syntax.c 98.8 KB
Newer Older
Richard M. Stallman's avatar
Richard M. Stallman committed
1
/* GNU Emacs routines to deal with syntax tables; also word and list parsing.
2
   Copyright (C) 1985, 1987, 1993, 1994, 1995, 1997, 1998, 1999, 2001,
Glenn Morris's avatar
Glenn Morris committed
3
                 2002, 2003, 2004, 2005, 2006, 2007, 2008, 2009, 2010
4
                 Free Software Foundation, Inc.
Richard M. Stallman's avatar
Richard M. Stallman committed
5 6 7

This file is part of GNU Emacs.

8
GNU Emacs is free software: you can redistribute it and/or modify
Richard M. Stallman's avatar
Richard M. Stallman committed
9
it under the terms of the GNU General Public License as published by
10 11
the Free Software Foundation, either version 3 of the License, or
(at your option) any later version.
Richard M. Stallman's avatar
Richard M. Stallman committed
12 13 14 15 16 17 18

GNU Emacs is distributed in the hope that it will be useful,
but WITHOUT ANY WARRANTY; without even the implied warranty of
MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the
GNU General Public License for more details.

You should have received a copy of the GNU General Public License
19
along with GNU Emacs.  If not, see <http://www.gnu.org/licenses/>.  */
Richard M. Stallman's avatar
Richard M. Stallman committed
20 21


22
#include <config.h>
Richard M. Stallman's avatar
Richard M. Stallman committed
23
#include <ctype.h>
24
#include <setjmp.h>
Richard M. Stallman's avatar
Richard M. Stallman committed
25 26 27
#include "lisp.h"
#include "commands.h"
#include "buffer.h"
28
#include "character.h"
Stefan Monnier's avatar
Stefan Monnier committed
29
#include "keymap.h"
Richard M. Stallman's avatar
Richard M. Stallman committed
30
#include "regex.h"
31 32 33 34

/* Make syntax table lookup grant data in gl_state.  */
#define SYNTAX_ENTRY_VIA_PROPERTY

Richard M. Stallman's avatar
Richard M. Stallman committed
35
#include "syntax.h"
36 37 38 39 40 41 42 43
#include "intervals.h"

/* We use these constants in place for comment-style and
   string-ender-char to distinguish  comments/strings started by
   comment_fence and string_fence codes.  */

#define ST_COMMENT_STYLE (256 + 1)
#define ST_STRING_STYLE (256 + 2)
Karl Heuer's avatar
Karl Heuer committed
44
#include "category.h"
Richard M. Stallman's avatar
Richard M. Stallman committed
45

46
Lisp_Object Qsyntax_table_p, Qsyntax_table, Qscan_error;
Richard M. Stallman's avatar
Richard M. Stallman committed
47 48

int words_include_escapes;
49
int parse_sexp_lookup_properties;
Richard M. Stallman's avatar
Richard M. Stallman committed
50

51 52 53
/* Nonzero means `scan-sexps' treat all multibyte characters as symbol.  */
int multibyte_syntax_as_symbol;

54 55 56 57 58
/* Used as a temporary in SYNTAX_ENTRY and other macros in syntax.h,
   if not compiled with GCC.  No need to mark it, since it is used
   only very temporarily.  */
Lisp_Object syntax_temp;

59 60 61 62 63 64
/* Non-zero means an open parenthesis in column 0 is always considered
   to be the start of a defun.  Zero means an open parenthesis in
   column 0 has no special meaning.  */

int open_paren_in_column_0_is_defun_start;

65 66 67 68
/* This is the internal form of the parse state used in parse-partial-sexp.  */

struct lisp_parse_state
  {
69 70 71 72 73 74 75 76 77 78 79 80 81 82
    int depth;	   /* Depth at end of parsing.  */
    int instring;  /* -1 if not within string, else desired terminator.  */
    int incomment; /* -1 if in unnestable comment else comment nesting */
    int comstyle;  /* comment style a=0, or b=1, or ST_COMMENT_STYLE.  */
    int quoted;	   /* Nonzero if just after an escape char at end of parsing */
    int mindepth;  /* Minimum depth seen while scanning.  */
    /* Char number of most recent start-of-expression at current level */
    EMACS_INT thislevelstart;
    /* Char number of start of containing expression */
    EMACS_INT prevlevelstart;
    EMACS_INT location;	     /* Char number at which parsing stopped.  */
    EMACS_INT comstr_start;  /* Position of last comment/string starter.  */
    Lisp_Object levelstarts; /* Char numbers of starts-of-expression
				of levels (starting from outermost).  */
83 84
  };

85 86 87
/* These variables are a cache for finding the start of a defun.
   find_start_pos is the place for which the defun start was found.
   find_start_value is the defun start position found for it.
88
   find_start_value_byte is the corresponding byte position.
89 90 91 92
   find_start_buffer is the buffer it was found in.
   find_start_begv is the BEGV value when it was found.
   find_start_modiff is the value of MODIFF when it was found.  */

93 94 95
static EMACS_INT find_start_pos;
static EMACS_INT find_start_value;
static EMACS_INT find_start_value_byte;
96
static struct buffer *find_start_buffer;
97
static EMACS_INT find_start_begv;
98
static int find_start_modiff;
99 100


101 102 103 104 105 106 107
static Lisp_Object skip_chars (int, Lisp_Object, Lisp_Object, int);
static Lisp_Object skip_syntaxes (int, Lisp_Object, Lisp_Object);
static Lisp_Object scan_lists (EMACS_INT, EMACS_INT, EMACS_INT, int);
static void scan_sexps_forward (struct lisp_parse_state *,
                                EMACS_INT, EMACS_INT, EMACS_INT, int,
                                int, Lisp_Object, int);
static int in_classes (int, Lisp_Object);
108 109 110 111


struct gl_state_s gl_state;		/* Global state of syntax parser.  */

112
INTERVAL interval_of (int, Lisp_Object);
113 114 115
#define INTERVALS_AT_ONCE 10		/* 1 + max-number of intervals
					   to scan to property-change.  */

116 117
/* Update gl_state to an appropriate interval which contains CHARPOS.  The
   sign of COUNT give the relative position of CHARPOS wrt the previously
118
   valid interval.  If INIT, only [be]_property fields of gl_state are
119
   valid at start, the rest is filled basing on OBJECT.
120

121
   `gl_state.*_i' are the intervals, and CHARPOS is further in the search
122 123
   direction than the intervals - or in an interval.  We update the
   current syntax-table basing on the property of this interval, and
124
   update the interval to start further than CHARPOS - or be
125
   NULL_INTERVAL.  We also update lim_property to be the next value of
126
   charpos to call this subroutine again - or be before/after the
127 128 129
   start/end of OBJECT.  */

void
130
update_syntax_table (int charpos, int count, int init, Lisp_Object object)
131 132
{
  Lisp_Object tmp_table;
133
  int cnt = 0, invalidate = 1;
134
  INTERVAL i;
135 136 137

  if (init)
    {
138
      gl_state.old_prop = Qnil;
139 140
      gl_state.start = gl_state.b_property;
      gl_state.stop = gl_state.e_property;
141 142
      i = interval_of (charpos, object);
      gl_state.backward_i = gl_state.forward_i = i;
143 144 145
      invalidate = 0;
      if (NULL_INTERVAL_P (i))
	return;
146
      /* interval_of updates only ->position of the return value, so
147
	 update the parents manually to speed up update_interval.  */
148
      while (!NULL_PARENT (i))
149 150
	{
	  if (AM_RIGHT_CHILD (i))
151
	    INTERVAL_PARENT (i)->position = i->position
152
	      - LEFT_TOTAL_LENGTH (i) + TOTAL_LENGTH (i) /* right end */
153 154
	      - TOTAL_LENGTH (INTERVAL_PARENT (i))
	      + LEFT_TOTAL_LENGTH (INTERVAL_PARENT (i));
155
	  else
156
	    INTERVAL_PARENT (i)->position = i->position - LEFT_TOTAL_LENGTH (i)
157
	      + TOTAL_LENGTH (i);
158
	  i = INTERVAL_PARENT (i);
159 160
	}
      i = gl_state.forward_i;
161
      gl_state.b_property = i->position - gl_state.offset;
162
      gl_state.e_property = INTERVAL_LAST_POS (i) - gl_state.offset;
163 164
      goto update;
    }
165
  i = count > 0 ? gl_state.forward_i : gl_state.backward_i;
166

167
  /* We are guaranteed to be called with CHARPOS either in i,
168
     or further off.  */
169 170
  if (NULL_INTERVAL_P (i))
    error ("Error in syntax_table logic for to-the-end intervals");
171
  else if (charpos < i->position)		/* Move left.  */
172 173
    {
      if (count > 0)
174
	error ("Error in syntax_table logic for intervals <-");
175
      /* Update the interval.  */
176
      i = update_interval (i, charpos);
177
      if (INTERVAL_LAST_POS (i) != gl_state.b_property)
178 179 180
	{
	  invalidate = 0;
	  gl_state.forward_i = i;
181
	  gl_state.e_property = INTERVAL_LAST_POS (i) - gl_state.offset;
182
	}
183
    }
184
  else if (charpos >= INTERVAL_LAST_POS (i)) /* Move right.  */
185 186
    {
      if (count < 0)
187
	error ("Error in syntax_table logic for intervals ->");
188
      /* Update the interval.  */
189
      i = update_interval (i, charpos);
190
      if (i->position != gl_state.e_property)
191 192 193
	{
	  invalidate = 0;
	  gl_state.backward_i = i;
194
	  gl_state.b_property = i->position - gl_state.offset;
195 196 197 198 199 200 201 202
	}
    }

  update:
  tmp_table = textget (i->plist, Qsyntax_table);

  if (invalidate)
    invalidate = !EQ (tmp_table, gl_state.old_prop); /* Need to invalidate? */
Juanma Barranquero's avatar
Juanma Barranquero committed
203

204 205 206
  if (invalidate)		/* Did not get to adjacent interval.  */
    {				/* with the same table => */
				/* invalidate the old range.  */
207 208 209
      if (count > 0)
	{
	  gl_state.backward_i = i;
210 211 212
	  gl_state.b_property = i->position - gl_state.offset;
	}
      else
213
	{
214
	  gl_state.forward_i = i;
215
	  gl_state.e_property = INTERVAL_LAST_POS (i) - gl_state.offset;
216 217
	}
    }
218

219
  if (!EQ (tmp_table, gl_state.old_prop))
220
    {
221 222 223 224 225
      gl_state.current_syntax_table = tmp_table;
      gl_state.old_prop = tmp_table;
      if (EQ (Fsyntax_table_p (tmp_table), Qt))
	{
	  gl_state.use_global = 0;
Juanma Barranquero's avatar
Juanma Barranquero committed
226
	}
227 228 229 230 231
      else if (CONSP (tmp_table))
	{
	  gl_state.use_global = 1;
	  gl_state.global_code = tmp_table;
	}
Juanma Barranquero's avatar
Juanma Barranquero committed
232
      else
233 234 235 236
	{
	  gl_state.use_global = 0;
	  gl_state.current_syntax_table = current_buffer->syntax_table;
	}
237 238 239 240 241 242 243
    }

  while (!NULL_INTERVAL_P (i))
    {
      if (cnt && !EQ (tmp_table, textget (i->plist, Qsyntax_table)))
	{
	  if (count > 0)
244 245 246 247 248 249
	    {
	      gl_state.e_property = i->position - gl_state.offset;
	      gl_state.forward_i = i;
	    }
	  else
	    {
250 251
	      gl_state.b_property
		= i->position + LENGTH (i) - gl_state.offset;
252 253 254
	      gl_state.backward_i = i;
	    }
	  return;
255
	}
Juanma Barranquero's avatar
Juanma Barranquero committed
256
      else if (cnt == INTERVALS_AT_ONCE)
257 258
	{
	  if (count > 0)
259
	    {
260 261 262 263 264 265
	      gl_state.e_property
		= i->position + LENGTH (i) - gl_state.offset
		/* e_property at EOB is not set to ZV but to ZV+1, so that
		   we can do INC(from);UPDATE_SYNTAX_TABLE_FORWARD without
		   having to check eob between the two.  */
		+ (NULL_INTERVAL_P (next_interval (i)) ? 1 : 0);
266 267 268 269 270 271 272 273
	      gl_state.forward_i = i;
	    }
	  else
	    {
	      gl_state.b_property = i->position - gl_state.offset;
	      gl_state.backward_i = i;
	    }
	  return;
274 275 276 277
	}
      cnt++;
      i = count > 0 ? next_interval (i) : previous_interval (i);
    }
278 279 280 281 282
  eassert (NULL_INTERVAL_P (i)); /* This property goes to the end.  */
  if (count > 0)
    gl_state.e_property = gl_state.stop;
  else
    gl_state.b_property = gl_state.start;
283 284
}

285 286 287
/* Returns TRUE if char at CHARPOS is quoted.
   Global syntax-table data should be set up already to be good at CHARPOS
   or after.  On return global syntax data is good for lookup at CHARPOS. */
288 289

static int
290
char_quoted (EMACS_INT charpos, EMACS_INT bytepos)
291 292
{
  register enum syntaxcode code;
293
  register EMACS_INT beg = BEGV;
294
  register int quoted = 0;
295
  EMACS_INT orig = charpos;
296

297
  while (charpos > beg)
298
    {
299
      int c;
300
      DEC_BOTH (charpos, bytepos);
301

302
      UPDATE_SYNTAX_TABLE_BACKWARD (charpos);
303
      c = FETCH_CHAR_AS_MULTIBYTE (bytepos);
304
      code = SYNTAX (c);
305 306 307 308
      if (! (code == Scharquote || code == Sescape))
	break;

      quoted = !quoted;
309
    }
310 311

  UPDATE_SYNTAX_TABLE (orig);
312 313
  return quoted;
}
314 315 316 317

/* Return the bytepos one character after BYTEPOS.
   We assume that BYTEPOS is not at the end of the buffer.  */

318
INLINE EMACS_INT
319
inc_bytepos (EMACS_INT bytepos)
320
{
321 322 323
  if (NILP (current_buffer->enable_multibyte_characters))
    return bytepos + 1;

324 325 326 327 328 329 330
  INC_POS (bytepos);
  return bytepos;
}

/* Return the bytepos one character before BYTEPOS.
   We assume that BYTEPOS is not at the start of the buffer.  */

331
INLINE EMACS_INT
332
dec_bytepos (EMACS_INT bytepos)
333
{
334 335 336
  if (NILP (current_buffer->enable_multibyte_characters))
    return bytepos - 1;

337 338 339
  DEC_POS (bytepos);
  return bytepos;
}
340

Juanma Barranquero's avatar
Juanma Barranquero committed
341
/* Return a defun-start position before POS and not too far before.
Richard M. Stallman's avatar
Richard M. Stallman committed
342 343 344
   It should be the last one before POS, or nearly the last.

   When open_paren_in_column_0_is_defun_start is nonzero,
345
   only the beginning of the buffer is treated as a defun-start.
Richard M. Stallman's avatar
Richard M. Stallman committed
346 347 348 349

   We record the information about where the scan started
   and what its result was, so that another call in the same area
   can return the same value very quickly.
350 351 352 353

   There is no promise at which position the global syntax data is
   valid on return from the subroutine, so the caller should explicitly
   update the global data.  */
354

355
static EMACS_INT
356
find_defun_start (EMACS_INT pos, EMACS_INT pos_byte)
357
{
358
  EMACS_INT opoint = PT, opoint_byte = PT_BYTE;
359

360 361 362 363 364 365
  if (!open_paren_in_column_0_is_defun_start)
    {
      find_start_value_byte = BEGV_BYTE;
      return BEGV;
    }

366 367 368 369 370 371 372 373 374 375 376 377
  /* Use previous finding, if it's valid and applies to this inquiry.  */
  if (current_buffer == find_start_buffer
      /* Reuse the defun-start even if POS is a little farther on.
	 POS might be in the next defun, but that's ok.
	 Our value may not be the best possible, but will still be usable.  */
      && pos <= find_start_pos + 1000
      && pos >= find_start_value
      && BEGV == find_start_begv
      && MODIFF == find_start_modiff)
    return find_start_value;

  /* Back up to start of line.  */
378
  scan_newline (pos, pos_byte, BEGV, BEGV_BYTE, -1, 1);
379

380 381 382
  /* We optimize syntax-table lookup for rare updates.  Thus we accept
     only those `^\s(' which are good in global _and_ text-property
     syntax-tables.  */
383
  SETUP_BUFFER_SYNTAX_TABLE ();
384
  while (PT > BEGV)
385
    {
386 387
      int c;

388 389
      /* Open-paren at start of line means we may have found our
	 defun-start.  */
390
      c = FETCH_CHAR_AS_MULTIBYTE (PT_BYTE);
391
      if (SYNTAX (c) == Sopen)
392
	{
393
	  SETUP_SYNTAX_TABLE (PT + 1, -1);	/* Try again... */
394
	  c = FETCH_CHAR_AS_MULTIBYTE (PT_BYTE);
395
	  if (SYNTAX (c) == Sopen)
396 397
	    break;
	  /* Now fallback to the default value.  */
398
	  SETUP_BUFFER_SYNTAX_TABLE ();
399
	}
400 401
      /* Move to beg of previous line.  */
      scan_newline (PT, PT_BYTE, BEGV, BEGV_BYTE, -2, 1);
402 403 404
    }

  /* Record what we found, for the next try.  */
405 406
  find_start_value = PT;
  find_start_value_byte = PT_BYTE;
407 408 409 410 411
  find_start_buffer = current_buffer;
  find_start_modiff = MODIFF;
  find_start_begv = BEGV;
  find_start_pos = pos;

412 413
  TEMP_SET_PT_BOTH (opoint, opoint_byte);

414 415 416
  return find_start_value;
}

417 418 419
/* Return the SYNTAX_COMEND_FIRST of the character before POS, POS_BYTE.  */

static int
420
prev_char_comend_first (int pos, int pos_byte)
421 422 423 424 425 426 427 428 429 430 431 432 433
{
  int c, val;

  DEC_BOTH (pos, pos_byte);
  UPDATE_SYNTAX_TABLE_BACKWARD (pos);
  c = FETCH_CHAR (pos_byte);
  val = SYNTAX_COMEND_FIRST (c);
  UPDATE_SYNTAX_TABLE_FORWARD (pos + 1);
  return val;
}

/* Return the SYNTAX_COMSTART_FIRST of the character before POS, POS_BYTE.  */

434 435 436 437 438
/* static int
 * prev_char_comstart_first (pos, pos_byte)
 *      int pos, pos_byte;
 * {
 *   int c, val;
Juanma Barranquero's avatar
Juanma Barranquero committed
439
 *
440 441 442 443 444 445 446
 *   DEC_BOTH (pos, pos_byte);
 *   UPDATE_SYNTAX_TABLE_BACKWARD (pos);
 *   c = FETCH_CHAR (pos_byte);
 *   val = SYNTAX_COMSTART_FIRST (c);
 *   UPDATE_SYNTAX_TABLE_FORWARD (pos + 1);
 *   return val;
 * } */
447

448 449 450 451 452 453 454 455 456
/* Checks whether charpos FROM is at the end of a comment.
   FROM_BYTE is the bytepos corresponding to FROM.
   Do not move back before STOP.

   Return a positive value if we find a comment ending at FROM/FROM_BYTE;
   return -1 otherwise.

   If successful, store the charpos of the comment's beginning
   into *CHARPOS_PTR, and the bytepos into *BYTEPOS_PTR.
457 458 459

   Global syntax data remains valid for backward search starting at
   the returned value (or at FROM, if the search was not successful).  */
460 461

static int
462
back_comment (EMACS_INT from, EMACS_INT from_byte, EMACS_INT stop, int comnested, int comstyle, EMACS_INT *charpos_ptr, EMACS_INT *bytepos_ptr)
463 464 465 466 467 468 469 470 471 472
{
  /* Look back, counting the parity of string-quotes,
     and recording the comment-starters seen.
     When we reach a safe place, assume that's not in a string;
     then step the main scan to the earliest comment-starter seen
     an even number of string quotes away from the safe place.

     OFROM[I] is position of the earliest comment-starter seen
     which is I+2X quotes from the comment-end.
     PARITY is current parity of quotes from the comment end.  */
473
  int string_style = -1;	/* Presumed outside of any string. */
474
  int string_lossage = 0;
475
  /* Not a real lossage: indicates that we have passed a matching comment
Juanma Barranquero's avatar
Juanma Barranquero committed
476
     starter plus a non-matching comment-ender, meaning that any matching
477 478 479 480
     comment-starter we might see later could be a false positive (hidden
     inside another comment).
     Test case:  { a (* b } c (* d *) */
  int comment_lossage = 0;
481 482 483 484
  EMACS_INT comment_end = from;
  EMACS_INT comment_end_byte = from_byte;
  EMACS_INT comstart_pos = 0;
  EMACS_INT comstart_byte;
485 486
  /* Place where the containing defun starts,
     or 0 if we didn't come across it yet.  */
487 488
  EMACS_INT defun_start = 0;
  EMACS_INT defun_start_byte = 0;
489
  register enum syntaxcode code;
490
  int nesting = 1;		/* current comment nesting */
491
  int c;
492 493 494 495 496
  int syntax = 0;

  /* FIXME: A }} comment-ender style leads to incorrect behavior
     in the case of {{ c }}} because we ignore the last two chars which are
     assumed to be comment-enders although they aren't.  */
497 498 499 500 501

  /* At beginning of range to scan, we're outside of strings;
     that determines quote parity to the comment-end.  */
  while (from != stop)
    {
502 503
      int temp_byte, prev_syntax;
      int com2start, com2end;
504

505
      /* Move back and examine a character.  */
506
      DEC_BOTH (from, from_byte);
507 508
      UPDATE_SYNTAX_TABLE_BACKWARD (from);

509
      prev_syntax = syntax;
510
      c = FETCH_CHAR_AS_MULTIBYTE (from_byte);
511
      syntax = SYNTAX_WITH_FLAGS (c);
512 513
      code = SYNTAX (c);

514 515 516 517 518 519 520 521 522 523 524 525 526 527 528 529 530 531 532 533 534
      /* Check for 2-char comment markers.  */
      com2start = (SYNTAX_FLAGS_COMSTART_FIRST (syntax)
		   && SYNTAX_FLAGS_COMSTART_SECOND (prev_syntax)
		   && comstyle == SYNTAX_FLAGS_COMMENT_STYLE (prev_syntax)
		   && (SYNTAX_FLAGS_COMMENT_NESTED (prev_syntax)
		       || SYNTAX_FLAGS_COMMENT_NESTED (syntax)) == comnested);
      com2end = (SYNTAX_FLAGS_COMEND_FIRST (syntax)
		 && SYNTAX_FLAGS_COMEND_SECOND (prev_syntax));

      /* Nasty cases with overlapping 2-char comment markers:
	 - snmp-mode: -- c -- foo -- c --
	              --- c --
		      ------ c --
	 - c-mode:    *||*
		      |* *|* *|
		      |*| |* |*|
		      ///   */

      /* If a 2-char comment sequence partly overlaps with another,
	 we don't try to be clever.  */
      if (from > stop && (com2end || com2start))
535
	{
536 537 538
	  int next = from, next_byte = from_byte, next_c, next_syntax;
	  DEC_BOTH (next, next_byte);
	  UPDATE_SYNTAX_TABLE_BACKWARD (next);
539
	  next_c = FETCH_CHAR_AS_MULTIBYTE (next_byte);
540 541 542 543 544 545 546 547 548 549
	  next_syntax = SYNTAX_WITH_FLAGS (next_c);
	  if (((com2start || comnested)
	       && SYNTAX_FLAGS_COMEND_SECOND (syntax)
	       && SYNTAX_FLAGS_COMEND_FIRST (next_syntax))
	      || ((com2end || comnested)
		  && SYNTAX_FLAGS_COMSTART_SECOND (syntax)
		  && comstyle == SYNTAX_FLAGS_COMMENT_STYLE (syntax)
		  && SYNTAX_FLAGS_COMSTART_FIRST (next_syntax)))
	    goto lossage;
	  /* UPDATE_SYNTAX_TABLE_FORWARD (next + 1); */
550
	}
551 552 553 554 555 556 557 558 559 560 561 562 563 564 565 566 567

      if (com2start && comstart_pos == 0)
	/* We're looking at a comment starter.  But it might be a comment
	   ender as well (see snmp-mode).  The first time we see one, we
	   need to consider it as a comment starter,
	   and the subsequent times as a comment ender.  */
	com2end = 0;

      /* Turn a 2-char comment sequences into the appropriate syntax.  */
      if (com2end)
	code = Sendcomment;
      else if (com2start)
	code = Scomment;
      /* Ignore comment starters of a different style.  */
      else if (code == Scomment
	       && (comstyle != SYNTAX_FLAGS_COMMENT_STYLE (syntax)
		   || SYNTAX_FLAGS_COMMENT_NESTED (syntax) != comnested))
568
	continue;
569

570 571
      /* Ignore escaped characters, except comment-enders.  */
      if (code != Sendcomment && char_quoted (from, from_byte))
572 573
	continue;

574
      switch (code)
575
	{
576 577 578 579 580 581 582 583 584 585 586 587 588 589
	case Sstring_fence:
	case Scomment_fence:
	  c = (code == Sstring_fence ? ST_STRING_STYLE : ST_COMMENT_STYLE);
	case Sstring:
	  /* Track parity of quotes.  */
	  if (string_style == -1)
	    /* Entering a string.  */
	    string_style = c;
	  else if (string_style == c)
	    /* Leaving the string.  */
	    string_style = -1;
	  else
	    /* If we have two kinds of string delimiters.
	       There's no way to grok this scanning backwards.  */
590
	    string_lossage = 1;
591
	  break;
Juanma Barranquero's avatar
Juanma Barranquero committed
592

593 594
	case Scomment:
	  /* We've already checked that it is the relevant comstyle.  */
595
	  if (string_style != -1 || comment_lossage || string_lossage)
596 597 598 599
	    /* There are odd string quotes involved, so let's be careful.
	       Test case in Pascal: " { " a { " } */
	    goto lossage;

600 601 602 603 604 605 606
	  if (!comnested)
	    {
	      /* Record best comment-starter so far.  */
	      comstart_pos = from;
	      comstart_byte = from_byte;
	    }
	  else if (--nesting <= 0)
607 608 609 610 611 612
	    /* nested comments have to be balanced, so we don't need to
	       keep looking for earlier ones.  We use here the same (slightly
	       incorrect) reasoning as below:  since it is followed by uniform
	       paired string quotes, this comment-start has to be outside of
	       strings, else the comment-end itself would be inside a string. */
	    goto done;
613 614
	  break;

615
	case Sendcomment:
616
	  if (SYNTAX_FLAGS_COMMENT_STYLE (syntax) == comstyle
617
	      && ((com2end && SYNTAX_FLAGS_COMMENT_NESTED (prev_syntax))
618
		  || SYNTAX_FLAGS_COMMENT_NESTED (syntax)) == comnested)
619 620 621 622 623 624 625 626 627
	    /* This is the same style of comment ender as ours. */
	    {
	      if (comnested)
		nesting++;
	      else
		/* Anything before that can't count because it would match
		   this comment-ender rather than ours.  */
		from = stop;	/* Break out of the loop.  */
	    }
628 629 630 631 632 633 634 635 636
	  else if (comstart_pos != 0 || c != '\n')
	    /* We're mixing comment styles here, so we'd better be careful.
	       The (comstart_pos != 0 || c != '\n') check is not quite correct
	       (we should just always set comment_lossage), but removing it
	       would imply that any multiline comment in C would go through
	       lossage, which seems overkill.
	       The failure should only happen in the rare cases such as
	         { (* } *)   */
	    comment_lossage = 1;
637
	  break;
638

639 640 641 642 643 644 645 646 647 648 649
	case Sopen:
	  /* Assume a defun-start point is outside of strings.  */
	  if (open_paren_in_column_0_is_defun_start
	      && (from == stop
		  || (temp_byte = dec_bytepos (from_byte),
		      FETCH_CHAR (temp_byte) == '\n')))
	    {
	      defun_start = from;
	      defun_start_byte = from_byte;
	      from = stop;	/* Break out of the loop.  */
	    }
650
	  break;
651 652

	default:
653
	  break;
654
	}
655 656 657 658 659
    }

  if (comstart_pos == 0)
    {
      from = comment_end;
660
      from_byte = comment_end_byte;
661 662
      UPDATE_SYNTAX_TABLE_FORWARD (comment_end - 1);
    }
663 664 665
  /* If comstart_pos is set and we get here (ie. didn't jump to `lossage'
     or `done'), then we've found the beginning of the non-nested comment.  */
  else if (1)	/* !comnested */
666 667
    {
      from = comstart_pos;
668
      from_byte = comstart_byte;
669
      UPDATE_SYNTAX_TABLE_FORWARD (from - 1);
670 671 672
    }
  else
    {
673 674
      struct lisp_parse_state state;
    lossage:
675 676
      /* We had two kinds of string delimiters mixed up
	 together.  Decode this going forwards.
677
	 Scan fwd from a known safe place (beginning-of-defun)
678 679
	 to the one in question; this records where we
	 last passed a comment starter.  */
680 681 682 683 684 685
      /* If we did not already find the defun start, find it now.  */
      if (defun_start == 0)
	{
	  defun_start = find_defun_start (comment_end, comment_end_byte);
	  defun_start_byte = find_start_value_byte;
	}
686
      do
687
	{
688 689 690 691 692 693 694 695 696 697 698 699 700 701 702 703 704 705 706 707 708 709
	  scan_sexps_forward (&state,
			      defun_start, defun_start_byte,
			      comment_end, -10000, 0, Qnil, 0);
	  defun_start = comment_end;
	  if (state.incomment == (comnested ? 1 : -1)
	      && state.comstyle == comstyle)
	    from = state.comstr_start;
	  else
	    {
	      from = comment_end;
	      if (state.incomment)
		/* If comment_end is inside some other comment, maybe ours
		   is nested, so we need to try again from within the
		   surrounding comment.  Example: { a (* " *)  */
		{
		  /* FIXME: We should advance by one or two chars. */
		  defun_start = state.comstr_start + 2;
		  defun_start_byte = CHAR_TO_BYTE (defun_start);
		}
	    }
	} while (defun_start < comment_end);

710
      from_byte = CHAR_TO_BYTE (from);
711 712
      UPDATE_SYNTAX_TABLE_FORWARD (from - 1);
    }
Juanma Barranquero's avatar
Juanma Barranquero committed
713

714
 done:
715 716 717
  *charpos_ptr = from;
  *bytepos_ptr = from_byte;

718
  return (from == comment_end) ? -1 : from;
719 720
}

Richard M. Stallman's avatar
Richard M. Stallman committed
721
DEFUN ("syntax-table-p", Fsyntax_table_p, Ssyntax_table_p, 1, 1, 0,
722 723
       doc: /* Return t if OBJECT is a syntax table.
Currently, any char-table counts as a syntax table.  */)
Dan Nicolaescu's avatar
Dan Nicolaescu committed
724
  (Lisp_Object object)
Richard M. Stallman's avatar
Richard M. Stallman committed
725
{
726
  if (CHAR_TABLE_P (object)
727
      && EQ (XCHAR_TABLE (object)->purpose, Qsyntax_table))
Richard M. Stallman's avatar
Richard M. Stallman committed
728 729 730 731
    return Qt;
  return Qnil;
}

732
static void
733
check_syntax_table (Lisp_Object obj)
Richard M. Stallman's avatar
Richard M. Stallman committed
734
{
735 736
  CHECK_TYPE (CHAR_TABLE_P (obj) && EQ (XCHAR_TABLE (obj)->purpose, Qsyntax_table),
	      Qsyntax_table_p, obj);
Juanma Barranquero's avatar
Juanma Barranquero committed
737
}
Richard M. Stallman's avatar
Richard M. Stallman committed
738 739

DEFUN ("syntax-table", Fsyntax_table, Ssyntax_table, 0, 0, 0,
740 741
       doc: /* Return the current syntax table.
This is the one specified by the current buffer.  */)
Dan Nicolaescu's avatar
Dan Nicolaescu committed
742
  (void)
Richard M. Stallman's avatar
Richard M. Stallman committed
743 744 745 746 747 748
{
  return current_buffer->syntax_table;
}

DEFUN ("standard-syntax-table", Fstandard_syntax_table,
   Sstandard_syntax_table, 0, 0, 0,
749 750
       doc: /* Return the standard syntax table.
This is the one used for new buffers.  */)
Dan Nicolaescu's avatar
Dan Nicolaescu committed
751
  (void)
Richard M. Stallman's avatar
Richard M. Stallman committed
752 753 754 755 756
{
  return Vstandard_syntax_table;
}

DEFUN ("copy-syntax-table", Fcopy_syntax_table, Scopy_syntax_table, 0, 1, 0,
757 758
       doc: /* Construct a new syntax table and return it.
It is a copy of the TABLE, which defaults to the standard syntax table.  */)
Dan Nicolaescu's avatar
Dan Nicolaescu committed
759
  (Lisp_Object table)
Richard M. Stallman's avatar
Richard M. Stallman committed
760
{
761 762
  Lisp_Object copy;

Jim Blandy's avatar
Jim Blandy committed
763
  if (!NILP (table))
764 765 766 767 768
    check_syntax_table (table);
  else
    table = Vstandard_syntax_table;

  copy = Fcopy_sequence (table);
769 770 771 772 773 774 775 776 777 778

  /* Only the standard syntax table should have a default element.
     Other syntax tables should inherit from parents instead.  */
  XCHAR_TABLE (copy)->defalt = Qnil;

  /* Copied syntax tables should all have parents.
     If we copied one with no parent, such as the standard syntax table,
     use the standard syntax table as the copy's parent.  */
  if (NILP (XCHAR_TABLE (copy)->parent))
    Fset_char_table_parent (copy, Vstandard_syntax_table);
779
  return copy;
Richard M. Stallman's avatar
Richard M. Stallman committed
780 781 782
}

DEFUN ("set-syntax-table", Fset_syntax_table, Sset_syntax_table, 1, 1, 0,
783 784
       doc: /* Select a new syntax table for the current buffer.
One argument, a syntax table.  */)
Dan Nicolaescu's avatar
Dan Nicolaescu committed
785
  (Lisp_Object table)
Richard M. Stallman's avatar
Richard M. Stallman committed
786
{
787
  int idx;
788
  check_syntax_table (table);
Richard M. Stallman's avatar
Richard M. Stallman committed
789 790
  current_buffer->syntax_table = table;
  /* Indicate that this buffer now has a specified syntax table.  */
Gerd Moellmann's avatar
Gerd Moellmann committed
791 792
  idx = PER_BUFFER_VAR_IDX (syntax_table);
  SET_PER_BUFFER_VALUE_P (current_buffer, idx, 1);
Richard M. Stallman's avatar
Richard M. Stallman committed
793 794 795 796 797
  return table;
}

/* Convert a letter which signifies a syntax code
 into the code it signifies.
798
 This is used by modify-syntax-entry, and other things.  */
Richard M. Stallman's avatar
Richard M. Stallman committed
799 800 801 802 803 804

unsigned char syntax_spec_code[0400] =
  { 0377, 0377, 0377, 0377, 0377, 0377, 0377, 0377,
    0377, 0377, 0377, 0377, 0377, 0377, 0377, 0377,
    0377, 0377, 0377, 0377, 0377, 0377, 0377, 0377,
    0377, 0377, 0377, 0377, 0377, 0377, 0377, 0377,
805
    (char) Swhitespace, (char) Scomment_fence, (char) Sstring, 0377,
Richard M. Stallman's avatar
Richard M. Stallman committed
806 807 808 809 810 811
        (char) Smath, 0377, 0377, (char) Squote,
    (char) Sopen, (char) Sclose, 0377, 0377,
	0377, (char) Swhitespace, (char) Spunct, (char) Scharquote,
    0377, 0377, 0377, 0377, 0377, 0377, 0377, 0377,
    0377, 0377, 0377, 0377,
	(char) Scomment, 0377, (char) Sendcomment, 0377,
812
    (char) Sinherit, 0377, 0377, 0377, 0377, 0377, 0377, 0377,   /* @, A ... */
Richard M. Stallman's avatar
Richard M. Stallman committed
813 814 815 816 817 818
    0377, 0377, 0377, 0377, 0377, 0377, 0377, 0377,
    0377, 0377, 0377, 0377, 0377, 0377, 0377, (char) Sword,
    0377, 0377, 0377, 0377, (char) Sescape, 0377, 0377, (char) Ssymbol,
    0377, 0377, 0377, 0377, 0377, 0377, 0377, 0377,   /* `, a, ... */
    0377, 0377, 0377, 0377, 0377, 0377, 0377, 0377,
    0377, 0377, 0377, 0377, 0377, 0377, 0377, (char) Sword,
819
    0377, 0377, 0377, 0377, (char) Sstring_fence, 0377, 0377, 0377
Richard M. Stallman's avatar
Richard M. Stallman committed
820 821
  };

822
/* Indexed by syntax code, give the letter that describes it.  */
Richard M. Stallman's avatar
Richard M. Stallman committed
823

824
char syntax_code_spec[16] =
Richard M. Stallman's avatar
Richard M. Stallman committed
825
  {
826 827
    ' ', '.', 'w', '_', '(', ')', '\'', '\"', '$', '\\', '/', '<', '>', '@',
    '!', '|'
Richard M. Stallman's avatar
Richard M. Stallman committed
828
  };
Karl Heuer's avatar
Karl Heuer committed
829 830 831 832 833 834 835 836

/* Indexed by syntax code, give the object (cons of syntax code and
   nil) to be stored in syntax table.  Since these objects can be
   shared among syntax tables, we generate them in advance.  By
   sharing objects, the function `describe-syntax' can give a more
   compact listing.  */
static Lisp_Object Vsyntax_code_object;

Richard M. Stallman's avatar
Richard M. Stallman committed
837 838

DEFUN ("char-syntax", Fchar_syntax, Schar_syntax, 1, 1, 0,
839
       doc: /* Return the syntax code of CHARACTER, described by a character.
840 841
For example, if CHARACTER is a word constituent, the
character `w' (119) is returned.
842 843
The characters that correspond to various syntax codes
are listed in the documentation of `modify-syntax-entry'.  */)
Dan Nicolaescu's avatar
Dan Nicolaescu committed
844
  (Lisp_Object character)
Richard M. Stallman's avatar
Richard M. Stallman committed
845
{
846
  int char_int;
847
  CHECK_CHARACTER (character);
848
  char_int = XINT (character);
849
  SETUP_BUFFER_SYNTAX_TABLE ();
850
  return make_number (syntax_code_spec[(int) SYNTAX (char_int)]);
851 852 853
}

DEFUN ("matching-paren", Fmatching_paren, Smatching_paren, 1, 1, 0,
854
       doc: /* Return the matching parenthesis of CHARACTER, or nil if none.  */)
Dan Nicolaescu's avatar
Dan Nicolaescu committed
855
  (Lisp_Object character)
856
{
857
  int char_int, code;
858
  CHECK_NUMBER (character);
859
  char_int = XINT (character);
860
  SETUP_BUFFER_SYNTAX_TABLE ();
861
  code = SYNTAX (char_int);
862
  if (code == Sopen || code == Sclose)
863
    return SYNTAX_MATCH (char_int);
864
  return Qnil;
Richard M. Stallman's avatar
Richard M. Stallman committed
865 866
}

867
DEFUN ("string-to-syntax", Fstring_to_syntax, Sstring_to_syntax, 1, 1, 0,
868 869 870
       doc: /* Convert a syntax specification STRING into syntax cell form.
STRING should be a string as it is allowed as argument of
`modify-syntax-entry'.  Value is the equivalent cons cell
871
\(CODE . MATCHING-CHAR) that can be used as value of a `syntax-table'
872
text property.  */)
Dan Nicolaescu's avatar
Dan Nicolaescu committed
873
  (Lisp_Object string)
Richard M. Stallman's avatar
Richard M. Stallman committed
874
{
875
  register const unsigned char *p;
Richard M. Stallman's avatar
Richard M. Stallman committed
876
  register enum syntaxcode code;
877
  int val;
878
  Lisp_Object match;
Richard M. Stallman's avatar
Richard M. Stallman committed
879

880
  CHECK_STRING (string);
Richard M. Stallman's avatar
Richard M. Stallman committed
881

882
  p = SDATA (string);
Richard M. Stallman's avatar
Richard M. Stallman committed
883 884
  code = (enum syntaxcode) syntax_spec_code[*p++];
  if (((int) code & 0377) == 0377)
885
    error ("Invalid syntax description letter: %c", p[-1]);
Richard M. Stallman's avatar
Richard M. Stallman committed
886

887
  if (code == Sinherit)
888
    return Qnil;
889 890

  if (*p)
891
    {
Karl Heuer's avatar
Karl Heuer committed
892
      int len;
893
      int character = STRING_CHAR_AND_LENGTH (p, len);
Karl Heuer's avatar
Karl Heuer committed
894
      XSETINT (match, character);
895 896
      if (XFASTINT (match) == ' ')
	match = Qnil;
Karl Heuer's avatar
Karl Heuer committed
897
      p += len;
898 899
    }
  else
900
    match = Qnil;
Richard M. Stallman's avatar
Richard M. Stallman committed
901

902
  val = (int) code;
Richard M. Stallman's avatar
Richard M. Stallman committed
903 904 905 906
  while (*p)
    switch (*p++)
      {
      case '1':
907
	val |= 1 << 16;
Richard M. Stallman's avatar
Richard M. Stallman committed
908 909 910
	break;

      case '2':
911
	val |= 1 << 17;
Richard M. Stallman's avatar
Richard M. Stallman committed
912 913 914
	break;

      case '3':
915
	val |= 1 << 18;
Richard M. Stallman's avatar
Richard M. Stallman committed
916 917 918
	break;

      case '4':
919
	val |= 1 << 19;
Richard M. Stallman's avatar
Richard M. Stallman committed
920 921 922
	break;

      case 'p':
923
	val |= 1 << 20;
Richard M. Stallman's avatar
Richard M. Stallman committed
924
	break;
925 926

      case 'b':
927
	val |= 1 << 21;
928
	break;
929 930 931 932

      case 'n':
	val |= 1 << 22;
	break;
Richard M. Stallman's avatar
Richard M. Stallman committed
933
      }
Juanma Barranquero's avatar
Juanma Barranquero committed
934

Karl Heuer's avatar
Karl Heuer committed
935
  if (val < XVECTOR (Vsyntax_code_object)->size && NILP (match))
936
    return XVECTOR (Vsyntax_code_object)->contents[val];
Karl Heuer's avatar
Karl Heuer committed
937 938
  else
    /* Since we can't use a shared object, let's make a new one.  */
939 940 941
    return Fcons (make_number (val), match);
}

942
/* I really don't know why this is interactive
Stefan Monnier's avatar
Stefan Monnier committed
943
   help-form should at least be made useful whilst reading the second arg.  */
Juanma Barranquero's avatar
Juanma Barranquero committed
944
DEFUN ("modify-syntax-entry", Fmodify_syntax_entry, Smodify_syntax_entry, 2, 3,
945
  "cSet syntax for character: \nsSet syntax for %s to: ",
946
       doc: /* Set syntax for character CHAR according to string NEWENTRY.
947
The syntax is changed only for table SYNTAX-TABLE, which defaults to
948
 the current buffer's syntax table.
949
CHAR may be a cons (MIN . MAX), in which case, syntaxes of all characters
Juanma Barranquero's avatar