syntax.c 103 KB
Newer Older
Richard M. Stallman's avatar
Richard M. Stallman committed
1
/* GNU Emacs routines to deal with syntax tables; also word and list parsing.
2
   Copyright (C) 1985, 1987, 1993-1995, 1997-1999, 2001-2012
3
                 Free Software Foundation, Inc.
Richard M. Stallman's avatar
Richard M. Stallman committed
4 5 6

This file is part of GNU Emacs.

7
GNU Emacs is free software: you can redistribute it and/or modify
Richard M. Stallman's avatar
Richard M. Stallman committed
8
it under the terms of the GNU General Public License as published by
9 10
the Free Software Foundation, either version 3 of the License, or
(at your option) any later version.
Richard M. Stallman's avatar
Richard M. Stallman committed
11 12 13 14 15 16 17

GNU Emacs is distributed in the hope that it will be useful,
but WITHOUT ANY WARRANTY; without even the implied warranty of
MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the
GNU General Public License for more details.

You should have received a copy of the GNU General Public License
18
along with GNU Emacs.  If not, see <http://www.gnu.org/licenses/>.  */
Richard M. Stallman's avatar
Richard M. Stallman committed
19 20


21
#include <config.h>
22 23

#include <sys/types.h>
24
#include <setjmp.h>
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 32 33 34

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

Richard M. Stallman's avatar
Richard M. Stallman committed
35
#include "syntax.h"
36
#include "intervals.h"
37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65 66 67 68 69 70 71 72 73 74 75 76 77 78 79 80 81 82 83 84 85 86 87 88 89 90
#include "category.h"

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

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

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

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

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

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

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

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

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

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

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

#define SYNTAX_COMEND_FIRST(c) \
  (SYNTAX_FLAGS_COMEND_FIRST (SYNTAX_WITH_FLAGS (c)))
#define SYNTAX_PREFIX(c) (SYNTAX_FLAGS_PREFIX (SYNTAX_WITH_FLAGS (c)))
91 92 93 94 95

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

96 97
#define ST_COMMENT_STYLE (256 + 1)
#define ST_STRING_STYLE (256 + 2)
Richard M. Stallman's avatar
Richard M. Stallman committed
98

99 100
static Lisp_Object Qsyntax_table_p;
static Lisp_Object Qsyntax_table, Qscan_error;
Richard M. Stallman's avatar
Richard M. Stallman committed
101

102
#ifndef __GNUC__
103 104 105 106
/* Used as a temporary in SYNTAX_ENTRY and other macros in syntax.h,
   if not compiled with GCC.  No need to mark it, since it is used
   only very temporarily.  */
Lisp_Object syntax_temp;
107
#endif
108

109 110 111 112
/* This is the internal form of the parse state used in parse-partial-sexp.  */

struct lisp_parse_state
  {
113
    EMACS_INT depth;	/* Depth at end of parsing.  */
114
    int instring;  /* -1 if not within string, else desired terminator.  */
115
    EMACS_INT incomment; /* -1 if in unnestable comment else comment nesting */
116 117
    int comstyle;  /* comment style a=0, or b=1, or ST_COMMENT_STYLE.  */
    int quoted;	   /* Nonzero if just after an escape char at end of parsing */
118
    EMACS_INT mindepth;	/* Minimum depth seen while scanning.  */
119
    /* Char number of most recent start-of-expression at current level */
120
    ptrdiff_t thislevelstart;
121
    /* Char number of start of containing expression */
122 123 124
    ptrdiff_t prevlevelstart;
    ptrdiff_t location;	     /* Char number at which parsing stopped.  */
    ptrdiff_t comstr_start;  /* Position of last comment/string starter.  */
125 126
    Lisp_Object levelstarts; /* Char numbers of starts-of-expression
				of levels (starting from outermost).  */
127 128
  };

129 130 131
/* 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.
132
   find_start_value_byte is the corresponding byte position.
133 134 135 136
   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.  */

137 138 139
static ptrdiff_t find_start_pos;
static ptrdiff_t find_start_value;
static ptrdiff_t find_start_value_byte;
140
static struct buffer *find_start_buffer;
141 142
static ptrdiff_t find_start_begv;
static EMACS_INT find_start_modiff;
143 144


145 146 147 148
static Lisp_Object skip_chars (int, Lisp_Object, Lisp_Object, int);
static Lisp_Object skip_syntaxes (int, Lisp_Object, Lisp_Object);
static Lisp_Object scan_lists (EMACS_INT, EMACS_INT, EMACS_INT, int);
static void scan_sexps_forward (struct lisp_parse_state *,
149
                                ptrdiff_t, ptrdiff_t, ptrdiff_t, EMACS_INT,
150 151
                                int, Lisp_Object, int);
static int in_classes (int, Lisp_Object);
Paul Eggert's avatar
Paul Eggert committed
152 153 154 155 156 157 158

/* This setter is used only in this file, so it can be private.  */
static inline void
bset_syntax_table (struct buffer *b, Lisp_Object val)
{
  b->INTERNAL_FIELD (syntax_table) = val;
}
159

160 161 162 163 164
/* Whether the syntax of the character C has the prefix flag set.  */
int syntax_prefix_flag_p (int c)
{
  return SYNTAX_PREFIX (c);
}
165 166 167 168 169 170

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

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

171 172
/* Update gl_state to an appropriate interval which contains CHARPOS.  The
   sign of COUNT give the relative position of CHARPOS wrt the previously
173
   valid interval.  If INIT, only [be]_property fields of gl_state are
174
   valid at start, the rest is filled basing on OBJECT.
175

176
   `gl_state.*_i' are the intervals, and CHARPOS is further in the search
177 178
   direction than the intervals - or in an interval.  We update the
   current syntax-table basing on the property of this interval, and
179
   update the interval to start further than CHARPOS - or be
Dmitry Antipov's avatar
Dmitry Antipov committed
180
   NULL.  We also update lim_property to be the next value of
181
   charpos to call this subroutine again - or be before/after the
182 183 184
   start/end of OBJECT.  */

void
185
update_syntax_table (ptrdiff_t charpos, EMACS_INT count, int init,
186
		     Lisp_Object object)
187 188
{
  Lisp_Object tmp_table;
189 190
  unsigned cnt = 0;
  int invalidate = 1;
191
  INTERVAL i;
192 193 194

  if (init)
    {
195
      gl_state.old_prop = Qnil;
196 197
      gl_state.start = gl_state.b_property;
      gl_state.stop = gl_state.e_property;
198 199
      i = interval_of (charpos, object);
      gl_state.backward_i = gl_state.forward_i = i;
200
      invalidate = 0;
Dmitry Antipov's avatar
Dmitry Antipov committed
201
      if (!i)
202
	return;
203
      /* interval_of updates only ->position of the return value, so
204
	 update the parents manually to speed up update_interval.  */
205
      while (!NULL_PARENT (i))
206 207
	{
	  if (AM_RIGHT_CHILD (i))
208
	    INTERVAL_PARENT (i)->position = i->position
209
	      - LEFT_TOTAL_LENGTH (i) + TOTAL_LENGTH (i) /* right end */
210 211
	      - TOTAL_LENGTH (INTERVAL_PARENT (i))
	      + LEFT_TOTAL_LENGTH (INTERVAL_PARENT (i));
212
	  else
213
	    INTERVAL_PARENT (i)->position = i->position - LEFT_TOTAL_LENGTH (i)
214
	      + TOTAL_LENGTH (i);
215
	  i = INTERVAL_PARENT (i);
216 217
	}
      i = gl_state.forward_i;
218
      gl_state.b_property = i->position - gl_state.offset;
219
      gl_state.e_property = INTERVAL_LAST_POS (i) - gl_state.offset;
220 221
      goto update;
    }
222
  i = count > 0 ? gl_state.forward_i : gl_state.backward_i;
223

224
  /* We are guaranteed to be called with CHARPOS either in i,
225
     or further off.  */
Dmitry Antipov's avatar
Dmitry Antipov committed
226
  if (!i)
227
    error ("Error in syntax_table logic for to-the-end intervals");
228
  else if (charpos < i->position)		/* Move left.  */
229 230
    {
      if (count > 0)
231
	error ("Error in syntax_table logic for intervals <-");
232
      /* Update the interval.  */
233
      i = update_interval (i, charpos);
234
      if (INTERVAL_LAST_POS (i) != gl_state.b_property)
235 236 237
	{
	  invalidate = 0;
	  gl_state.forward_i = i;
238
	  gl_state.e_property = INTERVAL_LAST_POS (i) - gl_state.offset;
239
	}
240
    }
241
  else if (charpos >= INTERVAL_LAST_POS (i)) /* Move right.  */
242 243
    {
      if (count < 0)
244
	error ("Error in syntax_table logic for intervals ->");
245
      /* Update the interval.  */
246
      i = update_interval (i, charpos);
247
      if (i->position != gl_state.e_property)
248 249 250
	{
	  invalidate = 0;
	  gl_state.backward_i = i;
251
	  gl_state.b_property = i->position - gl_state.offset;
252 253 254 255 256 257 258 259
	}
    }

  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
260

261 262 263
  if (invalidate)		/* Did not get to adjacent interval.  */
    {				/* with the same table => */
				/* invalidate the old range.  */
264 265 266
      if (count > 0)
	{
	  gl_state.backward_i = i;
267 268 269
	  gl_state.b_property = i->position - gl_state.offset;
	}
      else
270
	{
271
	  gl_state.forward_i = i;
272
	  gl_state.e_property = INTERVAL_LAST_POS (i) - gl_state.offset;
273 274
	}
    }
275

276
  if (!EQ (tmp_table, gl_state.old_prop))
277
    {
278 279 280 281 282
      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
283
	}
284 285 286 287 288
      else if (CONSP (tmp_table))
	{
	  gl_state.use_global = 1;
	  gl_state.global_code = tmp_table;
	}
Juanma Barranquero's avatar
Juanma Barranquero committed
289
      else
290 291
	{
	  gl_state.use_global = 0;
Tom Tromey's avatar
Tom Tromey committed
292
	  gl_state.current_syntax_table = BVAR (current_buffer, syntax_table);
293
	}
294 295
    }

Dmitry Antipov's avatar
Dmitry Antipov committed
296
  while (i)
297 298 299 300
    {
      if (cnt && !EQ (tmp_table, textget (i->plist, Qsyntax_table)))
	{
	  if (count > 0)
301 302 303 304 305 306
	    {
	      gl_state.e_property = i->position - gl_state.offset;
	      gl_state.forward_i = i;
	    }
	  else
	    {
307 308
	      gl_state.b_property
		= i->position + LENGTH (i) - gl_state.offset;
309 310 311
	      gl_state.backward_i = i;
	    }
	  return;
312
	}
Juanma Barranquero's avatar
Juanma Barranquero committed
313
      else if (cnt == INTERVALS_AT_ONCE)
314 315
	{
	  if (count > 0)
316
	    {
317 318 319 320 321
	      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
322
		+ (next_interval (i) ? 0 : 1);
323 324 325 326 327 328 329 330
	      gl_state.forward_i = i;
	    }
	  else
	    {
	      gl_state.b_property = i->position - gl_state.offset;
	      gl_state.backward_i = i;
	    }
	  return;
331 332 333 334
	}
      cnt++;
      i = count > 0 ? next_interval (i) : previous_interval (i);
    }
Dmitry Antipov's avatar
Dmitry Antipov committed
335
  eassert (i == NULL); /* This property goes to the end.  */
336 337 338 339
  if (count > 0)
    gl_state.e_property = gl_state.stop;
  else
    gl_state.b_property = gl_state.start;
340 341
}

342 343 344
/* Returns TRUE if char at CHARPOS is quoted.
   Global syntax-table data should be set up already to be good at CHARPOS
   or after.  On return global syntax data is good for lookup at CHARPOS. */
345 346

static int
347
char_quoted (ptrdiff_t charpos, ptrdiff_t bytepos)
348 349
{
  register enum syntaxcode code;
350
  register ptrdiff_t beg = BEGV;
351
  register int quoted = 0;
352
  ptrdiff_t orig = charpos;
353

354
  while (charpos > beg)
355
    {
356
      int c;
357
      DEC_BOTH (charpos, bytepos);
358

359
      UPDATE_SYNTAX_TABLE_BACKWARD (charpos);
360
      c = FETCH_CHAR_AS_MULTIBYTE (bytepos);
361
      code = SYNTAX (c);
362 363 364 365
      if (! (code == Scharquote || code == Sescape))
	break;

      quoted = !quoted;
366
    }
367 368

  UPDATE_SYNTAX_TABLE (orig);
369 370
  return quoted;
}
371 372 373 374

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

375 376
static inline ptrdiff_t
dec_bytepos (ptrdiff_t bytepos)
377
{
Tom Tromey's avatar
Tom Tromey committed
378
  if (NILP (BVAR (current_buffer, enable_multibyte_characters)))
379 380
    return bytepos - 1;

381 382 383
  DEC_POS (bytepos);
  return bytepos;
}
384

Juanma Barranquero's avatar
Juanma Barranquero committed
385
/* Return a defun-start position before POS and not too far before.
Richard M. Stallman's avatar
Richard M. Stallman committed
386 387 388
   It should be the last one before POS, or nearly the last.

   When open_paren_in_column_0_is_defun_start is nonzero,
389
   only the beginning of the buffer is treated as a defun-start.
Richard M. Stallman's avatar
Richard M. Stallman committed
390 391 392 393

   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.
394 395 396 397

   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.  */
398

399 400
static ptrdiff_t
find_defun_start (ptrdiff_t pos, ptrdiff_t pos_byte)
401
{
402
  ptrdiff_t opoint = PT, opoint_byte = PT_BYTE;
403

404 405
  if (!open_paren_in_column_0_is_defun_start)
    {
406
      find_start_value = BEGV;
407
      find_start_value_byte = BEGV_BYTE;
408 409 410 411
      find_start_buffer = current_buffer;
      find_start_modiff = MODIFF;
      find_start_begv = BEGV;
      find_start_pos = pos;
412 413 414
      return BEGV;
    }

415 416 417 418 419 420 421 422 423 424 425 426
  /* Use previous finding, if it's valid and applies to this inquiry.  */
  if (current_buffer == find_start_buffer
      /* Reuse the defun-start even if POS is a little farther on.
	 POS might be in the next defun, but that's ok.
	 Our value may not be the best possible, but will still be usable.  */
      && pos <= find_start_pos + 1000
      && pos >= find_start_value
      && BEGV == find_start_begv
      && MODIFF == find_start_modiff)
    return find_start_value;

  /* Back up to start of line.  */
427
  scan_newline (pos, pos_byte, BEGV, BEGV_BYTE, -1, 1);
428

429 430 431
  /* We optimize syntax-table lookup for rare updates.  Thus we accept
     only those `^\s(' which are good in global _and_ text-property
     syntax-tables.  */
432
  SETUP_BUFFER_SYNTAX_TABLE ();
433
  while (PT > BEGV)
434
    {
435 436
      int c;

437 438
      /* Open-paren at start of line means we may have found our
	 defun-start.  */
439
      c = FETCH_CHAR_AS_MULTIBYTE (PT_BYTE);
440
      if (SYNTAX (c) == Sopen)
441
	{
442
	  SETUP_SYNTAX_TABLE (PT + 1, -1);	/* Try again... */
443
	  c = FETCH_CHAR_AS_MULTIBYTE (PT_BYTE);
444
	  if (SYNTAX (c) == Sopen)
445 446
	    break;
	  /* Now fallback to the default value.  */
447
	  SETUP_BUFFER_SYNTAX_TABLE ();
448
	}
449 450
      /* Move to beg of previous line.  */
      scan_newline (PT, PT_BYTE, BEGV, BEGV_BYTE, -2, 1);
451 452 453
    }

  /* Record what we found, for the next try.  */
454 455
  find_start_value = PT;
  find_start_value_byte = PT_BYTE;
456 457 458 459 460
  find_start_buffer = current_buffer;
  find_start_modiff = MODIFF;
  find_start_begv = BEGV;
  find_start_pos = pos;

461 462
  TEMP_SET_PT_BOTH (opoint, opoint_byte);

463 464 465
  return find_start_value;
}

466 467 468
/* Return the SYNTAX_COMEND_FIRST of the character before POS, POS_BYTE.  */

static int
469
prev_char_comend_first (ptrdiff_t pos, ptrdiff_t pos_byte)
470 471 472 473 474 475 476 477 478 479 480 481 482
{
  int c, val;

  DEC_BOTH (pos, pos_byte);
  UPDATE_SYNTAX_TABLE_BACKWARD (pos);
  c = FETCH_CHAR (pos_byte);
  val = SYNTAX_COMEND_FIRST (c);
  UPDATE_SYNTAX_TABLE_FORWARD (pos + 1);
  return val;
}

/* Return the SYNTAX_COMSTART_FIRST of the character before POS, POS_BYTE.  */

483 484 485 486 487
/* static int
 * prev_char_comstart_first (pos, pos_byte)
 *      int pos, pos_byte;
 * {
 *   int c, val;
Juanma Barranquero's avatar
Juanma Barranquero committed
488
 *
489 490 491 492 493 494 495
 *   DEC_BOTH (pos, pos_byte);
 *   UPDATE_SYNTAX_TABLE_BACKWARD (pos);
 *   c = FETCH_CHAR (pos_byte);
 *   val = SYNTAX_COMSTART_FIRST (c);
 *   UPDATE_SYNTAX_TABLE_FORWARD (pos + 1);
 *   return val;
 * } */
496

497 498 499 500 501 502 503 504 505
/* Checks whether charpos FROM is at the end of a comment.
   FROM_BYTE is the bytepos corresponding to FROM.
   Do not move back before STOP.

   Return a positive value if we find a comment ending at FROM/FROM_BYTE;
   return -1 otherwise.

   If successful, store the charpos of the comment's beginning
   into *CHARPOS_PTR, and the bytepos into *BYTEPOS_PTR.
506 507 508

   Global syntax data remains valid for backward search starting at
   the returned value (or at FROM, if the search was not successful).  */
509 510

static int
511
back_comment (ptrdiff_t from, ptrdiff_t from_byte, ptrdiff_t stop, int comnested, int comstyle, ptrdiff_t *charpos_ptr, ptrdiff_t *bytepos_ptr)
512 513 514 515 516 517 518 519 520 521
{
  /* 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.  */
522
  int string_style = -1;	/* Presumed outside of any string. */
523
  int string_lossage = 0;
524
  /* Not a real lossage: indicates that we have passed a matching comment
Juanma Barranquero's avatar
Juanma Barranquero committed
525
     starter plus a non-matching comment-ender, meaning that any matching
526 527 528 529
     comment-starter we might see later could be a false positive (hidden
     inside another comment).
     Test case:  { a (* b } c (* d *) */
  int comment_lossage = 0;
530 531 532 533
  ptrdiff_t comment_end = from;
  ptrdiff_t comment_end_byte = from_byte;
  ptrdiff_t comstart_pos = 0;
  ptrdiff_t comstart_byte IF_LINT (= 0);
534 535
  /* Place where the containing defun starts,
     or 0 if we didn't come across it yet.  */
536 537
  ptrdiff_t defun_start = 0;
  ptrdiff_t defun_start_byte = 0;
538
  register enum syntaxcode code;
539
  int nesting = 1;		/* current comment nesting */
540
  int c;
541 542 543 544 545
  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.  */
546 547 548 549 550

  /* At beginning of range to scan, we're outside of strings;
     that determines quote parity to the comment-end.  */
  while (from != stop)
    {
551
      ptrdiff_t temp_byte;
552
      int prev_syntax, com2start, com2end;
553
      int comstart;
554

555
      /* Move back and examine a character.  */
556
      DEC_BOTH (from, from_byte);
557 558
      UPDATE_SYNTAX_TABLE_BACKWARD (from);

559
      prev_syntax = syntax;
560
      c = FETCH_CHAR_AS_MULTIBYTE (from_byte);
561
      syntax = SYNTAX_WITH_FLAGS (c);
562 563
      code = SYNTAX (c);

564 565 566
      /* Check for 2-char comment markers.  */
      com2start = (SYNTAX_FLAGS_COMSTART_FIRST (syntax)
		   && SYNTAX_FLAGS_COMSTART_SECOND (prev_syntax)
567 568
		   && (comstyle
		       == SYNTAX_FLAGS_COMMENT_STYLE (prev_syntax, syntax))
569 570 571 572
		   && (SYNTAX_FLAGS_COMMENT_NESTED (prev_syntax)
		       || SYNTAX_FLAGS_COMMENT_NESTED (syntax)) == comnested);
      com2end = (SYNTAX_FLAGS_COMEND_FIRST (syntax)
		 && SYNTAX_FLAGS_COMEND_SECOND (prev_syntax));
573
      comstart = (com2start || code == Scomment);
574

575 576 577 578 579 580 581 582 583 584
      /* 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,
585 586 587
	 we don't try to be clever.  E.g. |*| in C, or }% in modes that
	 have %..\n and %{..}%.  */
      if (from > stop && (com2end || comstart))
588
	{
589
	  ptrdiff_t next = from, next_byte = from_byte;
590
	  int next_c, next_syntax;
591 592
	  DEC_BOTH (next, next_byte);
	  UPDATE_SYNTAX_TABLE_BACKWARD (next);
593
	  next_c = FETCH_CHAR_AS_MULTIBYTE (next_byte);
594
	  next_syntax = SYNTAX_WITH_FLAGS (next_c);
595
	  if (((comstart || comnested)
596 597 598 599
	       && SYNTAX_FLAGS_COMEND_SECOND (syntax)
	       && SYNTAX_FLAGS_COMEND_FIRST (next_syntax))
	      || ((com2end || comnested)
		  && SYNTAX_FLAGS_COMSTART_SECOND (syntax)
600 601
		  && (comstyle
		      == SYNTAX_FLAGS_COMMENT_STYLE (syntax, prev_syntax))
602 603 604
		  && SYNTAX_FLAGS_COMSTART_FIRST (next_syntax)))
	    goto lossage;
	  /* UPDATE_SYNTAX_TABLE_FORWARD (next + 1); */
605
	}
606 607 608 609 610 611 612 613 614 615 616 617 618 619 620

      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
621
	       && (comstyle != SYNTAX_FLAGS_COMMENT_STYLE (syntax, 0)
622
		   || SYNTAX_FLAGS_COMMENT_NESTED (syntax) != comnested))
623
	continue;
624

625 626
      /* Ignore escaped characters, except comment-enders.  */
      if (code != Sendcomment && char_quoted (from, from_byte))
627 628
	continue;

629
      switch (code)
630
	{
631 632 633 634 635 636 637 638 639 640 641 642 643 644
	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.  */
645
	    string_lossage = 1;
646
	  break;
Juanma Barranquero's avatar
Juanma Barranquero committed
647

648 649
	case Scomment:
	  /* We've already checked that it is the relevant comstyle.  */
650
	  if (string_style != -1 || comment_lossage || string_lossage)
651 652 653 654
	    /* There are odd string quotes involved, so let's be careful.
	       Test case in Pascal: " { " a { " } */
	    goto lossage;

655 656 657 658 659 660 661
	  if (!comnested)
	    {
	      /* Record best comment-starter so far.  */
	      comstart_pos = from;
	      comstart_byte = from_byte;
	    }
	  else if (--nesting <= 0)
662 663 664 665 666 667
	    /* 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;
668 669
	  break;

670
	case Sendcomment:
671
	  if (SYNTAX_FLAGS_COMMENT_STYLE (syntax, 0) == comstyle
672
	      && ((com2end && SYNTAX_FLAGS_COMMENT_NESTED (prev_syntax))
673
		  || SYNTAX_FLAGS_COMMENT_NESTED (syntax)) == comnested)
674 675 676 677 678 679 680 681 682
	    /* 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.  */
	    }
683 684 685 686 687 688 689 690 691
	  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;
692
	  break;
693

694 695 696 697 698 699 700 701 702 703 704
	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.  */
	    }
705
	  break;
706 707

	default:
708
	  break;
709
	}
710 711 712 713 714
    }

  if (comstart_pos == 0)
    {
      from = comment_end;
715
      from_byte = comment_end_byte;
716 717
      UPDATE_SYNTAX_TABLE_FORWARD (comment_end - 1);
    }
718 719 720
  /* 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 */
721 722
    {
      from = comstart_pos;
723
      from_byte = comstart_byte;
724
      UPDATE_SYNTAX_TABLE_FORWARD (from - 1);
725 726 727
    }
  else
    {
728 729
      struct lisp_parse_state state;
    lossage:
730 731
      /* We had two kinds of string delimiters mixed up
	 together.  Decode this going forwards.
732
	 Scan fwd from a known safe place (beginning-of-defun)
733 734
	 to the one in question; this records where we
	 last passed a comment starter.  */
735 736 737 738 739 740
      /* 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;
	}
741
      do
742
	{
743 744
	  scan_sexps_forward (&state,
			      defun_start, defun_start_byte,
745 746
			      comment_end, TYPE_MINIMUM (EMACS_INT),
			      0, Qnil, 0);
747 748 749 750 751 752 753 754 755 756 757 758 759 760 761 762 763 764 765
	  defun_start = comment_end;
	  if (state.incomment == (comnested ? 1 : -1)
	      && state.comstyle == comstyle)
	    from = state.comstr_start;
	  else
	    {
	      from = comment_end;
	      if (state.incomment)
		/* If comment_end is inside some other comment, maybe ours
		   is nested, so we need to try again from within the
		   surrounding comment.  Example: { a (* " *)  */
		{
		  /* FIXME: We should advance by one or two chars. */
		  defun_start = state.comstr_start + 2;
		  defun_start_byte = CHAR_TO_BYTE (defun_start);
		}
	    }
	} while (defun_start < comment_end);

766
      from_byte = CHAR_TO_BYTE (from);
767 768
      UPDATE_SYNTAX_TABLE_FORWARD (from - 1);
    }
Juanma Barranquero's avatar
Juanma Barranquero committed
769

770
 done:
771 772 773
  *charpos_ptr = from;
  *bytepos_ptr = from_byte;

774
  return (from == comment_end) ? -1 : from;
775 776
}

Richard M. Stallman's avatar
Richard M. Stallman committed
777
DEFUN ("syntax-table-p", Fsyntax_table_p, Ssyntax_table_p, 1, 1, 0,
778 779
       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
780
  (Lisp_Object object)
Richard M. Stallman's avatar
Richard M. Stallman committed
781
{
782
  if (CHAR_TABLE_P (object)
783
      && EQ (XCHAR_TABLE (object)->purpose, Qsyntax_table))
Richard M. Stallman's avatar
Richard M. Stallman committed
784 785 786 787
    return Qt;
  return Qnil;
}

788
static void
789
check_syntax_table (Lisp_Object obj)
Richard M. Stallman's avatar
Richard M. Stallman committed
790
{
791 792
  CHECK_TYPE (CHAR_TABLE_P (obj) && EQ (XCHAR_TABLE (obj)->purpose, Qsyntax_table),
	      Qsyntax_table_p, obj);
Juanma Barranquero's avatar
Juanma Barranquero committed
793
}
Richard M. Stallman's avatar
Richard M. Stallman committed
794 795

DEFUN ("syntax-table", Fsyntax_table, Ssyntax_table, 0, 0, 0,
796 797
       doc: /* Return the current syntax table.
This is the one specified by the current buffer.  */)
Dan Nicolaescu's avatar
Dan Nicolaescu committed
798
  (void)
Richard M. Stallman's avatar
Richard M. Stallman committed
799
{
Tom Tromey's avatar
Tom Tromey committed
800
  return BVAR (current_buffer, syntax_table);
Richard M. Stallman's avatar
Richard M. Stallman committed
801 802 803 804
}

DEFUN ("standard-syntax-table", Fstandard_syntax_table,
   Sstandard_syntax_table, 0, 0, 0,
805 806
       doc: /* Return the standard syntax table.
This is the one used for new buffers.  */)
Dan Nicolaescu's avatar
Dan Nicolaescu committed
807
  (void)
Richard M. Stallman's avatar
Richard M. Stallman committed
808 809 810 811 812
{
  return Vstandard_syntax_table;
}

DEFUN ("copy-syntax-table", Fcopy_syntax_table, Scopy_syntax_table, 0, 1, 0,
813 814
       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
815
  (Lisp_Object table)
Richard M. Stallman's avatar
Richard M. Stallman committed
816
{
817 818
  Lisp_Object copy;

Jim Blandy's avatar
Jim Blandy committed
819
  if (!NILP (table))
820 821 822 823 824
    check_syntax_table (table);
  else
    table = Vstandard_syntax_table;

  copy = Fcopy_sequence (table);
825 826 827

  /* 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
828
  set_char_table_defalt (copy, Qnil);
829 830 831 832 833 834

  /* 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);
835
  return copy;
Richard M. Stallman's avatar
Richard M. Stallman committed
836 837 838
}

DEFUN ("set-syntax-table", Fset_syntax_table, Sset_syntax_table, 1, 1, 0,
839 840
       doc: /* Select a new syntax table for the current buffer.
One argument, a syntax table.  */)
Dan Nicolaescu's avatar
Dan Nicolaescu committed
841
  (Lisp_Object table)
Richard M. Stallman's avatar
Richard M. Stallman committed
842
{
843
  int idx;
844
  check_syntax_table (table);
Paul Eggert's avatar
Paul Eggert committed
845
  bset_syntax_table (current_buffer, table);
Richard M. Stallman's avatar
Richard M. Stallman committed
846
  /* Indicate that this buffer now has a specified syntax table.  */
Gerd Moellmann's avatar
Gerd Moellmann committed
847 848
  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
849 850 851 852 853
  return table;
}

/* Convert a letter which signifies a syntax code
 into the code it signifies.
854
 This is used by modify-syntax-entry, and other things.  */
Richard M. Stallman's avatar
Richard M. Stallman committed
855 856 857 858 859 860

unsigned char syntax_spec_code[0400] =
  { 0377, 0377, 0377, 0377, 0377, 0377, 0377, 0377,
    0377, 0377, 0377, 0377, 0377, 0377, 0377, 0377,
    0377, 0377, 0377, 0377, 0377, 0377, 0377, 0377,
    0377, 0377, 0377, 0377, 0377, 0377, 0377, 0377,
861
    (char) Swhitespace, (char) Scomment_fence, (char) Sstring, 0377,
Richard M. Stallman's avatar
Richard M. Stallman committed
862 863 864 865 866 867
        (char) Smath, 0377, 0377, (char) Squote,
    (char) Sopen, (char) Sclose, 0377, 0377,
	0377, (char) Swhitespace, (char) Spunct, (char) Scharquote,
    0377, 0377, 0377, 0377, 0377, 0377, 0377, 0377,
    0377, 0377, 0377, 0377,
	(char) Scomment, 0377, (char) Sendcomment, 0377,
868
    (char) Sinherit, 0377, 0377, 0377, 0377, 0377, 0377, 0377,   /* @, A ... */
Richard M. Stallman's avatar
Richard M. Stallman committed
869 870 871 872 873 874
    0377, 0377, 0377, 0377, 0377, 0377, 0377, 0377,
    0377, 0377, 0377, 0377, 0377, 0377, 0377, (char) Sword,
    0377, 0377, 0377, 0377, (char) Sescape, 0377, 0377, (char) Ssymbol,
    0377, 0377, 0377, 0377, 0377, 0377, 0377, 0377,   /* `, a, ... */
    0377, 0377, 0377, 0377, 0377, 0377, 0377, 0377,
    0377, 0377, 0377, 0377, 0377, 0377, 0377, (char) Sword,
875
    0377, 0377, 0377, 0377, (char) Sstring_fence, 0377, 0377, 0377
Richard M. Stallman's avatar
Richard M. Stallman committed
876 877
  };

878
/* Indexed by syntax code, give the letter that describes it.  */
Richard M. Stallman's avatar
Richard M. Stallman committed
879

880
char syntax_code_spec[16] =
Richard M. Stallman's avatar
Richard M. Stallman committed
881
  {
882 883
    ' ', '.', 'w', '_', '(', ')', '\'', '\"', '$', '\\', '/', '<', '>', '@',
    '!', '|'
Richard M. Stallman's avatar
Richard M. Stallman committed
884
  };
Karl Heuer's avatar
Karl Heuer committed
885 886 887 888 889 890 891 892

/* Indexed by syntax code, give the object (cons of syntax code and
   nil) to be stored in syntax table.  Since these objects can be
   shared among syntax tables, we generate them in advance.  By
   sharing objects, the function `describe-syntax' can give a more
   compact listing.  */
static Lisp_Object Vsyntax_code_object;

Richard M. Stallman's avatar
Richard M. Stallman committed
893 894

DEFUN ("char-syntax", Fchar_syntax, Schar_syntax, 1, 1, 0,
895
       doc: /* Return the syntax code of CHARACTER, described by a character.
896 897
For example, if CHARACTER is a word constituent, the
character `w' (119) is returned.
898 899
The characters that correspond to various syntax codes
are listed in the documentation of `modify-syntax-entry'.  */)
Dan Nicolaescu's avatar
Dan Nicolaescu committed
900
  (Lisp_Object character)
Richard M. Stallman's avatar
Richard M. Stallman committed
901
{
902
  int char_int;
903
  CHECK_CHARACTER (character);
904
  char_int = XINT (character);
905
  SETUP_BUFFER_SYNTAX_TABLE ();
906
  return make_number (syntax_code_spec[(int) SYNTAX (char_int)]);
907 908 909
}

DEFUN ("matching-paren", Fmatching_paren, Smatching_paren, 1, 1, 0,
910
       doc: /* Return the matching parenthesis of CHARACTER, or nil if none.  */)
Dan Nicolaescu's avatar
Dan Nicolaescu committed
911
  (Lisp_Object character)
912
{
913
  int char_int, code;
914
  CHECK_NUMBER (character);
915
  char_int = XINT (character);
916
  SETUP_BUFFER_SYNTAX_TABLE ();
917
  code = SYNTAX (char_int);
918
  if (code == Sopen || code == Sclose)
919
    return SYNTAX_MATCH (char_int);
920
  return Qnil;
Richard M. Stallman's avatar
Richard M. Stallman committed
921 922
}

923
DEFUN ("string-to-syntax", Fstring_to_syntax, Sstring_to_syntax, 1, 1, 0,
924 925 926
       doc: /* Convert a syntax specification STRING into syntax cell form.
STRING should be a string as it is allowed as argument of
`modify-syntax-entry'.  Value is the equivalent cons cell
927
\(CODE . MATCHING-CHAR) that can be used as value of a `syntax-table'
928
text property.  */)
Dan Nicolaescu's avatar
Dan Nicolaescu committed
929
  (Lisp_Object string)
Richard M. Stallman's avatar
Richard M. Stallman committed
930
{
931
  register const unsigned char *p;
Richard M. Stallman's avatar
Richard M. Stallman committed
932
  register enum syntaxcode code;
933
  int val;
934
  Lisp_Object match;
Richard M. Stallman's avatar
Richard M. Stallman committed
935

936
  CHECK_STRING (string);
Richard M. Stallman's avatar
Richard M. Stallman committed
937

938
  p = SDATA (string);
Richard M. Stallman's avatar
Richard M. Stallman committed
939 940
  code = (enum syntaxcode) syntax_spec_code[*p++];
  if (((int) code & 0377) == 0377)
941
    error ("Invalid syntax description letter: %c", p[-1]);
Richard M. Stallman's avatar
Richard M. Stallman committed
942