Commit | Line | Data |
---|---|---|
8489eb67 | 1 | /* GNU Emacs routines to deal with syntax tables; also word and list parsing. |
ba318903 | 2 | Copyright (C) 1985, 1987, 1993-1995, 1997-1999, 2001-2014 Free |
ab422c4d | 3 | Software Foundation, Inc. |
8489eb67 RS |
4 | |
5 | This file is part of GNU Emacs. | |
6 | ||
9ec0b715 | 7 | GNU Emacs is free software: you can redistribute it and/or modify |
8489eb67 | 8 | it under the terms of the GNU General Public License as published by |
9ec0b715 GM |
9 | the Free Software Foundation, either version 3 of the License, or |
10 | (at your option) any later version. | |
8489eb67 RS |
11 | |
12 | GNU Emacs is distributed in the hope that it will be useful, | |
13 | but WITHOUT ANY WARRANTY; without even the implied warranty of | |
14 | MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the | |
15 | GNU General Public License for more details. | |
16 | ||
17 | You should have received a copy of the GNU General Public License | |
9ec0b715 | 18 | along with GNU Emacs. If not, see <http://www.gnu.org/licenses/>. */ |
8489eb67 RS |
19 | |
20 | ||
18160b98 | 21 | #include <config.h> |
1b2de274 | 22 | |
1b2de274 | 23 | #include <sys/types.h> |
0328b6de | 24 | |
8489eb67 RS |
25 | #include "lisp.h" |
26 | #include "commands.h" | |
5c7b02ab | 27 | #include "character.h" |
e5560ff7 | 28 | #include "buffer.h" |
e35f6ff7 | 29 | #include "keymap.h" |
a1bc88d4 | 30 | #include "regex.h" |
195d1361 | 31 | |
8489eb67 | 32 | #include "syntax.h" |
195d1361 | 33 | #include "intervals.h" |
c5683ceb SM |
34 | #include "category.h" |
35 | ||
c6cfd910 PE |
36 | /* Make syntax table lookup grant data in gl_state. */ |
37 | #define SYNTAX(c) syntax_property (c, 1) | |
38 | #define SYNTAX_ENTRY(c) syntax_property_entry (c, 1) | |
39 | #define SYNTAX_WITH_FLAGS(c) syntax_property_with_flags (c, 1) | |
40 | ||
1fc71008 | 41 | /* Eight single-bit flags have the following meanings: |
c5683ceb SM |
42 | 1. This character is the first of a two-character comment-start sequence. |
43 | 2. This character is the second of a two-character comment-start sequence. | |
44 | 3. This character is the first of a two-character comment-end sequence. | |
45 | 4. This character is the second of a two-character comment-end sequence. | |
46 | 5. This character is a prefix, for backward-prefix-chars. | |
47 | 6. The char is part of a delimiter for comments of style "b". | |
48 | 7. This character is part of a nestable comment sequence. | |
49 | 8. The char is part of a delimiter for comments of style "c". | |
50 | Note that any two-character sequence whose first character has flag 1 | |
51 | and whose second character has flag 2 will be interpreted as a comment start. | |
52 | ||
1fc71008 | 53 | Bits 6 and 8 discriminate among different comment styles. |
c5683ceb | 54 | Languages such as C++ allow two orthogonal syntax start/end pairs |
1fc71008 | 55 | and bit 6 determines whether a comment-end or Scommentend |
c5683ceb SM |
56 | ends style a or b. Comment markers can start style a, b, c, or bc. |
57 | Style a is always the default. | |
1fc71008 | 58 | For 2-char comment markers, the style b flag is looked up only on the second |
c5683ceb | 59 | char of the comment marker and on the first char of the comment ender. |
1fc71008 PE |
60 | For style c (like the nested flag), the flag can be placed on any of |
61 | the chars. */ | |
c5683ceb | 62 | |
45b683a1 | 63 | /* These functions extract specific flags from an integer |
c5683ceb SM |
64 | that holds the syntax code and the flags. */ |
65 | ||
45b683a1 PE |
66 | static bool |
67 | SYNTAX_FLAGS_COMSTART_FIRST (int flags) | |
68 | { | |
69 | return (flags >> 16) & 1; | |
70 | } | |
71 | static bool | |
72 | SYNTAX_FLAGS_COMSTART_SECOND (int flags) | |
73 | { | |
74 | return (flags >> 17) & 1; | |
75 | } | |
76 | static bool | |
77 | SYNTAX_FLAGS_COMEND_FIRST (int flags) | |
78 | { | |
79 | return (flags >> 18) & 1; | |
80 | } | |
81 | static bool | |
82 | SYNTAX_FLAGS_COMEND_SECOND (int flags) | |
83 | { | |
84 | return (flags >> 19) & 1; | |
85 | } | |
86 | static bool | |
87 | SYNTAX_FLAGS_PREFIX (int flags) | |
88 | { | |
89 | return (flags >> 20) & 1; | |
90 | } | |
91 | static bool | |
92 | SYNTAX_FLAGS_COMMENT_STYLEB (int flags) | |
93 | { | |
94 | return (flags >> 21) & 1; | |
95 | } | |
96 | static bool | |
97 | SYNTAX_FLAGS_COMMENT_STYLEC (int flags) | |
98 | { | |
99 | return (flags >> 23) & 1; | |
100 | } | |
101 | static int | |
102 | SYNTAX_FLAGS_COMMENT_STYLEC2 (int flags) | |
103 | { | |
104 | return (flags >> 22) & 2; /* SYNTAX_FLAGS_COMMENT_STYLEC (flags) * 2 */ | |
105 | } | |
106 | static bool | |
107 | SYNTAX_FLAGS_COMMENT_NESTED (int flags) | |
108 | { | |
109 | return (flags >> 22) & 1; | |
110 | } | |
c5683ceb | 111 | |
c5683ceb SM |
112 | /* FLAGS should be the flags of the main char of the comment marker, e.g. |
113 | the second for comstart and the first for comend. */ | |
45b683a1 PE |
114 | static int |
115 | SYNTAX_FLAGS_COMMENT_STYLE (int flags, int other_flags) | |
116 | { | |
117 | return (SYNTAX_FLAGS_COMMENT_STYLEB (flags) | |
118 | | SYNTAX_FLAGS_COMMENT_STYLEC2 (flags) | |
119 | | SYNTAX_FLAGS_COMMENT_STYLEC2 (other_flags)); | |
120 | } | |
c5683ceb | 121 | |
45b683a1 | 122 | /* Extract a particular flag for a given character. */ |
c5683ceb | 123 | |
45b683a1 PE |
124 | static bool |
125 | SYNTAX_COMEND_FIRST (int c) | |
126 | { | |
127 | return SYNTAX_FLAGS_COMEND_FIRST (SYNTAX_WITH_FLAGS (c)); | |
128 | } | |
195d1361 RS |
129 | |
130 | /* We use these constants in place for comment-style and | |
45b683a1 | 131 | string-ender-char to distinguish comments/strings started by |
195d1361 RS |
132 | comment_fence and string_fence codes. */ |
133 | ||
45b683a1 PE |
134 | enum |
135 | { | |
136 | ST_COMMENT_STYLE = 256 + 1, | |
137 | ST_STRING_STYLE = 256 + 2 | |
138 | }; | |
8489eb67 | 139 | |
955cbe7b PE |
140 | static Lisp_Object Qsyntax_table_p; |
141 | static Lisp_Object Qsyntax_table, Qscan_error; | |
8489eb67 | 142 | |
e5d4f4dc RS |
143 | /* This is the internal form of the parse state used in parse-partial-sexp. */ |
144 | ||
145 | struct lisp_parse_state | |
146 | { | |
d311d28c | 147 | EMACS_INT depth; /* Depth at end of parsing. */ |
6449674e | 148 | int instring; /* -1 if not within string, else desired terminator. */ |
d311d28c | 149 | EMACS_INT incomment; /* -1 if in unnestable comment else comment nesting */ |
6449674e | 150 | int comstyle; /* comment style a=0, or b=1, or ST_COMMENT_STYLE. */ |
1fc71008 | 151 | bool quoted; /* True if just after an escape char at end of parsing. */ |
d311d28c | 152 | EMACS_INT mindepth; /* Minimum depth seen while scanning. */ |
6449674e | 153 | /* Char number of most recent start-of-expression at current level */ |
d311d28c | 154 | ptrdiff_t thislevelstart; |
6449674e | 155 | /* Char number of start of containing expression */ |
d311d28c PE |
156 | ptrdiff_t prevlevelstart; |
157 | ptrdiff_t location; /* Char number at which parsing stopped. */ | |
8ca30920 | 158 | ptrdiff_t location_byte; /* Corresponding byte position. */ |
d311d28c | 159 | ptrdiff_t comstr_start; /* Position of last comment/string starter. */ |
6449674e SM |
160 | Lisp_Object levelstarts; /* Char numbers of starts-of-expression |
161 | of levels (starting from outermost). */ | |
e5d4f4dc RS |
162 | }; |
163 | \f | |
37bef230 RS |
164 | /* These variables are a cache for finding the start of a defun. |
165 | find_start_pos is the place for which the defun start was found. | |
166 | find_start_value is the defun start position found for it. | |
6a140a74 | 167 | find_start_value_byte is the corresponding byte position. |
37bef230 RS |
168 | find_start_buffer is the buffer it was found in. |
169 | find_start_begv is the BEGV value when it was found. | |
170 | find_start_modiff is the value of MODIFF when it was found. */ | |
171 | ||
d311d28c PE |
172 | static ptrdiff_t find_start_pos; |
173 | static ptrdiff_t find_start_value; | |
174 | static ptrdiff_t find_start_value_byte; | |
37bef230 | 175 | static struct buffer *find_start_buffer; |
d311d28c PE |
176 | static ptrdiff_t find_start_begv; |
177 | static EMACS_INT find_start_modiff; | |
6a140a74 RS |
178 | |
179 | ||
1fc71008 PE |
180 | static Lisp_Object skip_chars (bool, Lisp_Object, Lisp_Object, bool); |
181 | static Lisp_Object skip_syntaxes (bool, Lisp_Object, Lisp_Object); | |
182 | static Lisp_Object scan_lists (EMACS_INT, EMACS_INT, EMACS_INT, bool); | |
f57e2426 | 183 | static void scan_sexps_forward (struct lisp_parse_state *, |
d311d28c | 184 | ptrdiff_t, ptrdiff_t, ptrdiff_t, EMACS_INT, |
1fc71008 PE |
185 | bool, Lisp_Object, int); |
186 | static bool in_classes (int, Lisp_Object); | |
39eb03f1 PE |
187 | |
188 | /* This setter is used only in this file, so it can be private. */ | |
b0ab8123 | 189 | static void |
39eb03f1 PE |
190 | bset_syntax_table (struct buffer *b, Lisp_Object val) |
191 | { | |
192 | b->INTERNAL_FIELD (syntax_table) = val; | |
193 | } | |
195d1361 | 194 | \f |
c5683ceb | 195 | /* Whether the syntax of the character C has the prefix flag set. */ |
1fc71008 PE |
196 | bool |
197 | syntax_prefix_flag_p (int c) | |
c5683ceb | 198 | { |
45b683a1 | 199 | return SYNTAX_FLAGS_PREFIX (SYNTAX_WITH_FLAGS (c)); |
c5683ceb | 200 | } |
195d1361 RS |
201 | |
202 | struct gl_state_s gl_state; /* Global state of syntax parser. */ | |
203 | ||
45b683a1 | 204 | enum { INTERVALS_AT_ONCE = 10 }; /* 1 + max-number of intervals |
195d1361 RS |
205 | to scan to property-change. */ |
206 | ||
45b683a1 PE |
207 | /* Set the syntax entry VAL for char C in table TABLE. */ |
208 | ||
209 | static void | |
210 | SET_RAW_SYNTAX_ENTRY (Lisp_Object table, int c, Lisp_Object val) | |
211 | { | |
212 | CHAR_TABLE_SET (table, c, val); | |
213 | } | |
214 | ||
215 | /* Set the syntax entry VAL for char-range RANGE in table TABLE. | |
216 | RANGE is a cons (FROM . TO) specifying the range of characters. */ | |
217 | ||
218 | static void | |
219 | SET_RAW_SYNTAX_ENTRY_RANGE (Lisp_Object table, Lisp_Object range, | |
220 | Lisp_Object val) | |
221 | { | |
222 | Fset_char_table_range (table, range, val); | |
223 | } | |
224 | ||
225 | /* Extract the information from the entry for character C | |
226 | in the current syntax table. */ | |
227 | ||
228 | static Lisp_Object | |
229 | SYNTAX_MATCH (int c) | |
230 | { | |
231 | Lisp_Object ent = SYNTAX_ENTRY (c); | |
232 | return CONSP (ent) ? XCDR (ent) : Qnil; | |
233 | } | |
234 | ||
235 | /* This should be called with FROM at the start of forward | |
236 | search, or after the last position of the backward search. It | |
237 | makes sure that the first char is picked up with correct table, so | |
238 | one does not need to call UPDATE_SYNTAX_TABLE immediately after the | |
239 | call. | |
240 | Sign of COUNT gives the direction of the search. | |
241 | */ | |
242 | ||
243 | static void | |
244 | SETUP_SYNTAX_TABLE (ptrdiff_t from, ptrdiff_t count) | |
245 | { | |
246 | SETUP_BUFFER_SYNTAX_TABLE (); | |
247 | gl_state.b_property = BEGV; | |
248 | gl_state.e_property = ZV + 1; | |
249 | gl_state.object = Qnil; | |
250 | gl_state.offset = 0; | |
251 | if (parse_sexp_lookup_properties) | |
252 | if (count > 0 || from > BEGV) | |
253 | update_syntax_table (count > 0 ? from : from - 1, count, 1, Qnil); | |
254 | } | |
255 | ||
256 | /* Same as above, but in OBJECT. If OBJECT is nil, use current buffer. | |
257 | If it is t (which is only used in fast_c_string_match_ignore_case), | |
258 | ignore properties altogether. | |
259 | ||
260 | This is meant for regex.c to use. For buffers, regex.c passes arguments | |
261 | to the UPDATE_SYNTAX_TABLE functions which are relative to BEGV. | |
262 | So if it is a buffer, we set the offset field to BEGV. */ | |
263 | ||
264 | void | |
265 | SETUP_SYNTAX_TABLE_FOR_OBJECT (Lisp_Object object, | |
266 | ptrdiff_t from, ptrdiff_t count) | |
267 | { | |
268 | SETUP_BUFFER_SYNTAX_TABLE (); | |
269 | gl_state.object = object; | |
270 | if (BUFFERP (gl_state.object)) | |
271 | { | |
272 | struct buffer *buf = XBUFFER (gl_state.object); | |
273 | gl_state.b_property = 1; | |
274 | gl_state.e_property = BUF_ZV (buf) - BUF_BEGV (buf) + 1; | |
275 | gl_state.offset = BUF_BEGV (buf) - 1; | |
276 | } | |
277 | else if (NILP (gl_state.object)) | |
278 | { | |
279 | gl_state.b_property = 1; | |
280 | gl_state.e_property = ZV - BEGV + 1; | |
281 | gl_state.offset = BEGV - 1; | |
282 | } | |
283 | else if (EQ (gl_state.object, Qt)) | |
284 | { | |
285 | gl_state.b_property = 0; | |
286 | gl_state.e_property = PTRDIFF_MAX; | |
287 | gl_state.offset = 0; | |
288 | } | |
289 | else | |
290 | { | |
291 | gl_state.b_property = 0; | |
292 | gl_state.e_property = 1 + SCHARS (gl_state.object); | |
293 | gl_state.offset = 0; | |
294 | } | |
295 | if (parse_sexp_lookup_properties) | |
296 | update_syntax_table (from + gl_state.offset - (count <= 0), | |
297 | count, 1, gl_state.object); | |
298 | } | |
299 | ||
6a140a74 RS |
300 | /* Update gl_state to an appropriate interval which contains CHARPOS. The |
301 | sign of COUNT give the relative position of CHARPOS wrt the previously | |
195d1361 | 302 | valid interval. If INIT, only [be]_property fields of gl_state are |
6a140a74 | 303 | valid at start, the rest is filled basing on OBJECT. |
195d1361 | 304 | |
6a140a74 | 305 | `gl_state.*_i' are the intervals, and CHARPOS is further in the search |
195d1361 RS |
306 | direction than the intervals - or in an interval. We update the |
307 | current syntax-table basing on the property of this interval, and | |
6a140a74 | 308 | update the interval to start further than CHARPOS - or be |
77c7bcb1 | 309 | NULL. We also update lim_property to be the next value of |
6a140a74 | 310 | charpos to call this subroutine again - or be before/after the |
195d1361 RS |
311 | start/end of OBJECT. */ |
312 | ||
313 | void | |
1fc71008 | 314 | update_syntax_table (ptrdiff_t charpos, EMACS_INT count, bool init, |
4f3a2f8d | 315 | Lisp_Object object) |
195d1361 RS |
316 | { |
317 | Lisp_Object tmp_table; | |
1fc71008 PE |
318 | int cnt = 0; |
319 | bool invalidate = 1; | |
5a774522 | 320 | INTERVAL i; |
195d1361 RS |
321 | |
322 | if (init) | |
323 | { | |
bb0de084 | 324 | gl_state.old_prop = Qnil; |
195d1361 RS |
325 | gl_state.start = gl_state.b_property; |
326 | gl_state.stop = gl_state.e_property; | |
bb0de084 SM |
327 | i = interval_of (charpos, object); |
328 | gl_state.backward_i = gl_state.forward_i = i; | |
195d1361 | 329 | invalidate = 0; |
77c7bcb1 | 330 | if (!i) |
195d1361 | 331 | return; |
f902a008 | 332 | /* interval_of updates only ->position of the return value, so |
d80f4cc7 | 333 | update the parents manually to speed up update_interval. */ |
bb0de084 | 334 | while (!NULL_PARENT (i)) |
d80f4cc7 RS |
335 | { |
336 | if (AM_RIGHT_CHILD (i)) | |
439d5cb4 | 337 | INTERVAL_PARENT (i)->position = i->position |
d80f4cc7 | 338 | - LEFT_TOTAL_LENGTH (i) + TOTAL_LENGTH (i) /* right end */ |
439d5cb4 KR |
339 | - TOTAL_LENGTH (INTERVAL_PARENT (i)) |
340 | + LEFT_TOTAL_LENGTH (INTERVAL_PARENT (i)); | |
d80f4cc7 | 341 | else |
439d5cb4 | 342 | INTERVAL_PARENT (i)->position = i->position - LEFT_TOTAL_LENGTH (i) |
d80f4cc7 | 343 | + TOTAL_LENGTH (i); |
439d5cb4 | 344 | i = INTERVAL_PARENT (i); |
d80f4cc7 RS |
345 | } |
346 | i = gl_state.forward_i; | |
bb0de084 | 347 | gl_state.b_property = i->position - gl_state.offset; |
ee0cdb48 | 348 | gl_state.e_property = INTERVAL_LAST_POS (i) - gl_state.offset; |
195d1361 RS |
349 | goto update; |
350 | } | |
5a774522 | 351 | i = count > 0 ? gl_state.forward_i : gl_state.backward_i; |
195d1361 | 352 | |
423e705d | 353 | /* We are guaranteed to be called with CHARPOS either in i, |
6a140a74 | 354 | or further off. */ |
77c7bcb1 | 355 | if (!i) |
195d1361 | 356 | error ("Error in syntax_table logic for to-the-end intervals"); |
6a140a74 | 357 | else if (charpos < i->position) /* Move left. */ |
195d1361 RS |
358 | { |
359 | if (count > 0) | |
f902a008 | 360 | error ("Error in syntax_table logic for intervals <-"); |
195d1361 | 361 | /* Update the interval. */ |
6a140a74 | 362 | i = update_interval (i, charpos); |
bb0de084 | 363 | if (INTERVAL_LAST_POS (i) != gl_state.b_property) |
195d1361 RS |
364 | { |
365 | invalidate = 0; | |
195d1361 | 366 | gl_state.forward_i = i; |
ee0cdb48 | 367 | gl_state.e_property = INTERVAL_LAST_POS (i) - gl_state.offset; |
195d1361 | 368 | } |
423e705d | 369 | } |
6a140a74 | 370 | else if (charpos >= INTERVAL_LAST_POS (i)) /* Move right. */ |
195d1361 RS |
371 | { |
372 | if (count < 0) | |
f902a008 | 373 | error ("Error in syntax_table logic for intervals ->"); |
195d1361 | 374 | /* Update the interval. */ |
6a140a74 | 375 | i = update_interval (i, charpos); |
bb0de084 | 376 | if (i->position != gl_state.e_property) |
195d1361 RS |
377 | { |
378 | invalidate = 0; | |
195d1361 | 379 | gl_state.backward_i = i; |
bb0de084 | 380 | gl_state.b_property = i->position - gl_state.offset; |
195d1361 RS |
381 | } |
382 | } | |
195d1361 RS |
383 | |
384 | update: | |
385 | tmp_table = textget (i->plist, Qsyntax_table); | |
386 | ||
387 | if (invalidate) | |
388 | invalidate = !EQ (tmp_table, gl_state.old_prop); /* Need to invalidate? */ | |
7d0393cf | 389 | |
423e705d SM |
390 | if (invalidate) /* Did not get to adjacent interval. */ |
391 | { /* with the same table => */ | |
392 | /* invalidate the old range. */ | |
195d1361 RS |
393 | if (count > 0) |
394 | { | |
395 | gl_state.backward_i = i; | |
bb0de084 SM |
396 | gl_state.b_property = i->position - gl_state.offset; |
397 | } | |
398 | else | |
195d1361 | 399 | { |
bb0de084 | 400 | gl_state.forward_i = i; |
ee0cdb48 | 401 | gl_state.e_property = INTERVAL_LAST_POS (i) - gl_state.offset; |
195d1361 RS |
402 | } |
403 | } | |
37bef230 | 404 | |
bb0de084 | 405 | if (!EQ (tmp_table, gl_state.old_prop)) |
195d1361 | 406 | { |
bb0de084 SM |
407 | gl_state.current_syntax_table = tmp_table; |
408 | gl_state.old_prop = tmp_table; | |
409 | if (EQ (Fsyntax_table_p (tmp_table), Qt)) | |
410 | { | |
411 | gl_state.use_global = 0; | |
7d0393cf | 412 | } |
bb0de084 SM |
413 | else if (CONSP (tmp_table)) |
414 | { | |
415 | gl_state.use_global = 1; | |
416 | gl_state.global_code = tmp_table; | |
417 | } | |
7d0393cf | 418 | else |
bb0de084 SM |
419 | { |
420 | gl_state.use_global = 0; | |
4b4deea2 | 421 | gl_state.current_syntax_table = BVAR (current_buffer, syntax_table); |
bb0de084 | 422 | } |
195d1361 RS |
423 | } |
424 | ||
77c7bcb1 | 425 | while (i) |
195d1361 RS |
426 | { |
427 | if (cnt && !EQ (tmp_table, textget (i->plist, Qsyntax_table))) | |
428 | { | |
429 | if (count > 0) | |
bb0de084 SM |
430 | { |
431 | gl_state.e_property = i->position - gl_state.offset; | |
432 | gl_state.forward_i = i; | |
433 | } | |
434 | else | |
435 | { | |
5a774522 SM |
436 | gl_state.b_property |
437 | = i->position + LENGTH (i) - gl_state.offset; | |
bb0de084 SM |
438 | gl_state.backward_i = i; |
439 | } | |
440 | return; | |
195d1361 | 441 | } |
7d0393cf | 442 | else if (cnt == INTERVALS_AT_ONCE) |
195d1361 RS |
443 | { |
444 | if (count > 0) | |
bb0de084 | 445 | { |
5a774522 SM |
446 | gl_state.e_property |
447 | = i->position + LENGTH (i) - gl_state.offset | |
448 | /* e_property at EOB is not set to ZV but to ZV+1, so that | |
449 | we can do INC(from);UPDATE_SYNTAX_TABLE_FORWARD without | |
450 | having to check eob between the two. */ | |
77c7bcb1 | 451 | + (next_interval (i) ? 0 : 1); |
bb0de084 SM |
452 | gl_state.forward_i = i; |
453 | } | |
454 | else | |
455 | { | |
456 | gl_state.b_property = i->position - gl_state.offset; | |
457 | gl_state.backward_i = i; | |
458 | } | |
459 | return; | |
195d1361 RS |
460 | } |
461 | cnt++; | |
462 | i = count > 0 ? next_interval (i) : previous_interval (i); | |
463 | } | |
77c7bcb1 | 464 | eassert (i == NULL); /* This property goes to the end. */ |
bb0de084 SM |
465 | if (count > 0) |
466 | gl_state.e_property = gl_state.stop; | |
467 | else | |
468 | gl_state.b_property = gl_state.start; | |
195d1361 RS |
469 | } |
470 | \f | |
1fc71008 | 471 | /* Returns true if char at CHARPOS is quoted. |
6a140a74 RS |
472 | Global syntax-table data should be set up already to be good at CHARPOS |
473 | or after. On return global syntax data is good for lookup at CHARPOS. */ | |
195d1361 | 474 | |
1fc71008 | 475 | static bool |
d311d28c | 476 | char_quoted (ptrdiff_t charpos, ptrdiff_t bytepos) |
195d1361 | 477 | { |
1fc71008 PE |
478 | enum syntaxcode code; |
479 | ptrdiff_t beg = BEGV; | |
480 | bool quoted = 0; | |
d311d28c | 481 | ptrdiff_t orig = charpos; |
6a140a74 | 482 | |
02d8b017 | 483 | while (charpos > beg) |
195d1361 | 484 | { |
ab229fdd | 485 | int c; |
02d8b017 | 486 | DEC_BOTH (charpos, bytepos); |
ab229fdd | 487 | |
6a140a74 | 488 | UPDATE_SYNTAX_TABLE_BACKWARD (charpos); |
327719ee | 489 | c = FETCH_CHAR_AS_MULTIBYTE (bytepos); |
ab229fdd | 490 | code = SYNTAX (c); |
6a140a74 RS |
491 | if (! (code == Scharquote || code == Sescape)) |
492 | break; | |
493 | ||
6a140a74 | 494 | quoted = !quoted; |
195d1361 | 495 | } |
6a140a74 RS |
496 | |
497 | UPDATE_SYNTAX_TABLE (orig); | |
195d1361 RS |
498 | return quoted; |
499 | } | |
6a140a74 | 500 | |
6a140a74 RS |
501 | /* Return the bytepos one character before BYTEPOS. |
502 | We assume that BYTEPOS is not at the start of the buffer. */ | |
503 | ||
b0ab8123 | 504 | static ptrdiff_t |
d311d28c | 505 | dec_bytepos (ptrdiff_t bytepos) |
6a140a74 | 506 | { |
4b4deea2 | 507 | if (NILP (BVAR (current_buffer, enable_multibyte_characters))) |
ef316cf0 RS |
508 | return bytepos - 1; |
509 | ||
6a140a74 RS |
510 | DEC_POS (bytepos); |
511 | return bytepos; | |
512 | } | |
195d1361 | 513 | \f |
2b34df4e | 514 | /* Return a defun-start position before POS and not too far before. |
7cc80f0a RS |
515 | It should be the last one before POS, or nearly the last. |
516 | ||
517 | When open_paren_in_column_0_is_defun_start is nonzero, | |
1fd1cc2f | 518 | only the beginning of the buffer is treated as a defun-start. |
7cc80f0a RS |
519 | |
520 | We record the information about where the scan started | |
521 | and what its result was, so that another call in the same area | |
522 | can return the same value very quickly. | |
195d1361 RS |
523 | |
524 | There is no promise at which position the global syntax data is | |
525 | valid on return from the subroutine, so the caller should explicitly | |
526 | update the global data. */ | |
37bef230 | 527 | |
d311d28c PE |
528 | static ptrdiff_t |
529 | find_defun_start (ptrdiff_t pos, ptrdiff_t pos_byte) | |
37bef230 | 530 | { |
d311d28c | 531 | ptrdiff_t opoint = PT, opoint_byte = PT_BYTE; |
37bef230 | 532 | |
1fd1cc2f RS |
533 | if (!open_paren_in_column_0_is_defun_start) |
534 | { | |
6e20a0d4 | 535 | find_start_value = BEGV; |
1fd1cc2f | 536 | find_start_value_byte = BEGV_BYTE; |
6e20a0d4 CY |
537 | find_start_buffer = current_buffer; |
538 | find_start_modiff = MODIFF; | |
539 | find_start_begv = BEGV; | |
540 | find_start_pos = pos; | |
1fd1cc2f RS |
541 | return BEGV; |
542 | } | |
543 | ||
37bef230 RS |
544 | /* Use previous finding, if it's valid and applies to this inquiry. */ |
545 | if (current_buffer == find_start_buffer | |
546 | /* Reuse the defun-start even if POS is a little farther on. | |
547 | POS might be in the next defun, but that's ok. | |
548 | Our value may not be the best possible, but will still be usable. */ | |
549 | && pos <= find_start_pos + 1000 | |
550 | && pos >= find_start_value | |
551 | && BEGV == find_start_begv | |
552 | && MODIFF == find_start_modiff) | |
553 | return find_start_value; | |
554 | ||
555 | /* Back up to start of line. */ | |
6a140a74 | 556 | scan_newline (pos, pos_byte, BEGV, BEGV_BYTE, -1, 1); |
37bef230 | 557 | |
195d1361 RS |
558 | /* We optimize syntax-table lookup for rare updates. Thus we accept |
559 | only those `^\s(' which are good in global _and_ text-property | |
560 | syntax-tables. */ | |
d48cd3f4 | 561 | SETUP_BUFFER_SYNTAX_TABLE (); |
1fd1cc2f | 562 | while (PT > BEGV) |
37bef230 | 563 | { |
ab229fdd AS |
564 | int c; |
565 | ||
1fd1cc2f RS |
566 | /* Open-paren at start of line means we may have found our |
567 | defun-start. */ | |
327719ee | 568 | c = FETCH_CHAR_AS_MULTIBYTE (PT_BYTE); |
ab229fdd | 569 | if (SYNTAX (c) == Sopen) |
195d1361 | 570 | { |
1fd1cc2f | 571 | SETUP_SYNTAX_TABLE (PT + 1, -1); /* Try again... */ |
327719ee | 572 | c = FETCH_CHAR_AS_MULTIBYTE (PT_BYTE); |
ab229fdd | 573 | if (SYNTAX (c) == Sopen) |
1fd1cc2f RS |
574 | break; |
575 | /* Now fallback to the default value. */ | |
d48cd3f4 | 576 | SETUP_BUFFER_SYNTAX_TABLE (); |
195d1361 | 577 | } |
1fd1cc2f RS |
578 | /* Move to beg of previous line. */ |
579 | scan_newline (PT, PT_BYTE, BEGV, BEGV_BYTE, -2, 1); | |
37bef230 RS |
580 | } |
581 | ||
582 | /* Record what we found, for the next try. */ | |
6a140a74 RS |
583 | find_start_value = PT; |
584 | find_start_value_byte = PT_BYTE; | |
37bef230 RS |
585 | find_start_buffer = current_buffer; |
586 | find_start_modiff = MODIFF; | |
587 | find_start_begv = BEGV; | |
588 | find_start_pos = pos; | |
589 | ||
6a140a74 RS |
590 | TEMP_SET_PT_BOTH (opoint, opoint_byte); |
591 | ||
37bef230 RS |
592 | return find_start_value; |
593 | } | |
594 | \f | |
f902a008 RS |
595 | /* Return the SYNTAX_COMEND_FIRST of the character before POS, POS_BYTE. */ |
596 | ||
1fc71008 | 597 | static bool |
d311d28c | 598 | prev_char_comend_first (ptrdiff_t pos, ptrdiff_t pos_byte) |
f902a008 | 599 | { |
1fc71008 PE |
600 | int c; |
601 | bool val; | |
f902a008 RS |
602 | |
603 | DEC_BOTH (pos, pos_byte); | |
604 | UPDATE_SYNTAX_TABLE_BACKWARD (pos); | |
605 | c = FETCH_CHAR (pos_byte); | |
606 | val = SYNTAX_COMEND_FIRST (c); | |
607 | UPDATE_SYNTAX_TABLE_FORWARD (pos + 1); | |
608 | return val; | |
609 | } | |
610 | ||
1fc71008 | 611 | /* Check whether charpos FROM is at the end of a comment. |
6a140a74 RS |
612 | FROM_BYTE is the bytepos corresponding to FROM. |
613 | Do not move back before STOP. | |
614 | ||
1fc71008 | 615 | Return true if we find a comment ending at FROM/FROM_BYTE. |
6a140a74 RS |
616 | |
617 | If successful, store the charpos of the comment's beginning | |
618 | into *CHARPOS_PTR, and the bytepos into *BYTEPOS_PTR. | |
527a32d9 KH |
619 | |
620 | Global syntax data remains valid for backward search starting at | |
621 | the returned value (or at FROM, if the search was not successful). */ | |
195d1361 | 622 | |
1fc71008 PE |
623 | static bool |
624 | back_comment (ptrdiff_t from, ptrdiff_t from_byte, ptrdiff_t stop, | |
625 | bool comnested, int comstyle, ptrdiff_t *charpos_ptr, | |
626 | ptrdiff_t *bytepos_ptr) | |
195d1361 RS |
627 | { |
628 | /* Look back, counting the parity of string-quotes, | |
629 | and recording the comment-starters seen. | |
630 | When we reach a safe place, assume that's not in a string; | |
631 | then step the main scan to the earliest comment-starter seen | |
632 | an even number of string quotes away from the safe place. | |
633 | ||
634 | OFROM[I] is position of the earliest comment-starter seen | |
635 | which is I+2X quotes from the comment-end. | |
636 | PARITY is current parity of quotes from the comment end. */ | |
3ee5041c | 637 | int string_style = -1; /* Presumed outside of any string. */ |
1fc71008 | 638 | bool string_lossage = 0; |
f7c436c1 | 639 | /* Not a real lossage: indicates that we have passed a matching comment |
7d0393cf | 640 | starter plus a non-matching comment-ender, meaning that any matching |
f7c436c1 SM |
641 | comment-starter we might see later could be a false positive (hidden |
642 | inside another comment). | |
643 | Test case: { a (* b } c (* d *) */ | |
1fc71008 | 644 | bool comment_lossage = 0; |
d311d28c PE |
645 | ptrdiff_t comment_end = from; |
646 | ptrdiff_t comment_end_byte = from_byte; | |
647 | ptrdiff_t comstart_pos = 0; | |
648 | ptrdiff_t comstart_byte IF_LINT (= 0); | |
eb35b628 RS |
649 | /* Place where the containing defun starts, |
650 | or 0 if we didn't come across it yet. */ | |
d311d28c PE |
651 | ptrdiff_t defun_start = 0; |
652 | ptrdiff_t defun_start_byte = 0; | |
1fc71008 PE |
653 | enum syntaxcode code; |
654 | ptrdiff_t nesting = 1; /* current comment nesting */ | |
ea315ed6 | 655 | int c; |
3f679f55 SM |
656 | int syntax = 0; |
657 | ||
658 | /* FIXME: A }} comment-ender style leads to incorrect behavior | |
659 | in the case of {{ c }}} because we ignore the last two chars which are | |
660 | assumed to be comment-enders although they aren't. */ | |
195d1361 RS |
661 | |
662 | /* At beginning of range to scan, we're outside of strings; | |
663 | that determines quote parity to the comment-end. */ | |
664 | while (from != stop) | |
665 | { | |
d311d28c | 666 | ptrdiff_t temp_byte; |
1fc71008 PE |
667 | int prev_syntax; |
668 | bool com2start, com2end, comstart; | |
6a140a74 | 669 | |
195d1361 | 670 | /* Move back and examine a character. */ |
6a140a74 | 671 | DEC_BOTH (from, from_byte); |
195d1361 RS |
672 | UPDATE_SYNTAX_TABLE_BACKWARD (from); |
673 | ||
3f679f55 | 674 | prev_syntax = syntax; |
b7dbcc19 | 675 | c = FETCH_CHAR_AS_MULTIBYTE (from_byte); |
3f679f55 | 676 | syntax = SYNTAX_WITH_FLAGS (c); |
195d1361 RS |
677 | code = SYNTAX (c); |
678 | ||
3f679f55 SM |
679 | /* Check for 2-char comment markers. */ |
680 | com2start = (SYNTAX_FLAGS_COMSTART_FIRST (syntax) | |
681 | && SYNTAX_FLAGS_COMSTART_SECOND (prev_syntax) | |
c5683ceb SM |
682 | && (comstyle |
683 | == SYNTAX_FLAGS_COMMENT_STYLE (prev_syntax, syntax)) | |
3f679f55 SM |
684 | && (SYNTAX_FLAGS_COMMENT_NESTED (prev_syntax) |
685 | || SYNTAX_FLAGS_COMMENT_NESTED (syntax)) == comnested); | |
686 | com2end = (SYNTAX_FLAGS_COMEND_FIRST (syntax) | |
687 | && SYNTAX_FLAGS_COMEND_SECOND (prev_syntax)); | |
fbb3da77 | 688 | comstart = (com2start || code == Scomment); |
01f44d5a | 689 | |
3f679f55 SM |
690 | /* Nasty cases with overlapping 2-char comment markers: |
691 | - snmp-mode: -- c -- foo -- c -- | |
692 | --- c -- | |
693 | ------ c -- | |
694 | - c-mode: *||* | |
695 | |* *|* *| | |
696 | |*| |* |*| | |
697 | /// */ | |
698 | ||
699 | /* If a 2-char comment sequence partly overlaps with another, | |
fbb3da77 SM |
700 | we don't try to be clever. E.g. |*| in C, or }% in modes that |
701 | have %..\n and %{..}%. */ | |
702 | if (from > stop && (com2end || comstart)) | |
195d1361 | 703 | { |
d311d28c | 704 | ptrdiff_t next = from, next_byte = from_byte; |
4f3a2f8d | 705 | int next_c, next_syntax; |
3f679f55 SM |
706 | DEC_BOTH (next, next_byte); |
707 | UPDATE_SYNTAX_TABLE_BACKWARD (next); | |
b7dbcc19 | 708 | next_c = FETCH_CHAR_AS_MULTIBYTE (next_byte); |
3f679f55 | 709 | next_syntax = SYNTAX_WITH_FLAGS (next_c); |
fbb3da77 | 710 | if (((comstart || comnested) |
3f679f55 SM |
711 | && SYNTAX_FLAGS_COMEND_SECOND (syntax) |
712 | && SYNTAX_FLAGS_COMEND_FIRST (next_syntax)) | |
713 | || ((com2end || comnested) | |
714 | && SYNTAX_FLAGS_COMSTART_SECOND (syntax) | |
c5683ceb SM |
715 | && (comstyle |
716 | == SYNTAX_FLAGS_COMMENT_STYLE (syntax, prev_syntax)) | |
3f679f55 SM |
717 | && SYNTAX_FLAGS_COMSTART_FIRST (next_syntax))) |
718 | goto lossage; | |
719 | /* UPDATE_SYNTAX_TABLE_FORWARD (next + 1); */ | |
f902a008 | 720 | } |
3f679f55 SM |
721 | |
722 | if (com2start && comstart_pos == 0) | |
723 | /* We're looking at a comment starter. But it might be a comment | |
724 | ender as well (see snmp-mode). The first time we see one, we | |
725 | need to consider it as a comment starter, | |
726 | and the subsequent times as a comment ender. */ | |
727 | com2end = 0; | |
728 | ||
729 | /* Turn a 2-char comment sequences into the appropriate syntax. */ | |
730 | if (com2end) | |
731 | code = Sendcomment; | |
732 | else if (com2start) | |
733 | code = Scomment; | |
734 | /* Ignore comment starters of a different style. */ | |
735 | else if (code == Scomment | |
c5683ceb | 736 | && (comstyle != SYNTAX_FLAGS_COMMENT_STYLE (syntax, 0) |
3f679f55 | 737 | || SYNTAX_FLAGS_COMMENT_NESTED (syntax) != comnested)) |
1aa963c8 | 738 | continue; |
195d1361 | 739 | |
9828a477 KH |
740 | /* Ignore escaped characters, except comment-enders. */ |
741 | if (code != Sendcomment && char_quoted (from, from_byte)) | |
195d1361 RS |
742 | continue; |
743 | ||
3ee5041c | 744 | switch (code) |
195d1361 | 745 | { |
3ee5041c SM |
746 | case Sstring_fence: |
747 | case Scomment_fence: | |
748 | c = (code == Sstring_fence ? ST_STRING_STYLE : ST_COMMENT_STYLE); | |
749 | case Sstring: | |
750 | /* Track parity of quotes. */ | |
751 | if (string_style == -1) | |
752 | /* Entering a string. */ | |
753 | string_style = c; | |
754 | else if (string_style == c) | |
755 | /* Leaving the string. */ | |
756 | string_style = -1; | |
757 | else | |
758 | /* If we have two kinds of string delimiters. | |
759 | There's no way to grok this scanning backwards. */ | |
195d1361 | 760 | string_lossage = 1; |
3ee5041c | 761 | break; |
7d0393cf | 762 | |
3ee5041c SM |
763 | case Scomment: |
764 | /* We've already checked that it is the relevant comstyle. */ | |
f7c436c1 | 765 | if (string_style != -1 || comment_lossage || string_lossage) |
3ee5041c SM |
766 | /* There are odd string quotes involved, so let's be careful. |
767 | Test case in Pascal: " { " a { " } */ | |
768 | goto lossage; | |
769 | ||
f7c436c1 SM |
770 | if (!comnested) |
771 | { | |
772 | /* Record best comment-starter so far. */ | |
773 | comstart_pos = from; | |
774 | comstart_byte = from_byte; | |
775 | } | |
776 | else if (--nesting <= 0) | |
95ff8dfc RS |
777 | /* nested comments have to be balanced, so we don't need to |
778 | keep looking for earlier ones. We use here the same (slightly | |
779 | incorrect) reasoning as below: since it is followed by uniform | |
780 | paired string quotes, this comment-start has to be outside of | |
781 | strings, else the comment-end itself would be inside a string. */ | |
782 | goto done; | |
3ee5041c SM |
783 | break; |
784 | ||
02010917 | 785 | case Sendcomment: |
c5683ceb | 786 | if (SYNTAX_FLAGS_COMMENT_STYLE (syntax, 0) == comstyle |
d070eb22 | 787 | && ((com2end && SYNTAX_FLAGS_COMMENT_NESTED (prev_syntax)) |
3f679f55 | 788 | || SYNTAX_FLAGS_COMMENT_NESTED (syntax)) == comnested) |
02010917 SM |
789 | /* This is the same style of comment ender as ours. */ |
790 | { | |
791 | if (comnested) | |
792 | nesting++; | |
793 | else | |
794 | /* Anything before that can't count because it would match | |
795 | this comment-ender rather than ours. */ | |
796 | from = stop; /* Break out of the loop. */ | |
797 | } | |
f7c436c1 SM |
798 | else if (comstart_pos != 0 || c != '\n') |
799 | /* We're mixing comment styles here, so we'd better be careful. | |
800 | The (comstart_pos != 0 || c != '\n') check is not quite correct | |
801 | (we should just always set comment_lossage), but removing it | |
802 | would imply that any multiline comment in C would go through | |
803 | lossage, which seems overkill. | |
804 | The failure should only happen in the rare cases such as | |
805 | { (* } *) */ | |
806 | comment_lossage = 1; | |
02010917 | 807 | break; |
195d1361 | 808 | |
02010917 SM |
809 | case Sopen: |
810 | /* Assume a defun-start point is outside of strings. */ | |
811 | if (open_paren_in_column_0_is_defun_start | |
812 | && (from == stop | |
813 | || (temp_byte = dec_bytepos (from_byte), | |
814 | FETCH_CHAR (temp_byte) == '\n'))) | |
815 | { | |
816 | defun_start = from; | |
817 | defun_start_byte = from_byte; | |
818 | from = stop; /* Break out of the loop. */ | |
819 | } | |
eb35b628 | 820 | break; |
02010917 SM |
821 | |
822 | default: | |
f7c436c1 | 823 | break; |
eb35b628 | 824 | } |
195d1361 RS |
825 | } |
826 | ||
827 | if (comstart_pos == 0) | |
828 | { | |
829 | from = comment_end; | |
6a140a74 | 830 | from_byte = comment_end_byte; |
195d1361 RS |
831 | UPDATE_SYNTAX_TABLE_FORWARD (comment_end - 1); |
832 | } | |
f7c436c1 SM |
833 | /* If comstart_pos is set and we get here (ie. didn't jump to `lossage' |
834 | or `done'), then we've found the beginning of the non-nested comment. */ | |
835 | else if (1) /* !comnested */ | |
195d1361 RS |
836 | { |
837 | from = comstart_pos; | |
6a140a74 | 838 | from_byte = comstart_byte; |
789f3320 | 839 | UPDATE_SYNTAX_TABLE_FORWARD (from - 1); |
195d1361 RS |
840 | } |
841 | else | |
842 | { | |
3ee5041c SM |
843 | struct lisp_parse_state state; |
844 | lossage: | |
195d1361 RS |
845 | /* We had two kinds of string delimiters mixed up |
846 | together. Decode this going forwards. | |
02010917 | 847 | Scan fwd from a known safe place (beginning-of-defun) |
195d1361 RS |
848 | to the one in question; this records where we |
849 | last passed a comment starter. */ | |
eb35b628 RS |
850 | /* If we did not already find the defun start, find it now. */ |
851 | if (defun_start == 0) | |
852 | { | |
853 | defun_start = find_defun_start (comment_end, comment_end_byte); | |
854 | defun_start_byte = find_start_value_byte; | |
855 | } | |
02010917 | 856 | do |
195d1361 | 857 | { |
02010917 SM |
858 | scan_sexps_forward (&state, |
859 | defun_start, defun_start_byte, | |
d311d28c PE |
860 | comment_end, TYPE_MINIMUM (EMACS_INT), |
861 | 0, Qnil, 0); | |
02010917 SM |
862 | defun_start = comment_end; |
863 | if (state.incomment == (comnested ? 1 : -1) | |
864 | && state.comstyle == comstyle) | |
865 | from = state.comstr_start; | |
866 | else | |
867 | { | |
868 | from = comment_end; | |
869 | if (state.incomment) | |
870 | /* If comment_end is inside some other comment, maybe ours | |
871 | is nested, so we need to try again from within the | |
872 | surrounding comment. Example: { a (* " *) */ | |
873 | { | |
874 | /* FIXME: We should advance by one or two chars. */ | |
875 | defun_start = state.comstr_start + 2; | |
876 | defun_start_byte = CHAR_TO_BYTE (defun_start); | |
877 | } | |
878 | } | |
879 | } while (defun_start < comment_end); | |
880 | ||
6a140a74 | 881 | from_byte = CHAR_TO_BYTE (from); |
195d1361 RS |
882 | UPDATE_SYNTAX_TABLE_FORWARD (from - 1); |
883 | } | |
7d0393cf | 884 | |
95ff8dfc | 885 | done: |
6a140a74 RS |
886 | *charpos_ptr = from; |
887 | *bytepos_ptr = from_byte; | |
888 | ||
1fc71008 | 889 | return from != comment_end; |
195d1361 RS |
890 | } |
891 | \f | |
8489eb67 | 892 | DEFUN ("syntax-table-p", Fsyntax_table_p, Ssyntax_table_p, 1, 1, 0, |
fdb82f93 PJ |
893 | doc: /* Return t if OBJECT is a syntax table. |
894 | Currently, any char-table counts as a syntax table. */) | |
5842a27b | 895 | (Lisp_Object object) |
8489eb67 | 896 | { |
2203e1e8 | 897 | if (CHAR_TABLE_P (object) |
e704cb4b | 898 | && EQ (XCHAR_TABLE (object)->purpose, Qsyntax_table)) |
8489eb67 RS |
899 | return Qt; |
900 | return Qnil; | |
901 | } | |
902 | ||
8ea151b2 | 903 | static void |
971de7fb | 904 | check_syntax_table (Lisp_Object obj) |
8489eb67 | 905 | { |
47f5f6ae KS |
906 | CHECK_TYPE (CHAR_TABLE_P (obj) && EQ (XCHAR_TABLE (obj)->purpose, Qsyntax_table), |
907 | Qsyntax_table_p, obj); | |
7d0393cf | 908 | } |
8489eb67 | 909 | |
8489eb67 | 910 | DEFUN ("syntax-table", Fsyntax_table, Ssyntax_table, 0, 0, 0, |
fdb82f93 PJ |
911 | doc: /* Return the current syntax table. |
912 | This is the one specified by the current buffer. */) | |
5842a27b | 913 | (void) |
8489eb67 | 914 | { |
4b4deea2 | 915 | return BVAR (current_buffer, syntax_table); |
8489eb67 RS |
916 | } |
917 | ||
918 | DEFUN ("standard-syntax-table", Fstandard_syntax_table, | |
919 | Sstandard_syntax_table, 0, 0, 0, | |
fdb82f93 PJ |
920 | doc: /* Return the standard syntax table. |
921 | This is the one used for new buffers. */) | |
5842a27b | 922 | (void) |
8489eb67 RS |
923 | { |
924 | return Vstandard_syntax_table; | |
925 | } | |
926 | ||
927 | DEFUN ("copy-syntax-table", Fcopy_syntax_table, Scopy_syntax_table, 0, 1, 0, | |
fdb82f93 PJ |
928 | doc: /* Construct a new syntax table and return it. |
929 | It is a copy of the TABLE, which defaults to the standard syntax table. */) | |
5842a27b | 930 | (Lisp_Object table) |
8489eb67 | 931 | { |
8ea151b2 RS |
932 | Lisp_Object copy; |
933 | ||
265a9e55 | 934 | if (!NILP (table)) |
8ea151b2 RS |
935 | check_syntax_table (table); |
936 | else | |
937 | table = Vstandard_syntax_table; | |
938 | ||
939 | copy = Fcopy_sequence (table); | |
0f867324 RS |
940 | |
941 | /* Only the standard syntax table should have a default element. | |
942 | Other syntax tables should inherit from parents instead. */ | |
742af32f | 943 | set_char_table_defalt (copy, Qnil); |
0f867324 RS |
944 | |
945 | /* Copied syntax tables should all have parents. | |
946 | If we copied one with no parent, such as the standard syntax table, | |
947 | use the standard syntax table as the copy's parent. */ | |
948 | if (NILP (XCHAR_TABLE (copy)->parent)) | |
949 | Fset_char_table_parent (copy, Vstandard_syntax_table); | |
8ea151b2 | 950 | return copy; |
8489eb67 RS |
951 | } |
952 | ||
953 | DEFUN ("set-syntax-table", Fset_syntax_table, Sset_syntax_table, 1, 1, 0, | |
fdb82f93 PJ |
954 | doc: /* Select a new syntax table for the current buffer. |
955 | One argument, a syntax table. */) | |
5842a27b | 956 | (Lisp_Object table) |
8489eb67 | 957 | { |
4e4dfa78 | 958 | int idx; |
8ea151b2 | 959 | check_syntax_table (table); |
39eb03f1 | 960 | bset_syntax_table (current_buffer, table); |
8489eb67 | 961 | /* Indicate that this buffer now has a specified syntax table. */ |
f6cd0527 GM |
962 | idx = PER_BUFFER_VAR_IDX (syntax_table); |
963 | SET_PER_BUFFER_VALUE_P (current_buffer, idx, 1); | |
8489eb67 RS |
964 | return table; |
965 | } | |
966 | \f | |
967 | /* Convert a letter which signifies a syntax code | |
968 | into the code it signifies. | |
195d1361 | 969 | This is used by modify-syntax-entry, and other things. */ |
8489eb67 | 970 | |
1fc71008 | 971 | unsigned char const syntax_spec_code[0400] = |
8489eb67 RS |
972 | { 0377, 0377, 0377, 0377, 0377, 0377, 0377, 0377, |
973 | 0377, 0377, 0377, 0377, 0377, 0377, 0377, 0377, | |
974 | 0377, 0377, 0377, 0377, 0377, 0377, 0377, 0377, | |
975 | 0377, 0377, 0377, 0377, 0377, 0377, 0377, 0377, | |
1fc71008 PE |
976 | Swhitespace, Scomment_fence, Sstring, 0377, Smath, 0377, 0377, Squote, |
977 | Sopen, Sclose, 0377, 0377, 0377, Swhitespace, Spunct, Scharquote, | |
8489eb67 | 978 | 0377, 0377, 0377, 0377, 0377, 0377, 0377, 0377, |
1fc71008 PE |
979 | 0377, 0377, 0377, 0377, Scomment, 0377, Sendcomment, 0377, |
980 | Sinherit, 0377, 0377, 0377, 0377, 0377, 0377, 0377, /* @, A ... */ | |
8489eb67 | 981 | 0377, 0377, 0377, 0377, 0377, 0377, 0377, 0377, |
1fc71008 PE |
982 | 0377, 0377, 0377, 0377, 0377, 0377, 0377, Sword, |
983 | 0377, 0377, 0377, 0377, Sescape, 0377, 0377, Ssymbol, | |
8489eb67 RS |
984 | 0377, 0377, 0377, 0377, 0377, 0377, 0377, 0377, /* `, a, ... */ |
985 | 0377, 0377, 0377, 0377, 0377, 0377, 0377, 0377, | |
1fc71008 PE |
986 | 0377, 0377, 0377, 0377, 0377, 0377, 0377, Sword, |
987 | 0377, 0377, 0377, 0377, Sstring_fence, 0377, 0377, 0377 | |
8489eb67 RS |
988 | }; |
989 | ||
195d1361 | 990 | /* Indexed by syntax code, give the letter that describes it. */ |
8489eb67 | 991 | |
1fc71008 | 992 | char const syntax_code_spec[16] = |
8489eb67 | 993 | { |
195d1361 RS |
994 | ' ', '.', 'w', '_', '(', ')', '\'', '\"', '$', '\\', '/', '<', '>', '@', |
995 | '!', '|' | |
8489eb67 | 996 | }; |
93da5fff KH |
997 | |
998 | /* Indexed by syntax code, give the object (cons of syntax code and | |
999 | nil) to be stored in syntax table. Since these objects can be | |
1000 | shared among syntax tables, we generate them in advance. By | |
1001 | sharing objects, the function `describe-syntax' can give a more | |
1002 | compact listing. */ | |
1003 | static Lisp_Object Vsyntax_code_object; | |
1004 | ||
8489eb67 RS |
1005 | \f |
1006 | DEFUN ("char-syntax", Fchar_syntax, Schar_syntax, 1, 1, 0, | |
fdb82f93 | 1007 | doc: /* Return the syntax code of CHARACTER, described by a character. |
e2b6daf4 CY |
1008 | For example, if CHARACTER is a word constituent, the |
1009 | character `w' (119) is returned. | |
fdb82f93 PJ |
1010 | The characters that correspond to various syntax codes |
1011 | are listed in the documentation of `modify-syntax-entry'. */) | |
5842a27b | 1012 | (Lisp_Object character) |
8489eb67 | 1013 | { |
8ea151b2 | 1014 | int char_int; |
774b9a60 | 1015 | CHECK_CHARACTER (character); |
2203e1e8 | 1016 | char_int = XINT (character); |
d48cd3f4 | 1017 | SETUP_BUFFER_SYNTAX_TABLE (); |
1fc71008 | 1018 | return make_number (syntax_code_spec[SYNTAX (char_int)]); |
beefa22e RS |
1019 | } |
1020 | ||
1021 | DEFUN ("matching-paren", Fmatching_paren, Smatching_paren, 1, 1, 0, | |
fdb82f93 | 1022 | doc: /* Return the matching parenthesis of CHARACTER, or nil if none. */) |
5842a27b | 1023 | (Lisp_Object character) |
beefa22e | 1024 | { |
1fc71008 PE |
1025 | int char_int; |
1026 | enum syntaxcode code; | |
1027 | CHECK_CHARACTER (character); | |
2203e1e8 | 1028 | char_int = XINT (character); |
d48cd3f4 | 1029 | SETUP_BUFFER_SYNTAX_TABLE (); |
8ea151b2 | 1030 | code = SYNTAX (char_int); |
a8bd7cd8 | 1031 | if (code == Sopen || code == Sclose) |
2e34157c | 1032 | return SYNTAX_MATCH (char_int); |
beefa22e | 1033 | return Qnil; |
8489eb67 RS |
1034 | } |
1035 | ||
c65adb44 | 1036 | DEFUN ("string-to-syntax", Fstring_to_syntax, Sstring_to_syntax, 1, 1, 0, |
6dcef6ec CY |
1037 | doc: /* Convert a syntax descriptor STRING into a raw syntax descriptor. |
1038 | STRING should be a string of the form allowed as argument of | |
1039 | `modify-syntax-entry'. The return value is a raw syntax descriptor: a | |
1040 | cons cell \(CODE . MATCHING-CHAR) which can be used, for example, as | |
1041 | the value of a `syntax-table' text property. */) | |
5842a27b | 1042 | (Lisp_Object string) |
8489eb67 | 1043 | { |
1fc71008 | 1044 | const unsigned char *p; |
247e20a8 | 1045 | int val; |
8ea151b2 | 1046 | Lisp_Object match; |
8489eb67 | 1047 | |
b7826503 | 1048 | CHECK_STRING (string); |
8489eb67 | 1049 | |
d5db4077 | 1050 | p = SDATA (string); |
1fc71008 PE |
1051 | val = syntax_spec_code[*p++]; |
1052 | if (val == 0377) | |
b764a653 | 1053 | error ("Invalid syntax description letter: %c", p[-1]); |
8489eb67 | 1054 | |
1fc71008 | 1055 | if (val == Sinherit) |
c65adb44 | 1056 | return Qnil; |
8ea151b2 RS |
1057 | |
1058 | if (*p) | |
d1be9f0f | 1059 | { |
93da5fff | 1060 | int len; |
62a6e103 | 1061 | int character = STRING_CHAR_AND_LENGTH (p, len); |
93da5fff | 1062 | XSETINT (match, character); |
d1be9f0f RS |
1063 | if (XFASTINT (match) == ' ') |
1064 | match = Qnil; | |
93da5fff | 1065 | p += len; |
d1be9f0f RS |
1066 | } |
1067 | else | |
8ea151b2 | 1068 | match = Qnil; |
8489eb67 | 1069 | |
8489eb67 RS |
1070 | while (*p) |
1071 | switch (*p++) | |
1072 | { | |
1073 | case '1': | |
247e20a8 | 1074 | val |= 1 << 16; |
8489eb67 RS |
1075 | break; |
1076 | ||
1077 | case '2': | |
247e20a8 | 1078 | val |= 1 << 17; |
8489eb67 RS |
1079 | break; |
1080 | ||
1081 | case '3': | |
247e20a8 | 1082 | val |= 1 << 18; |
8489eb67 RS |
1083 | break; |
1084 | ||
1085 | case '4': | |
247e20a8 | 1086 | val |= 1 << 19; |
8489eb67 RS |
1087 | break; |
1088 | ||
1089 | case 'p': | |
247e20a8 | 1090 | val |= 1 << 20; |
8489eb67 | 1091 | break; |
e5d4f4dc RS |
1092 | |
1093 | case 'b': | |
247e20a8 | 1094 | val |= 1 << 21; |
e5d4f4dc | 1095 | break; |
95ff8dfc RS |
1096 | |
1097 | case 'n': | |
1098 | val |= 1 << 22; | |
1099 | break; | |
c5683ceb SM |
1100 | |
1101 | case 'c': | |
1102 | val |= 1 << 23; | |
1103 | break; | |
8489eb67 | 1104 | } |
7d0393cf | 1105 | |
77b37c05 | 1106 | if (val < ASIZE (Vsyntax_code_object) && NILP (match)) |
28be1ada | 1107 | return AREF (Vsyntax_code_object, val); |
93da5fff KH |
1108 | else |
1109 | /* Since we can't use a shared object, let's make a new one. */ | |
c65adb44 SM |
1110 | return Fcons (make_number (val), match); |
1111 | } | |
1112 | ||
fdb82f93 | 1113 | /* I really don't know why this is interactive |
2c9e1900 | 1114 | help-form should at least be made useful whilst reading the second arg. */ |
7d0393cf | 1115 | DEFUN ("modify-syntax-entry", Fmodify_syntax_entry, Smodify_syntax_entry, 2, 3, |
c65adb44 | 1116 | "cSet syntax for character: \nsSet syntax for %s to: ", |
a50a10a0 | 1117 | doc: /* Set syntax for character CHAR according to string NEWENTRY. |
d7ee9fab | 1118 | The syntax is changed only for table SYNTAX-TABLE, which defaults to |
fdb82f93 | 1119 | the current buffer's syntax table. |
5c7b02ab | 1120 | CHAR may be a cons (MIN . MAX), in which case, syntaxes of all characters |
fedc6ab5 | 1121 | in the range MIN to MAX are changed. |
fdb82f93 PJ |
1122 | The first character of NEWENTRY should be one of the following: |
1123 | Space or - whitespace syntax. w word constituent. | |
1124 | _ symbol constituent. . punctuation. | |
1125 | ( open-parenthesis. ) close-parenthesis. | |
1126 | " string quote. \\ escape. | |
1127 | $ paired delimiter. ' expression quote or prefix operator. | |
1128 | < comment starter. > comment ender. | |
02676e5d | 1129 | / character-quote. @ inherit from parent table. |
fdb82f93 PJ |
1130 | | generic string fence. ! generic comment fence. |
1131 | ||
1132 | Only single-character comment start and end sequences are represented thus. | |
1133 | Two-character sequences are represented as described below. | |
1134 | The second character of NEWENTRY is the matching parenthesis, | |
1135 | used only if the first character is `(' or `)'. | |
1136 | Any additional characters are flags. | |
1137 | Defined flags are the characters 1, 2, 3, 4, b, p, and n. | |
a50a10a0 PJ |
1138 | 1 means CHAR is the start of a two-char comment start sequence. |
1139 | 2 means CHAR is the second character of such a sequence. | |
1140 | 3 means CHAR is the start of a two-char comment end sequence. | |
1141 | 4 means CHAR is the second character of such a sequence. | |
fdb82f93 | 1142 | |
c5683ceb | 1143 | There can be several orthogonal comment sequences. This is to support |
fdb82f93 PJ |
1144 | language modes such as C++. By default, all comment sequences are of style |
1145 | a, but you can set the comment sequence style to b (on the second character | |
c5683ceb SM |
1146 | of a comment-start, and the first character of a comment-end sequence) and/or |
1147 | c (on any of its chars) using this flag: | |
a50a10a0 | 1148 | b means CHAR is part of comment sequence b. |
c5683ceb | 1149 | c means CHAR is part of comment sequence c. |
a50a10a0 | 1150 | n means CHAR is part of a nestable comment sequence. |
fdb82f93 | 1151 | |
a50a10a0 | 1152 | p means CHAR is a prefix character for `backward-prefix-chars'; |
fdb82f93 | 1153 | such characters are treated as whitespace when they occur |
a50a10a0 | 1154 | between expressions. |
fedc6ab5 | 1155 | usage: (modify-syntax-entry CHAR NEWENTRY &optional SYNTAX-TABLE) */) |
5842a27b | 1156 | (Lisp_Object c, Lisp_Object newentry, Lisp_Object syntax_table) |
c65adb44 | 1157 | { |
5c7b02ab KH |
1158 | if (CONSP (c)) |
1159 | { | |
8f924df7 KH |
1160 | CHECK_CHARACTER_CAR (c); |
1161 | CHECK_CHARACTER_CDR (c); | |
5c7b02ab KH |
1162 | } |
1163 | else | |
1164 | CHECK_CHARACTER (c); | |
c65adb44 SM |
1165 | |
1166 | if (NILP (syntax_table)) | |
4b4deea2 | 1167 | syntax_table = BVAR (current_buffer, syntax_table); |
c65adb44 SM |
1168 | else |
1169 | check_syntax_table (syntax_table); | |
8489eb67 | 1170 | |
5c7b02ab KH |
1171 | newentry = Fstring_to_syntax (newentry); |
1172 | if (CONSP (c)) | |
1173 | SET_RAW_SYNTAX_ENTRY_RANGE (syntax_table, c, newentry); | |
1174 | else | |
1175 | SET_RAW_SYNTAX_ENTRY (syntax_table, XINT (c), newentry); | |
e5b94d44 CY |
1176 | |
1177 | /* We clear the regexp cache, since character classes can now have | |
1178 | different values from those in the compiled regexps.*/ | |
1179 | clear_regexp_cache (); | |
1180 | ||
8489eb67 RS |
1181 | return Qnil; |
1182 | } | |
1183 | \f | |
1184 | /* Dump syntax table to buffer in human-readable format */ | |
1185 | ||
62abe9cb SM |
1186 | DEFUN ("internal-describe-syntax-value", Finternal_describe_syntax_value, |
1187 | Sinternal_describe_syntax_value, 1, 1, 0, | |
1188 | doc: /* Insert a description of the internal syntax description SYNTAX at point. */) | |
5842a27b | 1189 | (Lisp_Object syntax) |
8489eb67 | 1190 | { |
1fc71008 PE |
1191 | int code, syntax_code; |
1192 | bool start1, start2, end1, end2, prefix, comstyleb, comstylec, comnested; | |
8489eb67 | 1193 | char str[2]; |
62abe9cb | 1194 | Lisp_Object first, match_lisp, value = syntax; |
8489eb67 | 1195 | |
8ea151b2 RS |
1196 | if (NILP (value)) |
1197 | { | |
62abe9cb SM |
1198 | insert_string ("default"); |
1199 | return syntax; | |
8ea151b2 RS |
1200 | } |
1201 | ||
908b7fea KH |
1202 | if (CHAR_TABLE_P (value)) |
1203 | { | |
62abe9cb SM |
1204 | insert_string ("deeper char-table ..."); |
1205 | return syntax; | |
908b7fea KH |
1206 | } |
1207 | ||
8ea151b2 RS |
1208 | if (!CONSP (value)) |
1209 | { | |
62abe9cb SM |
1210 | insert_string ("invalid"); |
1211 | return syntax; | |
8ea151b2 RS |
1212 | } |
1213 | ||
c1d497be KR |
1214 | first = XCAR (value); |
1215 | match_lisp = XCDR (value); | |
8ea151b2 | 1216 | |
d311d28c | 1217 | if (!INTEGERP (first) || !(NILP (match_lisp) || CHARACTERP (match_lisp))) |
8489eb67 | 1218 | { |
62abe9cb SM |
1219 | insert_string ("invalid"); |
1220 | return syntax; | |
8489eb67 RS |
1221 | } |
1222 | ||
d311d28c | 1223 | syntax_code = XINT (first) & INT_MAX; |
1fc71008 | 1224 | code = syntax_code & 0377; |
c5683ceb SM |
1225 | start1 = SYNTAX_FLAGS_COMSTART_FIRST (syntax_code); |
1226 | start2 = SYNTAX_FLAGS_COMSTART_SECOND (syntax_code);; | |
1227 | end1 = SYNTAX_FLAGS_COMEND_FIRST (syntax_code); | |
1228 | end2 = SYNTAX_FLAGS_COMEND_SECOND (syntax_code); | |
1229 | prefix = SYNTAX_FLAGS_PREFIX (syntax_code); | |
1230 | comstyleb = SYNTAX_FLAGS_COMMENT_STYLEB (syntax_code); | |
1231 | comstylec = SYNTAX_FLAGS_COMMENT_STYLEC (syntax_code); | |
1232 | comnested = SYNTAX_FLAGS_COMMENT_NESTED (syntax_code); | |
8489eb67 | 1233 | |
1fc71008 | 1234 | if (Smax <= code) |
8489eb67 RS |
1235 | { |
1236 | insert_string ("invalid"); | |
62abe9cb | 1237 | return syntax; |
8489eb67 | 1238 | } |
8489eb67 | 1239 | |
1fc71008 | 1240 | str[0] = syntax_code_spec[code], str[1] = 0; |
8489eb67 RS |
1241 | insert (str, 1); |
1242 | ||
93da5fff KH |
1243 | if (NILP (match_lisp)) |
1244 | insert (" ", 1); | |
1245 | else | |
1246 | insert_char (XINT (match_lisp)); | |
8489eb67 | 1247 | |
8489eb67 RS |
1248 | if (start1) |
1249 | insert ("1", 1); | |
1250 | if (start2) | |
1251 | insert ("2", 1); | |
1252 | ||
1253 | if (end1) | |
1254 | insert ("3", 1); | |
1255 | if (end2) | |
1256 | insert ("4", 1); | |
1257 | ||
1258 | if (prefix) | |
1259 | insert ("p", 1); | |
c5683ceb | 1260 | if (comstyleb) |
e5d4f4dc | 1261 | insert ("b", 1); |
c5683ceb SM |
1262 | if (comstylec) |
1263 | insert ("c", 1); | |
e6365855 SM |
1264 | if (comnested) |
1265 | insert ("n", 1); | |
8489eb67 RS |
1266 | |
1267 | insert_string ("\twhich means: "); | |
1268 | ||
7393bcbb | 1269 | switch (code) |
8489eb67 RS |
1270 | { |
1271 | case Swhitespace: | |
1272 | insert_string ("whitespace"); break; | |
1273 | case Spunct: | |
1274 | insert_string ("punctuation"); break; | |
1275 | case Sword: | |
1276 | insert_string ("word"); break; | |
1277 | case Ssymbol: | |
1278 | insert_string ("symbol"); break; | |
1279 | case Sopen: | |
1280 | insert_string ("open"); break; | |
1281 | case Sclose: | |
1282 | insert_string ("close"); break; | |
1283 | case Squote: | |
d671382e | 1284 | insert_string ("prefix"); break; |
8489eb67 RS |
1285 | case Sstring: |
1286 | insert_string ("string"); break; | |
1287 | case Smath: | |
1288 | insert_string ("math"); break; | |
1289 | case Sescape: | |
1290 | insert_string ("escape"); break; | |
1291 | case Scharquote: | |
1292 | insert_string ("charquote"); break; | |
1293 | case Scomment: | |
1294 | insert_string ("comment"); break; | |
1295 | case Sendcomment: | |
1296 | insert_string ("endcomment"); break; | |
d671382e SM |
1297 | case Sinherit: |
1298 | insert_string ("inherit"); break; | |
1299 | case Scomment_fence: | |
1300 | insert_string ("comment fence"); break; | |
1301 | case Sstring_fence: | |
1302 | insert_string ("string fence"); break; | |
8489eb67 RS |
1303 | default: |
1304 | insert_string ("invalid"); | |
62abe9cb | 1305 | return syntax; |
8489eb67 RS |
1306 | } |
1307 | ||
8ea151b2 | 1308 | if (!NILP (match_lisp)) |
8489eb67 RS |
1309 | { |
1310 | insert_string (", matches "); | |
8ea151b2 | 1311 | insert_char (XINT (match_lisp)); |
8489eb67 RS |
1312 | } |
1313 | ||
1314 | if (start1) | |
1315 | insert_string (",\n\t is the first character of a comment-start sequence"); | |
1316 | if (start2) | |
1317 | insert_string (",\n\t is the second character of a comment-start sequence"); | |
1318 | ||
1319 | if (end1) | |
1320 | insert_string (",\n\t is the first character of a comment-end sequence"); | |
1321 | if (end2) | |
1322 | insert_string (",\n\t is the second character of a comment-end sequence"); | |
c5683ceb | 1323 | if (comstyleb) |
e5d4f4dc | 1324 | insert_string (" (comment style b)"); |
c5683ceb SM |
1325 | if (comstylec) |
1326 | insert_string (" (comment style c)"); | |
e6365855 SM |
1327 | if (comnested) |
1328 | insert_string (" (nestable)"); | |
e5d4f4dc | 1329 | |
8489eb67 RS |
1330 | if (prefix) |
1331 | insert_string (",\n\t is a prefix character for `backward-prefix-chars'"); | |
1332 | ||
62abe9cb SM |
1333 | return syntax; |
1334 | } | |
8489eb67 RS |
1335 | \f |
1336 | /* Return the position across COUNT words from FROM. | |
1337 | If that many words cannot be found before the end of the buffer, return 0. | |
1338 | COUNT negative means scan backward and stop at word beginning. */ | |
1339 | ||
d311d28c PE |
1340 | ptrdiff_t |
1341 | scan_words (register ptrdiff_t from, register EMACS_INT count) | |
8489eb67 | 1342 | { |
d311d28c PE |
1343 | register ptrdiff_t beg = BEGV; |
1344 | register ptrdiff_t end = ZV; | |
1345 | register ptrdiff_t from_byte = CHAR_TO_BYTE (from); | |
93da5fff KH |
1346 | register enum syntaxcode code; |
1347 | int ch0, ch1; | |
fca8fe46 | 1348 | Lisp_Object func, pos; |
8489eb67 RS |
1349 | |
1350 | immediate_quit = 1; | |
1351 | QUIT; | |
1352 | ||
195d1361 RS |
1353 | SETUP_SYNTAX_TABLE (from, count); |
1354 | ||
8489eb67 RS |
1355 | while (count > 0) |
1356 | { | |
1357 | while (1) | |
1358 | { | |
1359 | if (from == end) | |
1360 | { | |
1361 | immediate_quit = 0; | |
1362 | return 0; | |
1363 | } | |
195d1361 | 1364 | UPDATE_SYNTAX_TABLE_FORWARD (from); |
b7dbcc19 | 1365 | ch0 = FETCH_CHAR_AS_MULTIBYTE (from_byte); |
93da5fff | 1366 | code = SYNTAX (ch0); |
6a140a74 | 1367 | INC_BOTH (from, from_byte); |
8489eb67 RS |
1368 | if (words_include_escapes |
1369 | && (code == Sescape || code == Scharquote)) | |
1370 | break; | |
1371 | if (code == Sword) | |
1372 | break; | |
8489eb67 | 1373 | } |
93da5fff KH |
1374 | /* Now CH0 is a character which begins a word and FROM is the |
1375 | position of the next character. */ | |
8f924df7 | 1376 | func = CHAR_TABLE_REF (Vfind_word_boundary_function_table, ch0); |
869bb237 | 1377 | if (! NILP (Ffboundp (func))) |
8489eb67 | 1378 | { |
869bb237 | 1379 | pos = call2 (func, make_number (from - 1), make_number (end)); |
17f85215 | 1380 | if (INTEGERP (pos) && from < XINT (pos) && XINT (pos) <= ZV) |
8f924df7 KH |
1381 | { |
1382 | from = XINT (pos); | |
1383 | from_byte = CHAR_TO_BYTE (from); | |
1384 | } | |
8489eb67 | 1385 | } |
869bb237 | 1386 | else |
8cb8232a | 1387 | { |
8cb8232a KH |
1388 | while (1) |
1389 | { | |
1390 | if (from == end) break; | |
1391 | UPDATE_SYNTAX_TABLE_FORWARD (from); | |
b7dbcc19 | 1392 | ch1 = FETCH_CHAR_AS_MULTIBYTE (from_byte); |
8cb8232a KH |
1393 | code = SYNTAX (ch1); |
1394 | if ((code != Sword | |
1395 | && (! words_include_escapes | |
1396 | || (code != Sescape && code != Scharquote))) | |
2b70a11c | 1397 | || word_boundary_p (ch0, ch1)) |
869bb237 | 1398 | break; |
8cb8232a KH |
1399 | INC_BOTH (from, from_byte); |
1400 | ch0 = ch1; | |
1401 | } | |
8489eb67 RS |
1402 | } |
1403 | count--; | |
1404 | } | |
1405 | while (count < 0) | |
1406 | { | |
1407 | while (1) | |
1408 | { | |
1409 | if (from == beg) | |
1410 | { | |
1411 | immediate_quit = 0; | |
1412 | return 0; | |
1413 | } | |
6a140a74 | 1414 | DEC_BOTH (from, from_byte); |
195d1361 | 1415 | UPDATE_SYNTAX_TABLE_BACKWARD (from); |
b7dbcc19 | 1416 | ch1 = FETCH_CHAR_AS_MULTIBYTE (from_byte); |
93da5fff | 1417 | code = SYNTAX (ch1); |
8489eb67 RS |
1418 | if (words_include_escapes |
1419 | && (code == Sescape || code == Scharquote)) | |
1420 | break; | |
1421 | if (code == Sword) | |
1422 | break; | |
8489eb67 | 1423 | } |
93da5fff KH |
1424 | /* Now CH1 is a character which ends a word and FROM is the |
1425 | position of it. */ | |
8f924df7 | 1426 | func = CHAR_TABLE_REF (Vfind_word_boundary_function_table, ch1); |
869bb237 | 1427 | if (! NILP (Ffboundp (func))) |
8f924df7 | 1428 | { |
869bb237 | 1429 | pos = call2 (func, make_number (from), make_number (beg)); |
17f85215 | 1430 | if (INTEGERP (pos) && BEGV <= XINT (pos) && XINT (pos) < from) |
8f924df7 KH |
1431 | { |
1432 | from = XINT (pos); | |
1433 | from_byte = CHAR_TO_BYTE (from); | |
1434 | } | |
8489eb67 | 1435 | } |
869bb237 | 1436 | else |
8489eb67 | 1437 | { |
8cb8232a KH |
1438 | while (1) |
1439 | { | |
8cb8232a KH |
1440 | if (from == beg) |
1441 | break; | |
262be72a | 1442 | DEC_BOTH (from, from_byte); |
8cb8232a | 1443 | UPDATE_SYNTAX_TABLE_BACKWARD (from); |
262be72a | 1444 | ch0 = FETCH_CHAR_AS_MULTIBYTE (from_byte); |
8cb8232a KH |
1445 | code = SYNTAX (ch0); |
1446 | if ((code != Sword | |
1447 | && (! words_include_escapes | |
1448 | || (code != Sescape && code != Scharquote))) | |
2b70a11c | 1449 | || word_boundary_p (ch0, ch1)) |
262be72a MB |
1450 | { |
1451 | INC_BOTH (from, from_byte); | |
1452 | break; | |
1453 | } | |
8cb8232a KH |
1454 | ch1 = ch0; |
1455 | } | |
8489eb67 RS |
1456 | } |
1457 | count++; | |
1458 | } | |
1459 | ||
1460 | immediate_quit = 0; | |
1461 | ||
1462 | return from; | |
1463 | } | |
1464 | ||
a7ca3326 | 1465 | DEFUN ("forward-word", Fforward_word, Sforward_word, 0, 1, "^p", |
fdb82f93 | 1466 | doc: /* Move point forward ARG words (backward if ARG is negative). |
f73f4ce6 | 1467 | If ARG is omitted or nil, move point forward one word. |
fdb82f93 PJ |
1468 | Normally returns t. |
1469 | If an edge of the buffer or a field boundary is reached, point is left there | |
1470 | and the function returns nil. Field boundaries are not noticed if | |
1471 | `inhibit-field-text-motion' is non-nil. */) | |
5842a27b | 1472 | (Lisp_Object arg) |
8489eb67 | 1473 | { |
c096ae4d | 1474 | Lisp_Object tmp; |
d311d28c | 1475 | ptrdiff_t orig_val, val; |
8489eb67 | 1476 | |
839966f3 KH |
1477 | if (NILP (arg)) |
1478 | XSETFASTINT (arg, 1); | |
1479 | else | |
1480 | CHECK_NUMBER (arg); | |
1481 | ||
1482 | val = orig_val = scan_words (PT, XINT (arg)); | |
2c6ea900 | 1483 | if (! orig_val) |
839966f3 | 1484 | val = XINT (arg) > 0 ? ZV : BEGV; |
5878ee6f | 1485 | |
b8855607 | 1486 | /* Avoid jumping out of an input field. */ |
c096ae4d SM |
1487 | tmp = Fconstrain_to_field (make_number (val), make_number (PT), |
1488 | Qt, Qnil, Qnil); | |
1489 | val = XFASTINT (tmp); | |
7d0393cf | 1490 | |
8489eb67 | 1491 | SET_PT (val); |
e6d8341f | 1492 | return val == orig_val ? Qt : Qnil; |
8489eb67 RS |
1493 | } |
1494 | \f | |
a7ca3326 | 1495 | DEFUN ("skip-chars-forward", Fskip_chars_forward, Sskip_chars_forward, 1, 2, 0, |
fdb82f93 PJ |
1496 | doc: /* Move point forward, stopping before a char not in STRING, or at pos LIM. |
1497 | STRING is like the inside of a `[...]' in a regular expression | |
1498 | except that `]' is never special and `\\' quotes `^', `-' or `\\' | |
7087d5e9 | 1499 | (but not at the end of a range; quoting is never needed there). |
fdb82f93 PJ |
1500 | Thus, with arg "a-zA-Z", this skips letters stopping before first nonletter. |
1501 | With arg "^a-zA-Z", skips nonletters stopping before first letter. | |
a1bc88d4 RS |
1502 | Char classes, e.g. `[:alpha:]', are supported. |
1503 | ||
1504 | Returns the distance traveled, either zero or positive. */) | |
5842a27b | 1505 | (Lisp_Object string, Lisp_Object lim) |
195d1361 | 1506 | { |
327719ee | 1507 | return skip_chars (1, string, lim, 1); |
195d1361 RS |
1508 | } |
1509 | ||
a7ca3326 | 1510 | DEFUN ("skip-chars-backward", Fskip_chars_backward, Sskip_chars_backward, 1, 2, 0, |
fdb82f93 PJ |
1511 | doc: /* Move point backward, stopping after a char not in STRING, or at pos LIM. |
1512 | See `skip-chars-forward' for details. | |
1513 | Returns the distance traveled, either zero or negative. */) | |
5842a27b | 1514 | (Lisp_Object string, Lisp_Object lim) |
195d1361 | 1515 | { |
327719ee | 1516 | return skip_chars (0, string, lim, 1); |
195d1361 RS |
1517 | } |
1518 | ||
1519 | DEFUN ("skip-syntax-forward", Fskip_syntax_forward, Sskip_syntax_forward, 1, 2, 0, | |
fdb82f93 PJ |
1520 | doc: /* Move point forward across chars in specified syntax classes. |
1521 | SYNTAX is a string of syntax code characters. | |
1522 | Stop before a char whose syntax is not in SYNTAX, or at position LIM. | |
1523 | If SYNTAX starts with ^, skip characters whose syntax is NOT in SYNTAX. | |
1524 | This function returns the distance traveled, either zero or positive. */) | |
5842a27b | 1525 | (Lisp_Object syntax, Lisp_Object lim) |
195d1361 | 1526 | { |
b7dbcc19 | 1527 | return skip_syntaxes (1, syntax, lim); |
195d1361 RS |
1528 | } |
1529 | ||
1530 | DEFUN ("skip-syntax-backward", Fskip_syntax_backward, Sskip_syntax_backward, 1, 2, 0, | |
fdb82f93 PJ |
1531 | doc: /* Move point backward across chars in specified syntax classes. |
1532 | SYNTAX is a string of syntax code characters. | |
1533 | Stop on reaching a char whose syntax is not in SYNTAX, or at position LIM. | |
1534 | If SYNTAX starts with ^, skip characters whose syntax is NOT in SYNTAX. | |
1535 | This function returns the distance traveled, either zero or negative. */) | |
5842a27b | 1536 | (Lisp_Object syntax, Lisp_Object lim) |
195d1361 | 1537 | { |
b7dbcc19 | 1538 | return skip_syntaxes (0, syntax, lim); |
195d1361 RS |
1539 | } |
1540 | ||
6a140a74 | 1541 | static Lisp_Object |
1fc71008 PE |
1542 | skip_chars (bool forwardp, Lisp_Object string, Lisp_Object lim, |
1543 | bool handle_iso_classes) | |
195d1361 | 1544 | { |
1fc71008 PE |
1545 | int c; |
1546 | char fastmap[0400]; | |
b7dbcc19 | 1547 | /* Store the ranges of non-ASCII characters. */ |
4f63c6bb | 1548 | int *char_ranges IF_LINT (= NULL); |
7e68b0ea | 1549 | int n_char_ranges = 0; |
1fc71008 PE |
1550 | bool negate = 0; |
1551 | ptrdiff_t i, i_byte; | |
1552 | /* True if the current buffer is multibyte and the region contains | |
1553 | non-ASCII chars. */ | |
1554 | bool multibyte; | |
1555 | /* True if STRING is multibyte and it contains non-ASCII chars. */ | |
1556 | bool string_multibyte; | |
d311d28c | 1557 | ptrdiff_t size_byte; |
2e567bd3 | 1558 | const unsigned char *str; |
82d497fc | 1559 | int len; |
a1bc88d4 | 1560 | Lisp_Object iso_classes; |
195d1361 | 1561 | |
b7826503 | 1562 | CHECK_STRING (string); |
a1bc88d4 | 1563 | iso_classes = Qnil; |
82d497fc | 1564 | |
195d1361 RS |
1565 | if (NILP (lim)) |
1566 | XSETINT (lim, forwardp ? ZV : BEGV); | |
1567 | else | |
b7826503 | 1568 | CHECK_NUMBER_COERCE_MARKER (lim); |
195d1361 RS |
1569 | |
1570 | /* In any case, don't allow scan outside bounds of buffer. */ | |
195d1361 RS |
1571 | if (XINT (lim) > ZV) |
1572 | XSETFASTINT (lim, ZV); | |
1573 | if (XINT (lim) < BEGV) | |
1574 | XSETFASTINT (lim, BEGV); | |
1575 | ||
4b4deea2 | 1576 | multibyte = (!NILP (BVAR (current_buffer, enable_multibyte_characters)) |
92eaa22e | 1577 | && (XINT (lim) - PT != CHAR_TO_BYTE (XINT (lim)) - PT_BYTE)); |
8f924df7 | 1578 | string_multibyte = SBYTES (string) > SCHARS (string); |
b7dbcc19 | 1579 | |
72af86bd | 1580 | memset (fastmap, 0, sizeof fastmap); |
195d1361 | 1581 | |
8f924df7 KH |
1582 | str = SDATA (string); |
1583 | size_byte = SBYTES (string); | |
4101e6fe | 1584 | |
82d497fc | 1585 | i_byte = 0; |
13090112 | 1586 | if (i_byte < size_byte |
d5db4077 | 1587 | && SREF (string, 0) == '^') |
195d1361 | 1588 | { |
82d497fc | 1589 | negate = 1; i_byte++; |
195d1361 RS |
1590 | } |
1591 | ||
1592 | /* Find the characters specified and set their elements of fastmap. | |
b7dbcc19 | 1593 | Handle backslashes and ranges specially. |
195d1361 | 1594 | |
b7dbcc19 KH |
1595 | If STRING contains non-ASCII characters, setup char_ranges for |
1596 | them and use fastmap only for their leading codes. */ | |
4101e6fe | 1597 | |
b7dbcc19 | 1598 | if (! string_multibyte) |
195d1361 | 1599 | { |
1fc71008 | 1600 | bool string_has_eight_bit = 0; |
4101e6fe | 1601 | |
b7dbcc19 KH |
1602 | /* At first setup fastmap. */ |
1603 | while (i_byte < size_byte) | |
1604 | { | |
1605 | c = str[i_byte++]; | |
1606 | ||
a1bc88d4 RS |
1607 | if (handle_iso_classes && c == '[' |
1608 | && i_byte < size_byte | |
327719ee | 1609 | && str[i_byte] == ':') |
a1bc88d4 RS |
1610 | { |
1611 | const unsigned char *class_beg = str + i_byte + 1; | |
1612 | const unsigned char *class_end = class_beg; | |
b3bda4fd | 1613 | const unsigned char *class_limit = str + size_byte - 2; |
db9cd97a | 1614 | /* Leave room for the null. */ |
a1bc88d4 RS |
1615 | unsigned char class_name[CHAR_CLASS_MAX_LENGTH + 1]; |
1616 | re_wctype_t cc; | |
1617 | ||
1618 | if (class_limit - class_beg > CHAR_CLASS_MAX_LENGTH) | |
1619 | class_limit = class_beg + CHAR_CLASS_MAX_LENGTH; | |
1620 | ||
b3bda4fd KS |
1621 | while (class_end < class_limit |
1622 | && *class_end >= 'a' && *class_end <= 'z') | |
a1bc88d4 RS |
1623 | class_end++; |
1624 | ||
b3bda4fd KS |
1625 | if (class_end == class_beg |
1626 | || *class_end != ':' || class_end[1] != ']') | |
1627 | goto not_a_class_name; | |
a1bc88d4 | 1628 | |
72af86bd | 1629 | memcpy (class_name, class_beg, class_end - class_beg); |
a1bc88d4 RS |
1630 | class_name[class_end - class_beg] = 0; |
1631 | ||
1632 | cc = re_wctype (class_name); | |
1633 | if (cc == 0) | |
1634 | error ("Invalid ISO C character class"); | |
1635 | ||
1636 | iso_classes = Fcons (make_number (cc), iso_classes); | |
1637 | ||
1638 | i_byte = class_end + 2 - str; | |
1639 | continue; | |
1640 | } | |
1641 | ||
b3bda4fd | 1642 | not_a_class_name: |
b7dbcc19 KH |
1643 | if (c == '\\') |
1644 | { | |
13090112 | 1645 | if (i_byte == size_byte) |
4101e6fe RS |
1646 | break; |
1647 | ||
5c7b02ab | 1648 | c = str[i_byte++]; |
195d1361 | 1649 | } |
839966f3 KH |
1650 | /* Treat `-' as range character only if another character |
1651 | follows. */ | |
1652 | if (i_byte + 1 < size_byte | |
82d497fc | 1653 | && str[i_byte] == '-') |
195d1361 | 1654 | { |
1fc71008 | 1655 | int c2; |
4101e6fe RS |
1656 | |
1657 | /* Skip over the dash. */ | |
82d497fc | 1658 | i_byte++; |
4101e6fe | 1659 | |
4101e6fe | 1660 | /* Get the end of the range. */ |
5c7b02ab | 1661 | c2 = str[i_byte++]; |
b7dbcc19 KH |
1662 | if (c2 == '\\' |
1663 | && i_byte < size_byte) | |
1664 | c2 = str[i_byte++]; | |
1665 | ||
8f924df7 KH |
1666 | if (c <= c2) |
1667 | { | |
1fc71008 | 1668 | int lim2 = c2 + 1; |
1e69125e | 1669 | while (c < lim2) |
8f924df7 KH |
1670 | fastmap[c++] = 1; |
1671 | if (! ASCII_CHAR_P (c2)) | |
1672 | string_has_eight_bit = 1; | |
1673 | } | |
195d1361 RS |
1674 | } |
1675 | else | |
b7dbcc19 KH |
1676 | { |
1677 | fastmap[c] = 1; | |
1678 | if (! ASCII_CHAR_P (c)) | |
1679 | string_has_eight_bit = 1; | |
1680 | } | |
1681 | } | |
1682 | ||
1683 | /* If the current range is multibyte and STRING contains | |
1684 | eight-bit chars, arrange fastmap and setup char_ranges for | |
1685 | the corresponding multibyte chars. */ | |
1686 | if (multibyte && string_has_eight_bit) | |
1687 | { | |
178ba3e6 PE |
1688 | char *p1; |
1689 | char himap[0200 + 1]; | |
1690 | memcpy (himap, fastmap + 0200, 0200); | |
1691 | himap[0200] = 0; | |
72af86bd | 1692 | memset (fastmap + 0200, 0, 0200); |
38182d90 | 1693 | char_ranges = alloca (sizeof *char_ranges * 128 * 2); |
178ba3e6 PE |
1694 | i = 0; |
1695 | ||
1696 | while ((p1 = memchr (himap + i, 1, 0200 - i))) | |
b7dbcc19 | 1697 | { |
178ba3e6 PE |
1698 | /* Deduce the next range C..C2 from the next clump of 1s |
1699 | in HIMAP starting with &HIMAP[I]. HIMAP is the high | |
1700 | order half of the old FASTMAP. */ | |
1701 | int c2, leading_code; | |
1702 | i = p1 - himap; | |
1703 | c = BYTE8_TO_CHAR (i + 0200); | |
1704 | i += strlen (p1); | |
1705 | c2 = BYTE8_TO_CHAR (i + 0200 - 1); | |
1706 | ||
1707 | char_ranges[n_char_ranges++] = c; | |
1708 | char_ranges[n_char_ranges++] = c2; | |
1709 | leading_code = CHAR_LEADING_CODE (c); | |
1710 | memset (fastmap + leading_code, 1, | |
1711 | CHAR_LEADING_CODE (c2) - leading_code + 1); | |
b7dbcc19 | 1712 | } |
195d1361 RS |
1713 | } |
1714 | } | |
f003ca54 | 1715 | else /* STRING is multibyte */ |
b7dbcc19 | 1716 | { |
38182d90 | 1717 | char_ranges = alloca (sizeof *char_ranges * SCHARS (string) * 2); |
f003ca54 | 1718 | |
b7dbcc19 | 1719 | while (i_byte < size_byte) |
195d1361 | 1720 | { |
1fc71008 | 1721 | int leading_code = str[i_byte]; |
62a6e103 | 1722 | c = STRING_CHAR_AND_LENGTH (str + i_byte, len); |
b7dbcc19 | 1723 | i_byte += len; |
5c7b02ab | 1724 | |
327719ee MB |
1725 | if (handle_iso_classes && c == '[' |
1726 | && i_byte < size_byte | |
62a6e103 | 1727 | && STRING_CHAR (str + i_byte) == ':') |
327719ee MB |
1728 | { |
1729 | const unsigned char *class_beg = str + i_byte + 1; | |
1730 | const unsigned char *class_end = class_beg; | |
1731 | const unsigned char *class_limit = str + size_byte - 2; | |
1732 | /* Leave room for the null. */ | |
1733 | unsigned char class_name[CHAR_CLASS_MAX_LENGTH + 1]; | |
1734 | re_wctype_t cc; | |
1735 | ||
1736 | if (class_limit - class_beg > CHAR_CLASS_MAX_LENGTH) | |
1737 | class_limit = class_beg + CHAR_CLASS_MAX_LENGTH; | |
1738 | ||
1739 | while (class_end < class_limit | |
1740 | && *class_end >= 'a' && *class_end <= 'z') | |
1741 | class_end++; | |
1742 | ||
1743 | if (class_end == class_beg | |
1744 | || *class_end != ':' || class_end[1] != ']') | |
1745 | goto not_a_class_name_multibyte; | |
1746 | ||
72af86bd | 1747 | memcpy (class_name, class_beg, class_end - class_beg); |
327719ee MB |
1748 | class_name[class_end - class_beg] = 0; |
1749 | ||
1750 | cc = re_wctype (class_name); | |
1751 | if (cc == 0) | |
1752 | error ("Invalid ISO C character class"); | |
1753 | ||
1754 | iso_classes = Fcons (make_number (cc), iso_classes); | |
1755 | ||
1756 | i_byte = class_end + 2 - str; | |
1757 | continue; | |
1758 | } | |
1759 | ||
1760 | not_a_class_name_multibyte: | |
195d1361 RS |
1761 | if (c == '\\') |
1762 | { | |
13090112 | 1763 | if (i_byte == size_byte) |
4101e6fe RS |
1764 | break; |
1765 | ||
b7dbcc19 | 1766 | leading_code = str[i_byte]; |
62a6e103 | 1767 | c = STRING_CHAR_AND_LENGTH (str + i_byte, len); |
82d497fc | 1768 | i_byte += len; |
195d1361 | 1769 | } |
839966f3 KH |
1770 | /* Treat `-' as range character only if another character |
1771 | follows. */ | |
1772 | if (i_byte + 1 < size_byte | |
82d497fc | 1773 | && str[i_byte] == '-') |
195d1361 | 1774 | { |
1fc71008 | 1775 | int c2, leading_code2; |
4101e6fe RS |
1776 | |
1777 | /* Skip over the dash. */ | |
82d497fc | 1778 | i_byte++; |
4101e6fe | 1779 | |
4101e6fe | 1780 | /* Get the end of the range. */ |
b7dbcc19 | 1781 | leading_code2 = str[i_byte]; |
62a6e103 | 1782 | c2 = STRING_CHAR_AND_LENGTH (str + i_byte, len); |
82d497fc | 1783 | i_byte += len; |
7e68b0ea | 1784 | |
b7dbcc19 KH |
1785 | if (c2 == '\\' |
1786 | && i_byte < size_byte) | |
1787 | { | |
1788 | leading_code2 = str[i_byte]; | |
1fc71008 | 1789 | c2 = STRING_CHAR_AND_LENGTH (str + i_byte, len); |
b7dbcc19 KH |
1790 | i_byte += len; |
1791 | } | |
1792 | ||
f003ca54 KH |
1793 | if (c > c2) |
1794 | continue; | |
b7dbcc19 | 1795 | if (ASCII_CHAR_P (c)) |
06274af5 | 1796 | { |
5c7b02ab KH |
1797 | while (c <= c2 && c < 0x80) |
1798 | fastmap[c++] = 1; | |
b7dbcc19 KH |
1799 | leading_code = CHAR_LEADING_CODE (c); |
1800 | } | |
1801 | if (! ASCII_CHAR_P (c)) | |
1802 | { | |
1fc71008 | 1803 | int lim2 = leading_code2 + 1; |
1e69125e | 1804 | while (leading_code < lim2) |
b7dbcc19 KH |
1805 | fastmap[leading_code++] = 1; |
1806 | if (c <= c2) | |
e39091b6 | 1807 | { |
b7dbcc19 | 1808 | char_ranges[n_char_ranges++] = c; |
e39091b6 | 1809 | char_ranges[n_char_ranges++] = c2; |
06274af5 KH |
1810 | } |
1811 | } | |
195d1361 RS |
1812 | } |
1813 | else | |
7e68b0ea | 1814 | { |
b7dbcc19 | 1815 | if (ASCII_CHAR_P (c)) |
e39091b6 KH |
1816 | fastmap[c] = 1; |
1817 | else | |
7e68b0ea | 1818 | { |
b7dbcc19 | 1819 | fastmap[leading_code] = 1; |
4101e6fe RS |
1820 | char_ranges[n_char_ranges++] = c; |
1821 | char_ranges[n_char_ranges++] = c; | |
7e68b0ea RS |
1822 | } |
1823 | } | |
195d1361 | 1824 | } |
b7dbcc19 KH |
1825 | |
1826 | /* If the current range is unibyte and STRING contains non-ASCII | |
1827 | chars, arrange fastmap for the corresponding unibyte | |
1828 | chars. */ | |
1829 | ||
1830 | if (! multibyte && n_char_ranges > 0) | |
1831 | { | |
72af86bd | 1832 | memset (fastmap + 0200, 0, 0200); |
b7dbcc19 KH |
1833 | for (i = 0; i < n_char_ranges; i += 2) |
1834 | { | |
1835 | int c1 = char_ranges[i]; | |
1fc71008 | 1836 | int lim2 = char_ranges[i + 1] + 1; |
b7dbcc19 | 1837 | |
1e69125e | 1838 | for (; c1 < lim2; c1++) |
2afc21f5 SM |
1839 | { |
1840 | int b = CHAR_TO_BYTE_SAFE (c1); | |
1841 | if (b >= 0) | |
1842 | fastmap[b] = 1; | |
1843 | } | |
b7dbcc19 KH |
1844 | } |
1845 | } | |
195d1361 RS |
1846 | } |
1847 | ||
9690d026 | 1848 | /* If ^ was the first character, complement the fastmap. */ |
195d1361 | 1849 | if (negate) |
b7dbcc19 KH |
1850 | { |
1851 | if (! multibyte) | |
1852 | for (i = 0; i < sizeof fastmap; i++) | |
1853 | fastmap[i] ^= 1; | |
1854 | else | |
1855 | { | |
1856 | for (i = 0; i < 0200; i++) | |
1857 | fastmap[i] ^= 1; | |
1858 | /* All non-ASCII chars possibly match. */ | |
1859 | for (; i < sizeof fastmap; i++) | |
1860 | fastmap[i] = 1; | |
1861 | } | |
1862 | } | |
195d1361 RS |
1863 | |
1864 | { | |
d311d28c PE |
1865 | ptrdiff_t start_point = PT; |
1866 | ptrdiff_t pos = PT; | |
1867 | ptrdiff_t pos_byte = PT_BYTE; | |
9af7511a KH |
1868 | unsigned char *p = PT_ADDR, *endp, *stop; |
1869 | ||
1870 | if (forwardp) | |
1871 | { | |
1872 | endp = (XINT (lim) == GPT) ? GPT_ADDR : CHAR_POS_ADDR (XINT (lim)); | |
d7ee9fab | 1873 | stop = (pos < GPT && GPT < XINT (lim)) ? GPT_ADDR : endp; |
9af7511a KH |
1874 | } |
1875 | else | |
1876 | { | |
1877 | endp = CHAR_POS_ADDR (XINT (lim)); | |
d7ee9fab | 1878 | stop = (pos >= GPT && GPT > XINT (lim)) ? GAP_END_ADDR : endp; |
9af7511a | 1879 | } |
195d1361 RS |
1880 | |
1881 | immediate_quit = 1; | |
45b683a1 | 1882 | /* This code may look up syntax tables using functions that rely on the |
618db430 SM |
1883 | gl_state object. To make sure this object is not out of date, |
1884 | let's initialize it manually. | |
1885 | We ignore syntax-table text-properties for now, since that's | |
1886 | what we've done in the past. */ | |
d48cd3f4 | 1887 | SETUP_BUFFER_SYNTAX_TABLE (); |
b7dbcc19 | 1888 | if (forwardp) |
195d1361 | 1889 | { |
b7dbcc19 | 1890 | if (multibyte) |
8f924df7 | 1891 | while (1) |
b7dbcc19 | 1892 | { |
8f924df7 | 1893 | int nbytes; |
9af7511a | 1894 | |
8f924df7 | 1895 | if (p >= stop) |
9af7511a | 1896 | { |
8f924df7 | 1897 | if (p >= endp) |
9af7511a | 1898 | break; |
8f924df7 KH |
1899 | p = GAP_END_ADDR; |
1900 | stop = endp; | |
9af7511a | 1901 | } |
62a6e103 | 1902 | c = STRING_CHAR_AND_LENGTH (p, nbytes); |
327719ee | 1903 | if (! NILP (iso_classes) && in_classes (c, iso_classes)) |
9af7511a | 1904 | { |
327719ee | 1905 | if (negate) |
9af7511a | 1906 | break; |
327719ee MB |
1907 | else |
1908 | goto fwd_ok; | |
9af7511a | 1909 | } |
9af7511a | 1910 | |
8f924df7 | 1911 | if (! fastmap[*p]) |
b7dbcc19 KH |
1912 | break; |
1913 | if (! ASCII_CHAR_P (c)) | |
9af7511a | 1914 | { |
b7dbcc19 KH |
1915 | /* As we are looking at a multibyte character, we |
1916 | must look up the character in the table | |
1917 | CHAR_RANGES. If there's no data in the table, | |
1918 | that character is not what we want to skip. */ | |
1919 | ||
1920 | /* The following code do the right thing even if | |
1921 | n_char_ranges is zero (i.e. no data in | |
1922 | CHAR_RANGES). */ | |
1923 | for (i = 0; i < n_char_ranges; i += 2) | |
1924 | if (c >= char_ranges[i] && c <= char_ranges[i + 1]) | |
9af7511a | 1925 | break; |
b7dbcc19 KH |
1926 | if (!(negate ^ (i < n_char_ranges))) |
1927 | break; | |
9af7511a | 1928 | } |
327719ee | 1929 | fwd_ok: |
8f924df7 | 1930 | p += nbytes, pos++, pos_byte += nbytes; |
b7dbcc19 KH |
1931 | } |
1932 | else | |
8f924df7 KH |
1933 | while (1) |
1934 | { | |
1935 | if (p >= stop) | |
9af7511a | 1936 | { |
8f924df7 | 1937 | if (p >= endp) |
9af7511a | 1938 | break; |
8f924df7 KH |
1939 | p = GAP_END_ADDR; |
1940 | stop = endp; | |
9af7511a | 1941 | } |
327719ee MB |
1942 | |
1943 | if (!NILP (iso_classes) && in_classes (*p, iso_classes)) | |
1944 | { | |
1945 | if (negate) | |
1946 | break; | |
1947 | else | |
1948 | goto fwd_unibyte_ok; | |
1949 | } | |
1950 | ||
8f924df7 KH |
1951 | if (!fastmap[*p]) |
1952 | break; | |
327719ee | 1953 | fwd_unibyte_ok: |
8f924df7 KH |
1954 | p++, pos++, pos_byte++; |
1955 | } | |
195d1361 RS |
1956 | } |
1957 | else | |
1958 | { | |
b7dbcc19 | 1959 | if (multibyte) |
8f924df7 | 1960 | while (1) |
b7dbcc19 | 1961 | { |
8f924df7 | 1962 | unsigned char *prev_p; |
9af7511a | 1963 | |
8f924df7 | 1964 | if (p <= stop) |
9af7511a | 1965 | { |
8f924df7 | 1966 | if (p <= endp) |
9af7511a | 1967 | break; |
8f924df7 KH |
1968 | p = GPT_ADDR; |
1969 | stop = endp; | |
9af7511a | 1970 | } |
8f924df7 KH |
1971 | prev_p = p; |
1972 | while (--p >= stop && ! CHAR_HEAD_P (*p)); | |
62a6e103 | 1973 | c = STRING_CHAR (p); |
a1bc88d4 | 1974 | |
327719ee MB |
1975 | if (! NILP (iso_classes) && in_classes (c, iso_classes)) |
1976 | { | |
1977 | if (negate) | |
1978 | break; | |
9690d026 | 1979 | else |
327719ee | 1980 | goto back_ok; |
7e68b0ea | 1981 | } |
a1bc88d4 | 1982 | |
8f924df7 | 1983 | if (! fastmap[*p]) |
b7dbcc19 KH |
1984 | break; |
1985 | if (! ASCII_CHAR_P (c)) | |
7e68b0ea | 1986 | { |
b7dbcc19 KH |
1987 | /* See the comment in the previous similar code. */ |
1988 | for (i = 0; i < n_char_ranges; i += 2) | |
1989 | if (c >= char_ranges[i] && c <= char_ranges[i + 1]) | |
1990 | break; | |
1991 | if (!(negate ^ (i < n_char_ranges))) | |
1992 | break; | |
7e68b0ea | 1993 | } |
327719ee | 1994 | back_ok: |
8f924df7 | 1995 | pos--, pos_byte -= prev_p - p; |
b7dbcc19 KH |
1996 | } |
1997 | else | |
8f924df7 KH |
1998 | while (1) |
1999 | { | |
2000 | if (p <= stop) | |
9af7511a | 2001 | { |
8f924df7 | 2002 | if (p <= endp) |
9af7511a | 2003 | break; |
8f924df7 KH |
2004 | p = GPT_ADDR; |
2005 | stop = endp; | |
9af7511a | 2006 | } |
e39091b6 | 2007 | |
327719ee MB |
2008 | if (! NILP (iso_classes) && in_classes (p[-1], iso_classes)) |
2009 | { | |
2010 | if (negate) | |
2011 | break; | |
9af7511a | 2012 | else |
327719ee MB |
2013 | goto back_unibyte_ok; |
2014 | } | |
a1bc88d4 | 2015 | |
8f924df7 KH |
2016 | if (!fastmap[p[-1]]) |
2017 | break; | |
327719ee | 2018 | back_unibyte_ok: |
8f924df7 KH |
2019 | p--, pos--, pos_byte--; |
2020 | } | |
195d1361 | 2021 | } |
7e68b0ea | 2022 | |
b7dbcc19 KH |
2023 | SET_PT_BOTH (pos, pos_byte); |
2024 | immediate_quit = 0; | |
2025 | ||
2026 | return make_number (PT - start_point); | |
2027 | } | |
2028 | } | |
7e68b0ea | 2029 | |
b7dbcc19 KH |
2030 | |
2031 | static Lisp_Object | |
1fc71008 | 2032 | skip_syntaxes (bool forwardp, Lisp_Object string, Lisp_Object lim) |
b7dbcc19 | 2033 | { |
1fc71008 | 2034 | int c; |
b7dbcc19 | 2035 | unsigned char fastmap[0400]; |
1fc71008 PE |
2036 | bool negate = 0; |
2037 | ptrdiff_t i, i_byte; | |
2038 | bool multibyte; | |
d311d28c | 2039 | ptrdiff_t size_byte; |
b7dbcc19 KH |
2040 | unsigned char *str; |
2041 | ||
2042 | CHECK_STRING (string); | |
2043 | ||
2044 | if (NILP (lim)) | |
2045 | XSETINT (lim, forwardp ? ZV : BEGV); | |
2046 | else | |
2047 | CHECK_NUMBER_COERCE_MARKER (lim); | |
2048 | ||
2049 | /* In any case, don't allow scan outside bounds of buffer. */ | |
2050 | if (XINT (lim) > ZV) | |
2051 | XSETFASTINT (lim, ZV); | |
2052 | if (XINT (lim) < BEGV) | |
2053 | XSETFASTINT (lim, BEGV); | |
2054 | ||
ade8ee9e | 2055 | if (forwardp ? (PT >= XFASTINT (lim)) : (PT <= XFASTINT (lim))) |
49feb1cd | 2056 | return make_number (0); |
8c6e735b | 2057 | |
4b4deea2 | 2058 | multibyte = (!NILP (BVAR (current_buffer, enable_multibyte_characters)) |
92eaa22e | 2059 | && (XINT (lim) - PT != CHAR_TO_BYTE (XINT (lim)) - PT_BYTE)); |
b7dbcc19 | 2060 | |
72af86bd | 2061 | memset (fastmap, 0, sizeof fastmap); |
b7dbcc19 | 2062 | |
8f924df7 KH |
2063 | if (SBYTES (string) > SCHARS (string)) |
2064 | /* As this is very rare case (syntax spec is ASCII only), don't | |
2065 | consider efficiency. */ | |
bc796a59 KH |
2066 | string = string_make_unibyte (string); |
2067 | ||
8f924df7 KH |
2068 | str = SDATA (string); |
2069 | size_byte = SBYTES (string); | |
bc796a59 | 2070 | |
b7dbcc19 KH |
2071 | i_byte = 0; |
2072 | if (i_byte < size_byte | |
8f924df7 | 2073 | && SREF (string, 0) == '^') |
b7dbcc19 KH |
2074 | { |
2075 | negate = 1; i_byte++; | |
2076 | } | |
2077 | ||
b7dbcc19 KH |
2078 | /* Find the syntaxes specified and set their elements of fastmap. */ |
2079 | ||
2080 | while (i_byte < size_byte) | |
2081 | { | |
2082 | c = str[i_byte++]; | |
8f924df7 | 2083 | fastmap[syntax_spec_code[c]] = 1; |
b7dbcc19 | 2084 | } |
195d1361 | 2085 | |
9690d026 | 2086 | /* If ^ was the first character, complement the fastmap. */ |
195d1361 RS |
2087 | if (negate) |
2088 | for (i = 0; i < sizeof fastmap; i++) | |
9690d026 | 2089 | fastmap[i] ^= 1; |
195d1361 RS |
2090 | |
2091 | { | |
d311d28c PE |
2092 | ptrdiff_t start_point = PT; |
2093 | ptrdiff_t pos = PT; | |
2094 | ptrdiff_t pos_byte = PT_BYTE; | |
8f924df7 KH |
2095 | unsigned char *p = PT_ADDR, *endp, *stop; |
2096 | ||
2097 | if (forwardp) | |
2098 | { | |
2099 | endp = (XINT (lim) == GPT) ? GPT_ADDR : CHAR_POS_ADDR (XINT (lim)); | |
2100 | stop = (pos < GPT && GPT < XINT (lim)) ? GPT_ADDR : endp; | |
2101 | } | |
2102 | else | |
2103 | { | |
2104 | endp = CHAR_POS_ADDR (XINT (lim)); | |
2105 | stop = (pos >= GPT && GPT > XINT (lim)) ? GAP_END_ADDR : endp; | |
2106 | } | |
195d1361 RS |
2107 | |
2108 | immediate_quit = 1; | |
b7dbcc19 KH |
2109 | SETUP_SYNTAX_TABLE (pos, forwardp ? 1 : -1); |
2110 | if (forwardp) | |
195d1361 | 2111 | { |
b7dbcc19 | 2112 | if (multibyte) |
195d1361 | 2113 | { |
8f924df7 | 2114 | while (1) |
8c6e735b | 2115 | { |
8f924df7 KH |
2116 | int nbytes; |
2117 | ||
2118 | if (p >= stop) | |
2119 | { | |
2120 | if (p >= endp) | |
2121 | break; | |
2122 | p = GAP_END_ADDR; | |
2123 | stop = endp; | |
2124 | } | |
62a6e103 | 2125 | c = STRING_CHAR_AND_LENGTH (p, nbytes); |
1fc71008 | 2126 | if (! fastmap[SYNTAX (c)]) |
8c6e735b | 2127 | break; |
8f924df7 | 2128 | p += nbytes, pos++, pos_byte += nbytes; |
8c6e735b KH |
2129 | UPDATE_SYNTAX_TABLE_FORWARD (pos); |
2130 | } | |
195d1361 RS |
2131 | } |
2132 | else | |
2133 | { | |
8c6e735b | 2134 | while (1) |
7e68b0ea | 2135 | { |
8f924df7 KH |
2136 | if (p >= stop) |
2137 | { | |
2138 | if (p >= endp) | |
2139 | break; | |
2140 | p = GAP_END_ADDR; | |
2141 | stop = endp; | |
2142 | } | |
1fc71008 | 2143 | if (! fastmap[SYNTAX (*p)]) |
8c6e735b | 2144 | break; |
8f924df7 | 2145 | p++, pos++, pos_byte++; |
b7dbcc19 | 2146 | UPDATE_SYNTAX_TABLE_FORWARD (pos); |
195d1361 RS |
2147 | } |
2148 | } | |
2149 | } | |
2150 | else | |
2151 | { | |
b7dbcc19 | 2152 | if (multibyte) |
195d1361 | 2153 | { |
8c6e735b | 2154 | while (1) |
b7dbcc19 | 2155 | { |
8f924df7 KH |
2156 | unsigned char *prev_p; |
2157 | ||
2158 | if (p <= stop) | |
b7dbcc19 | 2159 | { |
8f924df7 KH |
2160 | if (p <= endp) |
2161 | break; | |
2162 | p = GPT_ADDR; | |
2163 | stop = endp; | |
b7dbcc19 | 2164 | } |
262be72a | 2165 | UPDATE_SYNTAX_TABLE_BACKWARD (pos - 1); |
8f924df7 KH |
2166 | prev_p = p; |
2167 | while (--p >= stop && ! CHAR_HEAD_P (*p)); | |
62a6e103 | 2168 | c = STRING_CHAR (p); |
1fc71008 | 2169 | if (! fastmap[SYNTAX (c)]) |
8f924df7 KH |
2170 | break; |
2171 | pos--, pos_byte -= prev_p - p; | |
b7dbcc19 | 2172 | } |
195d1361 RS |
2173 | } |
2174 | else | |
2175 | { | |
8c6e735b KH |
2176 | while (1) |
2177 | { | |
8f924df7 KH |
2178 | if (p <= stop) |
2179 | { | |
2180 | if (p <= endp) | |
2181 | break; | |
2182 | p = GPT_ADDR; | |
2183 | stop = endp; | |
2184 | } | |
102f6132 | 2185 | UPDATE_SYNTAX_TABLE_BACKWARD (pos - 1); |
1fc71008 | 2186 | if (! fastmap[SYNTAX (p[-1])]) |
8c6e735b | 2187 | break; |
8f924df7 | 2188 | p--, pos--, pos_byte--; |
8c6e735b | 2189 | } |
195d1361 RS |
2190 | } |
2191 | } | |
6a140a74 RS |
2192 | |
2193 | SET_PT_BOTH (pos, pos_byte); | |
195d1361 RS |
2194 | immediate_quit = 0; |
2195 | ||
2196 | return make_number (PT - start_point); | |
2197 | } | |
2198 | } | |
a1bc88d4 | 2199 | |
1fc71008 | 2200 | /* Return true if character C belongs to one of the ISO classes |
a1bc88d4 RS |
2201 | in the list ISO_CLASSES. Each class is represented by an |
2202 | integer which is its type according to re_wctype. */ | |
2203 | ||
1fc71008 | 2204 | static bool |
971de7fb | 2205 | in_classes (int c, Lisp_Object iso_classes) |
a1bc88d4 | 2206 | { |
1fc71008 | 2207 | bool fits_class = 0; |
a1bc88d4 | 2208 | |
618db430 | 2209 | while (CONSP (iso_classes)) |
a1bc88d4 RS |
2210 | { |
2211 | Lisp_Object elt; | |
2212 | elt = XCAR (iso_classes); | |
2213 | iso_classes = XCDR (iso_classes); | |
2214 | ||
2215 | if (re_iswctype (c, XFASTINT (elt))) | |
2216 | fits_class = 1; | |
2217 | } | |
2218 | ||
2219 | return fits_class; | |
2220 | } | |
195d1361 | 2221 | \f |
95ff8dfc RS |
2222 | /* Jump over a comment, assuming we are at the beginning of one. |
2223 | FROM is the current position. | |
2224 | FROM_BYTE is the bytepos corresponding to FROM. | |
2225 | Do not move past STOP (a charpos). | |
2226 | The comment over which we have to jump is of style STYLE | |
1fc71008 | 2227 | (either SYNTAX_FLAGS_COMMENT_STYLE (foo) or ST_COMMENT_STYLE). |
95ff8dfc RS |
2228 | NESTING should be positive to indicate the nesting at the beginning |
2229 | for nested comments and should be zero or negative else. | |
2230 | ST_COMMENT_STYLE cannot be nested. | |
2231 | PREV_SYNTAX is the SYNTAX_WITH_FLAGS of the previous character | |
2232 | (or 0 If the search cannot start in the middle of a two-character). | |
2233 | ||
1fc71008 | 2234 | If successful, return true and store the charpos of the comment's end |
95ff8dfc | 2235 | into *CHARPOS_PTR and the corresponding bytepos into *BYTEPOS_PTR. |
1fc71008 | 2236 | Else, return false and store the charpos STOP into *CHARPOS_PTR, the |
95ff8dfc RS |
2237 | corresponding bytepos into *BYTEPOS_PTR and the current nesting |
2238 | (as defined for state.incomment) in *INCOMMENT_PTR. | |
2239 | ||
2240 | The comment end is the last character of the comment rather than the | |
1fc71008 | 2241 | character just after the comment. |
95ff8dfc RS |
2242 | |
2243 | Global syntax data is assumed to initially be valid for FROM and | |
2244 | remains valid for forward search starting at the returned position. */ | |
2245 | ||
1fc71008 | 2246 | static bool |
d311d28c PE |
2247 | forw_comment (ptrdiff_t from, ptrdiff_t from_byte, ptrdiff_t stop, |
2248 | EMACS_INT nesting, int style, int prev_syntax, | |
2249 | ptrdiff_t *charpos_ptr, ptrdiff_t *bytepos_ptr, | |
2250 | EMACS_INT *incomment_ptr) | |
95ff8dfc RS |
2251 | { |
2252 | register int c, c1; | |
2253 | register enum syntaxcode code; | |
c5683ceb | 2254 | register int syntax, other_syntax; |
95ff8dfc RS |
2255 | |
2256 | if (nesting <= 0) nesting = -1; | |
2257 | ||
2258 | /* Enter the loop in the middle so that we find | |
2259 | a 2-char comment ender if we start in the middle of it. */ | |
2260 | syntax = prev_syntax; | |
2261 | if (syntax != 0) goto forw_incomment; | |
2262 | ||
2263 | while (1) | |
2264 | { | |
2265 | if (from == stop) | |
2266 | { | |
2267 | *incomment_ptr = nesting; | |
2268 | *charpos_ptr = from; | |
2269 | *bytepos_ptr = from_byte; | |
2270 | return 0; | |
2271 | } | |
8f924df7 | 2272 | c = FETCH_CHAR_AS_MULTIBYTE (from_byte); |
95ff8dfc RS |
2273 | syntax = SYNTAX_WITH_FLAGS (c); |
2274 | code = syntax & 0xff; | |
2275 | if (code == Sendcomment | |
c5683ceb | 2276 | && SYNTAX_FLAGS_COMMENT_STYLE (syntax, 0) == style |
abf8a9ff SM |
2277 | && (SYNTAX_FLAGS_COMMENT_NESTED (syntax) ? |
2278 | (nesting > 0 && --nesting == 0) : nesting < 0)) | |
95ff8dfc RS |
2279 | /* we have encountered a comment end of the same style |
2280 | as the comment sequence which began this comment | |
2281 | section */ | |
2282 | break; | |
2283 | if (code == Scomment_fence | |
2284 | && style == ST_COMMENT_STYLE) | |
2285 | /* we have encountered a comment end of the same style | |
2286 | as the comment sequence which began this comment | |
2287 | section. */ | |
2288 | break; | |
2289 | if (nesting > 0 | |
2290 | && code == Scomment | |
abf8a9ff | 2291 | && SYNTAX_FLAGS_COMMENT_NESTED (syntax) |
c5683ceb | 2292 | && SYNTAX_FLAGS_COMMENT_STYLE (syntax, 0) == style) |
95ff8dfc RS |
2293 | /* we have encountered a nested comment of the same style |
2294 | as the comment sequence which began this comment section */ | |
2295 | nesting++; | |
2296 | INC_BOTH (from, from_byte); | |
2297 | UPDATE_SYNTAX_TABLE_FORWARD (from); | |
7d0393cf | 2298 | |
95ff8dfc RS |
2299 | forw_incomment: |
2300 | if (from < stop && SYNTAX_FLAGS_COMEND_FIRST (syntax) | |
8f924df7 | 2301 | && (c1 = FETCH_CHAR_AS_MULTIBYTE (from_byte), |
c5683ceb SM |
2302 | other_syntax = SYNTAX_WITH_FLAGS (c1), |
2303 | SYNTAX_FLAGS_COMEND_SECOND (other_syntax)) | |
2304 | && SYNTAX_FLAGS_COMMENT_STYLE (syntax, other_syntax) == style | |
abf8a9ff | 2305 | && ((SYNTAX_FLAGS_COMMENT_NESTED (syntax) || |
c5683ceb SM |
2306 | SYNTAX_FLAGS_COMMENT_NESTED (other_syntax)) |
2307 | ? nesting > 0 : nesting < 0)) | |
4ffe723b GM |
2308 | { |
2309 | if (--nesting <= 0) | |
2310 | /* we have encountered a comment end of the same style | |
2311 | as the comment sequence which began this comment | |
2312 | section */ | |
2313 | break; | |
2314 | else | |
2315 | { | |
2316 | INC_BOTH (from, from_byte); | |
2317 | UPDATE_SYNTAX_TABLE_FORWARD (from); | |
2318 | } | |
2319 | } | |
95ff8dfc RS |
2320 | if (nesting > 0 |
2321 | && from < stop | |
2322 | && SYNTAX_FLAGS_COMSTART_FIRST (syntax) | |
8f924df7 | 2323 | && (c1 = FETCH_CHAR_AS_MULTIBYTE (from_byte), |
c5683ceb SM |
2324 | other_syntax = SYNTAX_WITH_FLAGS (c1), |
2325 | SYNTAX_FLAGS_COMMENT_STYLE (other_syntax, syntax) == style | |
2326 | && SYNTAX_FLAGS_COMSTART_SECOND (other_syntax)) | |
abf8a9ff | 2327 | && (SYNTAX_FLAGS_COMMENT_NESTED (syntax) || |
c5683ceb | 2328 | SYNTAX_FLAGS_COMMENT_NESTED (other_syntax))) |
95ff8dfc RS |
2329 | /* we have encountered a nested comment of the same style |
2330 | as the comment sequence which began this comment | |
2331 | section */ | |
2332 | { | |
2333 | INC_BOTH (from, from_byte); | |
2334 | UPDATE_SYNTAX_TABLE_FORWARD (from); | |
2335 | nesting++; | |
2336 | } | |
2337 | } | |
2338 | *charpos_ptr = from; | |
2339 | *bytepos_ptr = from_byte; | |
2340 | return 1; | |
2341 | } | |
2342 | ||
b3cfe0c8 | 2343 | DEFUN ("forward-comment", Fforward_comment, Sforward_comment, 1, 1, 0, |
177c0ea7 | 2344 | doc: /* |
ee648c11 | 2345 | Move forward across up to COUNT comments. If COUNT is negative, move backward. |
fdb82f93 PJ |
2346 | Stop scanning if we find something other than a comment or whitespace. |
2347 | Set point to where scanning stops. | |
ee648c11 | 2348 | If COUNT comments are found as expected, with nothing except whitespace |
fdb82f93 | 2349 | between them, return t; otherwise return nil. */) |
5842a27b | 2350 | (Lisp_Object count) |
b3cfe0c8 | 2351 | { |
1fc71008 PE |
2352 | ptrdiff_t from, from_byte, stop; |
2353 | int c, c1; | |
2354 | enum syntaxcode code; | |
b3cfe0c8 | 2355 | int comstyle = 0; /* style of comment encountered */ |
1fc71008 PE |
2356 | bool comnested = 0; /* whether the comment is nestable or not */ |
2357 | bool found; | |
2312c580 | 2358 | EMACS_INT count1; |
d311d28c | 2359 | ptrdiff_t out_charpos, out_bytepos; |
7c85f529 | 2360 | EMACS_INT dummy; |
840f481c | 2361 | |
b7826503 | 2362 | CHECK_NUMBER (count); |
840f481c | 2363 | count1 = XINT (count); |
195d1361 | 2364 | stop = count1 > 0 ? ZV : BEGV; |
b3cfe0c8 RS |
2365 | |
2366 | immediate_quit = 1; | |
2367 | QUIT; | |
2368 | ||
2369 | from = PT; | |
6a140a74 | 2370 | from_byte = PT_BYTE; |
b3cfe0c8 | 2371 | |
195d1361 | 2372 | SETUP_SYNTAX_TABLE (from, count1); |
840f481c | 2373 | while (count1 > 0) |
b3cfe0c8 | 2374 | { |
04882296 | 2375 | do |
b3cfe0c8 | 2376 | { |
1fc71008 PE |
2377 | bool comstart_first; |
2378 | int syntax, other_syntax; | |
f902a008 | 2379 | |
04882296 KH |
2380 | if (from == stop) |
2381 | { | |
ef316cf0 | 2382 | SET_PT_BOTH (from, from_byte); |
b7e6e612 | 2383 | immediate_quit = 0; |
04882296 KH |
2384 | return Qnil; |
2385 | } | |
b7dbcc19 | 2386 | c = FETCH_CHAR_AS_MULTIBYTE (from_byte); |
c5683ceb | 2387 | syntax = SYNTAX_WITH_FLAGS (c); |
b3cfe0c8 | 2388 | code = SYNTAX (c); |
c5683ceb SM |
2389 | comstart_first = SYNTAX_FLAGS_COMSTART_FIRST (syntax); |
2390 | comnested = SYNTAX_FLAGS_COMMENT_NESTED (syntax); | |
2391 | comstyle = SYNTAX_FLAGS_COMMENT_STYLE (syntax, 0); | |
6a140a74 | 2392 | INC_BOTH (from, from_byte); |
f902a008 | 2393 | UPDATE_SYNTAX_TABLE_FORWARD (from); |
f902a008 | 2394 | if (from < stop && comstart_first |
b7dbcc19 | 2395 | && (c1 = FETCH_CHAR_AS_MULTIBYTE (from_byte), |
c5683ceb SM |
2396 | other_syntax = SYNTAX_WITH_FLAGS (c1), |
2397 | SYNTAX_FLAGS_COMSTART_SECOND (other_syntax))) | |
b3cfe0c8 | 2398 | { |
7d0393cf | 2399 | /* We have encountered a comment start sequence and we |
7fc8191e | 2400 | are ignoring all text inside comments. We must record |
b3cfe0c8 RS |
2401 | the comment style this sequence begins so that later, |
2402 | only a comment end of the same style actually ends | |
7fc8191e | 2403 | the comment section. */ |
b3cfe0c8 | 2404 | code = Scomment; |
c5683ceb | 2405 | comstyle = SYNTAX_FLAGS_COMMENT_STYLE (other_syntax, syntax); |
1fc71008 | 2406 | comnested |= SYNTAX_FLAGS_COMMENT_NESTED (other_syntax); |
6a140a74 | 2407 | INC_BOTH (from, from_byte); |
95ff8dfc | 2408 | UPDATE_SYNTAX_TABLE_FORWARD (from); |
b3cfe0c8 | 2409 | } |
04882296 | 2410 | } |
abf8a9ff | 2411 | while (code == Swhitespace || (code == Sendcomment && c == '\n')); |
6a140a74 | 2412 | |
da71ab91 KH |
2413 | if (code == Scomment_fence) |
2414 | comstyle = ST_COMMENT_STYLE; | |
2415 | else if (code != Scomment) | |
04882296 KH |
2416 | { |
2417 | immediate_quit = 0; | |
6a140a74 | 2418 | DEC_BOTH (from, from_byte); |
ef316cf0 | 2419 | SET_PT_BOTH (from, from_byte); |
04882296 KH |
2420 | return Qnil; |
2421 | } | |
2422 | /* We're at the start of a comment. */ | |
95ff8dfc RS |
2423 | found = forw_comment (from, from_byte, stop, comnested, comstyle, 0, |
2424 | &out_charpos, &out_bytepos, &dummy); | |
2425 | from = out_charpos; from_byte = out_bytepos; | |
2426 | if (!found) | |
04882296 | 2427 | { |
95ff8dfc RS |
2428 | immediate_quit = 0; |
2429 | SET_PT_BOTH (from, from_byte); | |
2430 | return Qnil; | |
b3cfe0c8 | 2431 | } |
95ff8dfc RS |
2432 | INC_BOTH (from, from_byte); |
2433 | UPDATE_SYNTAX_TABLE_FORWARD (from); | |
04882296 | 2434 | /* We have skipped one comment. */ |
840f481c | 2435 | count1--; |
b3cfe0c8 RS |
2436 | } |
2437 | ||
840f481c | 2438 | while (count1 < 0) |
b3cfe0c8 | 2439 | { |
b9145dbb | 2440 | while (1) |
b3cfe0c8 | 2441 | { |
1fc71008 PE |
2442 | bool quoted; |
2443 | int syntax; | |
f902a008 | 2444 | |
b9145dbb RS |
2445 | if (from <= stop) |
2446 | { | |
6a140a74 | 2447 | SET_PT_BOTH (BEGV, BEGV_BYTE); |
b9145dbb RS |
2448 | immediate_quit = 0; |
2449 | return Qnil; | |
2450 | } | |
b3cfe0c8 | 2451 | |
6a140a74 | 2452 | DEC_BOTH (from, from_byte); |
f902a008 | 2453 | /* char_quoted does UPDATE_SYNTAX_TABLE_BACKWARD (from). */ |
6a140a74 | 2454 | quoted = char_quoted (from, from_byte); |
b7dbcc19 | 2455 | c = FETCH_CHAR_AS_MULTIBYTE (from_byte); |
c5683ceb | 2456 | syntax = SYNTAX_WITH_FLAGS (c); |
b3cfe0c8 RS |
2457 | code = SYNTAX (c); |
2458 | comstyle = 0; | |
c5683ceb | 2459 | comnested = SYNTAX_FLAGS_COMMENT_NESTED (syntax); |
7fc8191e | 2460 | if (code == Sendcomment) |
c5683ceb SM |
2461 | comstyle = SYNTAX_FLAGS_COMMENT_STYLE (syntax, 0); |
2462 | if (from > stop && SYNTAX_FLAGS_COMEND_SECOND (syntax) | |
f902a008 | 2463 | && prev_char_comend_first (from, from_byte) |
89c6809a | 2464 | && !char_quoted (from - 1, dec_bytepos (from_byte))) |
b3cfe0c8 | 2465 | { |
c5683ceb | 2466 | int other_syntax; |
7fc8191e | 2467 | /* We must record the comment style encountered so that |
b3cfe0c8 | 2468 | later, we can match only the proper comment begin |
7fc8191e | 2469 | sequence of the same style. */ |
6a140a74 | 2470 | DEC_BOTH (from, from_byte); |
f902a008 RS |
2471 | code = Sendcomment; |
2472 | /* Calling char_quoted, above, set up global syntax position | |
2473 | at the new value of FROM. */ | |
b7dbcc19 | 2474 | c1 = FETCH_CHAR_AS_MULTIBYTE (from_byte); |
c5683ceb SM |
2475 | other_syntax = SYNTAX_WITH_FLAGS (c1); |
2476 | comstyle = SYNTAX_FLAGS_COMMENT_STYLE (other_syntax, syntax); | |
1fc71008 | 2477 | comnested |= SYNTAX_FLAGS_COMMENT_NESTED (other_syntax); |
b3cfe0c8 RS |
2478 | } |
2479 | ||
195d1361 RS |
2480 | if (code == Scomment_fence) |
2481 | { | |
2482 | /* Skip until first preceding unquoted comment_fence. */ | |
1fc71008 | 2483 | bool fence_found = 0; |
d311d28c | 2484 | ptrdiff_t ini = from, ini_byte = from_byte; |
7d0393cf | 2485 | |
6a140a74 | 2486 | while (1) |
195d1361 | 2487 | { |
6a140a74 | 2488 | DEC_BOTH (from, from_byte); |
195d1361 | 2489 | UPDATE_SYNTAX_TABLE_BACKWARD (from); |
b7dbcc19 | 2490 | c = FETCH_CHAR_AS_MULTIBYTE (from_byte); |
6a140a74 | 2491 | if (SYNTAX (c) == Scomment_fence |
7d0393cf | 2492 | && !char_quoted (from, from_byte)) |
195d1361 | 2493 | { |
01f44d5a | 2494 | fence_found = 1; |
195d1361 RS |
2495 | break; |
2496 | } | |
fcdd4585 SM |
2497 | else if (from == stop) |
2498 | break; | |
195d1361 | 2499 | } |
01f44d5a | 2500 | if (fence_found == 0) |
195d1361 RS |
2501 | { |
2502 | from = ini; /* Set point to ini + 1. */ | |
6a140a74 | 2503 | from_byte = ini_byte; |
195d1361 RS |
2504 | goto leave; |
2505 | } | |
39c41ad4 SM |
2506 | else |
2507 | /* We have skipped one comment. */ | |
2508 | break; | |
195d1361 RS |
2509 | } |
2510 | else if (code == Sendcomment) | |
b3cfe0c8 | 2511 | { |
95ff8dfc | 2512 | found = back_comment (from, from_byte, stop, comnested, comstyle, |
6a140a74 | 2513 | &out_charpos, &out_bytepos); |
1fc71008 | 2514 | if (!found) |
1aa963c8 | 2515 | { |
abf8a9ff SM |
2516 | if (c == '\n') |
2517 | /* This end-of-line is not an end-of-comment. | |
2518 | Treat it like a whitespace. | |
2519 | CC-mode (and maybe others) relies on this behavior. */ | |
2520 | ; | |
2521 | else | |
2522 | { | |
2523 | /* Failure: we should go back to the end of this | |
2524 | not-quite-endcomment. */ | |
c5683ceb | 2525 | if (SYNTAX (c) != code) |
abf8a9ff SM |
2526 | /* It was a two-char Sendcomment. */ |
2527 | INC_BOTH (from, from_byte); | |
2528 | goto leave; | |
2529 | } | |
1aa963c8 | 2530 | } |
2b927d02 | 2531 | else |
abf8a9ff SM |
2532 | { |
2533 | /* We have skipped one comment. */ | |
2534 | from = out_charpos, from_byte = out_bytepos; | |
2535 | break; | |
2536 | } | |
b3cfe0c8 | 2537 | } |
bb0de084 | 2538 | else if (code != Swhitespace || quoted) |
b3cfe0c8 | 2539 | { |
195d1361 | 2540 | leave: |
b3cfe0c8 | 2541 | immediate_quit = 0; |
6a140a74 | 2542 | INC_BOTH (from, from_byte); |
ef316cf0 | 2543 | SET_PT_BOTH (from, from_byte); |
b3cfe0c8 RS |
2544 | return Qnil; |
2545 | } | |
2546 | } | |
2547 | ||
840f481c | 2548 | count1++; |
b3cfe0c8 RS |
2549 | } |
2550 | ||
ef316cf0 | 2551 | SET_PT_BOTH (from, from_byte); |
b3cfe0c8 RS |
2552 | immediate_quit = 0; |
2553 | return Qt; | |
2554 | } | |
2555 | \f | |
b7dbcc19 | 2556 | /* Return syntax code of character C if C is an ASCII character |
45b683a1 | 2557 | or if MULTIBYTE_SYMBOL_P is false. Otherwise, return Ssymbol. */ |
bd25db08 | 2558 | |
45b683a1 PE |
2559 | static enum syntaxcode |
2560 | syntax_multibyte (int c, bool multibyte_symbol_p) | |
2561 | { | |
2562 | return ASCII_CHAR_P (c) || !multibyte_symbol_p ? SYNTAX (c) : Ssymbol; | |
2563 | } | |
bd25db08 | 2564 | |
6a140a74 | 2565 | static Lisp_Object |
1fc71008 | 2566 | scan_lists (EMACS_INT from, EMACS_INT count, EMACS_INT depth, bool sexpflag) |
8489eb67 RS |
2567 | { |
2568 | Lisp_Object val; | |
1fc71008 PE |
2569 | ptrdiff_t stop = count > 0 ? ZV : BEGV; |
2570 | int c, c1; | |
93da5fff | 2571 | int stringterm; |
1fc71008 PE |
2572 | bool quoted; |
2573 | bool mathexit = 0; | |
45b683a1 | 2574 | enum syntaxcode code; |
d311d28c | 2575 | EMACS_INT min_depth = depth; /* Err out if depth gets less than this. */ |
e5d4f4dc | 2576 | int comstyle = 0; /* style of comment encountered */ |
1fc71008 | 2577 | bool comnested = 0; /* whether the comment is nestable or not */ |
d311d28c | 2578 | ptrdiff_t temp_pos; |
932e6895 | 2579 | EMACS_INT last_good = from; |
1fc71008 | 2580 | bool found; |
d311d28c PE |
2581 | ptrdiff_t from_byte; |
2582 | ptrdiff_t out_bytepos, out_charpos; | |
7c85f529 | 2583 | EMACS_INT dummy; |
1fc71008 | 2584 | bool multibyte_symbol_p = sexpflag && multibyte_syntax_as_symbol; |
8489eb67 RS |
2585 | |
2586 | if (depth > 0) min_depth = 0; | |
2587 | ||
c3907a7e RS |
2588 | if (from > ZV) from = ZV; |
2589 | if (from < BEGV) from = BEGV; | |
2590 | ||
2591 | from_byte = CHAR_TO_BYTE (from); | |
2592 | ||
8489eb67 RS |
2593 | immediate_quit = 1; |
2594 | QUIT; | |
2595 | ||
195d1361 | 2596 | SETUP_SYNTAX_TABLE (from, count); |
8489eb67 RS |
2597 | while (count > 0) |
2598 | { | |
8489eb67 RS |
2599 | while (from < stop) |
2600 | { | |
1fc71008 PE |
2601 | bool comstart_first, prefix; |
2602 | int syntax, other_syntax; | |
195d1361 | 2603 | UPDATE_SYNTAX_TABLE_FORWARD (from); |
b7dbcc19 | 2604 | c = FETCH_CHAR_AS_MULTIBYTE (from_byte); |
c5683ceb | 2605 | syntax = SYNTAX_WITH_FLAGS (c); |
45b683a1 | 2606 | code = syntax_multibyte (c, multibyte_symbol_p); |
c5683ceb SM |
2607 | comstart_first = SYNTAX_FLAGS_COMSTART_FIRST (syntax); |
2608 | comnested = SYNTAX_FLAGS_COMMENT_NESTED (syntax); | |
2609 | comstyle = SYNTAX_FLAGS_COMMENT_STYLE (syntax, 0); | |
2610 | prefix = SYNTAX_FLAGS_PREFIX (syntax); | |
7bf5e9e4 RS |
2611 | if (depth == min_depth) |
2612 | last_good = from; | |
6a140a74 | 2613 | INC_BOTH (from, from_byte); |
195d1361 | 2614 | UPDATE_SYNTAX_TABLE_FORWARD (from); |
f902a008 | 2615 | if (from < stop && comstart_first |
327719ee | 2616 | && (c = FETCH_CHAR_AS_MULTIBYTE (from_byte), |
c5683ceb SM |
2617 | other_syntax = SYNTAX_WITH_FLAGS (c), |
2618 | SYNTAX_FLAGS_COMSTART_SECOND (other_syntax)) | |
8489eb67 | 2619 | && parse_sexp_ignore_comments) |
e5d4f4dc | 2620 | { |
7d0393cf | 2621 | /* we have encountered a comment start sequence and we |
195d1361 | 2622 | are ignoring all text inside comments. We must record |
e5d4f4dc RS |
2623 | the comment style this sequence begins so that later, |
2624 | only a comment end of the same style actually ends | |
2625 | the comment section */ | |
2626 | code = Scomment; | |
c5683ceb | 2627 | comstyle = SYNTAX_FLAGS_COMMENT_STYLE (other_syntax, syntax); |
1fc71008 | 2628 | comnested |= SYNTAX_FLAGS_COMMENT_NESTED (other_syntax); |
6a140a74 | 2629 | INC_BOTH (from, from_byte); |
f902a008 | 2630 | UPDATE_SYNTAX_TABLE_FORWARD (from); |
e5d4f4dc | 2631 | } |
7d0393cf | 2632 | |
f902a008 | 2633 | if (prefix) |
8489eb67 RS |
2634 | continue; |
2635 | ||
7393bcbb | 2636 | switch (code) |
8489eb67 RS |
2637 | { |
2638 | case Sescape: | |
2639 | case Scharquote: | |
8b8bf8e6 SM |
2640 | if (from == stop) |
2641 | goto lose; | |
6a140a74 | 2642 | INC_BOTH (from, from_byte); |
8489eb67 RS |
2643 | /* treat following character as a word constituent */ |
2644 | case Sword: | |
2645 | case Ssymbol: | |
2646 | if (depth || !sexpflag) break; | |
195d1361 | 2647 | /* This word counts as a sexp; return at end of it. */ |
8489eb67 RS |
2648 | while (from < stop) |
2649 | { | |
195d1361 | 2650 | UPDATE_SYNTAX_TABLE_FORWARD (from); |
7c7ca3d2 | 2651 | |
b7dbcc19 | 2652 | c = FETCH_CHAR_AS_MULTIBYTE (from_byte); |
45b683a1 | 2653 | switch (syntax_multibyte (c, multibyte_symbol_p)) |
8489eb67 RS |
2654 | { |
2655 | case Scharquote: | |
2656 | case Sescape: | |
6a140a74 | 2657 | INC_BOTH (from, from_byte); |
8b8bf8e6 SM |
2658 | if (from == stop) |
2659 | goto lose; | |
8489eb67 RS |
2660 | break; |
2661 | case Sword: | |
2662 | case Ssymbol: | |
2663 | case Squote: | |
2664 | break; | |
2665 | default: | |
2666 | goto done; | |
2667 | } | |
6a140a74 | 2668 | INC_BOTH (from, from_byte); |
8489eb67 RS |
2669 | } |
2670 | goto done; | |
2671 | ||
195d1361 | 2672 | case Scomment_fence: |
95ff8dfc RS |
2673 | comstyle = ST_COMMENT_STYLE; |
2674 | /* FALLTHROUGH */ | |
2675 | case Scomment: | |
8489eb67 | 2676 | if (!parse_sexp_ignore_comments) break; |
95ff8dfc RS |
2677 | UPDATE_SYNTAX_TABLE_FORWARD (from); |
2678 | found = forw_comment (from, from_byte, stop, | |
2679 | comnested, comstyle, 0, | |
2680 | &out_charpos, &out_bytepos, &dummy); | |
2681 | from = out_charpos, from_byte = out_bytepos; | |
2682 | if (!found) | |
8489eb67 | 2683 | { |
95ff8dfc RS |
2684 | if (depth == 0) |
2685 | goto done; | |
2686 | goto lose; | |
8489eb67 | 2687 | } |
95ff8dfc RS |
2688 | INC_BOTH (from, from_byte); |
2689 | UPDATE_SYNTAX_TABLE_FORWARD (from); | |
8489eb67 RS |
2690 | break; |
2691 | ||
2692 | case Smath: | |
2693 | if (!sexpflag) | |
2694 | break; | |
b7dbcc19 | 2695 | if (from != stop && c == FETCH_CHAR_AS_MULTIBYTE (from_byte)) |
6a140a74 RS |
2696 | { |
2697 | INC_BOTH (from, from_byte); | |
2698 | } | |
8489eb67 RS |
2699 | if (mathexit) |
2700 | { | |
2701 | mathexit = 0; | |
2702 | goto close1; | |
2703 | } | |
2704 | mathexit = 1; | |
2705 | ||
2706 | case Sopen: | |
2707 | if (!++depth) goto done; | |
2708 | break; | |
2709 | ||
2710 | case Sclose: | |
2711 | close1: | |
2712 | if (!--depth) goto done; | |
2713 | if (depth < min_depth) | |
9471945b KS |
2714 | xsignal3 (Qscan_error, |
2715 | build_string ("Containing expression ends prematurely"), | |
2716 | make_number (last_good), make_number (from)); | |
8489eb67 RS |
2717 | break; |
2718 | ||
2719 | case Sstring: | |
195d1361 | 2720 | case Sstring_fence: |
ef316cf0 | 2721 | temp_pos = dec_bytepos (from_byte); |
b7dbcc19 | 2722 | stringterm = FETCH_CHAR_AS_MULTIBYTE (temp_pos); |
8489eb67 RS |
2723 | while (1) |
2724 | { | |
45b683a1 | 2725 | enum syntaxcode c_code; |
8b8bf8e6 SM |
2726 | if (from >= stop) |
2727 | goto lose; | |
195d1361 | 2728 | UPDATE_SYNTAX_TABLE_FORWARD (from); |
b7dbcc19 | 2729 | c = FETCH_CHAR_AS_MULTIBYTE (from_byte); |
45b683a1 | 2730 | c_code = syntax_multibyte (c, multibyte_symbol_p); |
42ebfa31 | 2731 | if (code == Sstring |
45b683a1 PE |
2732 | ? c == stringterm && c_code == Sstring |
2733 | : c_code == Sstring_fence) | |
195d1361 | 2734 | break; |
7c7ca3d2 | 2735 | |
1fc71008 | 2736 | switch (c_code) |
8489eb67 RS |
2737 | { |
2738 | case Scharquote: | |
2739 | case Sescape: | |
6a140a74 | 2740 | INC_BOTH (from, from_byte); |
8489eb67 | 2741 | } |
6a140a74 | 2742 | INC_BOTH (from, from_byte); |
8489eb67 | 2743 | } |
6a140a74 | 2744 | INC_BOTH (from, from_byte); |
8489eb67 RS |
2745 | if (!depth && sexpflag) goto done; |
2746 | break; | |
240c43e8 SM |
2747 | default: |
2748 | /* Ignore whitespace, punctuation, quote, endcomment. */ | |
2749 | break; | |
8489eb67 RS |
2750 | } |
2751 | } | |
2752 | ||
2753 | /* Reached end of buffer. Error if within object, return nil if between */ | |
8b8bf8e6 SM |
2754 | if (depth) |
2755 | goto lose; | |
8489eb67 RS |
2756 | |
2757 | immediate_quit = 0; | |
2758 | return Qnil; | |
2759 | ||
2760 | /* End of object reached */ | |
2761 | done: | |
2762 | count--; | |
2763 | } | |
2764 | ||
2765 | while (count < 0) | |
2766 | { | |
8489eb67 RS |
2767 | while (from > stop) |
2768 | { | |
c5683ceb | 2769 | int syntax; |
6a140a74 | 2770 | DEC_BOTH (from, from_byte); |
195d1361 | 2771 | UPDATE_SYNTAX_TABLE_BACKWARD (from); |
b7dbcc19 | 2772 | c = FETCH_CHAR_AS_MULTIBYTE (from_byte); |
c5683ceb | 2773 | syntax= SYNTAX_WITH_FLAGS (c); |
45b683a1 | 2774 | code = syntax_multibyte (c, multibyte_symbol_p); |
7bf5e9e4 RS |
2775 | if (depth == min_depth) |
2776 | last_good = from; | |
7fc8191e | 2777 | comstyle = 0; |
c5683ceb | 2778 | comnested = SYNTAX_FLAGS_COMMENT_NESTED (syntax); |
7fc8191e | 2779 | if (code == Sendcomment) |
c5683ceb SM |
2780 | comstyle = SYNTAX_FLAGS_COMMENT_STYLE (syntax, 0); |
2781 | if (from > stop && SYNTAX_FLAGS_COMEND_SECOND (syntax) | |
1674d9a2 | 2782 | && prev_char_comend_first (from, from_byte) |
8489eb67 | 2783 | && parse_sexp_ignore_comments) |
e5d4f4dc | 2784 | { |
f902a008 | 2785 | /* We must record the comment style encountered so that |
e5d4f4dc | 2786 | later, we can match only the proper comment begin |
f902a008 | 2787 | sequence of the same style. */ |
01f44d5a | 2788 | int c2, other_syntax; |
6a140a74 | 2789 | DEC_BOTH (from, from_byte); |
f902a008 RS |
2790 | UPDATE_SYNTAX_TABLE_BACKWARD (from); |
2791 | code = Sendcomment; | |
01f44d5a PE |
2792 | c2 = FETCH_CHAR_AS_MULTIBYTE (from_byte); |
2793 | other_syntax = SYNTAX_WITH_FLAGS (c2); | |
c5683ceb | 2794 | comstyle = SYNTAX_FLAGS_COMMENT_STYLE (other_syntax, syntax); |
1fc71008 | 2795 | comnested |= SYNTAX_FLAGS_COMMENT_NESTED (other_syntax); |
e5d4f4dc | 2796 | } |
7d0393cf | 2797 | |
9828a477 | 2798 | /* Quoting turns anything except a comment-ender |
e6365855 | 2799 | into a word character. Note that this cannot be true |
f902a008 | 2800 | if we decremented FROM in the if-statement above. */ |
9828a477 | 2801 | if (code != Sendcomment && char_quoted (from, from_byte)) |
240c43e8 SM |
2802 | { |
2803 | DEC_BOTH (from, from_byte); | |
2804 | code = Sword; | |
2805 | } | |
c5683ceb | 2806 | else if (SYNTAX_FLAGS_PREFIX (syntax)) |
8489eb67 RS |
2807 | continue; |
2808 | ||
7393bcbb | 2809 | switch (code) |
8489eb67 RS |
2810 | { |
2811 | case Sword: | |
2812 | case Ssymbol: | |
9828a477 KH |
2813 | case Sescape: |
2814 | case Scharquote: | |
8489eb67 | 2815 | if (depth || !sexpflag) break; |
195d1361 RS |
2816 | /* This word counts as a sexp; count object finished |
2817 | after passing it. */ | |
8489eb67 RS |
2818 | while (from > stop) |
2819 | { | |
6a140a74 | 2820 | temp_pos = from_byte; |
4b4deea2 | 2821 | if (! NILP (BVAR (current_buffer, enable_multibyte_characters))) |
ef316cf0 RS |
2822 | DEC_POS (temp_pos); |
2823 | else | |
2824 | temp_pos--; | |
6a140a74 | 2825 | UPDATE_SYNTAX_TABLE_BACKWARD (from - 1); |
b7dbcc19 | 2826 | c1 = FETCH_CHAR_AS_MULTIBYTE (temp_pos); |
9828a477 | 2827 | /* Don't allow comment-end to be quoted. */ |
45b683a1 | 2828 | if (syntax_multibyte (c1, multibyte_symbol_p) == Sendcomment) |
9828a477 | 2829 | goto done2; |
6a140a74 | 2830 | quoted = char_quoted (from - 1, temp_pos); |
8489eb67 | 2831 | if (quoted) |
93da5fff | 2832 | { |
6a140a74 | 2833 | DEC_BOTH (from, from_byte); |
ef316cf0 | 2834 | temp_pos = dec_bytepos (temp_pos); |
6a140a74 | 2835 | UPDATE_SYNTAX_TABLE_BACKWARD (from - 1); |
93da5fff | 2836 | } |
b7dbcc19 | 2837 | c1 = FETCH_CHAR_AS_MULTIBYTE (temp_pos); |
45b683a1 PE |
2838 | if (! quoted) |
2839 | switch (syntax_multibyte (c1, multibyte_symbol_p)) | |
2840 | { | |
2841 | case Sword: case Ssymbol: case Squote: break; | |
2842 | default: goto done2; | |
2843 | } | |
6a140a74 | 2844 | DEC_BOTH (from, from_byte); |
8489eb67 RS |
2845 | } |
2846 | goto done2; | |
2847 | ||
2848 | case Smath: | |
2849 | if (!sexpflag) | |
2850 | break; | |
ef316cf0 | 2851 | temp_pos = dec_bytepos (from_byte); |
6a140a74 | 2852 | UPDATE_SYNTAX_TABLE_BACKWARD (from - 1); |
b7dbcc19 | 2853 | if (from != stop && c == FETCH_CHAR_AS_MULTIBYTE (temp_pos)) |
6a140a74 | 2854 | DEC_BOTH (from, from_byte); |
8489eb67 RS |
2855 | if (mathexit) |
2856 | { | |
2857 | mathexit = 0; | |
2858 | goto open2; | |
2859 | } | |
2860 | mathexit = 1; | |
2861 | ||
2862 | case Sclose: | |
2863 | if (!++depth) goto done2; | |
2864 | break; | |
2865 | ||
2866 | case Sopen: | |
2867 | open2: | |
2868 | if (!--depth) goto done2; | |
2869 | if (depth < min_depth) | |
9471945b KS |
2870 | xsignal3 (Qscan_error, |
2871 | build_string ("Containing expression ends prematurely"), | |
2872 | make_number (last_good), make_number (from)); | |
8489eb67 RS |
2873 | break; |
2874 | ||
2875 | case Sendcomment: | |
2876 | if (!parse_sexp_ignore_comments) | |
2877 | break; | |
95ff8dfc | 2878 | found = back_comment (from, from_byte, stop, comnested, comstyle, |
6a140a74 | 2879 | &out_charpos, &out_bytepos); |
1fc71008 | 2880 | /* FIXME: if !found, it really wasn't a comment-end. |
1aa963c8 SM |
2881 | For single-char Sendcomment, we can't do much about it apart |
2882 | from skipping the char. | |
2883 | For 2-char endcomments, we could try again, taking both | |
2884 | chars as separate entities, but it's a lot of trouble | |
2885 | for very little gain, so we don't bother either. -sm */ | |
1fc71008 | 2886 | if (found) |
6a140a74 | 2887 | from = out_charpos, from_byte = out_bytepos; |
8489eb67 RS |
2888 | break; |
2889 | ||
195d1361 RS |
2890 | case Scomment_fence: |
2891 | case Sstring_fence: | |
2892 | while (1) | |
2893 | { | |
8b8bf8e6 SM |
2894 | if (from == stop) |
2895 | goto lose; | |
6b61353c | 2896 | DEC_BOTH (from, from_byte); |
195d1361 | 2897 | UPDATE_SYNTAX_TABLE_BACKWARD (from); |
45b683a1 PE |
2898 | if (!char_quoted (from, from_byte)) |
2899 | { | |
2900 | c = FETCH_CHAR_AS_MULTIBYTE (from_byte); | |
2901 | if (syntax_multibyte (c, multibyte_symbol_p) == code) | |
2902 | break; | |
2903 | } | |
195d1361 RS |
2904 | } |
2905 | if (code == Sstring_fence && !depth && sexpflag) goto done2; | |
2906 | break; | |
7d0393cf | 2907 | |
8489eb67 | 2908 | case Sstring: |
b7dbcc19 | 2909 | stringterm = FETCH_CHAR_AS_MULTIBYTE (from_byte); |
8489eb67 RS |
2910 | while (1) |
2911 | { | |
8b8bf8e6 SM |
2912 | if (from == stop) |
2913 | goto lose; | |
6b61353c KH |
2914 | DEC_BOTH (from, from_byte); |
2915 | UPDATE_SYNTAX_TABLE_BACKWARD (from); | |
45b683a1 PE |
2916 | if (!char_quoted (from, from_byte)) |
2917 | { | |
2918 | c = FETCH_CHAR_AS_MULTIBYTE (from_byte); | |
2919 | if (c == stringterm | |
2920 | && (syntax_multibyte (c, multibyte_symbol_p) | |
2921 | == Sstring)) | |
2922 | break; | |
2923 | } | |
8489eb67 | 2924 | } |
8489eb67 RS |
2925 | if (!depth && sexpflag) goto done2; |
2926 | break; | |
240c43e8 SM |
2927 | default: |
2928 | /* Ignore whitespace, punctuation, quote, endcomment. */ | |
2929 | break; | |
8489eb67 RS |
2930 | } |
2931 | } | |
2932 | ||
2933 | /* Reached start of buffer. Error if within object, return nil if between */ | |
8b8bf8e6 SM |
2934 | if (depth) |
2935 | goto lose; | |
8489eb67 RS |
2936 | |
2937 | immediate_quit = 0; | |
2938 | return Qnil; | |
2939 | ||
2940 | done2: | |
2941 | count++; | |
2942 | } | |
2943 | ||
2944 | ||
2945 | immediate_quit = 0; | |
1e142fb7 | 2946 | XSETFASTINT (val, from); |
8489eb67 RS |
2947 | return val; |
2948 | ||
2949 | lose: | |
9471945b KS |
2950 | xsignal3 (Qscan_error, |
2951 | build_string ("Unbalanced parentheses"), | |
2952 | make_number (last_good), make_number (from)); | |
8489eb67 RS |
2953 | } |
2954 | ||
8489eb67 | 2955 | DEFUN ("scan-lists", Fscan_lists, Sscan_lists, 3, 3, 0, |
fdb82f93 | 2956 | doc: /* Scan from character number FROM by COUNT lists. |
7b926f3f CY |
2957 | Scan forward if COUNT is positive, backward if COUNT is negative. |
2958 | Return the character number of the position thus found. | |
2959 | ||
2960 | A \"list", in this context, refers to a balanced parenthetical | |
2961 | grouping, as determined by the syntax table. | |
fdb82f93 | 2962 | |
7b926f3f CY |
2963 | If DEPTH is nonzero, treat that as the nesting depth of the starting |
2964 | point (i.e. the starting point is DEPTH parentheses deep). This | |
2965 | function scans over parentheses until the depth goes to zero COUNT | |
2966 | times. Hence, positive DEPTH moves out that number of levels of | |
221cd413 | 2967 | parentheses, while negative DEPTH moves to a deeper level. |
fdb82f93 PJ |
2968 | |
2969 | Comments are ignored if `parse-sexp-ignore-comments' is non-nil. | |
2970 | ||
7b926f3f CY |
2971 | If we reach the beginning or end of the accessible part of the buffer |
2972 | before we have scanned over COUNT lists, return nil if the depth at | |
2973 | that point is zero, and signal a error if the depth is nonzero. */) | |
5842a27b | 2974 | (Lisp_Object from, Lisp_Object count, Lisp_Object depth) |
8489eb67 | 2975 | { |
b7826503 PJ |
2976 | CHECK_NUMBER (from); |
2977 | CHECK_NUMBER (count); | |
2978 | CHECK_NUMBER (depth); | |
8489eb67 RS |
2979 | |
2980 | return scan_lists (XINT (from), XINT (count), XINT (depth), 0); | |
2981 | } | |
2982 | ||
2983 | DEFUN ("scan-sexps", Fscan_sexps, Sscan_sexps, 2, 2, 0, | |
fdb82f93 PJ |
2984 | doc: /* Scan from character number FROM by COUNT balanced expressions. |
2985 | If COUNT is negative, scan backwards. | |
2986 | Returns the character number of the position thus found. | |
2987 | ||
2988 | Comments are ignored if `parse-sexp-ignore-comments' is non-nil. | |
2989 | ||
2990 | If the beginning or end of (the accessible part of) the buffer is reached | |
2991 | in the middle of a parenthetical grouping, an error is signaled. | |
2992 | If the beginning or end is reached between groupings | |
2993 | but before count is used up, nil is returned. */) | |
5842a27b | 2994 | (Lisp_Object from, Lisp_Object count) |
8489eb67 | 2995 | { |
b7826503 PJ |
2996 | CHECK_NUMBER (from); |
2997 | CHECK_NUMBER (count); | |
8489eb67 RS |
2998 | |
2999 | return scan_lists (XINT (from), XINT (count), 0, 1); | |
3000 | } | |
3001 | ||
3002 | DEFUN ("backward-prefix-chars", Fbackward_prefix_chars, Sbackward_prefix_chars, | |
fdb82f93 PJ |
3003 | 0, 0, 0, |
3004 | doc: /* Move point backward over any number of chars with prefix syntax. | |
3005 | This includes chars with "quote" or "prefix" syntax (' or p). */) | |
5842a27b | 3006 | (void) |
8489eb67 | 3007 | { |
d311d28c PE |
3008 | ptrdiff_t beg = BEGV; |
3009 | ptrdiff_t opoint = PT; | |
3010 | ptrdiff_t opoint_byte = PT_BYTE; | |
3011 | ptrdiff_t pos = PT; | |
3012 | ptrdiff_t pos_byte = PT_BYTE; | |
93da5fff | 3013 | int c; |
93da5fff | 3014 | |
7d0393cf | 3015 | if (pos <= beg) |
195d1361 | 3016 | { |
f902a008 RS |
3017 | SET_PT_BOTH (opoint, opoint_byte); |
3018 | ||
3019 | return Qnil; | |
195d1361 | 3020 | } |
8489eb67 | 3021 | |
f902a008 RS |
3022 | SETUP_SYNTAX_TABLE (pos, -1); |
3023 | ||
6a140a74 RS |
3024 | DEC_BOTH (pos, pos_byte); |
3025 | ||
1fd3172d | 3026 | while (!char_quoted (pos, pos_byte) |
195d1361 | 3027 | /* Previous statement updates syntax table. */ |
b7dbcc19 | 3028 | && ((c = FETCH_CHAR_AS_MULTIBYTE (pos_byte), SYNTAX (c) == Squote) |
45b683a1 | 3029 | || syntax_prefix_flag_p (c))) |
93da5fff | 3030 | { |
1fd3172d RS |
3031 | opoint = pos; |
3032 | opoint_byte = pos_byte; | |
3033 | ||
3034 | if (pos + 1 > beg) | |
3035 | DEC_BOTH (pos, pos_byte); | |
93da5fff | 3036 | } |
8489eb67 | 3037 | |
6a140a74 | 3038 | SET_PT_BOTH (opoint, opoint_byte); |
8489eb67 RS |
3039 | |
3040 | return Qnil; | |
3041 | } | |
3042 | \f | |
6a140a74 | 3043 | /* Parse forward from FROM / FROM_BYTE to END, |
e5d4f4dc RS |
3044 | assuming that FROM has state OLDSTATE (nil means FROM is start of function), |
3045 | and return a description of the state of the parse at END. | |
1fc71008 | 3046 | If STOPBEFORE, stop at the start of an atom. |
644ea4df RS |
3047 | If COMMENTSTOP is 1, stop at the start of a comment. |
3048 | If COMMENTSTOP is -1, stop at the start or end of a comment, | |
3049 | after the beginning of a string, or after the end of a string. */ | |
8489eb67 | 3050 | |
340f92b5 | 3051 | static void |
d5a3eaaf | 3052 | scan_sexps_forward (struct lisp_parse_state *stateptr, |
d311d28c | 3053 | ptrdiff_t from, ptrdiff_t from_byte, ptrdiff_t end, |
1fc71008 | 3054 | EMACS_INT targetdepth, bool stopbefore, |
d5a3eaaf | 3055 | Lisp_Object oldstate, int commentstop) |
8489eb67 RS |
3056 | { |
3057 | struct lisp_parse_state state; | |
1fc71008 | 3058 | enum syntaxcode code; |
95ff8dfc | 3059 | int c1; |
1fc71008 | 3060 | bool comnested; |
d311d28c | 3061 | struct level { ptrdiff_t last, prev; }; |
8489eb67 | 3062 | struct level levelstart[100]; |
1fc71008 | 3063 | struct level *curlevel = levelstart; |
8489eb67 | 3064 | struct level *endlevel = levelstart + 100; |
1fc71008 | 3065 | EMACS_INT depth; /* Paren depth of current scanning location. |
8489eb67 RS |
3066 | level - levelstart equals this except |
3067 | when the depth becomes negative. */ | |
d311d28c | 3068 | EMACS_INT mindepth; /* Lowest DEPTH value seen. */ |
1fc71008 | 3069 | bool start_quoted = 0; /* True means starting after a char quote. */ |
8489eb67 | 3070 | Lisp_Object tem; |
d311d28c PE |
3071 | ptrdiff_t prev_from; /* Keep one character before FROM. */ |
3072 | ptrdiff_t prev_from_byte; | |
1fd3172d | 3073 | int prev_from_syntax; |
1fc71008 PE |
3074 | bool boundary_stop = commentstop == -1; |
3075 | bool nofence; | |
3076 | bool found; | |
d311d28c | 3077 | ptrdiff_t out_bytepos, out_charpos; |
7c7ca3d2 | 3078 | int temp; |
93da5fff KH |
3079 | |
3080 | prev_from = from; | |
6a140a74 RS |
3081 | prev_from_byte = from_byte; |
3082 | if (from != BEGV) | |
3083 | DEC_BOTH (prev_from, prev_from_byte); | |
93da5fff KH |
3084 | |
3085 | /* Use this macro instead of `from++'. */ | |
6a140a74 RS |
3086 | #define INC_FROM \ |
3087 | do { prev_from = from; \ | |
3088 | prev_from_byte = from_byte; \ | |
327719ee | 3089 | temp = FETCH_CHAR_AS_MULTIBYTE (prev_from_byte); \ |
ab229fdd | 3090 | prev_from_syntax = SYNTAX_WITH_FLAGS (temp); \ |
ef316cf0 | 3091 | INC_BOTH (from, from_byte); \ |
3ceb4629 SM |
3092 | if (from < end) \ |
3093 | UPDATE_SYNTAX_TABLE_FORWARD (from); \ | |
6a140a74 | 3094 | } while (0) |
8489eb67 RS |
3095 | |
3096 | immediate_quit = 1; | |
3097 | QUIT; | |
3098 | ||
265a9e55 | 3099 | if (NILP (oldstate)) |
8489eb67 RS |
3100 | { |
3101 | depth = 0; | |
3102 | state.instring = -1; | |
3103 | state.incomment = 0; | |
195d1361 RS |
3104 | state.comstyle = 0; /* comment style a by default. */ |
3105 | state.comstr_start = -1; /* no comment/string seen. */ | |
8489eb67 RS |
3106 | } |
3107 | else | |
3108 | { | |
3109 | tem = Fcar (oldstate); | |
265a9e55 | 3110 | if (!NILP (tem)) |
8489eb67 RS |
3111 | depth = XINT (tem); |
3112 | else | |
3113 | depth = 0; | |
3114 | ||
3115 | oldstate = Fcdr (oldstate); | |
3116 | oldstate = Fcdr (oldstate); | |
3117 | oldstate = Fcdr (oldstate); | |
3118 | tem = Fcar (oldstate); | |
195d1361 | 3119 | /* Check whether we are inside string_fence-style string: */ |
7d0393cf | 3120 | state.instring = (!NILP (tem) |
d311d28c | 3121 | ? (CHARACTERP (tem) ? XFASTINT (tem) : ST_STRING_STYLE) |
e76f1c44 | 3122 | : -1); |
8489eb67 RS |
3123 | |
3124 | oldstate = Fcdr (oldstate); | |
3125 | tem = Fcar (oldstate); | |
e76f1c44 GM |
3126 | state.incomment = (!NILP (tem) |
3127 | ? (INTEGERP (tem) ? XINT (tem) : -1) | |
95ff8dfc | 3128 | : 0); |
8489eb67 RS |
3129 | |
3130 | oldstate = Fcdr (oldstate); | |
3131 | tem = Fcar (oldstate); | |
265a9e55 | 3132 | start_quoted = !NILP (tem); |
e5d4f4dc | 3133 | |
95ff8dfc | 3134 | /* if the eighth element of the list is nil, we are in comment |
195d1361 RS |
3135 | style a. If it is non-nil, we are in comment style b */ |
3136 | oldstate = Fcdr (oldstate); | |
e5d4f4dc | 3137 | oldstate = Fcdr (oldstate); |
195d1361 | 3138 | tem = Fcar (oldstate); |
c5683ceb SM |
3139 | state.comstyle = (NILP (tem) |
3140 | ? 0 | |
d311d28c PE |
3141 | : (RANGED_INTEGERP (0, tem, ST_COMMENT_STYLE) |
3142 | ? XINT (tem) | |
3143 | : ST_COMMENT_STYLE)); | |
195d1361 | 3144 | |
e5d4f4dc | 3145 | oldstate = Fcdr (oldstate); |
e5d4f4dc | 3146 | tem = Fcar (oldstate); |
d311d28c PE |
3147 | state.comstr_start = |
3148 | RANGED_INTEGERP (PTRDIFF_MIN, tem, PTRDIFF_MAX) ? XINT (tem) : -1; | |
1a102a58 RS |
3149 | oldstate = Fcdr (oldstate); |
3150 | tem = Fcar (oldstate); | |
3151 | while (!NILP (tem)) /* >= second enclosing sexps. */ | |
3152 | { | |
d311d28c PE |
3153 | Lisp_Object temhd = Fcar (tem); |
3154 | if (RANGED_INTEGERP (PTRDIFF_MIN, temhd, PTRDIFF_MAX)) | |
3155 | curlevel->last = XINT (temhd); | |
1a102a58 | 3156 | if (++curlevel == endlevel) |
02010917 | 3157 | curlevel--; /* error ("Nesting too deep for parser"); */ |
1a102a58 RS |
3158 | curlevel->prev = -1; |
3159 | curlevel->last = -1; | |
3160 | tem = Fcdr (tem); | |
3161 | } | |
8489eb67 RS |
3162 | } |
3163 | state.quoted = 0; | |
3164 | mindepth = depth; | |
3165 | ||
3166 | curlevel->prev = -1; | |
3167 | curlevel->last = -1; | |
3168 | ||
326b283b | 3169 | SETUP_SYNTAX_TABLE (prev_from, 1); |
ab229fdd AS |
3170 | temp = FETCH_CHAR (prev_from_byte); |
3171 | prev_from_syntax = SYNTAX_WITH_FLAGS (temp); | |
3ceb4629 | 3172 | UPDATE_SYNTAX_TABLE_FORWARD (from); |
326b283b | 3173 | |
195d1361 | 3174 | /* Enter the loop at a place appropriate for initial state. */ |
8489eb67 | 3175 | |
326b283b RS |
3176 | if (state.incomment) |
3177 | goto startincomment; | |
8489eb67 RS |
3178 | if (state.instring >= 0) |
3179 | { | |
195d1361 | 3180 | nofence = state.instring != ST_STRING_STYLE; |
326b283b RS |
3181 | if (start_quoted) |
3182 | goto startquotedinstring; | |
8489eb67 RS |
3183 | goto startinstring; |
3184 | } | |
326b283b RS |
3185 | else if (start_quoted) |
3186 | goto startquoted; | |
1fd3172d | 3187 | |
8489eb67 RS |
3188 | while (from < end) |
3189 | { | |
c5683ceb | 3190 | int syntax; |
93da5fff | 3191 | INC_FROM; |
1fd3172d | 3192 | code = prev_from_syntax & 0xff; |
4c920633 | 3193 | |
c5d4b96f SM |
3194 | if (from < end |
3195 | && SYNTAX_FLAGS_COMSTART_FIRST (prev_from_syntax) | |
3196 | && (c1 = FETCH_CHAR (from_byte), | |
c5683ceb SM |
3197 | syntax = SYNTAX_WITH_FLAGS (c1), |
3198 | SYNTAX_FLAGS_COMSTART_SECOND (syntax))) | |
c5d4b96f SM |
3199 | /* Duplicate code to avoid a complex if-expression |
3200 | which causes trouble for the SGI compiler. */ | |
95ff8dfc | 3201 | { |
c5d4b96f SM |
3202 | /* Record the comment style we have entered so that only |
3203 | the comment-end sequence of the same style actually | |
3204 | terminates the comment section. */ | |
c5683ceb SM |
3205 | state.comstyle |
3206 | = SYNTAX_FLAGS_COMMENT_STYLE (syntax, prev_from_syntax); | |
1fc71008 PE |
3207 | comnested = (SYNTAX_FLAGS_COMMENT_NESTED (prev_from_syntax) |
3208 | | SYNTAX_FLAGS_COMMENT_NESTED (syntax)); | |
c5d4b96f | 3209 | state.incomment = comnested ? 1 : -1; |
95ff8dfc | 3210 | state.comstr_start = prev_from; |
c5d4b96f SM |
3211 | INC_FROM; |
3212 | code = Scomment; | |
95ff8dfc | 3213 | } |
4c920633 | 3214 | else if (code == Scomment_fence) |
e5d4f4dc RS |
3215 | { |
3216 | /* Record the comment style we have entered so that only | |
3217 | the comment-end sequence of the same style actually | |
3218 | terminates the comment section. */ | |
95ff8dfc RS |
3219 | state.comstyle = ST_COMMENT_STYLE; |
3220 | state.incomment = -1; | |
195d1361 | 3221 | state.comstr_start = prev_from; |
e5d4f4dc | 3222 | code = Scomment; |
e5d4f4dc | 3223 | } |
c5d4b96f SM |
3224 | else if (code == Scomment) |
3225 | { | |
c5683ceb | 3226 | state.comstyle = SYNTAX_FLAGS_COMMENT_STYLE (prev_from_syntax, 0); |
c5d4b96f SM |
3227 | state.incomment = (SYNTAX_FLAGS_COMMENT_NESTED (prev_from_syntax) ? |
3228 | 1 : -1); | |
3229 | state.comstr_start = prev_from; | |
3230 | } | |
e5d4f4dc | 3231 | |
1fd3172d | 3232 | if (SYNTAX_FLAGS_PREFIX (prev_from_syntax)) |
8489eb67 | 3233 | continue; |
7393bcbb | 3234 | switch (code) |
8489eb67 RS |
3235 | { |
3236 | case Sescape: | |
3237 | case Scharquote: | |
3238 | if (stopbefore) goto stop; /* this arg means stop at sexp start */ | |
93da5fff | 3239 | curlevel->last = prev_from; |
8489eb67 RS |
3240 | startquoted: |
3241 | if (from == end) goto endquoted; | |
93da5fff | 3242 | INC_FROM; |
8489eb67 RS |
3243 | goto symstarted; |
3244 | /* treat following character as a word constituent */ | |
3245 | case Sword: | |
3246 | case Ssymbol: | |
3247 | if (stopbefore) goto stop; /* this arg means stop at sexp start */ | |
93da5fff | 3248 | curlevel->last = prev_from; |
8489eb67 RS |
3249 | symstarted: |
3250 | while (from < end) | |
3251 | { | |
1fc71008 | 3252 | int symchar = FETCH_CHAR_AS_MULTIBYTE (from_byte); |
45b683a1 | 3253 | switch (SYNTAX (symchar)) |
8489eb67 RS |
3254 | { |
3255 | case Scharquote: | |
3256 | case Sescape: | |
93da5fff | 3257 | INC_FROM; |
8489eb67 RS |
3258 | if (from == end) goto endquoted; |
3259 | break; | |
3260 | case Sword: | |
3261 | case Ssymbol: | |
3262 | case Squote: | |
3263 | break; | |
3264 | default: | |
3265 | goto symdone; | |
3266 | } | |
93da5fff | 3267 | INC_FROM; |
8489eb67 RS |
3268 | } |
3269 | symdone: | |
3270 | curlevel->prev = curlevel->last; | |
3271 | break; | |
3272 | ||
240c43e8 | 3273 | case Scomment_fence: /* Can't happen because it's handled above. */ |
8489eb67 | 3274 | case Scomment: |
195d1361 | 3275 | if (commentstop || boundary_stop) goto done; |
d355bd8a SM |
3276 | startincomment: |
3277 | /* The (from == BEGV) test was to enter the loop in the middle so | |
95ff8dfc | 3278 | that we find a 2-char comment ender even if we start in the |
e6365855 | 3279 | middle of it. We don't want to do that if we're just at the |
d355bd8a | 3280 | beginning of the comment (think of (*) ... (*)). */ |
95ff8dfc RS |
3281 | found = forw_comment (from, from_byte, end, |
3282 | state.incomment, state.comstyle, | |
e6365855 SM |
3283 | (from == BEGV || from < state.comstr_start + 3) |
3284 | ? 0 : prev_from_syntax, | |
95ff8dfc RS |
3285 | &out_charpos, &out_bytepos, &state.incomment); |
3286 | from = out_charpos; from_byte = out_bytepos; | |
3287 | /* Beware! prev_from and friends are invalid now. | |
3288 | Luckily, the `done' doesn't use them and the INC_FROM | |
3289 | sets them to a sane value without looking at them. */ | |
3290 | if (!found) goto done; | |
d355bd8a | 3291 | INC_FROM; |
8489eb67 | 3292 | state.incomment = 0; |
e5d4f4dc | 3293 | state.comstyle = 0; /* reset the comment style */ |
195d1361 | 3294 | if (boundary_stop) goto done; |
8489eb67 RS |
3295 | break; |
3296 | ||
3297 | case Sopen: | |
3298 | if (stopbefore) goto stop; /* this arg means stop at sexp start */ | |
3299 | depth++; | |
3300 | /* curlevel++->last ran into compiler bug on Apollo */ | |
93da5fff | 3301 | curlevel->last = prev_from; |
8489eb67 | 3302 | if (++curlevel == endlevel) |
02010917 | 3303 | curlevel--; /* error ("Nesting too deep for parser"); */ |
8489eb67 RS |
3304 | curlevel->prev = -1; |
3305 | curlevel->last = -1; | |
30844415 | 3306 | if (targetdepth == depth) goto done; |
8489eb67 RS |
3307 | break; |
3308 | ||
3309 | case Sclose: | |
3310 | depth--; | |
3311 | if (depth < mindepth) | |
3312 | mindepth = depth; | |
3313 | if (curlevel != levelstart) | |
3314 | curlevel--; | |
3315 | curlevel->prev = curlevel->last; | |
30844415 | 3316 | if (targetdepth == depth) goto done; |
8489eb67 RS |
3317 | break; |
3318 | ||
3319 | case Sstring: | |
195d1361 RS |
3320 | case Sstring_fence: |
3321 | state.comstr_start = from - 1; | |
8489eb67 | 3322 | if (stopbefore) goto stop; /* this arg means stop at sexp start */ |
93da5fff | 3323 | curlevel->last = prev_from; |
7d0393cf | 3324 | state.instring = (code == Sstring |
b7dbcc19 | 3325 | ? (FETCH_CHAR_AS_MULTIBYTE (prev_from_byte)) |
195d1361 RS |
3326 | : ST_STRING_STYLE); |
3327 | if (boundary_stop) goto done; | |
8489eb67 | 3328 | startinstring: |
195d1361 | 3329 | { |
644ea4df | 3330 | nofence = state.instring != ST_STRING_STYLE; |
7d0393cf | 3331 | |
644ea4df RS |
3332 | while (1) |
3333 | { | |
3334 | int c; | |
1fc71008 | 3335 | enum syntaxcode c_code; |
195d1361 | 3336 | |
644ea4df | 3337 | if (from >= end) goto done; |
b7dbcc19 | 3338 | c = FETCH_CHAR_AS_MULTIBYTE (from_byte); |
1fc71008 | 3339 | c_code = SYNTAX (c); |
ccf89641 | 3340 | |
1fc71008 | 3341 | /* Check C_CODE here so that if the char has |
ccf89641 KH |
3342 | a syntax-table property which says it is NOT |
3343 | a string character, it does not end the string. */ | |
1fc71008 | 3344 | if (nofence && c == state.instring && c_code == Sstring) |
ccf89641 KH |
3345 | break; |
3346 | ||
1fc71008 | 3347 | switch (c_code) |
644ea4df RS |
3348 | { |
3349 | case Sstring_fence: | |
3350 | if (!nofence) goto string_end; | |
3351 | break; | |
3352 | case Scharquote: | |
3353 | case Sescape: | |
3354 | INC_FROM; | |
3355 | startquotedinstring: | |
3356 | if (from >= end) goto endquoted; | |
195d1361 | 3357 | } |
644ea4df RS |
3358 | INC_FROM; |
3359 | } | |
195d1361 RS |
3360 | } |
3361 | string_end: | |
8489eb67 RS |
3362 | state.instring = -1; |
3363 | curlevel->prev = curlevel->last; | |
93da5fff | 3364 | INC_FROM; |
195d1361 | 3365 | if (boundary_stop) goto done; |
8489eb67 RS |
3366 | break; |
3367 | ||
3368 | case Smath: | |
240c43e8 SM |
3369 | /* FIXME: We should do something with it. */ |
3370 | break; | |
3371 | default: | |
3372 | /* Ignore whitespace, punctuation, quote, endcomment. */ | |
8489eb67 RS |
3373 | break; |
3374 | } | |
3375 | } | |
3376 | goto done; | |
3377 | ||
3378 | stop: /* Here if stopping before start of sexp. */ | |
93da5fff | 3379 | from = prev_from; /* We have just fetched the char that starts it; */ |
68643cde | 3380 | from_byte = prev_from_byte; |
8489eb67 RS |
3381 | goto done; /* but return the position before it. */ |
3382 | ||
3383 | endquoted: | |
3384 | state.quoted = 1; | |
3385 | done: | |
3386 | state.depth = depth; | |
3387 | state.mindepth = mindepth; | |
3388 | state.thislevelstart = curlevel->prev; | |
3389 | state.prevlevelstart | |
3390 | = (curlevel == levelstart) ? -1 : (curlevel - 1)->last; | |
3391 | state.location = from; | |
8ca30920 | 3392 | state.location_byte = from_byte; |
1a102a58 | 3393 | state.levelstarts = Qnil; |
eb4d412d PE |
3394 | while (curlevel > levelstart) |
3395 | state.levelstarts = Fcons (make_number ((--curlevel)->last), | |
3396 | state.levelstarts); | |
8489eb67 RS |
3397 | immediate_quit = 0; |
3398 | ||
e5d4f4dc | 3399 | *stateptr = state; |
8489eb67 RS |
3400 | } |
3401 | ||
c81a3712 | 3402 | DEFUN ("parse-partial-sexp", Fparse_partial_sexp, Sparse_partial_sexp, 2, 6, 0, |
fdb82f93 PJ |
3403 | doc: /* Parse Lisp syntax starting at FROM until TO; return status of parse at TO. |
3404 | Parsing stops at TO or when certain criteria are met; | |
3405 | point is set to where parsing stops. | |
3406 | If fifth arg OLDSTATE is omitted or nil, | |
3407 | parsing assumes that FROM is the beginning of a function. | |
0e6228bc | 3408 | Value is a list of elements describing final state of parsing: |
fdb82f93 PJ |
3409 | 0. depth in parens. |
3410 | 1. character address of start of innermost containing list; nil if none. | |
3411 | 2. character address of start of last complete sexp terminated. | |
3412 | 3. non-nil if inside a string. | |
3413 | (it is the character that will terminate the string, | |
3414 | or t if the string should be terminated by a generic string delimiter.) | |
7d0393cf | 3415 | 4. nil if outside a comment, t if inside a non-nestable comment, |
fdb82f93 PJ |
3416 | else an integer (the current comment nesting). |
3417 | 5. t if following a quote character. | |
3418 | 6. the minimum paren-depth encountered during this scan. | |
c5683ceb | 3419 | 7. style of comment, if any. |
fdb82f93 PJ |
3420 | 8. character address of start of comment or string; nil if not in one. |
3421 | 9. Intermediate data for continuation of parsing (subject to change). | |
3422 | If third arg TARGETDEPTH is non-nil, parsing stops if the depth | |
3423 | in parentheses becomes equal to TARGETDEPTH. | |
3424 | Fourth arg STOPBEFORE non-nil means stop when come to | |
3425 | any character that starts a sexp. | |
0e6228bc | 3426 | Fifth arg OLDSTATE is a list like what this function returns. |
fdb82f93 | 3427 | It is used to initialize the state of the parse. Elements number 1, 2, 6 |
d311d28c | 3428 | are ignored. |
fdb82f93 PJ |
3429 | Sixth arg COMMENTSTOP non-nil means stop at the start of a comment. |
3430 | If it is symbol `syntax-table', stop after the start of a comment or a | |
3431 | string, or after end of a comment or a string. */) | |
8ca30920 DA |
3432 | (Lisp_Object from, Lisp_Object to, Lisp_Object targetdepth, |
3433 | Lisp_Object stopbefore, Lisp_Object oldstate, Lisp_Object commentstop) | |
8489eb67 RS |
3434 | { |
3435 | struct lisp_parse_state state; | |
d311d28c | 3436 | EMACS_INT target; |
8489eb67 | 3437 | |
265a9e55 | 3438 | if (!NILP (targetdepth)) |
8489eb67 | 3439 | { |
b7826503 | 3440 | CHECK_NUMBER (targetdepth); |
8489eb67 RS |
3441 | target = XINT (targetdepth); |
3442 | } | |
3443 | else | |
d311d28c | 3444 | target = TYPE_MINIMUM (EMACS_INT); /* We won't reach this depth */ |
8489eb67 RS |
3445 | |
3446 | validate_region (&from, &to); | |
6a140a74 RS |
3447 | scan_sexps_forward (&state, XINT (from), CHAR_TO_BYTE (XINT (from)), |
3448 | XINT (to), | |
c81a3712 | 3449 | target, !NILP (stopbefore), oldstate, |
7d0393cf | 3450 | (NILP (commentstop) |
195d1361 | 3451 | ? 0 : (EQ (commentstop, Qsyntax_table) ? -1 : 1))); |
8489eb67 | 3452 | |
8ca30920 | 3453 | SET_PT_BOTH (state.location, state.location_byte); |
7d0393cf | 3454 | |
8489eb67 | 3455 | return Fcons (make_number (state.depth), |
c5683ceb SM |
3456 | Fcons (state.prevlevelstart < 0 |
3457 | ? Qnil : make_number (state.prevlevelstart), | |
3458 | Fcons (state.thislevelstart < 0 | |
3459 | ? Qnil : make_number (state.thislevelstart), | |
7d0393cf JB |
3460 | Fcons (state.instring >= 0 |
3461 | ? (state.instring == ST_STRING_STYLE | |
195d1361 | 3462 | ? Qt : make_number (state.instring)) : Qnil, |
95ff8dfc RS |
3463 | Fcons (state.incomment < 0 ? Qt : |
3464 | (state.incomment == 0 ? Qnil : | |
3465 | make_number (state.incomment)), | |
8489eb67 | 3466 | Fcons (state.quoted ? Qt : Qnil, |
bff37d2a | 3467 | Fcons (make_number (state.mindepth), |
7d0393cf | 3468 | Fcons ((state.comstyle |
bff37d2a | 3469 | ? (state.comstyle == ST_COMMENT_STYLE |
c5683ceb SM |
3470 | ? Qsyntax_table |
3471 | : make_number (state.comstyle)) | |
3472 | : Qnil), | |
0beaf54f DL |
3473 | Fcons (((state.incomment |
3474 | || (state.instring >= 0)) | |
bff37d2a RS |
3475 | ? make_number (state.comstr_start) |
3476 | : Qnil), | |
1a102a58 | 3477 | Fcons (state.levelstarts, Qnil)))))))))); |
8489eb67 RS |
3478 | } |
3479 | \f | |
dfcf069d | 3480 | void |
971de7fb | 3481 | init_syntax_once (void) |
8489eb67 | 3482 | { |
78f9a1f7 | 3483 | register int i, c; |
8ea151b2 | 3484 | Lisp_Object temp; |
8489eb67 | 3485 | |
5ebaddf5 | 3486 | /* This has to be done here, before we call Fmake_char_table. */ |
cd3520a4 | 3487 | DEFSYM (Qsyntax_table, "syntax-table"); |
5ebaddf5 | 3488 | |
c76826d9 DA |
3489 | /* This variable is DEFSYMed in alloc.c and not initialized yet, so |
3490 | intern it here. NOTE: you must guarantee that init_syntax_once | |
3491 | is called before all other users of this variable. */ | |
d67b4f80 | 3492 | Qchar_table_extra_slots = intern_c_string ("char-table-extra-slots"); |
5ebaddf5 | 3493 | |
93da5fff | 3494 | /* Create objects which can be shared among syntax tables. */ |
25721f5b DA |
3495 | Vsyntax_code_object = make_uninit_vector (Smax); |
3496 | for (i = 0; i < Smax; i++) | |
28be1ada | 3497 | ASET (Vsyntax_code_object, i, Fcons (make_number (i), Qnil)); |
93da5fff | 3498 | |
5ebaddf5 RS |
3499 | /* Now we are ready to set up this property, so we can |
3500 | create syntax tables. */ | |
3501 | Fput (Qsyntax_table, Qchar_table_extra_slots, make_number (0)); | |
3502 | ||
1fc71008 | 3503 | temp = AREF (Vsyntax_code_object, Swhitespace); |
8489eb67 | 3504 | |
5ebaddf5 | 3505 | Vstandard_syntax_table = Fmake_char_table (Qsyntax_table, temp); |
8489eb67 | 3506 | |
aa7b08b4 | 3507 | /* Control characters should not be whitespace. */ |
1fc71008 | 3508 | temp = AREF (Vsyntax_code_object, Spunct); |
aa7b08b4 RS |
3509 | for (i = 0; i <= ' ' - 1; i++) |
3510 | SET_RAW_SYNTAX_ENTRY (Vstandard_syntax_table, i, temp); | |
3511 | SET_RAW_SYNTAX_ENTRY (Vstandard_syntax_table, 0177, temp); | |
3512 | ||
3513 | /* Except that a few really are whitespace. */ | |
1fc71008 | 3514 | temp = AREF (Vsyntax_code_object, Swhitespace); |
aa7b08b4 RS |
3515 | SET_RAW_SYNTAX_ENTRY (Vstandard_syntax_table, ' ', temp); |
3516 | SET_RAW_SYNTAX_ENTRY (Vstandard_syntax_table, '\t', temp); | |
3517 | SET_RAW_SYNTAX_ENTRY (Vstandard_syntax_table, '\n', temp); | |
3518 | SET_RAW_SYNTAX_ENTRY (Vstandard_syntax_table, 015, temp); | |
3519 | SET_RAW_SYNTAX_ENTRY (Vstandard_syntax_table, 014, temp); | |
3520 | ||
1fc71008 | 3521 | temp = AREF (Vsyntax_code_object, Sword); |
8489eb67 | 3522 | for (i = 'a'; i <= 'z'; i++) |
8ea151b2 | 3523 | SET_RAW_SYNTAX_ENTRY (Vstandard_syntax_table, i, temp); |
8489eb67 | 3524 | for (i = 'A'; i <= 'Z'; i++) |
8ea151b2 | 3525 | SET_RAW_SYNTAX_ENTRY (Vstandard_syntax_table, i, temp); |
8489eb67 | 3526 | for (i = '0'; i <= '9'; i++) |
8ea151b2 RS |
3527 | SET_RAW_SYNTAX_ENTRY (Vstandard_syntax_table, i, temp); |
3528 | ||
3529 | SET_RAW_SYNTAX_ENTRY (Vstandard_syntax_table, '$', temp); | |
3530 | SET_RAW_SYNTAX_ENTRY (Vstandard_syntax_table, '%', temp); | |
3531 | ||
3532 | SET_RAW_SYNTAX_ENTRY (Vstandard_syntax_table, '(', | |
3533 | Fcons (make_number (Sopen), make_number (')'))); | |
3534 | SET_RAW_SYNTAX_ENTRY (Vstandard_syntax_table, ')', | |
3535 | Fcons (make_number (Sclose), make_number ('('))); | |
3536 | SET_RAW_SYNTAX_ENTRY (Vstandard_syntax_table, '[', | |
3537 | Fcons (make_number (Sopen), make_number (']'))); | |
3538 | SET_RAW_SYNTAX_ENTRY (Vstandard_syntax_table, ']', | |
3539 | Fcons (make_number (Sclose), make_number ('['))); | |
3540 | SET_RAW_SYNTAX_ENTRY (Vstandard_syntax_table, '{', | |
3541 | Fcons (make_number (Sopen), make_number ('}'))); | |
3542 | SET_RAW_SYNTAX_ENTRY (Vstandard_syntax_table, '}', | |
3543 | Fcons (make_number (Sclose), make_number ('{'))); | |
3544 | SET_RAW_SYNTAX_ENTRY (Vstandard_syntax_table, '"', | |
1fc71008 | 3545 | Fcons (make_number (Sstring), Qnil)); |
8ea151b2 | 3546 | SET_RAW_SYNTAX_ENTRY (Vstandard_syntax_table, '\\', |
1fc71008 | 3547 | Fcons (make_number (Sescape), Qnil)); |
8ea151b2 | 3548 | |
1fc71008 | 3549 | temp = AREF (Vsyntax_code_object, Ssymbol); |
8489eb67 | 3550 | for (i = 0; i < 10; i++) |
78f9a1f7 KH |
3551 | { |
3552 | c = "_-+*/&|<>="[i]; | |
3553 | SET_RAW_SYNTAX_ENTRY (Vstandard_syntax_table, c, temp); | |
3554 | } | |
8489eb67 | 3555 | |
1fc71008 | 3556 | temp = AREF (Vsyntax_code_object, Spunct); |
8489eb67 | 3557 | for (i = 0; i < 12; i++) |
78f9a1f7 KH |
3558 | { |
3559 | c = ".,;:?!#@~^'`"[i]; | |
3560 | SET_RAW_SYNTAX_ENTRY (Vstandard_syntax_table, c, temp); | |
3561 | } | |
bd25db08 KH |
3562 | |
3563 | /* All multibyte characters have syntax `word' by default. */ | |
1fc71008 | 3564 | temp = AREF (Vsyntax_code_object, Sword); |
5c7b02ab | 3565 | char_table_set_range (Vstandard_syntax_table, 0x80, MAX_CHAR, temp); |
8489eb67 RS |
3566 | } |
3567 | ||
dfcf069d | 3568 | void |
971de7fb | 3569 | syms_of_syntax (void) |
8489eb67 | 3570 | { |
cd3520a4 | 3571 | DEFSYM (Qsyntax_table_p, "syntax-table-p"); |
8489eb67 | 3572 | |
93da5fff KH |
3573 | staticpro (&Vsyntax_code_object); |
3574 | ||
a2994c46 KS |
3575 | staticpro (&gl_state.object); |
3576 | staticpro (&gl_state.global_code); | |
3577 | staticpro (&gl_state.current_syntax_table); | |
3578 | staticpro (&gl_state.old_prop); | |
3579 | ||
3580 | /* Defined in regex.c */ | |
3581 | staticpro (&re_match_object); | |
3582 | ||
cd3520a4 | 3583 | DEFSYM (Qscan_error, "scan-error"); |
7bf5e9e4 | 3584 | Fput (Qscan_error, Qerror_conditions, |
3438fe21 | 3585 | listn (CONSTYPE_PURE, 2, Qscan_error, Qerror)); |
7bf5e9e4 | 3586 | Fput (Qscan_error, Qerror_message, |
2a0213a6 | 3587 | build_pure_c_string ("Scan error")); |
7bf5e9e4 | 3588 | |
29208e82 | 3589 | DEFVAR_BOOL ("parse-sexp-ignore-comments", parse_sexp_ignore_comments, |
fdb82f93 | 3590 | doc: /* Non-nil means `forward-sexp', etc., should treat comments as whitespace. */); |
8489eb67 | 3591 | |
29208e82 | 3592 | DEFVAR_BOOL ("parse-sexp-lookup-properties", parse_sexp_lookup_properties, |
fdb82f93 PJ |
3593 | doc: /* Non-nil means `forward-sexp', etc., obey `syntax-table' property. |
3594 | Otherwise, that text property is simply ignored. | |
3595 | See the info node `(elisp)Syntax Properties' for a description of the | |
3596 | `syntax-table' property. */); | |
195d1361 | 3597 | |
8489eb67 | 3598 | words_include_escapes = 0; |
29208e82 | 3599 | DEFVAR_BOOL ("words-include-escapes", words_include_escapes, |
fdb82f93 | 3600 | doc: /* Non-nil means `forward-word', etc., should treat escape chars part of words. */); |
8489eb67 | 3601 | |
29208e82 | 3602 | DEFVAR_BOOL ("multibyte-syntax-as-symbol", multibyte_syntax_as_symbol, |
fdb82f93 | 3603 | doc: /* Non-nil means `scan-sexps' treats all multibyte characters as symbol. */); |
bd25db08 KH |
3604 | multibyte_syntax_as_symbol = 0; |
3605 | ||
f4ed767f | 3606 | DEFVAR_BOOL ("open-paren-in-column-0-is-defun-start", |
29208e82 | 3607 | open_paren_in_column_0_is_defun_start, |
fb7ada5f | 3608 | doc: /* Non-nil means an open paren in column 0 denotes the start of a defun. */); |
f4ed767f GM |
3609 | open_paren_in_column_0_is_defun_start = 1; |
3610 | ||
8f924df7 KH |
3611 | |
3612 | DEFVAR_LISP ("find-word-boundary-function-table", | |
29208e82 | 3613 | Vfind_word_boundary_function_table, |
869bb237 | 3614 | doc: /* |
8f924df7 | 3615 | Char table of functions to search for the word boundary. |
869bb237 KH |
3616 | Each function is called with two arguments; POS and LIMIT. |
3617 | POS and LIMIT are character positions in the current buffer. | |
3618 | ||
3619 | If POS is less than LIMIT, POS is at the first character of a word, | |
3620 | and the return value of a function is a position after the last | |
3621 | character of that word. | |
3622 | ||
3623 | If POS is not less than LIMIT, POS is at the last character of a word, | |
3624 | and the return value of a function is a position at the first | |
3625 | character of that word. | |
3626 | ||
3627 | In both cases, LIMIT bounds the search. */); | |
8f924df7 | 3628 | Vfind_word_boundary_function_table = Fmake_char_table (Qnil, Qnil); |
869bb237 | 3629 | |
8489eb67 RS |
3630 | defsubr (&Ssyntax_table_p); |
3631 | defsubr (&Ssyntax_table); | |
3632 | defsubr (&Sstandard_syntax_table); | |
3633 | defsubr (&Scopy_syntax_table); | |
3634 | defsubr (&Sset_syntax_table); | |
3635 | defsubr (&Schar_syntax); | |
beefa22e | 3636 | defsubr (&Smatching_paren); |
c65adb44 | 3637 | defsubr (&Sstring_to_syntax); |
8489eb67 | 3638 | defsubr (&Smodify_syntax_entry); |
62abe9cb | 3639 | defsubr (&Sinternal_describe_syntax_value); |
8489eb67 RS |
3640 | |
3641 | defsubr (&Sforward_word); | |
3642 | ||
195d1361 RS |
3643 | defsubr (&Sskip_chars_forward); |
3644 | defsubr (&Sskip_chars_backward); | |
3645 | defsubr (&Sskip_syntax_forward); | |
3646 | defsubr (&Sskip_syntax_backward); | |
3647 | ||
b3cfe0c8 | 3648 | defsubr (&Sforward_comment); |
8489eb67 RS |
3649 | defsubr (&Sscan_lists); |
3650 | defsubr (&Sscan_sexps); | |
3651 | defsubr (&Sbackward_prefix_chars); | |
3652 | defsubr (&Sparse_partial_sexp); | |
3653 | } |