syntax.c 103 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, 1993-1995, 1997-1999, 2001-2012
3
                 Free Software Foundation, Inc.
Richard M. Stallman's avatar
Richard M. Stallman committed
4 5 6

This file is part of GNU Emacs.

7
GNU Emacs is free software: you can redistribute it and/or modify
Richard M. Stallman's avatar
Richard M. Stallman committed
8
it under the terms of the GNU General Public License as published by
9 10
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
11 12 13 14 15 16 17

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
18
along with GNU Emacs.  If not, see <http://www.gnu.org/licenses/>.  */
Richard M. Stallman's avatar
Richard M. Stallman committed
19 20


21
#include <config.h>
22

Richard M. Stallman's avatar
Richard M. Stallman committed
23
#include <ctype.h>
24
#include <sys/types.h>
25
#include <setjmp.h>
Richard M. Stallman's avatar
Richard M. Stallman committed
26 27
#include "lisp.h"
#include "commands.h"
28
#include "character.h"
29
#include "buffer.h"
Stefan Monnier's avatar
Stefan Monnier committed
30
#include "keymap.h"
Richard M. Stallman's avatar
Richard M. Stallman committed
31
#include "regex.h"
32 33 34 35

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

Richard M. Stallman's avatar
Richard M. Stallman committed
36
#include "syntax.h"
37
#include "intervals.h"
38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65 66 67 68 69 70 71 72 73 74 75 76 77 78 79 80 81 82 83 84 85 86 87 88 89 90 91
#include "category.h"

/* Then there are seven single-bit flags that have the following meanings:
  1. This character is the first of a two-character comment-start sequence.
  2. This character is the second of a two-character comment-start sequence.
  3. This character is the first of a two-character comment-end sequence.
  4. This character is the second of a two-character comment-end sequence.
  5. This character is a prefix, for backward-prefix-chars.
  6. The char is part of a delimiter for comments of style "b".
  7. This character is part of a nestable comment sequence.
  8. The char is part of a delimiter for comments of style "c".
  Note that any two-character sequence whose first character has flag 1
  and whose second character has flag 2 will be interpreted as a comment start.

  bit 6 and 8 are used to discriminate between different comment styles.
  Languages such as C++ allow two orthogonal syntax start/end pairs
  and bit 6 is used to determine whether a comment-end or Scommentend
  ends style a or b.  Comment markers can start style a, b, c, or bc.
  Style a is always the default.
  For 2-char comment markers, the style b flag is only looked up on the second
  char of the comment marker and on the first char of the comment ender.
  For style c (like to for the nested flag), the flag can be placed on any
  one of the chars.
  */

/* These macros extract specific flags from an integer
   that holds the syntax code and the flags.  */

#define SYNTAX_FLAGS_COMSTART_FIRST(flags) (((flags) >> 16) & 1)

#define SYNTAX_FLAGS_COMSTART_SECOND(flags) (((flags) >> 17) & 1)

#define SYNTAX_FLAGS_COMEND_FIRST(flags) (((flags) >> 18) & 1)

#define SYNTAX_FLAGS_COMEND_SECOND(flags) (((flags) >> 19) & 1)

#define SYNTAX_FLAGS_PREFIX(flags) (((flags) >> 20) & 1)

#define SYNTAX_FLAGS_COMMENT_STYLEB(flags) (((flags) >> 21) & 1)
#define SYNTAX_FLAGS_COMMENT_STYLEC(flags) (((flags) >> 22) & 2)
/* FLAGS should be the flags of the main char of the comment marker, e.g.
   the second for comstart and the first for comend.  */
#define SYNTAX_FLAGS_COMMENT_STYLE(flags, other_flags) \
  (SYNTAX_FLAGS_COMMENT_STYLEB (flags) \
   | SYNTAX_FLAGS_COMMENT_STYLEC (flags) \
   | SYNTAX_FLAGS_COMMENT_STYLEC (other_flags))

#define SYNTAX_FLAGS_COMMENT_NESTED(flags) (((flags) >> 22) & 1)

/* These macros extract a particular flag for a given character.  */

#define SYNTAX_COMEND_FIRST(c) \
  (SYNTAX_FLAGS_COMEND_FIRST (SYNTAX_WITH_FLAGS (c)))
#define SYNTAX_PREFIX(c) (SYNTAX_FLAGS_PREFIX (SYNTAX_WITH_FLAGS (c)))
92 93 94 95 96

/* 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.  */

97 98
#define ST_COMMENT_STYLE (256 + 1)
#define ST_STRING_STYLE (256 + 2)
Richard M. Stallman's avatar
Richard M. Stallman committed
99

100 101
static Lisp_Object Qsyntax_table_p;
static Lisp_Object Qsyntax_table, Qscan_error;
Richard M. Stallman's avatar
Richard M. Stallman committed
102

103
#ifndef __GNUC__
104 105 106 107
/* 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;
108
#endif
109

110 111 112 113
/* This is the internal form of the parse state used in parse-partial-sexp.  */

struct lisp_parse_state
  {
114
    EMACS_INT depth;	/* Depth at end of parsing.  */
115
    int instring;  /* -1 if not within string, else desired terminator.  */
116
    EMACS_INT incomment; /* -1 if in unnestable comment else comment nesting */
117 118
    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 */
119
    EMACS_INT mindepth;	/* Minimum depth seen while scanning.  */
120
    /* Char number of most recent start-of-expression at current level */
121
    ptrdiff_t thislevelstart;
122
    /* Char number of start of containing expression */
123 124 125
    ptrdiff_t prevlevelstart;
    ptrdiff_t location;	     /* Char number at which parsing stopped.  */
    ptrdiff_t comstr_start;  /* Position of last comment/string starter.  */
126 127
    Lisp_Object levelstarts; /* Char numbers of starts-of-expression
				of levels (starting from outermost).  */
128 129
  };

130 131 132
/* 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.
133
   find_start_value_byte is the corresponding byte position.
134 135 136 137
   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.  */

138 139 140
static ptrdiff_t find_start_pos;
static ptrdiff_t find_start_value;
static ptrdiff_t find_start_value_byte;
141
static struct buffer *find_start_buffer;
142 143
static ptrdiff_t find_start_begv;
static EMACS_INT find_start_modiff;
144 145


146 147 148 149
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 *,
150
                                ptrdiff_t, ptrdiff_t, ptrdiff_t, EMACS_INT,
151 152
                                int, Lisp_Object, int);
static int in_classes (int, Lisp_Object);
153

154 155 156 157 158
/* Whether the syntax of the character C has the prefix flag set.  */
int syntax_prefix_flag_p (int c)
{
  return SYNTAX_PREFIX (c);
}
159 160 161 162 163 164

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

#define INTERVALS_AT_ONCE 10		/* 1 + max-number of intervals
					   to scan to property-change.  */

165 166
/* Update gl_state to an appropriate interval which contains CHARPOS.  The
   sign of COUNT give the relative position of CHARPOS wrt the previously
167
   valid interval.  If INIT, only [be]_property fields of gl_state are
168
   valid at start, the rest is filled basing on OBJECT.
169

170
   `gl_state.*_i' are the intervals, and CHARPOS is further in the search
171 172
   direction than the intervals - or in an interval.  We update the
   current syntax-table basing on the property of this interval, and
173
   update the interval to start further than CHARPOS - or be
174
   NULL_INTERVAL.  We also update lim_property to be the next value of
175
   charpos to call this subroutine again - or be before/after the
176 177 178
   start/end of OBJECT.  */

void
179
update_syntax_table (ptrdiff_t charpos, EMACS_INT count, int init,
180
		     Lisp_Object object)
181 182
{
  Lisp_Object tmp_table;
183 184
  unsigned cnt = 0;
  int invalidate = 1;
185
  INTERVAL i;
186 187 188

  if (init)
    {
189
      gl_state.old_prop = Qnil;
190 191
      gl_state.start = gl_state.b_property;
      gl_state.stop = gl_state.e_property;
192 193
      i = interval_of (charpos, object);
      gl_state.backward_i = gl_state.forward_i = i;
194 195 196
      invalidate = 0;
      if (NULL_INTERVAL_P (i))
	return;
197
      /* interval_of updates only ->position of the return value, so
198
	 update the parents manually to speed up update_interval.  */
199
      while (!NULL_PARENT (i))
200 201
	{
	  if (AM_RIGHT_CHILD (i))
202
	    INTERVAL_PARENT (i)->position = i->position
203
	      - LEFT_TOTAL_LENGTH (i) + TOTAL_LENGTH (i) /* right end */
204 205
	      - TOTAL_LENGTH (INTERVAL_PARENT (i))
	      + LEFT_TOTAL_LENGTH (INTERVAL_PARENT (i));
206
	  else
207
	    INTERVAL_PARENT (i)->position = i->position - LEFT_TOTAL_LENGTH (i)
208
	      + TOTAL_LENGTH (i);
209
	  i = INTERVAL_PARENT (i);
210 211
	}
      i = gl_state.forward_i;
212
      gl_state.b_property = i->position - gl_state.offset;
213
      gl_state.e_property = INTERVAL_LAST_POS (i) - gl_state.offset;
214 215
      goto update;
    }
216
  i = count > 0 ? gl_state.forward_i : gl_state.backward_i;
217

218
  /* We are guaranteed to be called with CHARPOS either in i,
219
     or further off.  */
220 221
  if (NULL_INTERVAL_P (i))
    error ("Error in syntax_table logic for to-the-end intervals");
222
  else if (charpos < i->position)		/* Move left.  */
223 224
    {
      if (count > 0)
225
	error ("Error in syntax_table logic for intervals <-");
226
      /* Update the interval.  */
227
      i = update_interval (i, charpos);
228
      if (INTERVAL_LAST_POS (i) != gl_state.b_property)
229 230 231
	{
	  invalidate = 0;
	  gl_state.forward_i = i;
232
	  gl_state.e_property = INTERVAL_LAST_POS (i) - gl_state.offset;
233
	}
234
    }
235
  else if (charpos >= INTERVAL_LAST_POS (i)) /* Move right.  */
236 237
    {
      if (count < 0)
238
	error ("Error in syntax_table logic for intervals ->");
239
      /* Update the interval.  */
240
      i = update_interval (i, charpos);
241
      if (i->position != gl_state.e_property)
242 243 244
	{
	  invalidate = 0;
	  gl_state.backward_i = i;
245
	  gl_state.b_property = i->position - gl_state.offset;
246 247 248 249 250 251 252 253
	}
    }

  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
254

255 256 257
  if (invalidate)		/* Did not get to adjacent interval.  */
    {				/* with the same table => */
				/* invalidate the old range.  */
258 259 260
      if (count > 0)
	{
	  gl_state.backward_i = i;
261 262 263
	  gl_state.b_property = i->position - gl_state.offset;
	}
      else
264
	{
265
	  gl_state.forward_i = i;
266
	  gl_state.e_property = INTERVAL_LAST_POS (i) - gl_state.offset;
267 268
	}
    }
269

270
  if (!EQ (tmp_table, gl_state.old_prop))
271
    {
272 273 274 275 276
      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
277
	}
278 279 280 281 282
      else if (CONSP (tmp_table))
	{
	  gl_state.use_global = 1;
	  gl_state.global_code = tmp_table;
	}
Juanma Barranquero's avatar
Juanma Barranquero committed
283
      else
284 285
	{
	  gl_state.use_global = 0;
Tom Tromey's avatar
Tom Tromey committed
286
	  gl_state.current_syntax_table = BVAR (current_buffer, syntax_table);
287
	}
288 289 290 291 292 293 294
    }

  while (!NULL_INTERVAL_P (i))
    {
      if (cnt && !EQ (tmp_table, textget (i->plist, Qsyntax_table)))
	{
	  if (count > 0)
295 296 297 298 299 300
	    {
	      gl_state.e_property = i->position - gl_state.offset;
	      gl_state.forward_i = i;
	    }
	  else
	    {
301 302
	      gl_state.b_property
		= i->position + LENGTH (i) - gl_state.offset;
303 304 305
	      gl_state.backward_i = i;
	    }
	  return;
306
	}
Juanma Barranquero's avatar
Juanma Barranquero committed
307
      else if (cnt == INTERVALS_AT_ONCE)
308 309
	{
	  if (count > 0)
310
	    {
311 312 313 314 315 316
	      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);
317 318 319 320 321 322 323 324
	      gl_state.forward_i = i;
	    }
	  else
	    {
	      gl_state.b_property = i->position - gl_state.offset;
	      gl_state.backward_i = i;
	    }
	  return;
325 326 327 328
	}
      cnt++;
      i = count > 0 ? next_interval (i) : previous_interval (i);
    }
329 330 331 332 333
  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;
334 335
}

336 337 338
/* 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. */
339 340

static int
341
char_quoted (ptrdiff_t charpos, ptrdiff_t bytepos)
342 343
{
  register enum syntaxcode code;
344
  register ptrdiff_t beg = BEGV;
345
  register int quoted = 0;
346
  ptrdiff_t orig = charpos;
347

348
  while (charpos > beg)
349
    {
350
      int c;
351
      DEC_BOTH (charpos, bytepos);
352

353
      UPDATE_SYNTAX_TABLE_BACKWARD (charpos);
354
      c = FETCH_CHAR_AS_MULTIBYTE (bytepos);
355
      code = SYNTAX (c);
356 357 358 359
      if (! (code == Scharquote || code == Sescape))
	break;

      quoted = !quoted;
360
    }
361 362

  UPDATE_SYNTAX_TABLE (orig);
363 364
  return quoted;
}
365 366 367 368

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

369 370
static inline ptrdiff_t
dec_bytepos (ptrdiff_t bytepos)
371
{
Tom Tromey's avatar
Tom Tromey committed
372
  if (NILP (BVAR (current_buffer, enable_multibyte_characters)))
373 374
    return bytepos - 1;

375 376 377
  DEC_POS (bytepos);
  return bytepos;
}
378

Juanma Barranquero's avatar
Juanma Barranquero committed
379
/* Return a defun-start position before POS and not too far before.
Richard M. Stallman's avatar
Richard M. Stallman committed
380 381 382
   It should be the last one before POS, or nearly the last.

   When open_paren_in_column_0_is_defun_start is nonzero,
383
   only the beginning of the buffer is treated as a defun-start.
Richard M. Stallman's avatar
Richard M. Stallman committed
384 385 386 387

   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.
388 389 390 391

   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.  */
392

393 394
static ptrdiff_t
find_defun_start (ptrdiff_t pos, ptrdiff_t pos_byte)
395
{
396
  ptrdiff_t opoint = PT, opoint_byte = PT_BYTE;
397

398 399
  if (!open_paren_in_column_0_is_defun_start)
    {
400
      find_start_value = BEGV;
401
      find_start_value_byte = BEGV_BYTE;
402 403 404 405
      find_start_buffer = current_buffer;
      find_start_modiff = MODIFF;
      find_start_begv = BEGV;
      find_start_pos = pos;
406 407 408
      return BEGV;
    }

409 410 411 412 413 414 415 416 417 418 419 420
  /* 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.  */
421
  scan_newline (pos, pos_byte, BEGV, BEGV_BYTE, -1, 1);
422

423 424 425
  /* We optimize syntax-table lookup for rare updates.  Thus we accept
     only those `^\s(' which are good in global _and_ text-property
     syntax-tables.  */
426
  SETUP_BUFFER_SYNTAX_TABLE ();
427
  while (PT > BEGV)
428
    {
429 430
      int c;

431 432
      /* Open-paren at start of line means we may have found our
	 defun-start.  */
433
      c = FETCH_CHAR_AS_MULTIBYTE (PT_BYTE);
434
      if (SYNTAX (c) == Sopen)
435
	{
436
	  SETUP_SYNTAX_TABLE (PT + 1, -1);	/* Try again... */
437
	  c = FETCH_CHAR_AS_MULTIBYTE (PT_BYTE);
438
	  if (SYNTAX (c) == Sopen)
439 440
	    break;
	  /* Now fallback to the default value.  */
441
	  SETUP_BUFFER_SYNTAX_TABLE ();
442
	}
443 444
      /* Move to beg of previous line.  */
      scan_newline (PT, PT_BYTE, BEGV, BEGV_BYTE, -2, 1);
445 446 447
    }

  /* Record what we found, for the next try.  */
448 449
  find_start_value = PT;
  find_start_value_byte = PT_BYTE;
450 451 452 453 454
  find_start_buffer = current_buffer;
  find_start_modiff = MODIFF;
  find_start_begv = BEGV;
  find_start_pos = pos;

455 456
  TEMP_SET_PT_BOTH (opoint, opoint_byte);

457 458 459
  return find_start_value;
}

460 461 462
/* Return the SYNTAX_COMEND_FIRST of the character before POS, POS_BYTE.  */

static int
463
prev_char_comend_first (ptrdiff_t pos, ptrdiff_t pos_byte)
464 465 466 467 468 469 470 471 472 473 474 475 476
{
  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.  */

477 478 479 480 481
/* static int
 * prev_char_comstart_first (pos, pos_byte)
 *      int pos, pos_byte;
 * {
 *   int c, val;
Juanma Barranquero's avatar
Juanma Barranquero committed
482
 *
483 484 485 486 487 488 489
 *   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;
 * } */
490

491 492 493 494 495 496 497 498 499
/* 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.
500 501 502

   Global syntax data remains valid for backward search starting at
   the returned value (or at FROM, if the search was not successful).  */
503 504

static int
505
back_comment (ptrdiff_t from, ptrdiff_t from_byte, ptrdiff_t stop, int comnested, int comstyle, ptrdiff_t *charpos_ptr, ptrdiff_t *bytepos_ptr)
506 507 508 509 510 511 512 513 514 515
{
  /* 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.  */
516
  int string_style = -1;	/* Presumed outside of any string. */
517
  int string_lossage = 0;
518
  /* Not a real lossage: indicates that we have passed a matching comment
Juanma Barranquero's avatar
Juanma Barranquero committed
519
     starter plus a non-matching comment-ender, meaning that any matching
520 521 522 523
     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;
524 525 526 527
  ptrdiff_t comment_end = from;
  ptrdiff_t comment_end_byte = from_byte;
  ptrdiff_t comstart_pos = 0;
  ptrdiff_t comstart_byte IF_LINT (= 0);
528 529
  /* Place where the containing defun starts,
     or 0 if we didn't come across it yet.  */
530 531
  ptrdiff_t defun_start = 0;
  ptrdiff_t defun_start_byte = 0;
532
  register enum syntaxcode code;
533
  int nesting = 1;		/* current comment nesting */
534
  int c;
535 536 537 538 539
  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.  */
540 541 542 543 544

  /* At beginning of range to scan, we're outside of strings;
     that determines quote parity to the comment-end.  */
  while (from != stop)
    {
545
      ptrdiff_t temp_byte;
546
      int prev_syntax, com2start, com2end;
547
      int comstart;
548

549
      /* Move back and examine a character.  */
550
      DEC_BOTH (from, from_byte);
551 552
      UPDATE_SYNTAX_TABLE_BACKWARD (from);

553
      prev_syntax = syntax;
554
      c = FETCH_CHAR_AS_MULTIBYTE (from_byte);
555
      syntax = SYNTAX_WITH_FLAGS (c);
556 557
      code = SYNTAX (c);

558 559 560
      /* Check for 2-char comment markers.  */
      com2start = (SYNTAX_FLAGS_COMSTART_FIRST (syntax)
		   && SYNTAX_FLAGS_COMSTART_SECOND (prev_syntax)
561 562
		   && (comstyle
		       == SYNTAX_FLAGS_COMMENT_STYLE (prev_syntax, syntax))
563 564 565 566
		   && (SYNTAX_FLAGS_COMMENT_NESTED (prev_syntax)
		       || SYNTAX_FLAGS_COMMENT_NESTED (syntax)) == comnested);
      com2end = (SYNTAX_FLAGS_COMEND_FIRST (syntax)
		 && SYNTAX_FLAGS_COMEND_SECOND (prev_syntax));
567
      comstart = (com2start || code == Scomment);
568

569 570 571 572 573 574 575 576 577 578
      /* 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,
579 580 581
	 we don't try to be clever.  E.g. |*| in C, or }% in modes that
	 have %..\n and %{..}%.  */
      if (from > stop && (com2end || comstart))
582
	{
583
	  ptrdiff_t next = from, next_byte = from_byte;
584
	  int next_c, next_syntax;
585 586
	  DEC_BOTH (next, next_byte);
	  UPDATE_SYNTAX_TABLE_BACKWARD (next);
587
	  next_c = FETCH_CHAR_AS_MULTIBYTE (next_byte);
588
	  next_syntax = SYNTAX_WITH_FLAGS (next_c);
589
	  if (((comstart || comnested)
590 591 592 593
	       && SYNTAX_FLAGS_COMEND_SECOND (syntax)
	       && SYNTAX_FLAGS_COMEND_FIRST (next_syntax))
	      || ((com2end || comnested)
		  && SYNTAX_FLAGS_COMSTART_SECOND (syntax)
594 595
		  && (comstyle
		      == SYNTAX_FLAGS_COMMENT_STYLE (syntax, prev_syntax))
596 597 598
		  && SYNTAX_FLAGS_COMSTART_FIRST (next_syntax)))
	    goto lossage;
	  /* UPDATE_SYNTAX_TABLE_FORWARD (next + 1); */
599
	}
600 601 602 603 604 605 606 607 608 609 610 611 612 613 614

      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
615
	       && (comstyle != SYNTAX_FLAGS_COMMENT_STYLE (syntax, 0)
616
		   || SYNTAX_FLAGS_COMMENT_NESTED (syntax) != comnested))
617
	continue;
618

619 620
      /* Ignore escaped characters, except comment-enders.  */
      if (code != Sendcomment && char_quoted (from, from_byte))
621 622
	continue;

623
      switch (code)
624
	{
625 626 627 628 629 630 631 632 633 634 635 636 637 638
	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.  */
639
	    string_lossage = 1;
640
	  break;
Juanma Barranquero's avatar
Juanma Barranquero committed
641

642 643
	case Scomment:
	  /* We've already checked that it is the relevant comstyle.  */
644
	  if (string_style != -1 || comment_lossage || string_lossage)
645 646 647 648
	    /* There are odd string quotes involved, so let's be careful.
	       Test case in Pascal: " { " a { " } */
	    goto lossage;

649 650 651 652 653 654 655
	  if (!comnested)
	    {
	      /* Record best comment-starter so far.  */
	      comstart_pos = from;
	      comstart_byte = from_byte;
	    }
	  else if (--nesting <= 0)
656 657 658 659 660 661
	    /* 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;
662 663
	  break;

664
	case Sendcomment:
665
	  if (SYNTAX_FLAGS_COMMENT_STYLE (syntax, 0) == comstyle
666
	      && ((com2end && SYNTAX_FLAGS_COMMENT_NESTED (prev_syntax))
667
		  || SYNTAX_FLAGS_COMMENT_NESTED (syntax)) == comnested)
668 669 670 671 672 673 674 675 676
	    /* 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.  */
	    }
677 678 679 680 681 682 683 684 685
	  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;
686
	  break;
687

688 689 690 691 692 693 694 695 696 697 698
	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.  */
	    }
699
	  break;
700 701

	default:
702
	  break;
703
	}
704 705 706 707 708
    }

  if (comstart_pos == 0)
    {
      from = comment_end;
709
      from_byte = comment_end_byte;
710 711
      UPDATE_SYNTAX_TABLE_FORWARD (comment_end - 1);
    }
712 713 714
  /* 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 */
715 716
    {
      from = comstart_pos;
717
      from_byte = comstart_byte;
718
      UPDATE_SYNTAX_TABLE_FORWARD (from - 1);
719 720 721
    }
  else
    {
722 723
      struct lisp_parse_state state;
    lossage:
724 725
      /* We had two kinds of string delimiters mixed up
	 together.  Decode this going forwards.
726
	 Scan fwd from a known safe place (beginning-of-defun)
727 728
	 to the one in question; this records where we
	 last passed a comment starter.  */
729 730 731 732 733 734
      /* 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;
	}
735
      do
736
	{
737 738
	  scan_sexps_forward (&state,
			      defun_start, defun_start_byte,
739 740
			      comment_end, TYPE_MINIMUM (EMACS_INT),
			      0, Qnil, 0);
741 742 743 744 745 746 747 748 749 750 751 752 753 754 755 756 757 758 759
	  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);

760
      from_byte = CHAR_TO_BYTE (from);
761 762
      UPDATE_SYNTAX_TABLE_FORWARD (from - 1);
    }
Juanma Barranquero's avatar
Juanma Barranquero committed
763

764
 done:
765 766 767
  *charpos_ptr = from;
  *bytepos_ptr = from_byte;

768
  return (from == comment_end) ? -1 : from;
769 770
}

Richard M. Stallman's avatar
Richard M. Stallman committed
771
DEFUN ("syntax-table-p", Fsyntax_table_p, Ssyntax_table_p, 1, 1, 0,
772 773
       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
774
  (Lisp_Object object)
Richard M. Stallman's avatar
Richard M. Stallman committed
775
{
776
  if (CHAR_TABLE_P (object)
777
      && EQ (XCHAR_TABLE (object)->purpose, Qsyntax_table))
Richard M. Stallman's avatar
Richard M. Stallman committed
778 779 780 781
    return Qt;
  return Qnil;
}

782
static void
783
check_syntax_table (Lisp_Object obj)
Richard M. Stallman's avatar
Richard M. Stallman committed
784
{
785 786
  CHECK_TYPE (CHAR_TABLE_P (obj) && EQ (XCHAR_TABLE (obj)->purpose, Qsyntax_table),
	      Qsyntax_table_p, obj);
Juanma Barranquero's avatar
Juanma Barranquero committed
787
}
Richard M. Stallman's avatar
Richard M. Stallman committed
788 789

DEFUN ("syntax-table", Fsyntax_table, Ssyntax_table, 0, 0, 0,
790 791
       doc: /* Return the current syntax table.
This is the one specified by the current buffer.  */)
Dan Nicolaescu's avatar
Dan Nicolaescu committed
792
  (void)
Richard M. Stallman's avatar
Richard M. Stallman committed
793
{
Tom Tromey's avatar
Tom Tromey committed
794
  return BVAR (current_buffer, syntax_table);
Richard M. Stallman's avatar
Richard M. Stallman committed
795 796 797 798
}

DEFUN ("standard-syntax-table", Fstandard_syntax_table,
   Sstandard_syntax_table, 0, 0, 0,
799 800
       doc: /* Return the standard syntax table.
This is the one used for new buffers.  */)
Dan Nicolaescu's avatar
Dan Nicolaescu committed
801
  (void)
Richard M. Stallman's avatar
Richard M. Stallman committed
802 803 804 805 806
{
  return Vstandard_syntax_table;
}

DEFUN ("copy-syntax-table", Fcopy_syntax_table, Scopy_syntax_table, 0, 1, 0,
807 808
       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
809
  (Lisp_Object table)
Richard M. Stallman's avatar
Richard M. Stallman committed
810
{
811 812
  Lisp_Object copy;

Jim Blandy's avatar
Jim Blandy committed
813
  if (!NILP (table))
814 815 816 817 818
    check_syntax_table (table);
  else
    table = Vstandard_syntax_table;

  copy = Fcopy_sequence (table);
819 820 821 822 823 824 825 826 827 828

  /* 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);
829
  return copy;
Richard M. Stallman's avatar
Richard M. Stallman committed
830 831 832
}

DEFUN ("set-syntax-table", Fset_syntax_table, Sset_syntax_table, 1, 1, 0,
833 834
       doc: /* Select a new syntax table for the current buffer.
One argument, a syntax table.  */)
Dan Nicolaescu's avatar
Dan Nicolaescu committed
835
  (Lisp_Object table)
Richard M. Stallman's avatar
Richard M. Stallman committed
836
{
837
  int idx;
838
  check_syntax_table (table);
Tom Tromey's avatar
Tom Tromey committed
839
  BVAR (current_buffer, syntax_table) = table;
Richard M. Stallman's avatar
Richard M. Stallman committed
840
  /* Indicate that this buffer now has a specified syntax table.  */
Gerd Moellmann's avatar
Gerd Moellmann committed
841 842
  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
843 844 845 846 847
  return table;
}

/* Convert a letter which signifies a syntax code
 into the code it signifies.
848
 This is used by modify-syntax-entry, and other things.  */
Richard M. Stallman's avatar
Richard M. Stallman committed
849 850 851 852 853 854

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,
855
    (char) Swhitespace, (char) Scomment_fence, (char) Sstring, 0377,
Richard M. Stallman's avatar
Richard M. Stallman committed
856 857 858 859 860 861
        (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,
862
    (char) Sinherit, 0377, 0377, 0377, 0377, 0377, 0377, 0377,   /* @, A ... */
Richard M. Stallman's avatar
Richard M. Stallman committed
863 864 865 866 867 868
    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,
869
    0377, 0377, 0377, 0377, (char) Sstring_fence, 0377, 0377, 0377
Richard M. Stallman's avatar
Richard M. Stallman committed
870 871
  };

872
/* Indexed by syntax code, give the letter that describes it.  */
Richard M. Stallman's avatar
Richard M. Stallman committed
873

874
char syntax_code_spec[16] =
Richard M. Stallman's avatar
Richard M. Stallman committed
875
  {
876 877
    ' ', '.', 'w', '_', '(', ')', '\'', '\"', '$', '\\', '/', '<', '>', '@',
    '!', '|'
Richard M. Stallman's avatar
Richard M. Stallman committed
878
  };
Karl Heuer's avatar
Karl Heuer committed
879 880 881 882 883 884 885 886

/* 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
887 888

DEFUN ("char-syntax", Fchar_syntax, Schar_syntax, 1, 1, 0,
889
       doc: /* Return the syntax code of CHARACTER, described by a character.
890 891
For example, if CHARACTER is a word constituent, the
character `w' (119) is returned.
892 893
The characters that correspond to various syntax codes
are listed in the documentation of `modify-syntax-entry'.  */)
Dan Nicolaescu's avatar
Dan Nicolaescu committed
894
  (Lisp_Object character)
Richard M. Stallman's avatar
Richard M. Stallman committed
895
{
896
  int char_int;
897
  CHECK_CHARACTER (character);
898
  char_int = XINT (character);
899
  SETUP_BUFFER_SYNTAX_TABLE ();
900
  return make_number (syntax_code_spec[(int) SYNTAX (char_int)]);
901 902 903
}

DEFUN ("matching-paren", Fmatching_paren, Smatching_paren, 1, 1, 0,
904
       doc: /* Return the matching parenthesis of CHARACTER, or nil if none.  */)
Dan Nicolaescu's avatar
Dan Nicolaescu committed
905
  (Lisp_Object character)
906
{
907
  int char_int, code;
908
  CHECK_NUMBER (character);
909
  char_int = XINT (character);
910
  SETUP_BUFFER_SYNTAX_TABLE ();
911
  code = SYNTAX (char_int);
912
  if (code == Sopen || code == Sclose)
913
    return SYNTAX_MATCH (char_int);
914
  return Qnil;
Richard M. Stallman's avatar
Richard M. Stallman committed
915 916
}

917
DEFUN ("string-to-syntax", Fstring_to_syntax, Sstring_to_syntax, 1, 1, 0,
918 919 920
       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
921
\(CODE . MATCHING-CHAR) that can be used as value of a `syntax-table'
922
text property.  */)
Dan Nicolaescu's avatar
Dan Nicolaescu committed
923
  (Lisp_Object string)
Richard M. Stallman's avatar
Richard M. Stallman committed
924
{
925
  register const unsigned char *p;
Richard M. Stallman's avatar
Richard M. Stallman committed
926
  register enum syntaxcode code;
927
  int val;
928
  Lisp_Object match;
Richard M. Stallman's avatar
Richard M. Stallman committed
929

930
  CHECK_STRING (string);
Richard M. Stallman's avatar
Richard M. Stallman committed
931

932
  p = SDATA (string);
Richard M. Stallman's avatar
Richard M. Stallman committed
933 934
  code = (enum syntaxcode) syntax_spec_code[*p++];
  if (((int) code & 0377) == 0377)
935
    error ("Invalid syntax description letter: %c", p[-1]);
Richard M. Stallman's avatar
Richard M. Stallman committed
936

937
  if (code == Sinherit)
938
    return Qnil;
939 940

  if (*p)
941
    {
Karl Heuer's avatar
Karl Heuer committed
942
      int len;
943
      int character = STRING_CHAR_AND_LENGTH (p, len);
Karl Heuer's avatar
Karl Heuer committed
944
      XSETINT (match, character);