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);
184
static void parse_sexp_propertize (ptrdiff_t charpos);
Paul Eggert's avatar
Paul Eggert committed
185 186

/* This setter is used only in this file, so it can be private.  */
187
static void
Paul Eggert's avatar
Paul Eggert committed
188 189
bset_syntax_table (struct buffer *b, Lisp_Object val)
{
190
  b->syntax_table_ = val;
Paul Eggert's avatar
Paul Eggert committed
191
}
192

193
/* Whether the syntax of the character C has the prefix flag set.  */
Paul Eggert's avatar
Paul Eggert committed
194 195
bool
syntax_prefix_flag_p (int c)
196
{
197
  return SYNTAX_FLAGS_PREFIX (SYNTAX_WITH_FLAGS (c));
198
}
199 200 201

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

202
enum { INTERVALS_AT_ONCE = 10 };	/* 1 + max-number of intervals
203 204
					   to scan to property-change.  */

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)
250
    {
251 252 253
      if (count > 0)
	update_syntax_table_forward (from, true, Qnil);
      else if (from > BEGV)
254
	{
255 256
	  update_syntax_table (from - 1, count, true, Qnil);
	  parse_sexp_propertize (from - 1);
257 258
	}
    }
259 260 261 262 263 264 265 266 267 268 269 270 271 272 273 274 275 276 277 278 279 280 281 282 283 284 285 286 287 288 289 290 291 292 293 294 295 296 297 298 299 300 301 302 303 304
}

/* 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);
}

305 306
/* Update gl_state to an appropriate interval which contains CHARPOS.  The
   sign of COUNT give the relative position of CHARPOS wrt the previously
307
   valid interval.  If INIT, only [be]_property fields of gl_state are
308
   valid at start, the rest is filled basing on OBJECT.
309

310
   `gl_state.*_i' are the intervals, and CHARPOS is further in the search
311 312
   direction than the intervals - or in an interval.  We update the
   current syntax-table basing on the property of this interval, and
313
   update the interval to start further than CHARPOS - or be
Dmitry Antipov's avatar
Dmitry Antipov committed
314
   NULL.  We also update lim_property to be the next value of
315
   charpos to call this subroutine again - or be before/after the
316 317 318
   start/end of OBJECT.  */

void
Paul Eggert's avatar
Paul Eggert committed
319
update_syntax_table (ptrdiff_t charpos, EMACS_INT count, bool init,
320
		     Lisp_Object object)
321 322
{
  Lisp_Object tmp_table;
Paul Eggert's avatar
Paul Eggert committed
323
  int cnt = 0;
324
  bool invalidate = true;
325
  INTERVAL i;
326 327 328

  if (init)
    {
329
      gl_state.old_prop = Qnil;
330 331
      gl_state.start = gl_state.b_property;
      gl_state.stop = gl_state.e_property;
332 333
      i = interval_of (charpos, object);
      gl_state.backward_i = gl_state.forward_i = i;
334
      invalidate = false;
Dmitry Antipov's avatar
Dmitry Antipov committed
335
      if (!i)
336
	return;
337
      /* interval_of updates only ->position of the return value, so
338
	 update the parents manually to speed up update_interval.  */
339
      while (!NULL_PARENT (i))
340 341
	{
	  if (AM_RIGHT_CHILD (i))
342
	    INTERVAL_PARENT (i)->position = i->position
343
	      - LEFT_TOTAL_LENGTH (i) + TOTAL_LENGTH (i) /* right end */
344 345
	      - TOTAL_LENGTH (INTERVAL_PARENT (i))
	      + LEFT_TOTAL_LENGTH (INTERVAL_PARENT (i));
346
	  else
347
	    INTERVAL_PARENT (i)->position = i->position - LEFT_TOTAL_LENGTH (i)
348
	      + TOTAL_LENGTH (i);
349
	  i = INTERVAL_PARENT (i);
350 351
	}
      i = gl_state.forward_i;
352
      gl_state.b_property = i->position - gl_state.offset;
353
      gl_state.e_property = INTERVAL_LAST_POS (i) - gl_state.offset;
354 355
      goto update;
    }
356
  i = count > 0 ? gl_state.forward_i : gl_state.backward_i;
357

358
  /* We are guaranteed to be called with CHARPOS either in i,
359
     or further off.  */
Dmitry Antipov's avatar
Dmitry Antipov committed
360
  if (!i)
361
    error ("Error in syntax_table logic for to-the-end intervals");
362
  else if (charpos < i->position)		/* Move left.  */
363 364
    {
      if (count > 0)
365
	error ("Error in syntax_table logic for intervals <-");
366
      /* Update the interval.  */
367
      i = update_interval (i, charpos);
368
      if (INTERVAL_LAST_POS (i) != gl_state.b_property)
369
	{
370
	  invalidate = false;
371
	  gl_state.forward_i = i;
372
	  gl_state.e_property = INTERVAL_LAST_POS (i) - gl_state.offset;
373
	}
374
    }
375
  else if (charpos >= INTERVAL_LAST_POS (i)) /* Move right.  */
376 377
    {
      if (count < 0)
378
	error ("Error in syntax_table logic for intervals ->");
379
      /* Update the interval.  */
380
      i = update_interval (i, charpos);
381
      if (i->position != gl_state.e_property)
382
	{
383
	  invalidate = false;
384
	  gl_state.backward_i = i;
385
	  gl_state.b_property = i->position - gl_state.offset;
386 387 388 389 390 391 392 393
	}
    }

  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
394

395 396 397
  if (invalidate)		/* Did not get to adjacent interval.  */
    {				/* with the same table => */
				/* invalidate the old range.  */
398 399 400
      if (count > 0)
	{
	  gl_state.backward_i = i;
401 402 403
	  gl_state.b_property = i->position - gl_state.offset;
	}
      else
404
	{
405
	  gl_state.forward_i = i;
406
	  gl_state.e_property = INTERVAL_LAST_POS (i) - gl_state.offset;
407 408
	}
    }
409

410
  if (!EQ (tmp_table, gl_state.old_prop))
411
    {
412 413 414 415 416
      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
417
	}
418 419 420 421 422
      else if (CONSP (tmp_table))
	{
	  gl_state.use_global = 1;
	  gl_state.global_code = tmp_table;
	}
Juanma Barranquero's avatar
Juanma Barranquero committed
423
      else
424 425
	{
	  gl_state.use_global = 0;
Tom Tromey's avatar
Tom Tromey committed
426
	  gl_state.current_syntax_table = BVAR (current_buffer, syntax_table);
427
	}
428 429
    }

Dmitry Antipov's avatar
Dmitry Antipov committed
430
  while (i)
431 432 433 434
    {
      if (cnt && !EQ (tmp_table, textget (i->plist, Qsyntax_table)))
	{
	  if (count > 0)
435 436 437 438 439 440
	    {
	      gl_state.e_property = i->position - gl_state.offset;
	      gl_state.forward_i = i;
	    }
	  else
	    {
441 442
	      gl_state.b_property
		= i->position + LENGTH (i) - gl_state.offset;
443 444 445
	      gl_state.backward_i = i;
	    }
	  return;
446
	}
Juanma Barranquero's avatar
Juanma Barranquero committed
447
      else if (cnt == INTERVALS_AT_ONCE)
448 449
	{
	  if (count > 0)
450
	    {
451 452 453 454 455
	      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
456
		+ (next_interval (i) ? 0 : 1);
457 458 459 460 461 462 463 464
	      gl_state.forward_i = i;
	    }
	  else
	    {
	      gl_state.b_property = i->position - gl_state.offset;
	      gl_state.backward_i = i;
	    }
	  return;
465 466 467 468
	}
      cnt++;
      i = count > 0 ? next_interval (i) : previous_interval (i);
    }
Dmitry Antipov's avatar
Dmitry Antipov committed
469
  eassert (i == NULL); /* This property goes to the end.  */
470
  if (count > 0)
471 472 473 474
    {
      gl_state.e_property = gl_state.stop;
      gl_state.forward_i = i;
    }
475 476
  else
    gl_state.b_property = gl_state.start;
477
}
478 479 480 481

static void
parse_sexp_propertize (ptrdiff_t charpos)
{
482 483 484 485 486 487 488 489 490 491 492 493 494 495 496 497 498 499 500 501
  EMACS_INT zv = ZV;
  if (syntax_propertize__done <= charpos
      && syntax_propertize__done < zv)
    {
      EMACS_INT modiffs = CHARS_MODIFF;
      safe_call1 (Qinternal__syntax_propertize,
		  make_number (min (zv, 1 + charpos)));
      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;
    }
502 503 504 505
}

void
update_syntax_table_forward (ptrdiff_t charpos, bool init,
506
			     Lisp_Object object)
507
{
508
  if (gl_state.e_property_truncated)
509
    {
510 511 512 513 514 515 516 517 518 519
      eassert (NILP (object));
      eassert (charpos >= gl_state.e_property);
      eassert (charpos >= syntax_propertize__done);
      parse_sexp_propertize (charpos);
    }
  else
    {
      update_syntax_table (charpos, 1, init, object);
      if (gl_state.e_property > syntax_propertize__done
	  && NILP (object))
520 521 522
	parse_sexp_propertize (charpos);
    }
}
523

Paul Eggert's avatar
Paul Eggert committed
524
/* Returns true if char at CHARPOS is quoted.
525
   Global syntax-table data should be set up already to be good at CHARPOS
526
   or after.  On return global syntax data is good for lookup at CHARPOS.  */
527

Paul Eggert's avatar
Paul Eggert committed
528
static bool
529
char_quoted (ptrdiff_t charpos, ptrdiff_t bytepos)
530
{
Paul Eggert's avatar
Paul Eggert committed
531 532 533
  enum syntaxcode code;
  ptrdiff_t beg = BEGV;
  bool quoted = 0;
534
  ptrdiff_t orig = charpos;
535

536
  while (charpos > beg)
537
    {
538
      int c;
539
      DEC_BOTH (charpos, bytepos);
540

541
      UPDATE_SYNTAX_TABLE_BACKWARD (charpos);
542
      c = FETCH_CHAR_AS_MULTIBYTE (bytepos);
543
      code = SYNTAX (c);
544 545 546 547
      if (! (code == Scharquote || code == Sescape))
	break;

      quoted = !quoted;
548
    }
549 550

  UPDATE_SYNTAX_TABLE (orig);
551 552
  return quoted;
}
553 554 555 556

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

557
static ptrdiff_t
558
dec_bytepos (ptrdiff_t bytepos)
559
{
Tom Tromey's avatar
Tom Tromey committed
560
  if (NILP (BVAR (current_buffer, enable_multibyte_characters)))
561 562
    return bytepos - 1;

563 564 565
  DEC_POS (bytepos);
  return bytepos;
}
566

Juanma Barranquero's avatar
Juanma Barranquero committed
567
/* Return a defun-start position before POS and not too far before.
Richard M. Stallman's avatar
Richard M. Stallman committed
568 569 570
   It should be the last one before POS, or nearly the last.

   When open_paren_in_column_0_is_defun_start is nonzero,
571
   only the beginning of the buffer is treated as a defun-start.
Richard M. Stallman's avatar
Richard M. Stallman committed
572 573 574 575

   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.
576 577 578 579

   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.  */
580

581 582
static ptrdiff_t
find_defun_start (ptrdiff_t pos, ptrdiff_t pos_byte)
583
{
584
  ptrdiff_t opoint = PT, opoint_byte = PT_BYTE;
585 586 587 588 589 590 591 592 593 594 595 596

  /* 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;

597 598 599 600 601 602 603
  if (!open_paren_in_column_0_is_defun_start)
    {
      find_start_value = BEGV;
      find_start_value_byte = BEGV_BYTE;
      goto found;
    }

604
  /* Back up to start of line.  */
605
  scan_newline (pos, pos_byte, BEGV, BEGV_BYTE, -1, 1);
606

607 608 609
  /* We optimize syntax-table lookup for rare updates.  Thus we accept
     only those `^\s(' which are good in global _and_ text-property
     syntax-tables.  */
610
  SETUP_BUFFER_SYNTAX_TABLE ();
611
  while (PT > BEGV)
612
    {
613 614
      int c;

615 616
      /* Open-paren at start of line means we may have found our
	 defun-start.  */
617
      c = FETCH_CHAR_AS_MULTIBYTE (PT_BYTE);
618
      if (SYNTAX (c) == Sopen)
619
	{
620
	  SETUP_SYNTAX_TABLE (PT + 1, -1);	/* Try again... */
621
	  c = FETCH_CHAR_AS_MULTIBYTE (PT_BYTE);
622
	  if (SYNTAX (c) == Sopen)
623 624
	    break;
	  /* Now fallback to the default value.  */
625
	  SETUP_BUFFER_SYNTAX_TABLE ();
626
	}
627 628
      /* Move to beg of previous line.  */
      scan_newline (PT, PT_BYTE, BEGV, BEGV_BYTE, -2, 1);
629 630 631
    }

  /* Record what we found, for the next try.  */
632 633
  find_start_value = PT;
  find_start_value_byte = PT_BYTE;
634 635 636
  TEMP_SET_PT_BOTH (opoint, opoint_byte);

 found:
637 638 639 640 641 642 643 644
  find_start_buffer = current_buffer;
  find_start_modiff = MODIFF;
  find_start_begv = BEGV;
  find_start_pos = pos;

  return find_start_value;
}

645 646
/* Return the SYNTAX_COMEND_FIRST of the character before POS, POS_BYTE.  */

Paul Eggert's avatar
Paul Eggert committed
647
static bool
648
prev_char_comend_first (ptrdiff_t pos, ptrdiff_t pos_byte)
649
{
Paul Eggert's avatar
Paul Eggert committed
650 651
  int c;
  bool val;
652 653 654 655 656 657 658 659 660

  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
661
/* Check whether charpos FROM is at the end of a comment.
662 663 664
   FROM_BYTE is the bytepos corresponding to FROM.
   Do not move back before STOP.

Paul Eggert's avatar
Paul Eggert committed
665
   Return true if we find a comment ending at FROM/FROM_BYTE.
666 667 668

   If successful, store the charpos of the comment's beginning
   into *CHARPOS_PTR, and the bytepos into *BYTEPOS_PTR.
669 670 671

   Global syntax data remains valid for backward search starting at
   the returned value (or at FROM, if the search was not successful).  */
672

Paul Eggert's avatar
Paul Eggert committed
673 674 675 676
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)
677 678 679 680 681 682 683 684 685 686
{
  /* 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.  */
687
  int string_style = -1;	/* Presumed outside of any string.  */
Paul Eggert's avatar
Paul Eggert committed
688
  bool string_lossage = 0;
689
  /* Not a real lossage: indicates that we have passed a matching comment
Juanma Barranquero's avatar
Juanma Barranquero committed
690
     starter plus a non-matching comment-ender, meaning that any matching
691 692 693
     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
694
  bool comment_lossage = 0;
695 696 697 698
  ptrdiff_t comment_end = from;
  ptrdiff_t comment_end_byte = from_byte;
  ptrdiff_t comstart_pos = 0;
  ptrdiff_t comstart_byte IF_LINT (= 0);
699 700
  /* Place where the containing defun starts,
     or 0 if we didn't come across it yet.  */
701 702
  ptrdiff_t defun_start = 0;
  ptrdiff_t defun_start_byte = 0;
Paul Eggert's avatar
Paul Eggert committed
703
  enum syntaxcode code;
704
  ptrdiff_t nesting = 1;		/* Current comment nesting.  */
705
  int c;
706 707 708 709 710
  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.  */
711 712 713 714 715

  /* At beginning of range to scan, we're outside of strings;
     that determines quote parity to the comment-end.  */
  while (from != stop)
    {
716
      ptrdiff_t temp_byte;
Paul Eggert's avatar
Paul Eggert committed
717 718
      int prev_syntax;
      bool com2start, com2end, comstart;
719

720
      /* Move back and examine a character.  */
721
      DEC_BOTH (from, from_byte);
722 723
      UPDATE_SYNTAX_TABLE_BACKWARD (from);

724
      prev_syntax = syntax;
725
      c = FETCH_CHAR_AS_MULTIBYTE (from_byte);
726
      syntax = SYNTAX_WITH_FLAGS (c);
727 728
      code = SYNTAX (c);

729 730 731
      /* Check for 2-char comment markers.  */
      com2start = (SYNTAX_FLAGS_COMSTART_FIRST (syntax)
		   && SYNTAX_FLAGS_COMSTART_SECOND (prev_syntax)
732 733
		   && (comstyle
		       == SYNTAX_FLAGS_COMMENT_STYLE (prev_syntax, syntax))
734 735 736 737
		   && (SYNTAX_FLAGS_COMMENT_NESTED (prev_syntax)
		       || SYNTAX_FLAGS_COMMENT_NESTED (syntax)) == comnested);
      com2end = (SYNTAX_FLAGS_COMEND_FIRST (syntax)
		 && SYNTAX_FLAGS_COMEND_SECOND (prev_syntax));
738
      comstart = (com2start || code == Scomment);
739

740 741 742 743 744 745 746 747 748 749
      /* 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,
750 751 752
	 we don't try to be clever.  E.g. |*| in C, or }% in modes that
	 have %..\n and %{..}%.  */
      if (from > stop && (com2end || comstart))
753
	{
754
	  ptrdiff_t next = from, next_byte = from_byte;
755
	  int next_c, next_syntax;
756 757
	  DEC_BOTH (next, next_byte);
	  UPDATE_SYNTAX_TABLE_BACKWARD (next);
758
	  next_c = FETCH_CHAR_AS_MULTIBYTE (next_byte);
759
	  next_syntax = SYNTAX_WITH_FLAGS (next_c);
760
	  if (((comstart || comnested)
761 762 763 764
	       && SYNTAX_FLAGS_COMEND_SECOND (syntax)
	       && SYNTAX_FLAGS_COMEND_FIRST (next_syntax))
	      || ((com2end || comnested)
		  && SYNTAX_FLAGS_COMSTART_SECOND (syntax)
765 766
		  && (comstyle
		      == SYNTAX_FLAGS_COMMENT_STYLE (syntax, prev_syntax))
767 768 769
		  && SYNTAX_FLAGS_COMSTART_FIRST (next_syntax)))
	    goto lossage;
	  /* UPDATE_SYNTAX_TABLE_FORWARD (next + 1); */
770
	}
771 772 773 774 775 776 777 778 779 780 781 782 783 784 785

      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
786
	       && (comstyle != SYNTAX_FLAGS_COMMENT_STYLE (syntax, 0)
787
		   || SYNTAX_FLAGS_COMMENT_NESTED (syntax) != comnested))
788
	continue;
789

790 791
      /* Ignore escaped characters, except comment-enders.  */
      if (code != Sendcomment && char_quoted (from, from_byte))
792 793
	continue;

794
      switch (code)
795
	{
796 797 798 799 800 801 802 803 804 805 806 807 808 809
	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.  */
810
	    string_lossage = 1;
811
	  break;
Juanma Barranquero's avatar
Juanma Barranquero committed
812

813 814
	case Scomment:
	  /* We've already checked that it is the relevant comstyle.  */
815
	  if (string_style != -1 || comment_lossage || string_lossage)
816 817 818 819
	    /* There are odd string quotes involved, so let's be careful.
	       Test case in Pascal: " { " a { " } */
	    goto lossage;

820 821 822 823 824 825 826
	  if (!comnested)
	    {
	      /* Record best comment-starter so far.  */
	      comstart_pos = from;
	      comstart_byte = from_byte;
	    }
	  else if (--nesting <= 0)
827 828 829 830 831 832
	    /* 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;
833 834
	  break;

835
	case Sendcomment:
836
	  if (SYNTAX_FLAGS_COMMENT_STYLE (syntax, 0) == comstyle
837
	      && ((com2end && SYNTAX_FLAGS_COMMENT_NESTED (prev_syntax))
838
		  || SYNTAX_FLAGS_COMMENT_NESTED (syntax)) == comnested)
839 840 841 842 843 844 845 846 847
	    /* 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.  */
	    }
848 849 850 851 852 853 854 855 856
	  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;
857
	  break;
858

859 860 861 862 863 864 865 866 867 868 869
	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.  */
	    }
870
	  break;
871 872

	default:
873
	  break;
874
	}
875 876 877 878 879
    }

  if (comstart_pos == 0)
    {
      from = comment_end;
880
      from_byte = comment_end_byte;
881
      UPDATE_SYNTAX_TABLE_FORWARD (comment_end);
882
    }
883 884 885
  /* 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 */
886 887
    {
      from = comstart_pos;
888
      from_byte = comstart_byte;
889
      UPDATE_SYNTAX_TABLE_FORWARD (from - 1);
890
    }
891
  else lossage:
892
    {
893
      struct lisp_parse_state state;
894
      bool adjusted = true;
895 896
      /* We had two kinds of string delimiters mixed up
	 together.  Decode this going forwards.
897
	 Scan fwd from a known safe place (beginning-of-defun)
898 899
	 to the one in question; this records where we
	 last passed a comment starter.  */
900 901 902 903 904
      /* 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;
905
	  adjusted = (defun_start > BEGV);
906
	}
907
      do
908
	{
909 910
	  scan_sexps_forward (&state,
			      defun_start, defun_start_byte,
911 912
			      comment_end, TYPE_MINIMUM (EMACS_INT),
			      0, Qnil, 0);
913
	  defun_start = comment_end;
914 915 916 917 918 919 920 921 922 923
	  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);
	    }

924 925 926 927 928 929 930 931 932 933 934
	  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 (* " *)  */
		{
935
		  /* FIXME: We should advance by one or two chars.  */
936 937 938 939 940 941
		  defun_start = state.comstr_start + 2;
		  defun_start_byte = CHAR_TO_BYTE (defun_start);
		}
	    }
	} while (defun_start < comment_end);

942
      from_byte = CHAR_TO_BYTE (from);
943 944
      UPDATE_SYNTAX_TABLE_FORWARD (from - 1);
    }
Juanma Barranquero's avatar
Juanma Barranquero committed
945

946
 done:
947 948 949
  *charpos_ptr = from;
  *bytepos_ptr = from_byte;

Paul Eggert's avatar
Paul Eggert committed
950
  return from != comment_end;
951 952
}

Richard M. Stallman's avatar
Richard M. Stallman committed
953
DEFUN ("syntax-table-p", Fsyntax_table_p, Ssyntax_table_p, 1, 1, 0,
954 955
       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
956
  (Lisp_Object object)
Richard M. Stallman's avatar
Richard M. Stallman committed
957
{
958
  if (CHAR_TABLE_P (object)
959
      && EQ (XCHAR_TABLE (object)->purpose, Qsyntax_table))
Richard M. Stallman's avatar
Richard M. Stallman committed
960 961 962 963
    return Qt;
  return Qnil;
}

964
static void
965
check_syntax_table (Lisp_Object obj)
Richard M. Stallman's avatar
Richard M. Stallman committed
966
{
967 968
  CHECK_TYPE (CHAR_TABLE_P (obj) && EQ (XCHAR_TABLE (obj)->purpose, Qsyntax_table),
	      Qsyntax_table_p, obj);
Juanma Barranquero's avatar
Juanma Barranquero committed
969
}
Richard M. Stallman's avatar
Richard M. Stallman committed
970 971

DEFUN ("syntax-table", Fsyntax_table, Ssyntax_table, 0, 0, 0,
972 973
       doc: /* Return the current syntax table.
This is the one specified by the current buffer.  */)
Dan Nicolaescu's avatar
Dan Nicolaescu committed
974
  (void)
Richard M. Stallman's avatar
Richard M. Stallman committed
975
{
Tom Tromey's avatar
Tom Tromey committed
976
  return BVAR (current_buffer, syntax_table);
Richard M. Stallman's avatar
Richard M. Stallman committed
977 978 979 980
}

DEFUN ("standard-syntax-table", Fstandard_syntax_table,
   Sstandard_syntax_table, 0, 0, 0,
981 982
       doc: /* Return the standard syntax table.
This is the one used for new buffers.  */)
Dan Nicolaescu's avatar
Dan Nicolaescu committed
983
  (void)
Richard M. Stallman's avatar
Richard M. Stallman committed
984 985 986 987 988
{
  return Vstandard_syntax_table;
}

DEFUN ("copy-syntax-table", Fcopy_syntax_table, Scopy_syntax_table, 0, 1, 0,
989 990
       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
991
  (Lisp_Object table)
Richard M. Stallman's avatar
Richard M. Stallman committed
992
{
993 994
  Lisp_Object copy;

Jim Blandy's avatar
Jim Blandy committed
995
  if (!NILP (table))
996 997 998 999 1000
    check_syntax_table (table);
  else
    table = Vstandard_syntax_table;

  copy = Fcopy_sequence (table);
1001 1002 1003

  /* 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
1004
  set_char_table_defalt (copy, Qnil);
1005 1006 1007 1008 1009 1010

  /* 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);
1011
  return copy;
Richard M. Stallman's avatar
Richard M. Stallman committed
1012 1013 1014
}

DEFUN ("set-syntax-table", Fset_syntax_table, Sset_syntax_table, 1, 1, 0,
1015 1016
       doc: /* Select a new syntax table for the current buffer.
One argument, a syntax table.  */)
Dan Nicolaescu's avatar
Dan Nicolaescu committed
1017
  (Lisp_Object table)
Richard M. Stallman's avatar
Richard M. Stallman committed
1018
{
1019
  int idx;
1020
  check_syntax_table (table);
Paul Eggert's avatar
Paul Eggert committed
1021
  bset_syntax_table (current_buffer, table);
Richard M. Stallman's avatar
Richard M. Stallman committed
1022
  /* Indicate that this buffer now has a specified syntax table.  */
Gerd Moellmann's avatar
Gerd Moellmann committed
1023 1024
  idx = PER_BUFFER_VAR_IDX (syntax_table);
  SET_PER_BUFFER_VALUE_P (current_buffer, idx, 1);
Richard M. Stallman's avatar
Richard M. Stallman committed
1025 1026 1027 1028 1029
  return table;
}

/* Convert a letter which signifies a syntax code
 into the code it signifies.
1030
 This is used by modify-syntax-entry, and other things.  */
Richard M. Stallman's avatar
Richard M. Stallman committed
1031

Paul Eggert's avatar
Paul Eggert committed
1032
unsigned char const syntax_spec_code[0400] =
Richard M. Stallman's avatar
Richard M. Stallman committed
1033 1034 1035 1036
  { 0377, 0377, 0377, 0377, 0377, 0377, 0377, 0377,
    0377, 0377, 0377, 0377, 0377, 0377, 0377, 0377,
    0377, 0377, 0377, 0377, 0377, 0377, 0377, 0377,
    0377, 0377, 0377, 0377, 0377, 0377, 0377, 0377,
Paul Eggert's avatar
Paul Eggert committed
1037 1038
    Swhitespace, Scomment_fence, Sstring, 0377, Smath, 0377, 0377, Squote,
    Sopen, Sclose, 0377, 0377, 0377, Swhitespace, Spunct, Scharquote,
Richard M. Stallman's avatar
Richard M. Stallman committed
1039
    0377, 0377, 0377, 0377, 0377, 0377, 0377, 0377,
Paul Eggert's avatar
Paul Eggert committed
1040 1041
    0377, 0377, 0377, 0377, Scomment, 0377, Sendcomment, 0377,
    Sinherit, 0377, 0377, 0377, 0377, 0377, 0377, 0377,   /* @, A ... */
Richard M. Stallman's avatar
Richard M. Stallman committed
1042
    0377, 0377, 0377, 0377, 0377, 0377, 0377, 0377,
Paul Eggert's avatar
Paul Eggert committed
1043 1044
    0377, 0377, 0377, 0377, 0377, 0377, 0377, Sword,
    0377, 0377, 0377, 0377, Sescape, 0377, 0377, Ssymbol,
Richard M. Stallman's avatar
Richard M. Stallman committed
1045 1046
    0377, 0377, 0377, 0377, 0377, 0377, 0377, 0377,   /* `, a, ... */
    0377, 0377, 0377, 0377, 0377, 0377, 0377, 0377,
Paul Eggert's avatar
Paul Eggert committed
1047 1048
    0377, 0377, 0377, 0377, 0377, 0377, 0377, Sword,
    0377, 0377, 0377, 0377, Sstring_fence, 0377, 0377, 0377
Richard M. Stallman's avatar
Richard M. Stallman committed
1049 1050
  };

1051
/* Indexed by syntax code, give the letter that describes it.  */
Richard M. Stallman's avatar
Richard M. Stallman committed
1052

Paul Eggert's avatar
Paul Eggert committed
1053
char const syntax_code_spec[16] =
Richard M. Stallman's avatar
Richard M. Stallman committed
1054
  {
1055 1056
    ' ', '.', 'w', '_', '(', ')', '\'', '\"', '$', '\\', '/', '<', '>', '@',
    '!', '|'
Richard M. Stallman's avatar
Richard M. Stallman committed
1057
  };
Karl Heuer's avatar
Karl Heuer committed
1058 1059 1060