syntax.c 106 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.
Paul Eggert's avatar
Paul Eggert committed
2
   Copyright (C) 1985, 1987, 1993-1995, 1997-1999, 2001-2015 Free
3
   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 23

#include <sys/types.h>
24

Richard M. Stallman's avatar
Richard M. Stallman committed
25 26
#include "lisp.h"
#include "commands.h"
27
#include "character.h"
28
#include "buffer.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

Richard M. Stallman's avatar
Richard M. Stallman committed
32
#include "syntax.h"
33
#include "intervals.h"
34 35
#include "category.h"

36 37 38 39 40
/* Make syntax table lookup grant data in gl_state.  */
#define SYNTAX(c) syntax_property (c, 1)
#define SYNTAX_ENTRY(c) syntax_property_entry (c, 1)
#define SYNTAX_WITH_FLAGS(c) syntax_property_with_flags (c, 1)

Paul Eggert's avatar
Paul Eggert committed
41
/* Eight single-bit flags have the following meanings:
42 43 44 45 46 47 48 49 50 51 52
  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.

Paul Eggert's avatar
Paul Eggert committed
53
  Bits 6 and 8 discriminate among different comment styles.
54
  Languages such as C++ allow two orthogonal syntax start/end pairs
Paul Eggert's avatar
Paul Eggert committed
55
  and bit 6 determines whether a comment-end or Scommentend
56 57
  ends style a or b.  Comment markers can start style a, b, c, or bc.
  Style a is always the default.
Paul Eggert's avatar
Paul Eggert committed
58
  For 2-char comment markers, the style b flag is looked up only on the second
59
  char of the comment marker and on the first char of the comment ender.
Paul Eggert's avatar
Paul Eggert committed
60 61
  For style c (like the nested flag), the flag can be placed on any of
  the chars.  */
62

63
/* These functions extract specific flags from an integer
64 65
   that holds the syntax code and the flags.  */

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 92 93 94 95 96 97 98 99 100 101 102 103 104 105 106 107 108 109 110
static bool
SYNTAX_FLAGS_COMSTART_FIRST (int flags)
{
  return (flags >> 16) & 1;
}
static bool
SYNTAX_FLAGS_COMSTART_SECOND (int flags)
{
  return (flags >> 17) & 1;
}
static bool
SYNTAX_FLAGS_COMEND_FIRST (int flags)
{
  return (flags >> 18) & 1;
}
static bool
SYNTAX_FLAGS_COMEND_SECOND (int flags)
{
  return (flags >> 19) & 1;
}
static bool
SYNTAX_FLAGS_PREFIX (int flags)
{
  return (flags >> 20) & 1;
}
static bool
SYNTAX_FLAGS_COMMENT_STYLEB (int flags)
{
  return (flags >> 21) & 1;
}
static bool
SYNTAX_FLAGS_COMMENT_STYLEC (int flags)
{
  return (flags >> 23) & 1;
}
static int
SYNTAX_FLAGS_COMMENT_STYLEC2 (int flags)
{
  return (flags >> 22) & 2; /* SYNTAX_FLAGS_COMMENT_STYLEC (flags) * 2 */
}
static bool
SYNTAX_FLAGS_COMMENT_NESTED (int flags)
{
  return (flags >> 22) & 1;
}
111 112 113

/* FLAGS should be the flags of the main char of the comment marker, e.g.
   the second for comstart and the first for comend.  */
114 115 116 117 118 119 120
static int
SYNTAX_FLAGS_COMMENT_STYLE (int flags, int other_flags)
{
  return (SYNTAX_FLAGS_COMMENT_STYLEB (flags)
	  | SYNTAX_FLAGS_COMMENT_STYLEC2 (flags)
	  | SYNTAX_FLAGS_COMMENT_STYLEC2 (other_flags));
}
121

122
/* Extract a particular flag for a given character.  */
123

124 125 126 127 128
static bool
SYNTAX_COMEND_FIRST (int c)
{
  return SYNTAX_FLAGS_COMEND_FIRST (SYNTAX_WITH_FLAGS (c));
}
129 130

/* We use these constants in place for comment-style and
131
   string-ender-char to distinguish comments/strings started by
132 133
   comment_fence and string_fence codes.  */

134 135 136 137 138
enum
  {
    ST_COMMENT_STYLE = 256 + 1,
    ST_STRING_STYLE = 256 + 2
  };
Richard M. Stallman's avatar
Richard M. Stallman committed
139

140 141 142 143
/* This is the internal form of the parse state used in parse-partial-sexp.  */

struct lisp_parse_state
  {
144
    EMACS_INT depth;	/* Depth at end of parsing.  */
145
    int instring;  /* -1 if not within string, else desired terminator.  */
146
    EMACS_INT incomment; /* -1 if in unnestable comment else comment nesting */
147
    int comstyle;  /* comment style a=0, or b=1, or ST_COMMENT_STYLE.  */
Paul Eggert's avatar
Paul Eggert committed
148
    bool quoted;   /* True if just after an escape char at end of parsing.  */
149
    EMACS_INT mindepth;	/* Minimum depth seen while scanning.  */
150
    /* Char number of most recent start-of-expression at current level */
151
    ptrdiff_t thislevelstart;
152
    /* Char number of start of containing expression */
153 154
    ptrdiff_t prevlevelstart;
    ptrdiff_t location;	     /* Char number at which parsing stopped.  */
155
    ptrdiff_t location_byte; /* Corresponding byte position.  */
156
    ptrdiff_t comstr_start;  /* Position of last comment/string starter.  */
157 158
    Lisp_Object levelstarts; /* Char numbers of starts-of-expression
				of levels (starting from outermost).  */
159 160
  };

161 162 163
/* 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.
164
   find_start_value_byte is the corresponding byte position.
165 166 167 168
   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.  */

169 170 171
static ptrdiff_t find_start_pos;
static ptrdiff_t find_start_value;
static ptrdiff_t find_start_value_byte;
172
static struct buffer *find_start_buffer;
173 174
static ptrdiff_t find_start_begv;
static EMACS_INT find_start_modiff;
175 176


Paul Eggert's avatar
Paul Eggert committed
177 178 179
static Lisp_Object skip_chars (bool, Lisp_Object, Lisp_Object, bool);
static Lisp_Object skip_syntaxes (bool, Lisp_Object, Lisp_Object);
static Lisp_Object scan_lists (EMACS_INT, EMACS_INT, EMACS_INT, bool);
180
static void scan_sexps_forward (struct lisp_parse_state *,
181
                                ptrdiff_t, ptrdiff_t, ptrdiff_t, EMACS_INT,
Paul Eggert's avatar
Paul Eggert committed
182 183
                                bool, Lisp_Object, int);
static bool in_classes (int, Lisp_Object);
Paul Eggert's avatar
Paul Eggert committed
184 185

/* This setter is used only in this file, so it can be private.  */
186
static void
Paul Eggert's avatar
Paul Eggert committed
187 188
bset_syntax_table (struct buffer *b, Lisp_Object val)
{
189
  b->syntax_table_ = val;
Paul Eggert's avatar
Paul Eggert committed
190
}
191

192
/* Whether the syntax of the character C has the prefix flag set.  */
Paul Eggert's avatar
Paul Eggert committed
193 194
bool
syntax_prefix_flag_p (int c)
195
{
196
  return SYNTAX_FLAGS_PREFIX (SYNTAX_WITH_FLAGS (c));
197
}
198 199 200

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

201
enum { INTERVALS_AT_ONCE = 10 };	/* 1 + max-number of intervals
202 203
					   to scan to property-change.  */

204 205 206 207 208 209 210 211 212 213 214 215 216 217 218 219 220 221 222 223 224 225 226 227 228 229 230 231 232 233 234 235 236 237 238 239 240 241 242 243 244 245 246 247 248
/* Set the syntax entry VAL for char C in table TABLE.  */

static void
SET_RAW_SYNTAX_ENTRY (Lisp_Object table, int c, Lisp_Object val)
{
  CHAR_TABLE_SET (table, c, val);
}

/* Set the syntax entry VAL for char-range RANGE in table TABLE.
   RANGE is a cons (FROM . TO) specifying the range of characters.  */

static void
SET_RAW_SYNTAX_ENTRY_RANGE (Lisp_Object table, Lisp_Object range,
			    Lisp_Object val)
{
  Fset_char_table_range (table, range, val);
}

/* Extract the information from the entry for character C
   in the current syntax table.  */

static Lisp_Object
SYNTAX_MATCH (int c)
{
  Lisp_Object ent = SYNTAX_ENTRY (c);
  return CONSP (ent) ? XCDR (ent) : Qnil;
}

/* This should be called with FROM at the start of forward
   search, or after the last position of the backward search.  It
   makes sure that the first char is picked up with correct table, so
   one does not need to call UPDATE_SYNTAX_TABLE immediately after the
   call.
   Sign of COUNT gives the direction of the search.
 */

static void
SETUP_SYNTAX_TABLE (ptrdiff_t from, ptrdiff_t count)
{
  SETUP_BUFFER_SYNTAX_TABLE ();
  gl_state.b_property = BEGV;
  gl_state.e_property = ZV + 1;
  gl_state.object = Qnil;
  gl_state.offset = 0;
  if (parse_sexp_lookup_properties)
249 250 251 252 253 254 255 256 257
    {
      if (count > 0 || from > BEGV)
	update_syntax_table (count > 0 ? from : from - 1, count, true, Qnil);
      if (gl_state.e_property > parse_sexp_propertize_done)
	{
	  gl_state.e_property = parse_sexp_propertize_done;
	  gl_state.e_property_truncated = true;
	}
    }
258 259 260 261 262 263 264 265 266 267 268 269 270 271 272 273 274 275 276 277 278 279 280 281 282 283 284 285 286 287 288 289 290 291 292 293 294 295 296 297 298 299 300 301 302 303
}

/* Same as above, but in OBJECT.  If OBJECT is nil, use current buffer.
   If it is t (which is only used in fast_c_string_match_ignore_case),
   ignore properties altogether.

   This is meant for regex.c to use.  For buffers, regex.c passes arguments
   to the UPDATE_SYNTAX_TABLE functions which are relative to BEGV.
   So if it is a buffer, we set the offset field to BEGV.  */

void
SETUP_SYNTAX_TABLE_FOR_OBJECT (Lisp_Object object,
			       ptrdiff_t from, ptrdiff_t count)
{
  SETUP_BUFFER_SYNTAX_TABLE ();
  gl_state.object = object;
  if (BUFFERP (gl_state.object))
    {
      struct buffer *buf = XBUFFER (gl_state.object);
      gl_state.b_property = 1;
      gl_state.e_property = BUF_ZV (buf) - BUF_BEGV (buf) + 1;
      gl_state.offset = BUF_BEGV (buf) - 1;
    }
  else if (NILP (gl_state.object))
    {
      gl_state.b_property = 1;
      gl_state.e_property = ZV - BEGV + 1;
      gl_state.offset = BEGV - 1;
    }
  else if (EQ (gl_state.object, Qt))
    {
      gl_state.b_property = 0;
      gl_state.e_property = PTRDIFF_MAX;
      gl_state.offset = 0;
    }
  else
    {
      gl_state.b_property = 0;
      gl_state.e_property = 1 + SCHARS (gl_state.object);
      gl_state.offset = 0;
    }
  if (parse_sexp_lookup_properties)
    update_syntax_table (from + gl_state.offset - (count <= 0),
			 count, 1, gl_state.object);
}

304 305
/* Update gl_state to an appropriate interval which contains CHARPOS.  The
   sign of COUNT give the relative position of CHARPOS wrt the previously
306
   valid interval.  If INIT, only [be]_property fields of gl_state are
307
   valid at start, the rest is filled basing on OBJECT.
308

309
   `gl_state.*_i' are the intervals, and CHARPOS is further in the search
310 311
   direction than the intervals - or in an interval.  We update the
   current syntax-table basing on the property of this interval, and
312
   update the interval to start further than CHARPOS - or be
Dmitry Antipov's avatar
Dmitry Antipov committed
313
   NULL.  We also update lim_property to be the next value of
314
   charpos to call this subroutine again - or be before/after the
315 316 317
   start/end of OBJECT.  */

void
Paul Eggert's avatar
Paul Eggert committed
318
update_syntax_table (ptrdiff_t charpos, EMACS_INT count, bool init,
319
		     Lisp_Object object)
320 321
{
  Lisp_Object tmp_table;
Paul Eggert's avatar
Paul Eggert committed
322
  int cnt = 0;
323
  bool invalidate = true;
324
  INTERVAL i;
325 326 327

  if (init)
    {
328
      gl_state.old_prop = Qnil;
329 330
      gl_state.start = gl_state.b_property;
      gl_state.stop = gl_state.e_property;
331 332
      i = interval_of (charpos, object);
      gl_state.backward_i = gl_state.forward_i = i;
333
      invalidate = false;
Dmitry Antipov's avatar
Dmitry Antipov committed
334
      if (!i)
335
	return;
336
      /* interval_of updates only ->position of the return value, so
337
	 update the parents manually to speed up update_interval.  */
338
      while (!NULL_PARENT (i))
339 340
	{
	  if (AM_RIGHT_CHILD (i))
341
	    INTERVAL_PARENT (i)->position = i->position
342
	      - LEFT_TOTAL_LENGTH (i) + TOTAL_LENGTH (i) /* right end */
343 344
	      - TOTAL_LENGTH (INTERVAL_PARENT (i))
	      + LEFT_TOTAL_LENGTH (INTERVAL_PARENT (i));
345
	  else
346
	    INTERVAL_PARENT (i)->position = i->position - LEFT_TOTAL_LENGTH (i)
347
	      + TOTAL_LENGTH (i);
348
	  i = INTERVAL_PARENT (i);
349 350
	}
      i = gl_state.forward_i;
351
      gl_state.b_property = i->position - gl_state.offset;
352
      gl_state.e_property = INTERVAL_LAST_POS (i) - gl_state.offset;
353 354
      goto update;
    }
355
  i = count > 0 ? gl_state.forward_i : gl_state.backward_i;
356

357
  /* We are guaranteed to be called with CHARPOS either in i,
358
     or further off.  */
Dmitry Antipov's avatar
Dmitry Antipov committed
359
  if (!i)
360
    error ("Error in syntax_table logic for to-the-end intervals");
361
  else if (charpos < i->position)		/* Move left.  */
362 363
    {
      if (count > 0)
364
	error ("Error in syntax_table logic for intervals <-");
365
      /* Update the interval.  */
366
      i = update_interval (i, charpos);
367
      if (INTERVAL_LAST_POS (i) != gl_state.b_property)
368
	{
369
	  invalidate = false;
370
	  gl_state.forward_i = i;
371
	  gl_state.e_property = INTERVAL_LAST_POS (i) - gl_state.offset;
372
	}
373
    }
374
  else if (charpos >= INTERVAL_LAST_POS (i)) /* Move right.  */
375 376
    {
      if (count < 0)
377
	error ("Error in syntax_table logic for intervals ->");
378
      /* Update the interval.  */
379
      i = update_interval (i, charpos);
380
      if (i->position != gl_state.e_property)
381
	{
382
	  invalidate = false;
383
	  gl_state.backward_i = i;
384
	  gl_state.b_property = i->position - gl_state.offset;
385 386 387 388 389 390 391 392
	}
    }

  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
393

394 395 396
  if (invalidate)		/* Did not get to adjacent interval.  */
    {				/* with the same table => */
				/* invalidate the old range.  */
397 398 399
      if (count > 0)
	{
	  gl_state.backward_i = i;
400 401 402
	  gl_state.b_property = i->position - gl_state.offset;
	}
      else
403
	{
404
	  gl_state.forward_i = i;
405
	  gl_state.e_property = INTERVAL_LAST_POS (i) - gl_state.offset;
406 407
	}
    }
408

409
  if (!EQ (tmp_table, gl_state.old_prop))
410
    {
411 412 413 414 415
      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
416
	}
417 418 419 420 421
      else if (CONSP (tmp_table))
	{
	  gl_state.use_global = 1;
	  gl_state.global_code = tmp_table;
	}
Juanma Barranquero's avatar
Juanma Barranquero committed
422
      else
423 424
	{
	  gl_state.use_global = 0;
Tom Tromey's avatar
Tom Tromey committed
425
	  gl_state.current_syntax_table = BVAR (current_buffer, syntax_table);
426
	}
427 428
    }

Dmitry Antipov's avatar
Dmitry Antipov committed
429
  while (i)
430 431 432 433
    {
      if (cnt && !EQ (tmp_table, textget (i->plist, Qsyntax_table)))
	{
	  if (count > 0)
434 435 436 437 438 439
	    {
	      gl_state.e_property = i->position - gl_state.offset;
	      gl_state.forward_i = i;
	    }
	  else
	    {
440 441
	      gl_state.b_property
		= i->position + LENGTH (i) - gl_state.offset;
442 443 444
	      gl_state.backward_i = i;
	    }
	  return;
445
	}
Juanma Barranquero's avatar
Juanma Barranquero committed
446
      else if (cnt == INTERVALS_AT_ONCE)
447 448
	{
	  if (count > 0)
449
	    {
450 451 452 453 454
	      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.  */
Dmitry Antipov's avatar
Dmitry Antipov committed
455
		+ (next_interval (i) ? 0 : 1);
456 457 458 459 460 461 462 463
	      gl_state.forward_i = i;
	    }
	  else
	    {
	      gl_state.b_property = i->position - gl_state.offset;
	      gl_state.backward_i = i;
	    }
	  return;
464 465 466 467
	}
      cnt++;
      i = count > 0 ? next_interval (i) : previous_interval (i);
    }
Dmitry Antipov's avatar
Dmitry Antipov committed
468
  eassert (i == NULL); /* This property goes to the end.  */
469
  if (count > 0)
470 471 472 473
    {
      gl_state.e_property = gl_state.stop;
      gl_state.forward_i = i;
    }
474 475
  else
    gl_state.b_property = gl_state.start;
476
}
477 478 479 480 481 482 483 484 485 486 487 488 489 490 491 492 493 494 495 496 497 498 499 500 501 502 503 504 505 506 507 508 509 510

static void
parse_sexp_propertize (ptrdiff_t charpos)
{
  EMACS_INT modiffs = CHARS_MODIFF;
  safe_call1 (Vparse_sexp_propertize_function,
	      make_number (1 + charpos));
  if (modiffs != CHARS_MODIFF)
    error ("parse-sexp-propertize-function modified the buffer!");
  if (parse_sexp_propertize_done <= charpos)
    error ("parse-sexp-propertize-function did not move"
	   " parse-sexp-propertize-done");
  SETUP_SYNTAX_TABLE (charpos, 1);
}

void
update_syntax_table_forward (ptrdiff_t charpos, bool init,
		     Lisp_Object object)
{
  if (!(gl_state.e_property_truncated))
    update_syntax_table (charpos, 1, init, object);
  if ((gl_state.e_property > parse_sexp_propertize_done
       || gl_state.e_property_truncated)
      && NILP (object))
    {
      if (parse_sexp_propertize_done > charpos)
	{
	  gl_state.e_property = parse_sexp_propertize_done;
	  gl_state.e_property_truncated = true;
	}
      else
	parse_sexp_propertize (charpos);
    }
}
511

Paul Eggert's avatar
Paul Eggert committed
512
/* Returns true if char at CHARPOS is quoted.
513
   Global syntax-table data should be set up already to be good at CHARPOS
514
   or after.  On return global syntax data is good for lookup at CHARPOS.  */
515

Paul Eggert's avatar
Paul Eggert committed
516
static bool
517
char_quoted (ptrdiff_t charpos, ptrdiff_t bytepos)
518
{
Paul Eggert's avatar
Paul Eggert committed
519 520 521
  enum syntaxcode code;
  ptrdiff_t beg = BEGV;
  bool quoted = 0;
522
  ptrdiff_t orig = charpos;
523

524
  while (charpos > beg)
525
    {
526
      int c;
527
      DEC_BOTH (charpos, bytepos);
528

529
      UPDATE_SYNTAX_TABLE_BACKWARD (charpos);
530
      c = FETCH_CHAR_AS_MULTIBYTE (bytepos);
531
      code = SYNTAX (c);
532 533 534 535
      if (! (code == Scharquote || code == Sescape))
	break;

      quoted = !quoted;
536
    }
537 538

  UPDATE_SYNTAX_TABLE (orig);
539 540
  return quoted;
}
541 542 543 544

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

545
static ptrdiff_t
546
dec_bytepos (ptrdiff_t bytepos)
547
{
Tom Tromey's avatar
Tom Tromey committed
548
  if (NILP (BVAR (current_buffer, enable_multibyte_characters)))
549 550
    return bytepos - 1;

551 552 553
  DEC_POS (bytepos);
  return bytepos;
}
554

Juanma Barranquero's avatar
Juanma Barranquero committed
555
/* Return a defun-start position before POS and not too far before.
Richard M. Stallman's avatar
Richard M. Stallman committed
556 557 558
   It should be the last one before POS, or nearly the last.

   When open_paren_in_column_0_is_defun_start is nonzero,
559
   only the beginning of the buffer is treated as a defun-start.
Richard M. Stallman's avatar
Richard M. Stallman committed
560 561 562 563

   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.
564 565 566 567

   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.  */
568

569 570
static ptrdiff_t
find_defun_start (ptrdiff_t pos, ptrdiff_t pos_byte)
571
{
572
  ptrdiff_t opoint = PT, opoint_byte = PT_BYTE;
573 574 575 576 577 578 579 580 581 582 583 584

  /* 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;

585 586 587 588 589 590 591
  if (!open_paren_in_column_0_is_defun_start)
    {
      find_start_value = BEGV;
      find_start_value_byte = BEGV_BYTE;
      goto found;
    }

592
  /* Back up to start of line.  */
593
  scan_newline (pos, pos_byte, BEGV, BEGV_BYTE, -1, 1);
594

595 596 597
  /* We optimize syntax-table lookup for rare updates.  Thus we accept
     only those `^\s(' which are good in global _and_ text-property
     syntax-tables.  */
598
  SETUP_BUFFER_SYNTAX_TABLE ();
599
  while (PT > BEGV)
600
    {
601 602
      int c;

603 604
      /* Open-paren at start of line means we may have found our
	 defun-start.  */
605
      c = FETCH_CHAR_AS_MULTIBYTE (PT_BYTE);
606
      if (SYNTAX (c) == Sopen)
607
	{
608
	  SETUP_SYNTAX_TABLE (PT + 1, -1);	/* Try again... */
609
	  c = FETCH_CHAR_AS_MULTIBYTE (PT_BYTE);
610
	  if (SYNTAX (c) == Sopen)
611 612
	    break;
	  /* Now fallback to the default value.  */
613
	  SETUP_BUFFER_SYNTAX_TABLE ();
614
	}
615 616
      /* Move to beg of previous line.  */
      scan_newline (PT, PT_BYTE, BEGV, BEGV_BYTE, -2, 1);
617 618 619
    }

  /* Record what we found, for the next try.  */
620 621
  find_start_value = PT;
  find_start_value_byte = PT_BYTE;
622 623 624
  TEMP_SET_PT_BOTH (opoint, opoint_byte);

 found:
625 626 627 628 629 630 631 632
  find_start_buffer = current_buffer;
  find_start_modiff = MODIFF;
  find_start_begv = BEGV;
  find_start_pos = pos;

  return find_start_value;
}

633 634
/* Return the SYNTAX_COMEND_FIRST of the character before POS, POS_BYTE.  */

Paul Eggert's avatar
Paul Eggert committed
635
static bool
636
prev_char_comend_first (ptrdiff_t pos, ptrdiff_t pos_byte)
637
{
Paul Eggert's avatar
Paul Eggert committed
638 639
  int c;
  bool val;
640 641 642 643 644 645 646 647 648

  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;
}

Paul Eggert's avatar
Paul Eggert committed
649
/* Check whether charpos FROM is at the end of a comment.
650 651 652
   FROM_BYTE is the bytepos corresponding to FROM.
   Do not move back before STOP.

Paul Eggert's avatar
Paul Eggert committed
653
   Return true if we find a comment ending at FROM/FROM_BYTE.
654 655 656

   If successful, store the charpos of the comment's beginning
   into *CHARPOS_PTR, and the bytepos into *BYTEPOS_PTR.
657 658 659

   Global syntax data remains valid for backward search starting at
   the returned value (or at FROM, if the search was not successful).  */
660

Paul Eggert's avatar
Paul Eggert committed
661 662 663 664
static bool
back_comment (ptrdiff_t from, ptrdiff_t from_byte, ptrdiff_t stop,
	      bool comnested, int comstyle, ptrdiff_t *charpos_ptr,
	      ptrdiff_t *bytepos_ptr)
665 666 667 668 669 670 671 672 673 674
{
  /* 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.  */
675
  int string_style = -1;	/* Presumed outside of any string.  */
Paul Eggert's avatar
Paul Eggert committed
676
  bool string_lossage = 0;
677
  /* Not a real lossage: indicates that we have passed a matching comment
Juanma Barranquero's avatar
Juanma Barranquero committed
678
     starter plus a non-matching comment-ender, meaning that any matching
679 680 681
     comment-starter we might see later could be a false positive (hidden
     inside another comment).
     Test case:  { a (* b } c (* d *) */
Paul Eggert's avatar
Paul Eggert committed
682
  bool comment_lossage = 0;
683 684 685 686
  ptrdiff_t comment_end = from;
  ptrdiff_t comment_end_byte = from_byte;
  ptrdiff_t comstart_pos = 0;
  ptrdiff_t comstart_byte IF_LINT (= 0);
687 688
  /* Place where the containing defun starts,
     or 0 if we didn't come across it yet.  */
689 690
  ptrdiff_t defun_start = 0;
  ptrdiff_t defun_start_byte = 0;
Paul Eggert's avatar
Paul Eggert committed
691
  enum syntaxcode code;
692
  ptrdiff_t nesting = 1;		/* Current comment nesting.  */
693
  int c;
694 695 696 697 698
  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.  */
699 700 701 702 703

  /* At beginning of range to scan, we're outside of strings;
     that determines quote parity to the comment-end.  */
  while (from != stop)
    {
704
      ptrdiff_t temp_byte;
Paul Eggert's avatar
Paul Eggert committed
705 706
      int prev_syntax;
      bool com2start, com2end, comstart;
707

708
      /* Move back and examine a character.  */
709
      DEC_BOTH (from, from_byte);
710 711
      UPDATE_SYNTAX_TABLE_BACKWARD (from);

712
      prev_syntax = syntax;
713
      c = FETCH_CHAR_AS_MULTIBYTE (from_byte);
714
      syntax = SYNTAX_WITH_FLAGS (c);
715 716
      code = SYNTAX (c);

717 718 719
      /* Check for 2-char comment markers.  */
      com2start = (SYNTAX_FLAGS_COMSTART_FIRST (syntax)
		   && SYNTAX_FLAGS_COMSTART_SECOND (prev_syntax)
720 721
		   && (comstyle
		       == SYNTAX_FLAGS_COMMENT_STYLE (prev_syntax, syntax))
722 723 724 725
		   && (SYNTAX_FLAGS_COMMENT_NESTED (prev_syntax)
		       || SYNTAX_FLAGS_COMMENT_NESTED (syntax)) == comnested);
      com2end = (SYNTAX_FLAGS_COMEND_FIRST (syntax)
		 && SYNTAX_FLAGS_COMEND_SECOND (prev_syntax));
726
      comstart = (com2start || code == Scomment);
727

728 729 730 731 732 733 734 735 736 737
      /* 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,
738 739 740
	 we don't try to be clever.  E.g. |*| in C, or }% in modes that
	 have %..\n and %{..}%.  */
      if (from > stop && (com2end || comstart))
741
	{
742
	  ptrdiff_t next = from, next_byte = from_byte;
743
	  int next_c, next_syntax;
744 745
	  DEC_BOTH (next, next_byte);
	  UPDATE_SYNTAX_TABLE_BACKWARD (next);
746
	  next_c = FETCH_CHAR_AS_MULTIBYTE (next_byte);
747
	  next_syntax = SYNTAX_WITH_FLAGS (next_c);
748
	  if (((comstart || comnested)
749 750 751 752
	       && SYNTAX_FLAGS_COMEND_SECOND (syntax)
	       && SYNTAX_FLAGS_COMEND_FIRST (next_syntax))
	      || ((com2end || comnested)
		  && SYNTAX_FLAGS_COMSTART_SECOND (syntax)
753 754
		  && (comstyle
		      == SYNTAX_FLAGS_COMMENT_STYLE (syntax, prev_syntax))
755 756 757
		  && SYNTAX_FLAGS_COMSTART_FIRST (next_syntax)))
	    goto lossage;
	  /* UPDATE_SYNTAX_TABLE_FORWARD (next + 1); */
758
	}
759 760 761 762 763 764 765 766 767 768 769 770 771 772 773

      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
774
	       && (comstyle != SYNTAX_FLAGS_COMMENT_STYLE (syntax, 0)
775
		   || SYNTAX_FLAGS_COMMENT_NESTED (syntax) != comnested))
776
	continue;
777

778 779
      /* Ignore escaped characters, except comment-enders.  */
      if (code != Sendcomment && char_quoted (from, from_byte))
780 781
	continue;

782
      switch (code)
783
	{
784 785 786 787 788 789 790 791 792 793 794 795 796 797
	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.  */
798
	    string_lossage = 1;
799
	  break;
Juanma Barranquero's avatar
Juanma Barranquero committed
800

801 802
	case Scomment:
	  /* We've already checked that it is the relevant comstyle.  */
803
	  if (string_style != -1 || comment_lossage || string_lossage)
804 805 806 807
	    /* There are odd string quotes involved, so let's be careful.
	       Test case in Pascal: " { " a { " } */
	    goto lossage;

808 809 810 811 812 813 814
	  if (!comnested)
	    {
	      /* Record best comment-starter so far.  */
	      comstart_pos = from;
	      comstart_byte = from_byte;
	    }
	  else if (--nesting <= 0)
815 816 817 818 819 820
	    /* 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;
821 822
	  break;

823
	case Sendcomment:
824
	  if (SYNTAX_FLAGS_COMMENT_STYLE (syntax, 0) == comstyle
825
	      && ((com2end && SYNTAX_FLAGS_COMMENT_NESTED (prev_syntax))
826
		  || SYNTAX_FLAGS_COMMENT_NESTED (syntax)) == comnested)
827 828 829 830 831 832 833 834 835
	    /* 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.  */
	    }
836 837 838 839 840 841 842 843 844
	  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;
845
	  break;
846

847 848 849 850 851 852 853 854 855 856 857
	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.  */
	    }
858
	  break;
859 860

	default:
861
	  break;
862
	}
863 864 865 866 867
    }

  if (comstart_pos == 0)
    {
      from = comment_end;
868
      from_byte = comment_end_byte;
869
      UPDATE_SYNTAX_TABLE_FORWARD (comment_end);
870
    }
871 872 873
  /* 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 */
874 875
    {
      from = comstart_pos;
876
      from_byte = comstart_byte;
877
      UPDATE_SYNTAX_TABLE_FORWARD (from - 1);
878
    }
879
  else lossage:
880
    {
881
      struct lisp_parse_state state;
882
      bool adjusted = true;
883 884
      /* We had two kinds of string delimiters mixed up
	 together.  Decode this going forwards.
885
	 Scan fwd from a known safe place (beginning-of-defun)
886 887
	 to the one in question; this records where we
	 last passed a comment starter.  */
888 889 890 891 892
      /* 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;
893
	  adjusted = (defun_start > BEGV);
894
	}
895
      do
896
	{
897 898
	  scan_sexps_forward (&state,
			      defun_start, defun_start_byte,
899 900
			      comment_end, TYPE_MINIMUM (EMACS_INT),
			      0, Qnil, 0);
901
	  defun_start = comment_end;
902 903 904 905 906 907 908 909 910 911
	  if (!adjusted)
	    {
	      adjusted = true;
	      find_start_value
		= CONSP (state.levelstarts) ? XINT (XCAR (state.levelstarts))
		: state.thislevelstart >= 0 ? state.thislevelstart
		: find_start_value;
	      find_start_value_byte = CHAR_TO_BYTE (find_start_value);
	    }

912 913 914 915 916 917 918 919 920 921 922
	  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 (* " *)  */
		{
923
		  /* FIXME: We should advance by one or two chars.  */
924 925 926 927 928 929
		  defun_start = state.comstr_start + 2;
		  defun_start_byte = CHAR_TO_BYTE (defun_start);
		}
	    }
	} while (defun_start < comment_end);

930
      from_byte = CHAR_TO_BYTE (from);
931 932
      UPDATE_SYNTAX_TABLE_FORWARD (from - 1);
    }
Juanma Barranquero's avatar
Juanma Barranquero committed
933

934
 done:
935 936 937
  *charpos_ptr = from;
  *bytepos_ptr = from_byte;

Paul Eggert's avatar
Paul Eggert committed
938
  return from != comment_end;
939 940
}

Richard M. Stallman's avatar
Richard M. Stallman committed
941
DEFUN ("syntax-table-p", Fsyntax_table_p, Ssyntax_table_p, 1, 1, 0,
942 943
       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
944
  (Lisp_Object object)
Richard M. Stallman's avatar
Richard M. Stallman committed
945
{
946
  if (CHAR_TABLE_P (object)
947
      && EQ (XCHAR_TABLE (object)->purpose, Qsyntax_table))
Richard M. Stallman's avatar
Richard M. Stallman committed
948 949 950 951
    return Qt;
  return Qnil;
}

952
static void
953
check_syntax_table (Lisp_Object obj)
Richard M. Stallman's avatar
Richard M. Stallman committed
954
{
955 956
  CHECK_TYPE (CHAR_TABLE_P (obj) && EQ (XCHAR_TABLE (obj)->purpose, Qsyntax_table),
	      Qsyntax_table_p, obj);
Juanma Barranquero's avatar
Juanma Barranquero committed
957
}
Richard M. Stallman's avatar
Richard M. Stallman committed
958 959

DEFUN ("syntax-table", Fsyntax_table, Ssyntax_table, 0, 0, 0,
960 961
       doc: /* Return the current syntax table.
This is the one specified by the current buffer.  */)
Dan Nicolaescu's avatar
Dan Nicolaescu committed
962
  (void)
Richard M. Stallman's avatar
Richard M. Stallman committed
963
{
Tom Tromey's avatar
Tom Tromey committed
964
  return BVAR (current_buffer, syntax_table);
Richard M. Stallman's avatar
Richard M. Stallman committed
965 966 967 968
}

DEFUN ("standard-syntax-table", Fstandard_syntax_table,
   Sstandard_syntax_table, 0, 0, 0,
969 970
       doc: /* Return the standard syntax table.
This is the one used for new buffers.  */)
Dan Nicolaescu's avatar
Dan Nicolaescu committed
971
  (void)
Richard M. Stallman's avatar
Richard M. Stallman committed
972 973 974 975 976
{
  return Vstandard_syntax_table;
}

DEFUN ("copy-syntax-table", Fcopy_syntax_table, Scopy_syntax_table, 0, 1, 0,
977 978
       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
979
  (Lisp_Object table)
Richard M. Stallman's avatar
Richard M. Stallman committed
980
{
981 982
  Lisp_Object copy;

Jim Blandy's avatar
Jim Blandy committed
983
  if (!NILP (table))
984 985 986 987 988
    check_syntax_table (table);
  else
    table = Vstandard_syntax_table;

  copy = Fcopy_sequence (table);
989 990 991

  /* Only the standard syntax table should have a default element.
     Other syntax tables should inherit from parents instead.  */
Paul Eggert's avatar
Paul Eggert committed
992
  set_char_table_defalt (copy, Qnil);
993 994 995 996 997 998

  /* 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);
999
  return copy;
Richard M. Stallman's avatar
Richard M. Stallman committed
1000 1001 1002
}

DEFUN ("set-syntax-table", Fset_syntax_table, Sset_syntax_table, 1, 1, 0,