syntax.c 102 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 3
   Copyright (C) 1985, 1987, 1993-1995, 1997-1999, 2001-2013 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

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
#include "category.h"

Paul Eggert's avatar
Paul Eggert committed
39
/* Eight single-bit flags have the following meanings:
40 41 42 43 44 45 46 47 48 49 50
  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
51
  Bits 6 and 8 discriminate among different comment styles.
52
  Languages such as C++ allow two orthogonal syntax start/end pairs
Paul Eggert's avatar
Paul Eggert committed
53
  and bit 6 determines whether a comment-end or Scommentend
54 55
  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
56
  For 2-char comment markers, the style b flag is looked up only on the second
57
  char of the comment marker and on the first char of the comment ender.
Paul Eggert's avatar
Paul Eggert committed
58 59
  For style c (like the nested flag), the flag can be placed on any of
  the chars.  */
60 61 62 63 64 65 66 67 68 69 70 71 72 73 74

/* 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)
Paul Eggert's avatar
Paul Eggert committed
75 76
#define SYNTAX_FLAGS_COMMENT_STYLEC(flags) (((flags) >> 23) & 1)
#define SYNTAX_FLAGS_COMMENT_STYLEC2(flags) (((flags) >> 22) & 2) /* C * 2 */
77 78 79 80
/* 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) \
Paul Eggert's avatar
Paul Eggert committed
81 82
   | SYNTAX_FLAGS_COMMENT_STYLEC2 (flags) \
   | SYNTAX_FLAGS_COMMENT_STYLEC2 (other_flags))
83 84 85 86 87 88 89 90

#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
    int comstyle;  /* comment style a=0, or b=1, or ST_COMMENT_STYLE.  */
Paul Eggert's avatar
Paul Eggert committed
117
    bool quoted;   /* True 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
    ptrdiff_t prevlevelstart;
    ptrdiff_t location;	     /* Char number at which parsing stopped.  */
124
    ptrdiff_t location_byte; /* Corresponding byte position.  */
125
    ptrdiff_t comstr_start;  /* Position of last comment/string starter.  */
126 127
    Lisp_Object levelstarts; /* Char numbers of starts-of-expression
				of levels (starting from outermost).  */
128 129
  };

130 131 132
/* 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.
133
   find_start_value_byte is the corresponding byte position.
134 135 136 137
   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.  */

138 139 140
static ptrdiff_t find_start_pos;
static ptrdiff_t find_start_value;
static ptrdiff_t find_start_value_byte;
141
static struct buffer *find_start_buffer;
142 143
static ptrdiff_t find_start_begv;
static EMACS_INT find_start_modiff;
144 145


Paul Eggert's avatar
Paul Eggert committed
146 147 148
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);
149
static void scan_sexps_forward (struct lisp_parse_state *,
150
                                ptrdiff_t, ptrdiff_t, ptrdiff_t, EMACS_INT,
Paul Eggert's avatar
Paul Eggert committed
151 152
                                bool, Lisp_Object, int);
static bool in_classes (int, Lisp_Object);
Paul Eggert's avatar
Paul Eggert committed
153 154

/* This setter is used only in this file, so it can be private.  */
155
static void
Paul Eggert's avatar
Paul Eggert committed
156 157 158 159
bset_syntax_table (struct buffer *b, Lisp_Object val)
{
  b->INTERNAL_FIELD (syntax_table) = val;
}
160

161
/* Whether the syntax of the character C has the prefix flag set.  */
Paul Eggert's avatar
Paul Eggert committed
162 163
bool
syntax_prefix_flag_p (int c)
164 165 166
{
  return SYNTAX_PREFIX (c);
}
167 168 169 170 171 172

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.  */

173 174
/* Update gl_state to an appropriate interval which contains CHARPOS.  The
   sign of COUNT give the relative position of CHARPOS wrt the previously
175
   valid interval.  If INIT, only [be]_property fields of gl_state are
176
   valid at start, the rest is filled basing on OBJECT.
177

178
   `gl_state.*_i' are the intervals, and CHARPOS is further in the search
179 180
   direction than the intervals - or in an interval.  We update the
   current syntax-table basing on the property of this interval, and
181
   update the interval to start further than CHARPOS - or be
Dmitry Antipov's avatar
Dmitry Antipov committed
182
   NULL.  We also update lim_property to be the next value of
183
   charpos to call this subroutine again - or be before/after the
184 185 186
   start/end of OBJECT.  */

void
Paul Eggert's avatar
Paul Eggert committed
187
update_syntax_table (ptrdiff_t charpos, EMACS_INT count, bool init,
188
		     Lisp_Object object)
189 190
{
  Lisp_Object tmp_table;
Paul Eggert's avatar
Paul Eggert committed
191 192
  int cnt = 0;
  bool invalidate = 1;
193
  INTERVAL i;
194 195 196

  if (init)
    {
197
      gl_state.old_prop = Qnil;
198 199
      gl_state.start = gl_state.b_property;
      gl_state.stop = gl_state.e_property;
200 201
      i = interval_of (charpos, object);
      gl_state.backward_i = gl_state.forward_i = i;
202
      invalidate = 0;
Dmitry Antipov's avatar
Dmitry Antipov committed
203
      if (!i)
204
	return;
205
      /* interval_of updates only ->position of the return value, so
206
	 update the parents manually to speed up update_interval.  */
207
      while (!NULL_PARENT (i))
208 209
	{
	  if (AM_RIGHT_CHILD (i))
210
	    INTERVAL_PARENT (i)->position = i->position
211
	      - LEFT_TOTAL_LENGTH (i) + TOTAL_LENGTH (i) /* right end */
212 213
	      - TOTAL_LENGTH (INTERVAL_PARENT (i))
	      + LEFT_TOTAL_LENGTH (INTERVAL_PARENT (i));
214
	  else
215
	    INTERVAL_PARENT (i)->position = i->position - LEFT_TOTAL_LENGTH (i)
216
	      + TOTAL_LENGTH (i);
217
	  i = INTERVAL_PARENT (i);
218 219
	}
      i = gl_state.forward_i;
220
      gl_state.b_property = i->position - gl_state.offset;
221
      gl_state.e_property = INTERVAL_LAST_POS (i) - gl_state.offset;
222 223
      goto update;
    }
224
  i = count > 0 ? gl_state.forward_i : gl_state.backward_i;
225

226
  /* We are guaranteed to be called with CHARPOS either in i,
227
     or further off.  */
Dmitry Antipov's avatar
Dmitry Antipov committed
228
  if (!i)
229
    error ("Error in syntax_table logic for to-the-end intervals");
230
  else if (charpos < i->position)		/* Move left.  */
231 232
    {
      if (count > 0)
233
	error ("Error in syntax_table logic for intervals <-");
234
      /* Update the interval.  */
235
      i = update_interval (i, charpos);
236
      if (INTERVAL_LAST_POS (i) != gl_state.b_property)
237 238 239
	{
	  invalidate = 0;
	  gl_state.forward_i = i;
240
	  gl_state.e_property = INTERVAL_LAST_POS (i) - gl_state.offset;
241
	}
242
    }
243
  else if (charpos >= INTERVAL_LAST_POS (i)) /* Move right.  */
244 245
    {
      if (count < 0)
246
	error ("Error in syntax_table logic for intervals ->");
247
      /* Update the interval.  */
248
      i = update_interval (i, charpos);
249
      if (i->position != gl_state.e_property)
250 251 252
	{
	  invalidate = 0;
	  gl_state.backward_i = i;
253
	  gl_state.b_property = i->position - gl_state.offset;
254 255 256 257 258 259 260 261
	}
    }

  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
262

263 264 265
  if (invalidate)		/* Did not get to adjacent interval.  */
    {				/* with the same table => */
				/* invalidate the old range.  */
266 267 268
      if (count > 0)
	{
	  gl_state.backward_i = i;
269 270 271
	  gl_state.b_property = i->position - gl_state.offset;
	}
      else
272
	{
273
	  gl_state.forward_i = i;
274
	  gl_state.e_property = INTERVAL_LAST_POS (i) - gl_state.offset;
275 276
	}
    }
277

278
  if (!EQ (tmp_table, gl_state.old_prop))
279
    {
280 281 282 283 284
      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
285
	}
286 287 288 289 290
      else if (CONSP (tmp_table))
	{
	  gl_state.use_global = 1;
	  gl_state.global_code = tmp_table;
	}
Juanma Barranquero's avatar
Juanma Barranquero committed
291
      else
292 293
	{
	  gl_state.use_global = 0;
Tom Tromey's avatar
Tom Tromey committed
294
	  gl_state.current_syntax_table = BVAR (current_buffer, syntax_table);
295
	}
296 297
    }

Dmitry Antipov's avatar
Dmitry Antipov committed
298
  while (i)
299 300 301 302
    {
      if (cnt && !EQ (tmp_table, textget (i->plist, Qsyntax_table)))
	{
	  if (count > 0)
303 304 305 306 307 308
	    {
	      gl_state.e_property = i->position - gl_state.offset;
	      gl_state.forward_i = i;
	    }
	  else
	    {
309 310
	      gl_state.b_property
		= i->position + LENGTH (i) - gl_state.offset;
311 312 313
	      gl_state.backward_i = i;
	    }
	  return;
314
	}
Juanma Barranquero's avatar
Juanma Barranquero committed
315
      else if (cnt == INTERVALS_AT_ONCE)
316 317
	{
	  if (count > 0)
318
	    {
319 320 321 322 323
	      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
324
		+ (next_interval (i) ? 0 : 1);
325 326 327 328 329 330 331 332
	      gl_state.forward_i = i;
	    }
	  else
	    {
	      gl_state.b_property = i->position - gl_state.offset;
	      gl_state.backward_i = i;
	    }
	  return;
333 334 335 336
	}
      cnt++;
      i = count > 0 ? next_interval (i) : previous_interval (i);
    }
Dmitry Antipov's avatar
Dmitry Antipov committed
337
  eassert (i == NULL); /* This property goes to the end.  */
338 339 340 341
  if (count > 0)
    gl_state.e_property = gl_state.stop;
  else
    gl_state.b_property = gl_state.start;
342 343
}

Paul Eggert's avatar
Paul Eggert committed
344
/* Returns true if char at CHARPOS is quoted.
345 346
   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. */
347

Paul Eggert's avatar
Paul Eggert committed
348
static bool
349
char_quoted (ptrdiff_t charpos, ptrdiff_t bytepos)
350
{
Paul Eggert's avatar
Paul Eggert committed
351 352 353
  enum syntaxcode code;
  ptrdiff_t beg = BEGV;
  bool quoted = 0;
354
  ptrdiff_t orig = charpos;
355

356
  while (charpos > beg)
357
    {
358
      int c;
359
      DEC_BOTH (charpos, bytepos);
360

361
      UPDATE_SYNTAX_TABLE_BACKWARD (charpos);
362
      c = FETCH_CHAR_AS_MULTIBYTE (bytepos);
363
      code = SYNTAX (c);
364 365 366 367
      if (! (code == Scharquote || code == Sescape))
	break;

      quoted = !quoted;
368
    }
369 370

  UPDATE_SYNTAX_TABLE (orig);
371 372
  return quoted;
}
373 374 375 376

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

377
static ptrdiff_t
378
dec_bytepos (ptrdiff_t bytepos)
379
{
Tom Tromey's avatar
Tom Tromey committed
380
  if (NILP (BVAR (current_buffer, enable_multibyte_characters)))
381 382
    return bytepos - 1;

383 384 385
  DEC_POS (bytepos);
  return bytepos;
}
386

Juanma Barranquero's avatar
Juanma Barranquero committed
387
/* Return a defun-start position before POS and not too far before.
Richard M. Stallman's avatar
Richard M. Stallman committed
388 389 390
   It should be the last one before POS, or nearly the last.

   When open_paren_in_column_0_is_defun_start is nonzero,
391
   only the beginning of the buffer is treated as a defun-start.
Richard M. Stallman's avatar
Richard M. Stallman committed
392 393 394 395

   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.
396 397 398 399

   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.  */
400

401 402
static ptrdiff_t
find_defun_start (ptrdiff_t pos, ptrdiff_t pos_byte)
403
{
404
  ptrdiff_t opoint = PT, opoint_byte = PT_BYTE;
405

406 407
  if (!open_paren_in_column_0_is_defun_start)
    {
408
      find_start_value = BEGV;
409
      find_start_value_byte = BEGV_BYTE;
410 411 412 413
      find_start_buffer = current_buffer;
      find_start_modiff = MODIFF;
      find_start_begv = BEGV;
      find_start_pos = pos;
414 415 416
      return BEGV;
    }

417 418 419 420 421 422 423 424 425 426 427 428
  /* 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.  */
429
  scan_newline (pos, pos_byte, BEGV, BEGV_BYTE, -1, 1);
430

431 432 433
  /* We optimize syntax-table lookup for rare updates.  Thus we accept
     only those `^\s(' which are good in global _and_ text-property
     syntax-tables.  */
434
  SETUP_BUFFER_SYNTAX_TABLE ();
435
  while (PT > BEGV)
436
    {
437 438
      int c;

439 440
      /* Open-paren at start of line means we may have found our
	 defun-start.  */
441
      c = FETCH_CHAR_AS_MULTIBYTE (PT_BYTE);
442
      if (SYNTAX (c) == Sopen)
443
	{
444
	  SETUP_SYNTAX_TABLE (PT + 1, -1);	/* Try again... */
445
	  c = FETCH_CHAR_AS_MULTIBYTE (PT_BYTE);
446
	  if (SYNTAX (c) == Sopen)
447 448
	    break;
	  /* Now fallback to the default value.  */
449
	  SETUP_BUFFER_SYNTAX_TABLE ();
450
	}
451 452
      /* Move to beg of previous line.  */
      scan_newline (PT, PT_BYTE, BEGV, BEGV_BYTE, -2, 1);
453 454 455
    }

  /* Record what we found, for the next try.  */
456 457
  find_start_value = PT;
  find_start_value_byte = PT_BYTE;
458 459 460 461 462
  find_start_buffer = current_buffer;
  find_start_modiff = MODIFF;
  find_start_begv = BEGV;
  find_start_pos = pos;

463 464
  TEMP_SET_PT_BOTH (opoint, opoint_byte);

465 466 467
  return find_start_value;
}

468 469
/* Return the SYNTAX_COMEND_FIRST of the character before POS, POS_BYTE.  */

Paul Eggert's avatar
Paul Eggert committed
470
static bool
471
prev_char_comend_first (ptrdiff_t pos, ptrdiff_t pos_byte)
472
{
Paul Eggert's avatar
Paul Eggert committed
473 474
  int c;
  bool val;
475 476 477 478 479 480 481 482 483

  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
484
/* Check whether charpos FROM is at the end of a comment.
485 486 487
   FROM_BYTE is the bytepos corresponding to FROM.
   Do not move back before STOP.

Paul Eggert's avatar
Paul Eggert committed
488
   Return true if we find a comment ending at FROM/FROM_BYTE.
489 490 491

   If successful, store the charpos of the comment's beginning
   into *CHARPOS_PTR, and the bytepos into *BYTEPOS_PTR.
492 493 494

   Global syntax data remains valid for backward search starting at
   the returned value (or at FROM, if the search was not successful).  */
495

Paul Eggert's avatar
Paul Eggert committed
496 497 498 499
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)
500 501 502 503 504 505 506 507 508 509
{
  /* 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.  */
510
  int string_style = -1;	/* Presumed outside of any string. */
Paul Eggert's avatar
Paul Eggert committed
511
  bool string_lossage = 0;
512
  /* Not a real lossage: indicates that we have passed a matching comment
Juanma Barranquero's avatar
Juanma Barranquero committed
513
     starter plus a non-matching comment-ender, meaning that any matching
514 515 516
     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
517
  bool comment_lossage = 0;
518 519 520 521
  ptrdiff_t comment_end = from;
  ptrdiff_t comment_end_byte = from_byte;
  ptrdiff_t comstart_pos = 0;
  ptrdiff_t comstart_byte IF_LINT (= 0);
522 523
  /* Place where the containing defun starts,
     or 0 if we didn't come across it yet.  */
524 525
  ptrdiff_t defun_start = 0;
  ptrdiff_t defun_start_byte = 0;
Paul Eggert's avatar
Paul Eggert committed
526 527
  enum syntaxcode code;
  ptrdiff_t nesting = 1;		/* current comment nesting */
528
  int c;
529 530 531 532 533
  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.  */
534 535 536 537 538

  /* At beginning of range to scan, we're outside of strings;
     that determines quote parity to the comment-end.  */
  while (from != stop)
    {
539
      ptrdiff_t temp_byte;
Paul Eggert's avatar
Paul Eggert committed
540 541
      int prev_syntax;
      bool com2start, com2end, comstart;
542

543
      /* Move back and examine a character.  */
544
      DEC_BOTH (from, from_byte);
545 546
      UPDATE_SYNTAX_TABLE_BACKWARD (from);

547
      prev_syntax = syntax;
548
      c = FETCH_CHAR_AS_MULTIBYTE (from_byte);
549
      syntax = SYNTAX_WITH_FLAGS (c);
550 551
      code = SYNTAX (c);

552 553 554
      /* Check for 2-char comment markers.  */
      com2start = (SYNTAX_FLAGS_COMSTART_FIRST (syntax)
		   && SYNTAX_FLAGS_COMSTART_SECOND (prev_syntax)
555 556
		   && (comstyle
		       == SYNTAX_FLAGS_COMMENT_STYLE (prev_syntax, syntax))
557 558 559 560
		   && (SYNTAX_FLAGS_COMMENT_NESTED (prev_syntax)
		       || SYNTAX_FLAGS_COMMENT_NESTED (syntax)) == comnested);
      com2end = (SYNTAX_FLAGS_COMEND_FIRST (syntax)
		 && SYNTAX_FLAGS_COMEND_SECOND (prev_syntax));
561
      comstart = (com2start || code == Scomment);
562

563 564 565 566 567 568 569 570 571 572
      /* 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,
573 574 575
	 we don't try to be clever.  E.g. |*| in C, or }% in modes that
	 have %..\n and %{..}%.  */
      if (from > stop && (com2end || comstart))
576
	{
577
	  ptrdiff_t next = from, next_byte = from_byte;
578
	  int next_c, next_syntax;
579 580
	  DEC_BOTH (next, next_byte);
	  UPDATE_SYNTAX_TABLE_BACKWARD (next);
581
	  next_c = FETCH_CHAR_AS_MULTIBYTE (next_byte);
582
	  next_syntax = SYNTAX_WITH_FLAGS (next_c);
583
	  if (((comstart || comnested)
584 585 586 587
	       && SYNTAX_FLAGS_COMEND_SECOND (syntax)
	       && SYNTAX_FLAGS_COMEND_FIRST (next_syntax))
	      || ((com2end || comnested)
		  && SYNTAX_FLAGS_COMSTART_SECOND (syntax)
588 589
		  && (comstyle
		      == SYNTAX_FLAGS_COMMENT_STYLE (syntax, prev_syntax))
590 591 592
		  && SYNTAX_FLAGS_COMSTART_FIRST (next_syntax)))
	    goto lossage;
	  /* UPDATE_SYNTAX_TABLE_FORWARD (next + 1); */
593
	}
594 595 596 597 598 599 600 601 602 603 604 605 606 607 608

      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
609
	       && (comstyle != SYNTAX_FLAGS_COMMENT_STYLE (syntax, 0)
610
		   || SYNTAX_FLAGS_COMMENT_NESTED (syntax) != comnested))
611
	continue;
612

613 614
      /* Ignore escaped characters, except comment-enders.  */
      if (code != Sendcomment && char_quoted (from, from_byte))
615 616
	continue;

617
      switch (code)
618
	{
619 620 621 622 623 624 625 626 627 628 629 630 631 632
	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.  */
633
	    string_lossage = 1;
634
	  break;
Juanma Barranquero's avatar
Juanma Barranquero committed
635

636 637
	case Scomment:
	  /* We've already checked that it is the relevant comstyle.  */
638
	  if (string_style != -1 || comment_lossage || string_lossage)
639 640 641 642
	    /* There are odd string quotes involved, so let's be careful.
	       Test case in Pascal: " { " a { " } */
	    goto lossage;

643 644 645 646 647 648 649
	  if (!comnested)
	    {
	      /* Record best comment-starter so far.  */
	      comstart_pos = from;
	      comstart_byte = from_byte;
	    }
	  else if (--nesting <= 0)
650 651 652 653 654 655
	    /* 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;
656 657
	  break;

658
	case Sendcomment:
659
	  if (SYNTAX_FLAGS_COMMENT_STYLE (syntax, 0) == comstyle
660
	      && ((com2end && SYNTAX_FLAGS_COMMENT_NESTED (prev_syntax))
661
		  || SYNTAX_FLAGS_COMMENT_NESTED (syntax)) == comnested)
662 663 664 665 666 667 668 669 670
	    /* 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.  */
	    }
671 672 673 674 675 676 677 678 679
	  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;
680
	  break;
681

682 683 684 685 686 687 688 689 690 691 692
	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.  */
	    }
693
	  break;
694 695

	default:
696
	  break;
697
	}
698 699 700 701 702
    }

  if (comstart_pos == 0)
    {
      from = comment_end;
703
      from_byte = comment_end_byte;
704 705
      UPDATE_SYNTAX_TABLE_FORWARD (comment_end - 1);
    }
706 707 708
  /* 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 */
709 710
    {
      from = comstart_pos;
711
      from_byte = comstart_byte;
712
      UPDATE_SYNTAX_TABLE_FORWARD (from - 1);
713 714 715
    }
  else
    {
716 717
      struct lisp_parse_state state;
    lossage:
718 719
      /* We had two kinds of string delimiters mixed up
	 together.  Decode this going forwards.
720
	 Scan fwd from a known safe place (beginning-of-defun)
721 722
	 to the one in question; this records where we
	 last passed a comment starter.  */
723 724 725 726 727 728
      /* 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;
	}
729
      do
730
	{
731 732
	  scan_sexps_forward (&state,
			      defun_start, defun_start_byte,
733 734
			      comment_end, TYPE_MINIMUM (EMACS_INT),
			      0, Qnil, 0);
735 736 737 738 739 740 741 742 743 744 745 746 747 748 749 750 751 752 753
	  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);

754
      from_byte = CHAR_TO_BYTE (from);
755 756
      UPDATE_SYNTAX_TABLE_FORWARD (from - 1);
    }
Juanma Barranquero's avatar
Juanma Barranquero committed
757

758
 done:
759 760 761
  *charpos_ptr = from;
  *bytepos_ptr = from_byte;

Paul Eggert's avatar
Paul Eggert committed
762
  return from != comment_end;
763 764
}

Richard M. Stallman's avatar
Richard M. Stallman committed
765
DEFUN ("syntax-table-p", Fsyntax_table_p, Ssyntax_table_p, 1, 1, 0,
766 767
       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
768
  (Lisp_Object object)
Richard M. Stallman's avatar
Richard M. Stallman committed
769
{
770
  if (CHAR_TABLE_P (object)
771
      && EQ (XCHAR_TABLE (object)->purpose, Qsyntax_table))
Richard M. Stallman's avatar
Richard M. Stallman committed
772 773 774 775
    return Qt;
  return Qnil;
}

776
static void
777
check_syntax_table (Lisp_Object obj)
Richard M. Stallman's avatar
Richard M. Stallman committed
778
{
779 780
  CHECK_TYPE (CHAR_TABLE_P (obj) && EQ (XCHAR_TABLE (obj)->purpose, Qsyntax_table),
	      Qsyntax_table_p, obj);
Juanma Barranquero's avatar
Juanma Barranquero committed
781
}
Richard M. Stallman's avatar
Richard M. Stallman committed
782 783

DEFUN ("syntax-table", Fsyntax_table, Ssyntax_table, 0, 0, 0,
784 785
       doc: /* Return the current syntax table.
This is the one specified by the current buffer.  */)
Dan Nicolaescu's avatar
Dan Nicolaescu committed
786
  (void)
Richard M. Stallman's avatar
Richard M. Stallman committed
787
{
Tom Tromey's avatar
Tom Tromey committed
788
  return BVAR (current_buffer, syntax_table);
Richard M. Stallman's avatar
Richard M. Stallman committed
789 790 791 792
}

DEFUN ("standard-syntax-table", Fstandard_syntax_table,
   Sstandard_syntax_table, 0, 0, 0,
793 794
       doc: /* Return the standard syntax table.
This is the one used for new buffers.  */)
Dan Nicolaescu's avatar
Dan Nicolaescu committed
795
  (void)
Richard M. Stallman's avatar
Richard M. Stallman committed
796 797 798 799 800
{
  return Vstandard_syntax_table;
}

DEFUN ("copy-syntax-table", Fcopy_syntax_table, Scopy_syntax_table, 0, 1, 0,
801 802
       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
803
  (Lisp_Object table)
Richard M. Stallman's avatar
Richard M. Stallman committed
804
{
805 806
  Lisp_Object copy;

Jim Blandy's avatar
Jim Blandy committed
807
  if (!NILP (table))
808 809 810 811 812
    check_syntax_table (table);
  else
    table = Vstandard_syntax_table;

  copy = Fcopy_sequence (table);
813 814 815

  /* 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
816
  set_char_table_defalt (copy, Qnil);
817 818 819 820 821 822

  /* 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);
823
  return copy;
Richard M. Stallman's avatar
Richard M. Stallman committed
824 825 826
}

DEFUN ("set-syntax-table", Fset_syntax_table, Sset_syntax_table, 1, 1, 0,
827 828
       doc: /* Select a new syntax table for the current buffer.
One argument, a syntax table.  */)
Dan Nicolaescu's avatar
Dan Nicolaescu committed
829
  (Lisp_Object table)
Richard M. Stallman's avatar
Richard M. Stallman committed
830
{
831
  int idx;
832
  check_syntax_table (table);
Paul Eggert's avatar
Paul Eggert committed
833
  bset_syntax_table (current_buffer, table);
Richard M. Stallman's avatar
Richard M. Stallman committed
834
  /* Indicate that this buffer now has a specified syntax table.  */
Gerd Moellmann's avatar
Gerd Moellmann committed
835 836
  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
837 838 839 840 841
  return table;
}

/* Convert a letter which signifies a syntax code
 into the code it signifies.
842
 This is used by modify-syntax-entry, and other things.  */
Richard M. Stallman's avatar
Richard M. Stallman committed
843

Paul Eggert's avatar
Paul Eggert committed
844
unsigned char const syntax_spec_code[0400] =
Richard M. Stallman's avatar
Richard M. Stallman committed
845 846 847 848
  { 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
849 850
    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
851
    0377, 0377, 0377, 0377, 0377, 0377, 0377, 0377,
Paul Eggert's avatar
Paul Eggert committed
852 853
    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
854
    0377, 0377, 0377, 0377, 0377, 0377, 0377, 0377,
Paul Eggert's avatar
Paul Eggert committed
855 856
    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
857 858
    0377, 0377, 0377, 0377, 0377, 0377, 0377, 0377,   /* `, a, ... */
    0377, 0377, 0377, 0377, 0377, 0377, 0377, 0377,
Paul Eggert's avatar
Paul Eggert committed
859 860
    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
861 862
  };

863
/* Indexed by syntax code, give the letter that describes it.  */
Richard M. Stallman's avatar
Richard M. Stallman committed
864

Paul Eggert's avatar
Paul Eggert committed
865
char const syntax_code_spec[16] =
Richard M. Stallman's avatar
Richard M. Stallman committed
866
  {
867 868
    ' ', '.', 'w', '_', '(', ')', '\'', '\"', '$', '\\', '/', '<', '>', '@',
    '!', '|'
Richard M. Stallman's avatar
Richard M. Stallman committed
869
  };
Karl Heuer's avatar
Karl Heuer committed
870 871 872 873 874 875 876 877

/* 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
878 879

DEFUN ("char-syntax", Fchar_syntax, Schar_syntax, 1, 1, 0,
880
       doc: /* Return the syntax code of CHARACTER, described by a character.
881 882
For example, if CHARACTER is a word constituent, the
character `w' (119) is returned.
883 884
The characters that correspond to various syntax codes
are listed in the documentation of `modify-syntax-entry'.  */)
Dan Nicolaescu's avatar
Dan Nicolaescu committed
885
  (Lisp_Object character)
Richard M. Stallman's avatar
Richard M. Stallman committed
886
{
887
  int char_int;
888
  CHECK_CHARACTER (character);
889
  char_int = XINT (character);
890
  SETUP_BUFFER_SYNTAX_TABLE ();
Paul Eggert's avatar
Paul Eggert committed
891
  return make_number (syntax_code_spec[SYNTAX (char_int)]);
892 893 894
}

DEFUN ("matching-paren", Fmatching_paren, Smatching_paren, 1, 1, 0,
895
       doc: /* Return the matching parenthesis of CHARACTER, or nil if none.  */)
Dan Nicolaescu's avatar
Dan Nicolaescu committed
896
  (Lisp_Object character)
897
{
Paul Eggert's avatar
Paul Eggert committed
898 899 900
  int char_int;
  enum syntaxcode code;
  CHECK_CHARACTER (character);
901
  char_int = XINT (character);
902
  SETUP_BUFFER_SYNTAX_TABLE ();
903
  code = SYNTAX (char_int);
904
  if (code == Sopen || code == Sclose)
905
    return SYNTAX_MATCH (char_int);
906
  return Qnil;
Richard M. Stallman's avatar
Richard M. Stallman committed
907 908
}

909
DEFUN ("string-to-syntax", Fstring_to_syntax, Sstring_to_syntax, 1, 1, 0,