined in output.c. | |
| 1380 | ||
| 837491d8 AD |
1381 | 2001-12-17 Akim Demaille <akim@epita.fr> |
| 1382 | ||
| 1383 | * src/output.c (action_row, save_column, default_goto) | |
| 1384 | (sort_actions, matching_state, pack_vector): Better variable | |
| 1385 | locality. | |
| 1386 | ||
| 796d61fb AD |
1387 | 2001-12-17 Akim Demaille <akim@epita.fr> |
| 1388 | ||
| 1389 | * src/output.c: Various formatting changes. | |
| 776209d6 | 1390 | |
| 64d15509 AD |
1391 | 2001-12-17 Akim Demaille <akim@epita.fr> |
| 1392 | ||
| 1393 | * src/files.c (output_files): Free the output_obstack. | |
| 1394 | * src/main.c (main): Call print and print_graph conditionally. | |
| 1395 | * src/print.c (print): Work unconditionally. | |
| 1396 | * src/print_graph.c (print_graph): Work unconditionally. | |
| 1397 | * src/conflicts.c (log_resolution): Output only if verbose_flag. | |
| 1398 | ||
| fbc8ecb7 MA |
1399 | 2001-12-16 Marc Autret <autret_m@epita.fr> |
| 1400 | ||
| 1401 | * src/output.c (actions_output): Fix. When we use %no-lines, | |
| 1402 | there is one less line per action. | |
| 1403 | ||
| f0440388 MA |
1404 | 2001-12-16 Marc Autret <autret_m@epita.fr> |
| 1405 | ||
| 1406 | * src/bison.simple: Remove a useless #line directive. | |
| 1407 | s/#line %%line %%skeleton/#line %%line "%%parser-file-name"/'. | |
| 1408 | * src/output.c (get_lines_number): New. | |
| 776209d6 | 1409 | (output_parser): Adjust, now takes care about the lines of a |
| f0440388 MA |
1410 | output muscles. |
| 1411 | Fix line numbering. | |
| 1412 | (actions_output): Computes the number of lines taken by actions. | |
| 1413 | (output_master_parser): Insert new skeleton which is the name of | |
| 1414 | the output parser file name. | |
| 1415 | ||
| a79986b8 MA |
1416 | 2001-12-15 Marc Autret <autret_m@epita.fr> |
| 1417 | ||
| 1418 | * src/bison.simple [YYERROR_VERBOSE]: Restore backward compatibility. | |
| 1419 | ||
| 4ec8e00f MA |
1420 | 2001-12-15 Marc Autret <autret_m@epita.fr> |
| 1421 | ||
| 1422 | * src/output.c (output_gram): Keep track of the hairy one. | |
| 1423 | ||
| 1a4648ff AD |
1424 | 2001-12-15 Akim Demaille <akim@epita.fr> |
| 1425 | ||
| 1426 | Make `make distcheck' work. | |
| 1427 | ||
| 1428 | * lib/Makefile.am (INCLUDES): Add top_srcdir/intl, since hash uses | |
| 1429 | system.h which uses libgettext.h. | |
| 1430 | ||
| 9c2c67e6 AD |
1431 | 2001-12-15 Akim Demaille <akim@epita.fr> |
| 1432 | ||
| 1433 | * src/nullable.c (set_nullable): Useless rules must be skipped, | |
| 1434 | otherwise, since we range over their symbols, we might look at a | |
| 1435 | nonterminal which no longer ``exists'', i.e., it is not counted in | |
| 1436 | `nvars', hence we overflow our arrays. | |
| 1437 | ||
| 93ede233 AD |
1438 | 2001-12-15 Akim Demaille <akim@epita.fr> |
| 1439 | ||
| 1440 | The header can also be produced directly, without any obstack! | |
| 1441 | Yahoo! | |
| 1442 | ||
| 1443 | * src/files.c, src/files.h (defines_obstack): Remove. | |
| 1444 | (compute_header_macro): Global. | |
| 1445 | (defines_obstack_save): Remove. | |
| 1446 | * src/reader.c (parse_union_decl): No longer output to | |
| 1447 | defines_obstack: its content can be found in the `stype' muscle | |
| 1448 | anyway. | |
| 1449 | (output_token_translations): Merge into... | |
| 1450 | (symbols_output): this. | |
| 1451 | Rename as... | |
| 1452 | (symbols_save): this. | |
| 1453 | (reader): Adjust. | |
| 1454 | * src/output.c (header_output): New. | |
| 1455 | (output): Call it. | |
| 1456 | ||
| 2666f928 AD |
1457 | 2001-12-15 Akim Demaille <akim@epita.fr> |
| 1458 | ||
| 1459 | * src/reader.c (parse_union_decl): Instead of handling two obstack | |
| 1460 | simultaneously, use one to define the `stype' muscle, and use the | |
| 1461 | value of the latter to fill defines_obstack. | |
| 1462 | (copy_comment): Remove. | |
| 1463 | (copy_comment2): Work for a single obstack. | |
| 1464 | Rename as... | |
| 1465 | (copy_comment): this. | |
| 1466 | ||
| 428046f8 AD |
1467 | 2001-12-15 Akim Demaille <akim@epita.fr> |
| 1468 | ||
| 1469 | * src/lex.c, src/lex.h (xgetc): No longer static. | |
| 1470 | * src/reader.c (parse_union_decl): Revamp. | |
| 1471 | ||
| ea52d706 AD |
1472 | 2001-12-15 Akim Demaille <akim@epita.fr> |
| 1473 | ||
| 1474 | Still making progress in separating Bison into (i) input, (ii) | |
| 1475 | process, (iii) output: now we can directly output the parser file | |
| 1476 | without using table_obstack at all. | |
| 1477 | ||
| 1478 | * src/files.c, src/files.h (table_obstack): Bye bye. | |
| 1479 | (parser_file_name): New. | |
| 1480 | * src/files.c (compute_output_file_names): Compute it. | |
| 1481 | * src/output.c (actions_output, output_parser) | |
| 1482 | (output_master_parser): To a file instead of an obstack. | |
| 1483 | ||
| 3f96f4dc AD |
1484 | 2001-12-15 Akim Demaille <akim@epita.fr> |
| 1485 | ||
| 1486 | Attach actions to rules, instead of pre-outputting them to | |
| 1487 | actions_obstack. | |
| 1488 | ||
| 1489 | * src/gram.h (rule_t): action and action_line are new members. | |
| 1490 | * src/reader.c (symbol_list): Likewise. | |
| 1491 | (copy_action): Save the actions within the rule. | |
| 1492 | (packgram): Save them in rule_table. | |
| 1493 | * src/output.c (actions_output): New. | |
| 1494 | (output_parser): Use it on `%%actions'. | |
| 1495 | (output_rule_data): Don't free rule_table. | |
| 1496 | (output): Do it. | |
| 1497 | (prepare): Don't save the `action' muscle. | |
| 1498 | * src/bison.simple: s/%%action/%%actions/. | |
| 1499 | ||
| 51576fb3 AD |
1500 | 2001-12-15 Akim Demaille <akim@epita.fr> |
| 1501 | ||
| 1502 | * src/reader.c (copy_action): When --yacc, don't append a `;' | |
| 1503 | to the user action: let it fail if lacking. | |
| dee049eb | 1504 | Suggested by Arnold Robbins and Tom Tromey. |
| 51576fb3 | 1505 | |
| 2648a72d AD |
1506 | 2001-12-14 Akim Demaille <akim@epita.fr> |
| 1507 | ||
| 1508 | * src/lex.c (literalchar): Simply return the char you decoded, non | |
| 1509 | longer mess around with obstacks and int pointers. | |
| 1510 | Adjust all callers. | |
| 1511 | ||
| 92790e5b AD |
1512 | 2001-12-14 Akim Demaille <akim@epita.fr> |
| 1513 | ||
| 1514 | * src/lex.c (literalchar): Don't escape the special characters, | |
| 1515 | just decode them, and keep them as char (before, eol was output as | |
| 1516 | the 2 char string `\n' etc.). | |
| 1517 | * src/output.c (output_rule_data): Use quotearg to output the | |
| 1518 | token strings. | |
| 1519 | ||
| 927c1557 PE |
1520 | 2001-12-13 Paul Eggert <eggert@twinsun.com> |
| 1521 | ||
| 1522 | * src/bison.simple (YYSIZE_T, YYSTACK_ALLOC, YYSTACK_FREE): | |
| 1523 | Do not infringe on the global user namespace when using C++. | |
| 1524 | (YYFPRINTF, YYSTDERR): New macros, needed for the above. | |
| 1525 | All uses of `fprintf' and `stderr' changed. | |
| 1526 | ||
| 1527 | * doc/bison.texinfo: Document YYFPRINTF, YYSTDERR. | |
| 1528 | ||
| ed8e1f68 AD |
1529 | 2001-12-13 Akim Demaille <akim@epita.fr> |
| 1530 | ||
| 1531 | The computation of nullable is broken: it doesn't handle empty | |
| 1532 | RHS's properly. | |
| 1533 | ||
| 1534 | * tests/torture.at (GNU AWK Grammar): New. | |
| 1535 | * tests/sets.at (Nullable): New. | |
| 1536 | * src/nullable.c (set_nullable): Instead of blindly looping over | |
| 1537 | `ritems', loop over the rules, and then over their rhs's. | |
| 1538 | ||
| 1539 | Work around Autotest bugs. | |
| 1540 | ||
| 1541 | * src/warshall.c (bitmatrix_print): Don't use `+--+' as table | |
| 1542 | frame, because Autotest understand lines starting with a `+' as | |
| 1543 | traces from the shell. Then, they are not processed properly. | |
| 1544 | Admittedly an Autotest bug, but we don't have time to wait for | |
| 1545 | Autotest to catch up. | |
| 1546 | * tests/regression.at (Broken Closure): Adjust to the new table | |
| 1547 | frames. | |
| 1548 | Move to... | |
| 1549 | * tests/sets.at: here. | |
| 1550 | ||
| cb581495 AD |
1551 | 2001-12-13 Akim Demaille <akim@epita.fr> |
| 1552 | ||
| 1553 | * src/closure.c (closure): Use nrules instead of playing tricks | |
| 1554 | with BITS_PER_WORD. | |
| 1555 | ||
| 2e729273 AD |
1556 | 2001-12-13 Akim Demaille <akim@epita.fr> |
| 1557 | ||
| 1558 | * src/print.c (print_actions): Output the handling of `$' as the | |
| 1559 | traces do: shifting the token EOF. Before EOF was treated as a | |
| 1560 | nonterminal. | |
| 1561 | * tests/regression.at: Adjust some tests. | |
| 1562 | * src/print_graph.c (print_core): Complete the set of items via | |
| 1563 | closure. The next-to-final and final states are still unsatisfying, | |
| 1564 | but that's to be addressed elsewhere. | |
| 1565 | No longer output the rule numbers, but do output the state number. | |
| 1566 | A single loop for the shifts + gotos is enough, but picked a | |
| 1567 | distinct color for each. | |
| 1568 | (print_graph): Initialize and finalize closure. | |
| 1569 | ||
| 107f7dfb AD |
1570 | 2001-12-13 Akim Demaille <akim@epita.fr> |
| 1571 | ||
| 1572 | * src/reader.c (readgram): Remove dead code, an strip useless | |
| 1573 | braces. | |
| 1574 | (get_type): Remove, unused. | |
| 1575 | ||
| 9b53a24f AD |
1576 | 2001-12-12 Akim Demaille <akim@epita.fr> |
| 1577 | ||
| 1578 | * src/complain.h, src/complain.c: Remove error_one_per_line, rely | |
| 1579 | on that of lib/error.c. | |
| 1580 | ||
| dbfb6dcd AD |
1581 | 2001-12-12 Akim Demaille <akim@epita.fr> |
| 1582 | ||
| 1583 | Some hosts don't like `/' in includes. | |
| 1584 | ||
| 1585 | * src/system.h: Include libgettext.h without qualifying the path. | |
| 1586 | * src/Makefile.am (INCLUDES): Add $(top_srcdir)/intl, remove | |
| 1587 | $(top_srcdir). | |
| 1588 | ||
| c25fb648 MA |
1589 | 2001-12-11 Marc Autret <autret_m@epita.fr> |
| 1590 | ||
| 1591 | * src/output.c (output_parser): Remove useless muscle. | |
| 1592 | ||
| 710ddc4f MA |
1593 | 2001-12-11 Marc Autret <autret_m@epita.fr> |
| 1594 | ||
| 1595 | * src/bison.simple: Remove #line just before %%epilogue. It | |
| 1596 | is now handled in ... | |
| 1597 | * src/reader.c (read_additionnal_code): Add the output of a | |
| 1598 | #line for the epilogue. | |
| 1599 | ||
| e83d80b8 MA |
1600 | 2001-12-10 Marc Autret <autret_m@epita.fr> |
| 1601 | ||
| 927c1557 | 1602 | * src/reader.c (copy_definition): Re-use CPP-outed code which |
| e83d80b8 MA |
1603 | replace precedent remove. |
| 1604 | * src/bison.simple: Remove #line before %%prologue because | |
| 1605 | %%input-line is wrong at this time. | |
| 1606 | ||
| 971d5158 MA |
1607 | 2001-12-10 Marc Autret <autret_m@epita.fr> |
| 1608 | ||
| 1609 | * src/reader.c (symbols_output): Clean up. | |
| 927c1557 | 1610 | * src/output.c (output_gram, output): Clean up. |
| 971d5158 | 1611 | |
| 5edafffd AD |
1612 | 2001-12-10 Akim Demaille <akim@epita.fr> |
| 1613 | ||
| 1614 | * src/lalr.c (initialize_lookaheads): New. Extracted from... | |
| 1615 | * src/LR0.c (set_state_table): here. | |
| 1616 | * src/lalr.c (lalr): Call it. | |
| 1617 | ||
| 0279f8e9 AD |
1618 | 2001-12-10 Akim Demaille <akim@epita.fr> |
| 1619 | ||
| 1620 | * src/state.h (shifts): Remove the `number' member: shifts are | |
| 1621 | attached to state, hence no longer need to be labelled with a | |
| 1622 | state number. | |
| 1623 | ||
| 190c4f5f AD |
1624 | 2001-12-10 Akim Demaille <akim@epita.fr> |
| 1625 | ||
| 1626 | Now that states have a complete set of members, the linked list of | |
| 1627 | shifts is useless: just fill directly the state's shifts member. | |
| 1628 | ||
| 1629 | * src/state.h (shifts): Remove the `next' member. | |
| 1630 | * src/LR0.c (first_state, last_state): Remove. | |
| 1631 | Adjust the callers. | |
| 1632 | (augment_automaton): Don't look for the shifts that must be added | |
| 1633 | a shift on EOF: it is those of the state we looked for! But now, | |
| 1634 | since shifts are attached, it is no longer needed to looking | |
| 1635 | merely by its id: its number. | |
| 1636 | ||
| 2a73b93d AD |
1637 | 2001-12-10 Akim Demaille <akim@epita.fr> |
| 1638 | ||
| 1639 | * src/LR0.c (augment_automaton): Better variable locality. | |
| 1640 | Remove an impossible branch: if there is a state corresponding to | |
| 1641 | the start symbol being shifted, then there is shift for the start | |
| 1642 | symbol from the initial state. | |
| 1643 | ||
| 74392f6a AD |
1644 | 2001-12-10 Akim Demaille <akim@epita.fr> |
| 1645 | ||
| 1646 | * src/LR0.c (augment_automaton): Call `insert_eof_shifting_state' | |
| 1647 | only when appropriate: when insert_start_shifting_state' is not | |
| 1648 | invoked. | |
| 1649 | * tests/regression.at (Rule Line Numbers): Adjust. | |
| 1650 | ||
| 37c82725 AD |
1651 | 2001-12-10 Akim Demaille <akim@epita.fr> |
| 1652 | ||
| 1653 | * src/LR0.c (augment_automaton): Now that all states have shifts, | |
| 1654 | merge the two cases addition shifts to the initial state. | |
| 1655 | ||
| 6a164e0c AD |
1656 | 2001-12-10 Akim Demaille <akim@epita.fr> |
| 1657 | ||
| 1658 | * src/lalr.c (set_state_table): Move to... | |
| 1659 | * src/LR0.c: here. | |
| 1660 | * src/lalr.c (lalr): Don't call it... | |
| 1661 | * src/LR0.c (generate_states): do it. | |
| 1662 | * src/LR0.h (first_state): Remove, only the table is used. | |
| 1663 | ||
| 7215de24 AD |
1664 | 2001-12-10 Akim Demaille <akim@epita.fr> |
| 1665 | ||
| 1666 | * src/LR0.h (first_shift, first_reduction): Remove. | |
| 1667 | * src/lalr.c: Don't use first_shift: find shifts through the | |
| 1668 | states. | |
| 1669 | ||
| 80e25d4d AD |
1670 | 2001-12-10 Akim Demaille <akim@epita.fr> |
| 1671 | ||
| 1672 | * src/LR0.c: Attach shifts to states as soon as they are | |
| 1673 | computed. | |
| 1674 | * src/lalr.c (set_state_table): Instead of assigning shifts to | |
| 1675 | state, just assert that the mapping was properly done. | |
| 1676 | ||
| 0ab3728b AD |
1677 | 2001-12-10 Akim Demaille <akim@epita.fr> |
| 1678 | ||
| 1679 | * src/LR0.c (insert_start_shift): Rename as... | |
| 1680 | (insert_start_shifting_state): this. | |
| 1681 | (insert_eof_shifting_state, insert_accepting_state): New. | |
| 1682 | (augment_automaton): Adjust. | |
| 1683 | Better locality of the variables. | |
| 1684 | When looking if the start_symbol is shifted from the initial | |
| 1685 | state, using `while (... symbol != start_symbol ...)' sounds | |
| 1686 | better than `while (... symbol < start_symbol ...)': If fail | |
| 1687 | to see how the order between symbols could be relevant! | |
| 1688 | ||
| 78af9bbc AD |
1689 | 2001-12-10 Akim Demaille <akim@epita.fr> |
| 1690 | ||
| 1691 | * src/getargs.h: Don't declare `spec_name_prefix' and | |
| 1692 | `spec_file_prefix', declared by src/files.h. | |
| 1693 | * src/files.c, src/files.h: Default for spec_name_prefix is "yy". | |
| 1694 | * src/muscle_tab.c (muscle_init): Default prefix to NULL. | |
| 1695 | * src/output.c (prepare): Adjust. | |
| 1696 | * src/reader.c (symbols_output): Likewise. | |
| 1697 | * src/vmsgetargs.c: Vaguely adjust, but who cares? | |
| 1698 | ||
| bdef2a41 AD |
1699 | 2001-12-10 Akim Demaille <akim@epita.fr> |
| 1700 | ||
| 1701 | * src/muscle_tab.c (muscle_init): NULL is a better default than | |
| 1702 | `"0"'. | |
| 1703 | ||
| 3735969c AD |
1704 | 2001-12-10 Akim Demaille <akim@epita.fr> |
| 1705 | ||
| 1706 | * src/reader.c (reader): Calling symbols_output once is enough. | |
| 1707 | ||
| 49701457 AD |
1708 | 2001-12-10 Akim Demaille <akim@epita.fr> |
| 1709 | ||
| 1710 | Now that states have a complete set of members, the linked list of | |
| 1711 | reductions is useless: just fill directly the state's reductions | |
| 1712 | member. | |
| 1713 | ||
| 1714 | * src/state.h (struct reductions): Remove member `number' and | |
| 1715 | `next'. | |
| 1716 | * src/LR0.c (first_reduction, last_reduction): Remove. | |
| 1717 | (save_reductions): Don't link the new reductions, store them in | |
| 1718 | this_state. | |
| 1719 | * src/lalr.c (set_state_table): No need to attach reductions to | |
| 1720 | states, it's already done. | |
| 1721 | * src/output.c (output_actions): No longer free the shifts, then | |
| 1722 | the reductions, then the states: free all the states and their | |
| 1723 | members. | |
| 1724 | ||
| 0edad749 AD |
1725 | 2001-12-10 Akim Demaille <akim@epita.fr> |
| 1726 | ||
| 1727 | * src/options.c (OPTN, DRTV, BOTH): New. | |
| 1728 | (option_table): Use them. | |
| 1729 | ||
| 0edad749 AD |
1730 | * src/muscle_tab.c: Don't include xalloc.h and string.h: that's |
| 1731 | the job of system.h. | |
| 1732 | * src/options.c: Don't include stdio.h and xalloc.h for the same | |
| 1733 | reasons. | |
| 1734 | ||
| 5449dd0f AD |
1735 | 2001-12-10 Akim Demaille <akim@epita.fr> |
| 1736 | ||
| 1737 | * src/output.c (output, prepare): Make sure the values of the | |
| 1738 | muscles `action' and `prologue' are 0-terminated. | |
| 1739 | ||
| a870c567 AD |
1740 | 2001-12-10 Akim Demaille <akim@epita.fr> |
| 1741 | ||
| 1742 | Clean up GCC warnings. | |
| 1743 | ||
| 1744 | * src/reader.c (copy_action): `buf' is not used. | |
| 1745 | (parse_skel_decl): Be static. | |
| 1746 | * src/muscle_tab.c (mhash1, mhash2, muscle_insert): Preserve `const'. | |
| 1747 | * src/options.h (create_long_option_table): Have a real prototype. | |
| 1748 | * lib/hash.c, lib/hash.h (hash_insert, hash_insert_at, hash_delete) | |
| 1749 | (hash_delete_at): Return const void *. | |
| 1750 | Adjust casts to preserve the const. | |
| 1751 | ||
| 80df8768 AD |
1752 | 2001-12-10 Akim Demaille <akim@epita.fr> |
| 1753 | ||
| 1754 | * configure.in: Require 2.52g. | |
| 1755 | M4 is not needed, but AUTOM4TE is. | |
| 1756 | * m4/m4.m4: Remove. | |
| 1757 | * tests/Makefile.am: Adjust. | |
| 1758 | ||
| f693ad14 AD |
1759 | 2001-12-10 Akim Demaille <akim@epita.fr> |
| 1760 | ||
| 1761 | One structure for states is enough, even though theoretically | |
| 1762 | there are LR(0) states and LALR(1) states. | |
| 1763 | ||
| 1764 | * src/lalr.h (state_t): Remove. | |
| 1765 | (state_table): Be state_t **, not state_t *. | |
| 1766 | * src/state.h (core, CORE_ALLOC): Rename as... | |
| 1767 | (state_t, STATE_ALLOC): this. | |
| 1768 | Add the LALR(1) members: shifts, reductions, errs. | |
| 1769 | * src/LR0.c (state_table): Rename as... | |
| 1770 | (state_hash): this, to avoid name clashes with the global | |
| 1771 | `state_table'. | |
| 1772 | * src/print_graph.c, src/LR0.c, src/LR0.h, src/conflicts.c | |
| 1773 | * src/lalr.c, src/lalr.h, src/output.c, src/print.c: Adjust. | |
| 1774 | ||
| 74ffbcb6 AD |
1775 | 2001-12-10 Akim Demaille <akim@epita.fr> |
| 1776 | ||
| 1777 | Bison dumps core on bash.y. | |
| 1778 | Reported by Pascal Bart. | |
| 1779 | ||
| 1780 | * src/warshall.c (bitmatrix_print): New. | |
| 1781 | (TC): Use it. | |
| 1782 | When performing a transitive closure R(i, j) && R(j, k) => R(i, k), | |
| 1783 | j must be the outer loop. | |
| 1784 | * tests/regression.at (Broken Closure): New. | |
| 1785 | ||
| 07708e19 AD |
1786 | 2001-12-05 Akim Demaille <akim@epita.fr> |
| 1787 | ||
| 1788 | * tests/atlocal.in (CPPFLAGS): Do not leave a space between -I and | |
| 1789 | its argument. | |
| 1790 |