syntax.c 111 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, 2001-2019 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 <https://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 "lisp.h"
24
#include "character.h"
25
#include "buffer.h"
26
#include "regex-emacs.h"
Richard M. Stallman's avatar
Richard M. Stallman committed
27
#include "syntax.h"
28
#include "intervals.h"
29 30
#include "category.h"

31 32 33 34 35
/* 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
36
/* Eight single-bit flags have the following meanings:
37 38 39 40 41 42 43 44 45 46 47
  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
48
  Bits 6 and 8 discriminate among different comment styles.
49
  Languages such as C++ allow two orthogonal syntax start/end pairs
Paul Eggert's avatar
Paul Eggert committed
50
  and bit 6 determines whether a comment-end or Scommentend
51 52
  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
53
  For 2-char comment markers, the style b flag is looked up only on the second
54
  char of the comment marker and on the first char of the comment ender.
Paul Eggert's avatar
Paul Eggert committed
55 56
  For style c (like the nested flag), the flag can be placed on any of
  the chars.  */
57

58
/* These functions extract specific flags from an integer
59 60
   that holds the syntax code and the flags.  */

61 62 63 64 65 66 67 68 69 70 71 72 73 74 75 76 77 78 79 80 81
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
82 83 84 85 86
SYNTAX_FLAGS_COMSTARTEND_FIRST (int flags)
{
  return (flags & 0x50000) != 0;
}
static bool
87 88 89 90 91 92 93 94 95 96 97 98 99 100 101 102 103 104 105 106 107 108 109 110
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
    int prev_syntax; /* Syntax of previous position scanned, when
                        that position (potentially) holds the first char
                        of a 2-char construct, i.e. comment delimiter
                        or Sescape, etc.  Smax otherwise. */
163 164
  };

165 166 167
/* 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.
168
   find_start_value_byte is the corresponding byte position.
169 170 171 172
   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.  */

173 174 175
static ptrdiff_t find_start_pos;
static ptrdiff_t find_start_value;
static ptrdiff_t find_start_value_byte;
176
static struct buffer *find_start_buffer;
177
static ptrdiff_t find_start_begv;
178
static modiff_count find_start_modiff;
179 180


Paul Eggert's avatar
Paul Eggert committed
181 182 183
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);
184
static void scan_sexps_forward (struct lisp_parse_state *,
185
                                ptrdiff_t, ptrdiff_t, ptrdiff_t, EMACS_INT,
186 187
                                bool, int);
static void internalize_parse_state (Lisp_Object, struct lisp_parse_state *);
Paul Eggert's avatar
Paul Eggert committed
188
static bool in_classes (int, Lisp_Object);
189
static void parse_sexp_propertize (ptrdiff_t charpos);
Paul Eggert's avatar
Paul Eggert committed
190 191

/* This setter is used only in this file, so it can be private.  */
192
static void
Paul Eggert's avatar
Paul Eggert committed
193 194
bset_syntax_table (struct buffer *b, Lisp_Object val)
{
195
  b->syntax_table_ = val;
Paul Eggert's avatar
Paul Eggert committed
196
}
197

198
/* Whether the syntax of the character C has the prefix flag set.  */
Paul Eggert's avatar
Paul Eggert committed
199 200
bool
syntax_prefix_flag_p (int c)
201
{
202
  return SYNTAX_FLAGS_PREFIX (SYNTAX_WITH_FLAGS (c));
203
}
204 205 206

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

207
enum { INTERVALS_AT_ONCE = 10 };	/* 1 + max-number of intervals
208 209
					   to scan to property-change.  */

210 211 212 213 214 215 216 217 218 219 220 221 222 223 224 225 226 227 228 229 230 231 232 233 234 235 236 237 238 239 240 241 242 243 244 245 246 247 248 249 250 251 252 253 254
/* 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)
255
    {
256
      if (count > 0)
257
	update_syntax_table_forward (from, true, Qnil);
258
      else if (from > BEGV)
259
	{
260 261
	  update_syntax_table (from - 1, count, true, Qnil);
	  parse_sexp_propertize (from - 1);
262 263
	}
    }
264 265 266 267 268 269
}

/* 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.

270 271 272 273
   This is meant for regex-emacs.c to use.  For buffers, regex-emacs.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.  */
274 275 276 277 278 279 280 281 282 283 284 285 286 287 288 289 290 291 292 293 294 295 296 297 298 299 300 301 302 303 304 305 306 307 308 309 310

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);
}

311 312
/* Update gl_state to an appropriate interval which contains CHARPOS.  The
   sign of COUNT give the relative position of CHARPOS wrt the previously
313
   valid interval.  If INIT, only [be]_property fields of gl_state are
314
   valid at start, the rest is filled basing on OBJECT.
315

316
   `gl_state.*_i' are the intervals, and CHARPOS is further in the search
317 318
   direction than the intervals - or in an interval.  We update the
   current syntax-table basing on the property of this interval, and
319
   update the interval to start further than CHARPOS - or be
Dmitry Antipov's avatar
Dmitry Antipov committed
320
   NULL.  We also update lim_property to be the next value of
321
   charpos to call this subroutine again - or be before/after the
322 323 324
   start/end of OBJECT.  */

void
Paul Eggert's avatar
Paul Eggert committed
325
update_syntax_table (ptrdiff_t charpos, EMACS_INT count, bool init,
326
		     Lisp_Object object)
327 328
{
  Lisp_Object tmp_table;
Paul Eggert's avatar
Paul Eggert committed
329
  int cnt = 0;
330
  bool invalidate = true;
331
  INTERVAL i;
332 333 334

  if (init)
    {
335
      gl_state.old_prop = Qnil;
336 337
      gl_state.start = gl_state.b_property;
      gl_state.stop = gl_state.e_property;
338 339
      i = interval_of (charpos, object);
      gl_state.backward_i = gl_state.forward_i = i;
340
      invalidate = false;
Dmitry Antipov's avatar
Dmitry Antipov committed
341
      if (!i)
342
	return;
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

static void
parse_sexp_propertize (ptrdiff_t charpos)
{
474 475 476 477
  EMACS_INT zv = ZV;
  if (syntax_propertize__done <= charpos
      && syntax_propertize__done < zv)
    {
478
      modiff_count modiffs = CHARS_MODIFF;
479
      safe_call1 (Qinternal__syntax_propertize,
480
		  make_fixnum (min (zv, 1 + charpos)));
481 482 483 484 485 486 487 488 489 490 491 492 493
      if (modiffs != CHARS_MODIFF)
	error ("parse-sexp-propertize-function modified the buffer!");
      if (syntax_propertize__done <= charpos
	  && syntax_propertize__done < zv)
	error ("parse-sexp-propertize-function did not move"
	       " syntax-propertize--done");
      SETUP_SYNTAX_TABLE (charpos, 1);
    }
  else if (gl_state.e_property > syntax_propertize__done)
    {
      gl_state.e_property = syntax_propertize__done;
      gl_state.e_property_truncated = true;
    }
Stefan Monnier's avatar
Stefan Monnier committed
494 495
  else if (gl_state.e_property_truncated
	   && gl_state.e_property < syntax_propertize__done)
496 497 498 499
    { /* When moving backward, e_property might be set without resetting
	 e_property_truncated, so the e_property_truncated flag may
	 occasionally be left raised spuriously.  This should be rare.  */
      gl_state.e_property_truncated = false;
500
      update_syntax_table_forward (charpos, false, Qnil);
501
    }
502 503 504
}

void
505
update_syntax_table_forward (ptrdiff_t charpos, bool init,
506
			     Lisp_Object object)
507
{
508
  if (gl_state.e_property_truncated)
509
    {
510 511
      eassert (NILP (object));
      eassert (charpos >= gl_state.e_property);
512
      parse_sexp_propertize (charpos);
513 514 515 516
    }
  else
    {
      update_syntax_table (charpos, 1, init, object);
517 518
      if (NILP (object) && gl_state.e_property > syntax_propertize__done)
	parse_sexp_propertize (charpos);
519 520
    }
}
521

Paul Eggert's avatar
Paul Eggert committed
522
/* Returns true if char at CHARPOS is quoted.
523
   Global syntax-table data should be set up already to be good at CHARPOS
524
   or after.  On return global syntax data is good for lookup at CHARPOS.  */
525

Paul Eggert's avatar
Paul Eggert committed
526
static bool
527
char_quoted (ptrdiff_t charpos, ptrdiff_t bytepos)
528
{
Paul Eggert's avatar
Paul Eggert committed
529 530 531
  enum syntaxcode code;
  ptrdiff_t beg = BEGV;
  bool quoted = 0;
532
  ptrdiff_t orig = charpos;
533

534
  while (charpos > beg)
535
    {
536
      int c;
537
      DEC_BOTH (charpos, bytepos);
538

539
      UPDATE_SYNTAX_TABLE_BACKWARD (charpos);
540
      c = FETCH_CHAR_AS_MULTIBYTE (bytepos);
541
      code = SYNTAX (c);
542 543 544 545
      if (! (code == Scharquote || code == Sescape))
	break;

      quoted = !quoted;
546
    }
547 548

  UPDATE_SYNTAX_TABLE (orig);
549 550
  return quoted;
}
551 552 553 554

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

555
static ptrdiff_t
556
dec_bytepos (ptrdiff_t bytepos)
557
{
Tom Tromey's avatar
Tom Tromey committed
558
  if (NILP (BVAR (current_buffer, enable_multibyte_characters)))
559 560
    return bytepos - 1;

561 562 563
  DEC_POS (bytepos);
  return bytepos;
}
564

Juanma Barranquero's avatar
Juanma Barranquero committed
565
/* Return a defun-start position before POS and not too far before.
Richard M. Stallman's avatar
Richard M. Stallman committed
566 567 568
   It should be the last one before POS, or nearly the last.

   When open_paren_in_column_0_is_defun_start is nonzero,
569
   only the beginning of the buffer is treated as a defun-start.
Richard M. Stallman's avatar
Richard M. Stallman committed
570 571 572 573

   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.
574 575 576 577

   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.  */
578

579 580
static ptrdiff_t
find_defun_start (ptrdiff_t pos, ptrdiff_t pos_byte)
581
{
582
  ptrdiff_t opoint = PT, opoint_byte = PT_BYTE;
583 584 585 586 587 588 589 590 591 592 593 594

  /* 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;

595 596
  if (!NILP (Vcomment_use_syntax_ppss))
    {
597
      modiff_count modiffs = CHARS_MODIFF;
598
      Lisp_Object ppss = call1 (Qsyntax_ppss, make_fixnum (pos));
599 600 601
      if (modiffs != CHARS_MODIFF)
	error ("syntax-ppss modified the buffer!");
      TEMP_SET_PT_BOTH (opoint, opoint_byte);
602
      Lisp_Object boc = Fnth (make_fixnum (8), ppss);
603
      if (FIXNUMP (boc))
604
        {
Tom Tromey's avatar
Tom Tromey committed
605
          find_start_value = XFIXNUM (boc);
606 607 608 609 610 611 612 613 614
          find_start_value_byte = CHAR_TO_BYTE (find_start_value);
        }
      else
        {
          find_start_value = pos;
          find_start_value_byte = pos_byte;
        }
      goto found;
    }
615 616 617 618 619 620 621
  if (!open_paren_in_column_0_is_defun_start)
    {
      find_start_value = BEGV;
      find_start_value_byte = BEGV_BYTE;
      goto found;
    }

622
  /* Back up to start of line.  */
623
  scan_newline (pos, pos_byte, BEGV, BEGV_BYTE, -1, 1);
624

625 626 627
  /* We optimize syntax-table lookup for rare updates.  Thus we accept
     only those `^\s(' which are good in global _and_ text-property
     syntax-tables.  */
628
  SETUP_BUFFER_SYNTAX_TABLE ();
629
  while (PT > BEGV)
630
    {
631 632
      /* Open-paren at start of line means we may have found our
	 defun-start.  */
Paul Eggert's avatar
Paul Eggert committed
633
      int c = FETCH_CHAR_AS_MULTIBYTE (PT_BYTE);
634
      if (SYNTAX (c) == Sopen)
635
	{
636
	  SETUP_SYNTAX_TABLE (PT + 1, -1);	/* Try again... */
637
	  c = FETCH_CHAR_AS_MULTIBYTE (PT_BYTE);
638
	  if (SYNTAX (c) == Sopen)
639 640
	    break;
	  /* Now fallback to the default value.  */
641
	  SETUP_BUFFER_SYNTAX_TABLE ();
642
	}
643 644
      /* Move to beg of previous line.  */
      scan_newline (PT, PT_BYTE, BEGV, BEGV_BYTE, -2, 1);
645 646 647
    }

  /* Record what we found, for the next try.  */
648 649
  find_start_value = PT;
  find_start_value_byte = PT_BYTE;
650 651 652
  TEMP_SET_PT_BOTH (opoint, opoint_byte);

 found:
653 654 655 656 657 658 659 660
  find_start_buffer = current_buffer;
  find_start_modiff = MODIFF;
  find_start_begv = BEGV;
  find_start_pos = pos;

  return find_start_value;
}

661 662
/* Return the SYNTAX_COMEND_FIRST of the character before POS, POS_BYTE.  */

Paul Eggert's avatar
Paul Eggert committed
663
static bool
664
prev_char_comend_first (ptrdiff_t pos, ptrdiff_t pos_byte)
665
{
Paul Eggert's avatar
Paul Eggert committed
666 667
  int c;
  bool val;
668 669 670 671 672 673 674 675 676

  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
677
/* Check whether charpos FROM is at the end of a comment.
678 679 680
   FROM_BYTE is the bytepos corresponding to FROM.
   Do not move back before STOP.

Paul Eggert's avatar
Paul Eggert committed
681
   Return true if we find a comment ending at FROM/FROM_BYTE.
682 683 684

   If successful, store the charpos of the comment's beginning
   into *CHARPOS_PTR, and the bytepos into *BYTEPOS_PTR.
685 686 687

   Global syntax data remains valid for backward search starting at
   the returned value (or at FROM, if the search was not successful).  */
688

Paul Eggert's avatar
Paul Eggert committed
689 690 691 692
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)
693 694 695 696 697 698 699 700 701 702
{
  /* 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.  */
703
  int string_style = -1;	/* Presumed outside of any string.  */
Paul Eggert's avatar
Paul Eggert committed
704
  bool string_lossage = 0;
705
  /* Not a real lossage: indicates that we have passed a matching comment
Juanma Barranquero's avatar
Juanma Barranquero committed
706
     starter plus a non-matching comment-ender, meaning that any matching
707 708 709
     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
710
  bool comment_lossage = 0;
711 712 713
  ptrdiff_t comment_end = from;
  ptrdiff_t comment_end_byte = from_byte;
  ptrdiff_t comstart_pos = 0;
714
  ptrdiff_t comstart_byte;
715 716
  /* Place where the containing defun starts,
     or 0 if we didn't come across it yet.  */
717 718
  ptrdiff_t defun_start = 0;
  ptrdiff_t defun_start_byte = 0;
Paul Eggert's avatar
Paul Eggert committed
719
  enum syntaxcode code;
720
  ptrdiff_t nesting = 1;		/* Current comment nesting.  */
721
  int c;
722
  int syntax = 0;
Paul Eggert's avatar
Paul Eggert committed
723
  unsigned short int quit_count = 0;
724 725 726 727

  /* 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.  */
728 729 730 731 732

  /* At beginning of range to scan, we're outside of strings;
     that determines quote parity to the comment-end.  */
  while (from != stop)
    {
733
      rarely_quit (++quit_count);
Paul Eggert's avatar
Paul Eggert committed
734

735
      ptrdiff_t temp_byte;
Paul Eggert's avatar
Paul Eggert committed
736 737
      int prev_syntax;
      bool com2start, com2end, comstart;
738

739
      /* Move back and examine a character.  */
740
      DEC_BOTH (from, from_byte);
741 742
      UPDATE_SYNTAX_TABLE_BACKWARD (from);

743
      prev_syntax = syntax;
744
      c = FETCH_CHAR_AS_MULTIBYTE (from_byte);
745
      syntax = SYNTAX_WITH_FLAGS (c);
746 747
      code = SYNTAX (c);

748 749 750
      /* Check for 2-char comment markers.  */
      com2start = (SYNTAX_FLAGS_COMSTART_FIRST (syntax)
		   && SYNTAX_FLAGS_COMSTART_SECOND (prev_syntax)
751 752
		   && (comstyle
		       == SYNTAX_FLAGS_COMMENT_STYLE (prev_syntax, syntax))
753 754 755 756
		   && (SYNTAX_FLAGS_COMMENT_NESTED (prev_syntax)
		       || SYNTAX_FLAGS_COMMENT_NESTED (syntax)) == comnested);
      com2end = (SYNTAX_FLAGS_COMEND_FIRST (syntax)
		 && SYNTAX_FLAGS_COMEND_SECOND (prev_syntax));
757
      comstart = (com2start || code == Scomment);
758

759 760 761 762 763 764 765 766 767 768
      /* 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,
769 770 771
	 we don't try to be clever.  E.g. |*| in C, or }% in modes that
	 have %..\n and %{..}%.  */
      if (from > stop && (com2end || comstart))
772
	{
773
	  ptrdiff_t next = from, next_byte = from_byte;
774
	  int next_c, next_syntax;
775 776
	  DEC_BOTH (next, next_byte);
	  UPDATE_SYNTAX_TABLE_BACKWARD (next);
777
	  next_c = FETCH_CHAR_AS_MULTIBYTE (next_byte);
778
	  next_syntax = SYNTAX_WITH_FLAGS (next_c);
779
	  if (((comstart || comnested)
780 781 782 783
	       && SYNTAX_FLAGS_COMEND_SECOND (syntax)
	       && SYNTAX_FLAGS_COMEND_FIRST (next_syntax))
	      || ((com2end || comnested)
		  && SYNTAX_FLAGS_COMSTART_SECOND (syntax)
784 785
		  && (comstyle
		      == SYNTAX_FLAGS_COMMENT_STYLE (syntax, prev_syntax))
786 787 788
		  && SYNTAX_FLAGS_COMSTART_FIRST (next_syntax)))
	    goto lossage;
	  /* UPDATE_SYNTAX_TABLE_FORWARD (next + 1); */
789
	}
790 791 792 793 794 795 796 797 798 799 800 801 802 803 804

      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
805
	       && (comstyle != SYNTAX_FLAGS_COMMENT_STYLE (syntax, 0)
806
		   || SYNTAX_FLAGS_COMMENT_NESTED (syntax) != comnested))
807
	continue;
808

809 810 811 812
      /* Ignore escaped characters, except comment-enders which cannot
         be escaped.  */
      if ((Vcomment_end_can_be_escaped || code != Sendcomment)
          && char_quoted (from, from_byte))
813 814
	continue;

815
      switch (code)
816
	{
817 818 819
	case Sstring_fence:
	case Scomment_fence:
	  c = (code == Sstring_fence ? ST_STRING_STYLE : ST_COMMENT_STYLE);
820
	  FALLTHROUGH;
821 822 823 824 825 826 827 828 829 830 831
	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.  */
832
	    string_lossage = 1;
833
	  break;
Juanma Barranquero's avatar
Juanma Barranquero committed
834

835 836
	case Scomment:
	  /* We've already checked that it is the relevant comstyle.  */
837
	  if (string_style != -1 || comment_lossage || string_lossage)
838 839 840 841
	    /* There are odd string quotes involved, so let's be careful.
	       Test case in Pascal: " { " a { " } */
	    goto lossage;

842 843 844 845 846 847 848
	  if (!comnested)
	    {
	      /* Record best comment-starter so far.  */
	      comstart_pos = from;
	      comstart_byte = from_byte;
	    }
	  else if (--nesting <= 0)
849 850 851 852 853 854
	    /* 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;
855 856
	  break;

857
	case Sendcomment:
858
	  if (SYNTAX_FLAGS_COMMENT_STYLE (syntax, 0) == comstyle
859
	      && ((com2end && SYNTAX_FLAGS_COMMENT_NESTED (prev_syntax))
860
		  || SYNTAX_FLAGS_COMMENT_NESTED (syntax)) == comnested)
861 862 863 864 865 866 867 868 869
	    /* 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.  */
	    }
870 871 872 873 874 875 876 877 878
	  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;
879
	  break;
880

881 882 883
	case Sopen:
	  /* Assume a defun-start point is outside of strings.  */
	  if (open_paren_in_column_0_is_defun_start
884
              && NILP (Vcomment_use_syntax_ppss)
885 886 887 888 889 890 891 892
	      && (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.  */
	    }
893
	  break;
894 895

	default:
896
	  break;
897
	}
898 899 900 901 902
    }

  if (comstart_pos == 0)
    {
      from = comment_end;
903
      from_byte = comment_end_byte;
904
      UPDATE_SYNTAX_TABLE_FORWARD (comment_end);
905
    }
906 907 908
  /* 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 */
909 910
    {
      from = comstart_pos;
911
      from_byte = comstart_byte;
912
      UPDATE_SYNTAX_TABLE_FORWARD (from - 1);
913
    }
914
  else lossage:
915
    {
916
      struct lisp_parse_state state;
917
      bool adjusted = true;
918 919
      /* We had two kinds of string delimiters mixed up
	 together.  Decode this going forwards.
920
	 Scan fwd from a known safe place (beginning-of-defun)
921 922
	 to the one in question; this records where we
	 last passed a comment starter.  */
923 924 925 926 927
      /* 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;
928
	  adjusted = (defun_start > BEGV);
929
	}
930
      do
931
	{
932
          internalize_parse_state (Qnil, &state);
933 934
	  scan_sexps_forward (&state,
			      defun_start, defun_start_byte,
935
			      comment_end, TYPE_MINIMUM (EMACS_INT),
936
			      0, 0);
937
	  defun_start = comment_end;
938 939 940 941
	  if (!adjusted)
	    {
	      adjusted = true;
	      find_start_value
Tom Tromey's avatar
Tom Tromey committed
942
		= CONSP (state.levelstarts) ? XFIXNUM (XCAR (state.levelstarts))
943 944 945 946 947
		: state.thislevelstart >= 0 ? state.thislevelstart
		: find_start_value;
	      find_start_value_byte = CHAR_TO_BYTE (find_start_value);
	    }

948 949 950 951 952 953 954 955 956 957 958
	  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 (* " *)  */
		{
959
		  /* FIXME: We should advance by one or two chars.  */
960 961 962 963
		  defun_start = state.comstr_start + 2;
		  defun_start_byte = CHAR_TO_BYTE (defun_start);
		}
	    }
964
	  rarely_quit (++quit_count);
Paul Eggert's avatar
Paul Eggert committed
965 966
	}
      while (defun_start < comment_end);
967

968
      from_byte = CHAR_TO_BYTE (from);
969 970
      UPDATE_SYNTAX_TABLE_FORWARD (from - 1);
    }
Juanma Barranquero's avatar
Juanma Barranquero committed
971

972
 done: