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.  */

204 205 206 207 208 209 210 211 212 213 214 215 216 217 218 219 220 221 222 223 224 225 226 227 228 229 230 231 232 233 234 235 236 237 238 239 240 241 242 243 244 245 246 247 248 249
/* 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)
    if (count > 0 || from > BEGV)
250
      update_syntax_table (count > 0 ? from : from - 1, count, true, Qnil);
251 252 253 254 255 256 257 258 259 260 261 262 263 264 265 266 267 268 269 270 271 272 273 274 275 276 277 278 279 280 281 282 283 284 285 286 287 288 289 290 291 292 293 294 295 296
}

/* 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);
}

297 298
/* Update gl_state to an appropriate interval which contains CHARPOS.  The
   sign of COUNT give the relative position of CHARPOS wrt the previously
299
   valid interval.  If INIT, only [be]_property fields of gl_state are
300
   valid at start, the rest is filled basing on OBJECT.
301

302
   `gl_state.*_i' are the intervals, and CHARPOS is further in the search
303 304
   direction than the intervals - or in an interval.  We update the
   current syntax-table basing on the property of this interval, and
305
   update the interval to start further than CHARPOS - or be
Dmitry Antipov's avatar
Dmitry Antipov committed
306
   NULL.  We also update lim_property to be the next value of
307
   charpos to call this subroutine again - or be before/after the
308 309 310
   start/end of OBJECT.  */

void
Paul Eggert's avatar
Paul Eggert committed
311
update_syntax_table (ptrdiff_t charpos, EMACS_INT count, bool init,
312
		     Lisp_Object object)
313 314
{
  Lisp_Object tmp_table;
Paul Eggert's avatar
Paul Eggert committed
315
  int cnt = 0;
316
  bool invalidate = true;
317
  INTERVAL i;
318 319 320

  if (init)
    {
321
      gl_state.old_prop = Qnil;
322 323
      gl_state.start = gl_state.b_property;
      gl_state.stop = gl_state.e_property;
324 325
      i = interval_of (charpos, object);
      gl_state.backward_i = gl_state.forward_i = i;
326
      invalidate = false;
Dmitry Antipov's avatar
Dmitry Antipov committed
327
      if (!i)
328
	return;
329
      /* interval_of updates only ->position of the return value, so
330
	 update the parents manually to speed up update_interval.  */
331
      while (!NULL_PARENT (i))
332 333
	{
	  if (AM_RIGHT_CHILD (i))
334
	    INTERVAL_PARENT (i)->position = i->position
335
	      - LEFT_TOTAL_LENGTH (i) + TOTAL_LENGTH (i) /* right end */
336 337
	      - TOTAL_LENGTH (INTERVAL_PARENT (i))
	      + LEFT_TOTAL_LENGTH (INTERVAL_PARENT (i));
338
	  else
339
	    INTERVAL_PARENT (i)->position = i->position - LEFT_TOTAL_LENGTH (i)
340
	      + TOTAL_LENGTH (i);
341
	  i = INTERVAL_PARENT (i);
342 343
	}
      i = gl_state.forward_i;
344
      gl_state.b_property = i->position - gl_state.offset;
345
      gl_state.e_property = INTERVAL_LAST_POS (i) - gl_state.offset;
346 347
      goto update;
    }
348
  i = count > 0 ? gl_state.forward_i : gl_state.backward_i;
349

350
  /* We are guaranteed to be called with CHARPOS either in i,
351
     or further off.  */
Dmitry Antipov's avatar
Dmitry Antipov committed
352
  if (!i)
353
    error ("Error in syntax_table logic for to-the-end intervals");
354
  else if (charpos < i->position)		/* Move left.  */
355 356
    {
      if (count > 0)
357
	error ("Error in syntax_table logic for intervals <-");
358
      /* Update the interval.  */
359
      i = update_interval (i, charpos);
360
      if (INTERVAL_LAST_POS (i) != gl_state.b_property)
361
	{
362
	  invalidate = false;
363
	  gl_state.forward_i = i;
364
	  gl_state.e_property = INTERVAL_LAST_POS (i) - gl_state.offset;
365
	}
366
    }
367
  else if (charpos >= INTERVAL_LAST_POS (i)) /* Move right.  */
368 369
    {
      if (count < 0)
370
	error ("Error in syntax_table logic for intervals ->");
371
      /* Update the interval.  */
372
      i = update_interval (i, charpos);
373
      if (i->position != gl_state.e_property)
374
	{
375
	  invalidate = false;
376
	  gl_state.backward_i = i;
377
	  gl_state.b_property = i->position - gl_state.offset;
378 379 380 381 382 383 384 385
	}
    }

  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
386

387 388 389
  if (invalidate)		/* Did not get to adjacent interval.  */
    {				/* with the same table => */
				/* invalidate the old range.  */
390 391 392
      if (count > 0)
	{
	  gl_state.backward_i = i;
393 394 395
	  gl_state.b_property = i->position - gl_state.offset;
	}
      else
396
	{
397
	  gl_state.forward_i = i;
398
	  gl_state.e_property = INTERVAL_LAST_POS (i) - gl_state.offset;
399 400
	}
    }
401

402
  if (!EQ (tmp_table, gl_state.old_prop))
403
    {
404 405 406 407 408
      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
409
	}
410 411 412 413 414
      else if (CONSP (tmp_table))
	{
	  gl_state.use_global = 1;
	  gl_state.global_code = tmp_table;
	}
Juanma Barranquero's avatar
Juanma Barranquero committed
415
      else
416 417
	{
	  gl_state.use_global = 0;
Tom Tromey's avatar
Tom Tromey committed
418
	  gl_state.current_syntax_table = BVAR (current_buffer, syntax_table);
419
	}
420 421
    }

Dmitry Antipov's avatar
Dmitry Antipov committed
422
  while (i)
423 424 425 426
    {
      if (cnt && !EQ (tmp_table, textget (i->plist, Qsyntax_table)))
	{
	  if (count > 0)
427 428 429 430 431 432
	    {
	      gl_state.e_property = i->position - gl_state.offset;
	      gl_state.forward_i = i;
	    }
	  else
	    {
433 434
	      gl_state.b_property
		= i->position + LENGTH (i) - gl_state.offset;
435 436 437
	      gl_state.backward_i = i;
	    }
	  return;
438
	}
Juanma Barranquero's avatar
Juanma Barranquero committed
439
      else if (cnt == INTERVALS_AT_ONCE)
440 441
	{
	  if (count > 0)
442
	    {
443 444 445 446 447
	      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
448
		+ (next_interval (i) ? 0 : 1);
449 450 451 452 453 454 455 456
	      gl_state.forward_i = i;
	    }
	  else
	    {
	      gl_state.b_property = i->position - gl_state.offset;
	      gl_state.backward_i = i;
	    }
	  return;
457 458 459 460
	}
      cnt++;
      i = count > 0 ? next_interval (i) : previous_interval (i);
    }
Dmitry Antipov's avatar
Dmitry Antipov committed
461
  eassert (i == NULL); /* This property goes to the end.  */
462
  if (count > 0)
463 464 465 466
    {
      gl_state.e_property = gl_state.stop;
      gl_state.forward_i = i;
    }
467 468
  else
    gl_state.b_property = gl_state.start;
469
}
470 471 472 473 474 475 476 477 478 479 480 481 482 483 484 485 486 487 488 489 490 491 492 493 494 495 496 497 498 499 500 501 502 503 504 505 506 507 508

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);
  if (gl_state.e_property > parse_sexp_propertize_done)
    {
      gl_state.e_property = parse_sexp_propertize_done;
      gl_state.e_property_truncated = true;
    }
}

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);
    }
}
509

Paul Eggert's avatar
Paul Eggert committed
510
/* Returns true if char at CHARPOS is quoted.
511
   Global syntax-table data should be set up already to be good at CHARPOS
512
   or after.  On return global syntax data is good for lookup at CHARPOS.  */
513

Paul Eggert's avatar
Paul Eggert committed
514
static bool
515
char_quoted (ptrdiff_t charpos, ptrdiff_t bytepos)
516
{
Paul Eggert's avatar
Paul Eggert committed
517 518 519
  enum syntaxcode code;
  ptrdiff_t beg = BEGV;
  bool quoted = 0;
520
  ptrdiff_t orig = charpos;
521

522
  while (charpos > beg)
523
    {
524
      int c;
525
      DEC_BOTH (charpos, bytepos);
526

527
      UPDATE_SYNTAX_TABLE_BACKWARD (charpos);
528
      c = FETCH_CHAR_AS_MULTIBYTE (bytepos);
529
      code = SYNTAX (c);
530 531 532 533
      if (! (code == Scharquote || code == Sescape))
	break;

      quoted = !quoted;
534
    }
535 536

  UPDATE_SYNTAX_TABLE (orig);
537 538
  return quoted;
}
539 540 541 542

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

543
static ptrdiff_t
544
dec_bytepos (ptrdiff_t bytepos)
545
{
Tom Tromey's avatar
Tom Tromey committed
546
  if (NILP (BVAR (current_buffer, enable_multibyte_characters)))
547 548
    return bytepos - 1;

549 550 551
  DEC_POS (bytepos);
  return bytepos;
}
552

Juanma Barranquero's avatar
Juanma Barranquero committed
553
/* Return a defun-start position before POS and not too far before.
Richard M. Stallman's avatar
Richard M. Stallman committed
554 555 556
   It should be the last one before POS, or nearly the last.

   When open_paren_in_column_0_is_defun_start is nonzero,
557
   only the beginning of the buffer is treated as a defun-start.
Richard M. Stallman's avatar
Richard M. Stallman committed
558 559 560 561

   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.
562 563 564 565

   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.  */
566

567 568
static ptrdiff_t
find_defun_start (ptrdiff_t pos, ptrdiff_t pos_byte)
569
{
570
  ptrdiff_t opoint = PT, opoint_byte = PT_BYTE;
571 572 573 574 575 576 577 578 579 580 581 582

  /* 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;

583 584 585 586 587 588 589
  if (!open_paren_in_column_0_is_defun_start)
    {
      find_start_value = BEGV;
      find_start_value_byte = BEGV_BYTE;
      goto found;
    }

590
  /* Back up to start of line.  */
591
  scan_newline (pos, pos_byte, BEGV, BEGV_BYTE, -1, 1);
592

593 594 595
  /* We optimize syntax-table lookup for rare updates.  Thus we accept
     only those `^\s(' which are good in global _and_ text-property
     syntax-tables.  */
596
  SETUP_BUFFER_SYNTAX_TABLE ();
597
  while (PT > BEGV)
598
    {
599 600
      int c;

601 602
      /* Open-paren at start of line means we may have found our
	 defun-start.  */
603
      c = FETCH_CHAR_AS_MULTIBYTE (PT_BYTE);
604
      if (SYNTAX (c) == Sopen)
605
	{
606
	  SETUP_SYNTAX_TABLE (PT + 1, -1);	/* Try again... */
607
	  c = FETCH_CHAR_AS_MULTIBYTE (PT_BYTE);
608
	  if (SYNTAX (c) == Sopen)
609 610
	    break;
	  /* Now fallback to the default value.  */
611
	  SETUP_BUFFER_SYNTAX_TABLE ();
612
	}
613 614
      /* Move to beg of previous line.  */
      scan_newline (PT, PT_BYTE, BEGV, BEGV_BYTE, -2, 1);
615 616 617
    }

  /* Record what we found, for the next try.  */
618 619
  find_start_value = PT;
  find_start_value_byte = PT_BYTE;
620 621 622
  TEMP_SET_PT_BOTH (opoint, opoint_byte);

 found:
623 624 625 626 627 628 629 630
  find_start_buffer = current_buffer;
  find_start_modiff = MODIFF;
  find_start_begv = BEGV;
  find_start_pos = pos;

  return find_start_value;
}

631 632
/* Return the SYNTAX_COMEND_FIRST of the character before POS, POS_BYTE.  */

Paul Eggert's avatar
Paul Eggert committed
633
static bool
634
prev_char_comend_first (ptrdiff_t pos, ptrdiff_t pos_byte)
635
{
Paul Eggert's avatar
Paul Eggert committed
636 637
  int c;
  bool val;
638 639 640 641 642 643 644 645 646

  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
647
/* Check whether charpos FROM is at the end of a comment.
648 649 650
   FROM_BYTE is the bytepos corresponding to FROM.
   Do not move back before STOP.

Paul Eggert's avatar
Paul Eggert committed
651
   Return true if we find a comment ending at FROM/FROM_BYTE.
652 653 654

   If successful, store the charpos of the comment's beginning
   into *CHARPOS_PTR, and the bytepos into *BYTEPOS_PTR.
655 656 657

   Global syntax data remains valid for backward search starting at
   the returned value (or at FROM, if the search was not successful).  */
658

Paul Eggert's avatar
Paul Eggert committed
659 660 661 662
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)
663 664 665 666 667 668 669 670 671 672
{
  /* 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.  */
673
  int string_style = -1;	/* Presumed outside of any string.  */
Paul Eggert's avatar
Paul Eggert committed
674
  bool string_lossage = 0;
675
  /* Not a real lossage: indicates that we have passed a matching comment
Juanma Barranquero's avatar
Juanma Barranquero committed
676
     starter plus a non-matching comment-ender, meaning that any matching
677 678 679
     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
680
  bool comment_lossage = 0;
681 682 683 684
  ptrdiff_t comment_end = from;
  ptrdiff_t comment_end_byte = from_byte;
  ptrdiff_t comstart_pos = 0;
  ptrdiff_t comstart_byte IF_LINT (= 0);
685 686
  /* Place where the containing defun starts,
     or 0 if we didn't come across it yet.  */
687 688
  ptrdiff_t defun_start = 0;
  ptrdiff_t defun_start_byte = 0;
Paul Eggert's avatar
Paul Eggert committed
689
  enum syntaxcode code;
690
  ptrdiff_t nesting = 1;		/* Current comment nesting.  */
691
  int c;
692 693 694 695 696
  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.  */
697 698 699 700 701

  /* At beginning of range to scan, we're outside of strings;
     that determines quote parity to the comment-end.  */
  while (from != stop)
    {
702
      ptrdiff_t temp_byte;
Paul Eggert's avatar
Paul Eggert committed
703 704
      int prev_syntax;
      bool com2start, com2end, comstart;
705

706
      /* Move back and examine a character.  */
707
      DEC_BOTH (from, from_byte);
708 709
      UPDATE_SYNTAX_TABLE_BACKWARD (from);

710
      prev_syntax = syntax;
711
      c = FETCH_CHAR_AS_MULTIBYTE (from_byte);
712
      syntax = SYNTAX_WITH_FLAGS (c);
713 714
      code = SYNTAX (c);

715 716 717
      /* Check for 2-char comment markers.  */
      com2start = (SYNTAX_FLAGS_COMSTART_FIRST (syntax)
		   && SYNTAX_FLAGS_COMSTART_SECOND (prev_syntax)
718 719
		   && (comstyle
		       == SYNTAX_FLAGS_COMMENT_STYLE (prev_syntax, syntax))
720 721 722 723
		   && (SYNTAX_FLAGS_COMMENT_NESTED (prev_syntax)
		       || SYNTAX_FLAGS_COMMENT_NESTED (syntax)) == comnested);
      com2end = (SYNTAX_FLAGS_COMEND_FIRST (syntax)
		 && SYNTAX_FLAGS_COMEND_SECOND (prev_syntax));
724
      comstart = (com2start || code == Scomment);
725

726 727 728 729 730 731 732 733 734 735
      /* 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,
736 737 738
	 we don't try to be clever.  E.g. |*| in C, or }% in modes that
	 have %..\n and %{..}%.  */
      if (from > stop && (com2end || comstart))
739
	{
740
	  ptrdiff_t next = from, next_byte = from_byte;
741
	  int next_c, next_syntax;
742 743
	  DEC_BOTH (next, next_byte);
	  UPDATE_SYNTAX_TABLE_BACKWARD (next);
744
	  next_c = FETCH_CHAR_AS_MULTIBYTE (next_byte);
745
	  next_syntax = SYNTAX_WITH_FLAGS (next_c);
746
	  if (((comstart || comnested)
747 748 749 750
	       && SYNTAX_FLAGS_COMEND_SECOND (syntax)
	       && SYNTAX_FLAGS_COMEND_FIRST (next_syntax))
	      || ((com2end || comnested)
		  && SYNTAX_FLAGS_COMSTART_SECOND (syntax)
751 752
		  && (comstyle
		      == SYNTAX_FLAGS_COMMENT_STYLE (syntax, prev_syntax))
753 754 755
		  && SYNTAX_FLAGS_COMSTART_FIRST (next_syntax)))
	    goto lossage;
	  /* UPDATE_SYNTAX_TABLE_FORWARD (next + 1); */
756
	}
757 758 759 760 761 762 763 764 765 766 767 768 769 770 771

      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
772
	       && (comstyle != SYNTAX_FLAGS_COMMENT_STYLE (syntax, 0)
773
		   || SYNTAX_FLAGS_COMMENT_NESTED (syntax) != comnested))
774
	continue;
775

776 777
      /* Ignore escaped characters, except comment-enders.  */
      if (code != Sendcomment && char_quoted (from, from_byte))
778 779
	continue;

780
      switch (code)
781
	{
782 783 784 785 786 787 788 789 790 791 792 793 794 795
	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.  */
796
	    string_lossage = 1;
797
	  break;
Juanma Barranquero's avatar
Juanma Barranquero committed
798

799 800
	case Scomment:
	  /* We've already checked that it is the relevant comstyle.  */
801
	  if (string_style != -1 || comment_lossage || string_lossage)
802 803 804 805
	    /* There are odd string quotes involved, so let's be careful.
	       Test case in Pascal: " { " a { " } */
	    goto lossage;

806 807 808 809 810 811 812
	  if (!comnested)
	    {
	      /* Record best comment-starter so far.  */
	      comstart_pos = from;
	      comstart_byte = from_byte;
	    }
	  else if (--nesting <= 0)
813 814 815 816 817 818
	    /* 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;
819 820
	  break;

821
	case Sendcomment:
822
	  if (SYNTAX_FLAGS_COMMENT_STYLE (syntax, 0) == comstyle
823
	      && ((com2end && SYNTAX_FLAGS_COMMENT_NESTED (prev_syntax))
824
		  || SYNTAX_FLAGS_COMMENT_NESTED (syntax)) == comnested)
825 826 827 828 829 830 831 832 833
	    /* 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.  */
	    }
834 835 836 837 838 839 840 841 842
	  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;
843
	  break;
844

845 846 847 848 849 850 851 852 853 854 855
	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.  */
	    }
856
	  break;
857 858

	default:
859
	  break;
860
	}
861 862 863 864 865
    }

  if (comstart_pos == 0)
    {
      from = comment_end;
866
      from_byte = comment_end_byte;
867
      UPDATE_SYNTAX_TABLE_FORWARD (comment_end);
868
    }
869 870 871
  /* 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 */
872 873
    {
      from = comstart_pos;
874
      from_byte = comstart_byte;
875
      UPDATE_SYNTAX_TABLE_FORWARD (from - 1);
876
    }
877
  else lossage:
878
    {
879
      struct lisp_parse_state state;
880
      bool adjusted = true;
881 882
      /* We had two kinds of string delimiters mixed up
	 together.  Decode this going forwards.
883
	 Scan fwd from a known safe place (beginning-of-defun)
884 885
	 to the one in question; this records where we
	 last passed a comment starter.  */
886 887 888 889 890
      /* 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;
891
	  adjusted = (defun_start > BEGV);
892
	}
893
      do
894
	{
895 896
	  scan_sexps_forward (&state,
			      defun_start, defun_start_byte,
897 898
			      comment_end, TYPE_MINIMUM (EMACS_INT),
			      0, Qnil, 0);
899
	  defun_start = comment_end;
900 901 902 903 904 905 906 907 908 909
	  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);
	    }

910 911 912 913 914 915 916 917 918 919 920
	  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 (* " *)  */
		{
921
		  /* FIXME: We should advance by one or two chars.  */
922 923 924 925 926 927
		  defun_start = state.comstr_start + 2;
		  defun_start_byte = CHAR_TO_BYTE (defun_start);
		}
	    }
	} while (defun_start < comment_end);

928
      from_byte = CHAR_TO_BYTE (from);
929 930
      UPDATE_SYNTAX_TABLE_FORWARD (from - 1);
    }
Juanma Barranquero's avatar
Juanma Barranquero committed
931

932
 done:
933 934 935
  *charpos_ptr = from;
  *bytepos_ptr = from_byte;

Paul Eggert's avatar
Paul Eggert committed
936
  return from != comment_end;
937 938
}

Richard M. Stallman's avatar
Richard M. Stallman committed
939
DEFUN ("syntax-table-p", Fsyntax_table_p, Ssyntax_table_p, 1, 1, 0,
940 941
       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
942
  (Lisp_Object object)
Richard M. Stallman's avatar
Richard M. Stallman committed
943
{
944
  if (CHAR_TABLE_P (object)
945
      && EQ (XCHAR_TABLE (object)->purpose, Qsyntax_table))
Richard M. Stallman's avatar
Richard M. Stallman committed
946 947 948 949
    return Qt;
  return Qnil;
}

950
static void
951
check_syntax_table (Lisp_Object obj)
Richard M. Stallman's avatar
Richard M. Stallman committed
952
{
953 954
  CHECK_TYPE (CHAR_TABLE_P (obj) && EQ (XCHAR_TABLE (obj)->purpose, Qsyntax_table),
	      Qsyntax_table_p, obj);
Juanma Barranquero's avatar
Juanma Barranquero committed
955
}
Richard M. Stallman's avatar
Richard M. Stallman committed
956 957

DEFUN ("syntax-table", Fsyntax_table, Ssyntax_table, 0, 0, 0,
958 959
       doc: /* Return the current syntax table.
This is the one specified by the current buffer.  */)
Dan Nicolaescu's avatar
Dan Nicolaescu committed
960
  (void)
Richard M. Stallman's avatar
Richard M. Stallman committed
961
{
Tom Tromey's avatar
Tom Tromey committed
962
  return BVAR (current_buffer, syntax_table);
Richard M. Stallman's avatar
Richard M. Stallman committed
963 964 965 966
}

DEFUN ("standard-syntax-table", Fstandard_syntax_table,
   Sstandard_syntax_table, 0, 0, 0,
967 968
       doc: /* Return the standard syntax table.
This is the one used for new buffers.  */)
Dan Nicolaescu's avatar
Dan Nicolaescu committed
969
  (void)
Richard M. Stallman's avatar
Richard M. Stallman committed
970 971 972 973 974
{
  return Vstandard_syntax_table;
}

DEFUN ("copy-syntax-table", Fcopy_syntax_table, Scopy_syntax_table, 0, 1, 0,
975 976
       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
977
  (Lisp_Object table)
Richard M. Stallman's avatar
Richard M. Stallman committed
978
{
979 980
  Lisp_Object copy;

Jim Blandy's avatar
Jim Blandy committed
981
  if (!NILP (table))
982 983 984 985 986
    check_syntax_table (table);
  else
    table = Vstandard_syntax_table;

  copy = Fcopy_sequence (table);
987 988 989

  /* 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
990
  set_char_table_defalt (copy, Qnil);
991 992 993 994 995 996

  /* 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);
997
  return copy;
Richard M. Stallman's avatar
Richard M. Stallman committed
998 999 1000
}

DEFUN ("set-syntax-table", Fset_syntax_table, Sset_syntax_table, 1, 1, 0,
1001 1002
       doc: /* Select a new syntax table for the current buffer.
One argument, a syntax table.  */)
Dan Nicolaescu's avatar