syntax.c 111 KB
Newer Older
Richard M. Stallman's avatar
Richard M. Stallman committed
1
/* GNU Emacs routines to deal with syntax tables; also word and list parsing.
Paul Eggert's avatar
Paul Eggert committed
2
   Copyright (C) 1985, 1987, 1993-1995, 1997-1999, 2001-2016 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
#include "lisp.h"
26
#include "character.h"
27
#include "buffer.h"
Richard M. Stallman's avatar
Richard M. Stallman committed
28
#include "regex.h"
Richard M. Stallman's avatar
Richard M. Stallman committed
29
#include "syntax.h"
30
#include "intervals.h"
31 32
#include "category.h"

33 34 35 36 37
/* 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
38
/* Eight single-bit flags have the following meanings:
39 40 41 42 43 44 45 46 47 48 49
  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
50
  Bits 6 and 8 discriminate among different comment styles.
51
  Languages such as C++ allow two orthogonal syntax start/end pairs
Paul Eggert's avatar
Paul Eggert committed
52
  and bit 6 determines whether a comment-end or Scommentend
53 54
  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
55
  For 2-char comment markers, the style b flag is looked up only on the second
56
  char of the comment marker and on the first char of the comment ender.
Paul Eggert's avatar
Paul Eggert committed
57 58
  For style c (like the nested flag), the flag can be placed on any of
  the chars.  */
59

60
/* These functions extract specific flags from an integer
61 62
   that holds the syntax code and the flags.  */

63 64 65 66 67 68 69 70 71 72 73 74 75 76 77 78 79 80 81 82 83
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
84 85 86 87 88
SYNTAX_FLAGS_COMSTARTEND_FIRST (int flags)
{
  return (flags & 0x50000) != 0;
}
static bool
89 90 91 92 93 94 95 96 97 98 99 100 101 102 103 104 105 106 107 108 109 110 111 112
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;
}
113 114 115

/* FLAGS should be the flags of the main char of the comment marker, e.g.
   the second for comstart and the first for comend.  */
116 117 118 119 120 121 122
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));
}
123

124
/* Extract a particular flag for a given character.  */
125

126 127 128 129 130
static bool
SYNTAX_COMEND_FIRST (int c)
{
  return SYNTAX_FLAGS_COMEND_FIRST (SYNTAX_WITH_FLAGS (c));
}
131 132

/* We use these constants in place for comment-style and
133
   string-ender-char to distinguish comments/strings started by
134 135
   comment_fence and string_fence codes.  */

136 137 138 139 140
enum
  {
    ST_COMMENT_STYLE = 256 + 1,
    ST_STRING_STYLE = 256 + 2
  };
Richard M. Stallman's avatar
Richard M. Stallman committed
141

142 143 144 145
/* This is the internal form of the parse state used in parse-partial-sexp.  */

struct lisp_parse_state
  {
146
    EMACS_INT depth;	/* Depth at end of parsing.  */
147
    int instring;  /* -1 if not within string, else desired terminator.  */
148
    EMACS_INT incomment; /* -1 if in unnestable comment else comment nesting */
149
    int comstyle;  /* comment style a=0, or b=1, or ST_COMMENT_STYLE.  */
Paul Eggert's avatar
Paul Eggert committed
150
    bool quoted;   /* True if just after an escape char at end of parsing.  */
151
    EMACS_INT mindepth;	/* Minimum depth seen while scanning.  */
152
    /* Char number of most recent start-of-expression at current level */
153
    ptrdiff_t thislevelstart;
154
    /* Char number of start of containing expression */
155 156
    ptrdiff_t prevlevelstart;
    ptrdiff_t location;	     /* Char number at which parsing stopped.  */
157
    ptrdiff_t location_byte; /* Corresponding byte position.  */
158
    ptrdiff_t comstr_start;  /* Position of last comment/string starter.  */
159 160
    Lisp_Object levelstarts; /* Char numbers of starts-of-expression
				of levels (starting from outermost).  */
161 162 163 164
    int prev_syntax; /* Syntax of previous position scanned, when
                        that position (potentially) holds the first char
                        of a 2-char construct, i.e. comment delimiter
                        or Sescape, etc.  Smax otherwise. */
165 166
  };

167 168 169
/* 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.
170
   find_start_value_byte is the corresponding byte position.
171 172 173 174
   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.  */

175 176 177
static ptrdiff_t find_start_pos;
static ptrdiff_t find_start_value;
static ptrdiff_t find_start_value_byte;
178
static struct buffer *find_start_buffer;
179 180
static ptrdiff_t find_start_begv;
static EMACS_INT find_start_modiff;
181 182


Paul Eggert's avatar
Paul Eggert committed
183 184 185
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);
186
static void scan_sexps_forward (struct lisp_parse_state *,
187
                                ptrdiff_t, ptrdiff_t, ptrdiff_t, EMACS_INT,
188 189
                                bool, int);
static void internalize_parse_state (Lisp_Object, struct lisp_parse_state *);
Paul Eggert's avatar
Paul Eggert committed
190
static bool in_classes (int, Lisp_Object);
191
static void parse_sexp_propertize (ptrdiff_t charpos);
Paul Eggert's avatar
Paul Eggert committed
192 193

/* This setter is used only in this file, so it can be private.  */
194
static void
Paul Eggert's avatar
Paul Eggert committed
195 196
bset_syntax_table (struct buffer *b, Lisp_Object val)
{
197
  b->syntax_table_ = val;
Paul Eggert's avatar
Paul Eggert committed
198
}
199

200
/* Whether the syntax of the character C has the prefix flag set.  */
Paul Eggert's avatar
Paul Eggert committed
201 202
bool
syntax_prefix_flag_p (int c)
203
{
204
  return SYNTAX_FLAGS_PREFIX (SYNTAX_WITH_FLAGS (c));
205
}
206 207 208

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

209
enum { INTERVALS_AT_ONCE = 10 };	/* 1 + max-number of intervals
210 211
					   to scan to property-change.  */

212 213 214 215 216 217 218 219 220 221 222 223 224 225 226 227 228 229 230 231 232 233 234 235 236 237 238 239 240 241 242 243 244 245 246 247 248 249 250 251 252 253 254 255 256
/* 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)
257
    {
258
      if (count > 0)
259
	update_syntax_table_forward (from, true, Qnil);
260
      else if (from > BEGV)
261
	{
262 263
	  update_syntax_table (from - 1, count, true, Qnil);
	  parse_sexp_propertize (from - 1);
264 265
	}
    }
266 267 268 269 270 271 272 273 274 275 276 277 278 279 280 281 282 283 284 285 286 287 288 289 290 291 292 293 294 295 296 297 298 299 300 301 302 303 304 305 306 307 308 309 310 311
}

/* 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);
}

312 313
/* Update gl_state to an appropriate interval which contains CHARPOS.  The
   sign of COUNT give the relative position of CHARPOS wrt the previously
314
   valid interval.  If INIT, only [be]_property fields of gl_state are
315
   valid at start, the rest is filled basing on OBJECT.
316

317
   `gl_state.*_i' are the intervals, and CHARPOS is further in the search
318 319
   direction than the intervals - or in an interval.  We update the
   current syntax-table basing on the property of this interval, and
320
   update the interval to start further than CHARPOS - or be
Dmitry Antipov's avatar
Dmitry Antipov committed
321
   NULL.  We also update lim_property to be the next value of
322
   charpos to call this subroutine again - or be before/after the
323 324 325
   start/end of OBJECT.  */

void
Paul Eggert's avatar
Paul Eggert committed
326
update_syntax_table (ptrdiff_t charpos, EMACS_INT count, bool init,
327
		     Lisp_Object object)
328 329
{
  Lisp_Object tmp_table;
Paul Eggert's avatar
Paul Eggert committed
330
  int cnt = 0;
331
  bool invalidate = true;
332
  INTERVAL i;
333 334 335

  if (init)
    {
336
      gl_state.old_prop = Qnil;
337 338
      gl_state.start = gl_state.b_property;
      gl_state.stop = gl_state.e_property;
339 340
      i = interval_of (charpos, object);
      gl_state.backward_i = gl_state.forward_i = i;
341
      invalidate = false;
Dmitry Antipov's avatar
Dmitry Antipov committed
342
      if (!i)
343
	return;
344
      /* interval_of updates only ->position of the return value, so
345
	 update the parents manually to speed up update_interval.  */
346
      while (!NULL_PARENT (i))
347 348
	{
	  if (AM_RIGHT_CHILD (i))
349
	    INTERVAL_PARENT (i)->position = i->position
350
	      - LEFT_TOTAL_LENGTH (i) + TOTAL_LENGTH (i) /* right end */
351 352
	      - TOTAL_LENGTH (INTERVAL_PARENT (i))
	      + LEFT_TOTAL_LENGTH (INTERVAL_PARENT (i));
353
	  else
354
	    INTERVAL_PARENT (i)->position = i->position - LEFT_TOTAL_LENGTH (i)
355
	      + TOTAL_LENGTH (i);
356
	  i = INTERVAL_PARENT (i);
357 358
	}
      i = gl_state.forward_i;
359
      gl_state.b_property = i->position - gl_state.offset;
360
      gl_state.e_property = INTERVAL_LAST_POS (i) - gl_state.offset;
361 362
      goto update;
    }
363
  i = count > 0 ? gl_state.forward_i : gl_state.backward_i;
364

365
  /* We are guaranteed to be called with CHARPOS either in i,
366
     or further off.  */
Dmitry Antipov's avatar
Dmitry Antipov committed
367
  if (!i)
368
    error ("Error in syntax_table logic for to-the-end intervals");
369
  else if (charpos < i->position)		/* Move left.  */
370 371
    {
      if (count > 0)
372
	error ("Error in syntax_table logic for intervals <-");
373
      /* Update the interval.  */
374
      i = update_interval (i, charpos);
375
      if (INTERVAL_LAST_POS (i) != gl_state.b_property)
376
	{
377
	  invalidate = false;
378
	  gl_state.forward_i = i;
379
	  gl_state.e_property = INTERVAL_LAST_POS (i) - gl_state.offset;
380
	}
381
    }
382
  else if (charpos >= INTERVAL_LAST_POS (i)) /* Move right.  */
383 384
    {
      if (count < 0)
385
	error ("Error in syntax_table logic for intervals ->");
386
      /* Update the interval.  */
387
      i = update_interval (i, charpos);
388
      if (i->position != gl_state.e_property)
389
	{
390
	  invalidate = false;
391
	  gl_state.backward_i = i;
392
	  gl_state.b_property = i->position - gl_state.offset;
393 394 395 396 397 398 399 400
	}
    }

  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
401

402 403 404
  if (invalidate)		/* Did not get to adjacent interval.  */
    {				/* with the same table => */
				/* invalidate the old range.  */
405 406 407
      if (count > 0)
	{
	  gl_state.backward_i = i;
408 409 410
	  gl_state.b_property = i->position - gl_state.offset;
	}
      else
411
	{
412
	  gl_state.forward_i = i;
413
	  gl_state.e_property = INTERVAL_LAST_POS (i) - gl_state.offset;
414 415
	}
    }
416

417
  if (!EQ (tmp_table, gl_state.old_prop))
418
    {
419 420 421 422 423
      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
424
	}
425 426 427 428 429
      else if (CONSP (tmp_table))
	{
	  gl_state.use_global = 1;
	  gl_state.global_code = tmp_table;
	}
Juanma Barranquero's avatar
Juanma Barranquero committed
430
      else
431 432
	{
	  gl_state.use_global = 0;
Tom Tromey's avatar
Tom Tromey committed
433
	  gl_state.current_syntax_table = BVAR (current_buffer, syntax_table);
434
	}
435 436
    }

Dmitry Antipov's avatar
Dmitry Antipov committed
437
  while (i)
438 439 440 441
    {
      if (cnt && !EQ (tmp_table, textget (i->plist, Qsyntax_table)))
	{
	  if (count > 0)
442 443 444 445 446 447
	    {
	      gl_state.e_property = i->position - gl_state.offset;
	      gl_state.forward_i = i;
	    }
	  else
	    {
448 449
	      gl_state.b_property
		= i->position + LENGTH (i) - gl_state.offset;
450 451 452
	      gl_state.backward_i = i;
	    }
	  return;
453
	}
Juanma Barranquero's avatar
Juanma Barranquero committed
454
      else if (cnt == INTERVALS_AT_ONCE)
455 456
	{
	  if (count > 0)
457
	    {
458 459 460 461 462
	      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
463
		+ (next_interval (i) ? 0 : 1);
464 465 466 467 468 469 470 471
	      gl_state.forward_i = i;
	    }
	  else
	    {
	      gl_state.b_property = i->position - gl_state.offset;
	      gl_state.backward_i = i;
	    }
	  return;
472 473 474 475
	}
      cnt++;
      i = count > 0 ? next_interval (i) : previous_interval (i);
    }
Dmitry Antipov's avatar
Dmitry Antipov committed
476
  eassert (i == NULL); /* This property goes to the end.  */
477
  if (count > 0)
478 479 480 481
    {
      gl_state.e_property = gl_state.stop;
      gl_state.forward_i = i;
    }
482 483
  else
    gl_state.b_property = gl_state.start;
484
}
485 486 487 488

static void
parse_sexp_propertize (ptrdiff_t charpos)
{
489 490 491 492 493 494 495 496 497 498 499 500 501 502 503 504 505 506 507 508
  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;
    }
Stefan Monnier's avatar
Stefan Monnier committed
509 510
  else if (gl_state.e_property_truncated
	   && gl_state.e_property < syntax_propertize__done)
511 512 513 514
    { /* When moving backward, e_property might be set without resetting
	 e_property_truncated, so the e_property_truncated flag may
	 occasionally be left raised spuriously.  This should be rare.  */
      gl_state.e_property_truncated = false;
515
      update_syntax_table_forward (charpos, false, Qnil);
516
    }
517 518 519
}

void
520
update_syntax_table_forward (ptrdiff_t charpos, bool init,
521
			     Lisp_Object object)
522
{
523
  if (gl_state.e_property_truncated)
524
    {
525 526
      eassert (NILP (object));
      eassert (charpos >= gl_state.e_property);
527
      parse_sexp_propertize (charpos);
528 529 530 531
    }
  else
    {
      update_syntax_table (charpos, 1, init, object);
532 533
      if (NILP (object) && gl_state.e_property > syntax_propertize__done)
	parse_sexp_propertize (charpos);
534 535
    }
}
536

Paul Eggert's avatar
Paul Eggert committed
537
/* Returns true if char at CHARPOS is quoted.
538
   Global syntax-table data should be set up already to be good at CHARPOS
539
   or after.  On return global syntax data is good for lookup at CHARPOS.  */
540

Paul Eggert's avatar
Paul Eggert committed
541
static bool
542
char_quoted (ptrdiff_t charpos, ptrdiff_t bytepos)
543
{
Paul Eggert's avatar
Paul Eggert committed
544 545 546
  enum syntaxcode code;
  ptrdiff_t beg = BEGV;
  bool quoted = 0;
547
  ptrdiff_t orig = charpos;
548

549
  while (charpos > beg)
550
    {
551
      int c;
552
      DEC_BOTH (charpos, bytepos);
553

554
      UPDATE_SYNTAX_TABLE_BACKWARD (charpos);
555
      c = FETCH_CHAR_AS_MULTIBYTE (bytepos);
556
      code = SYNTAX (c);
557 558 559 560
      if (! (code == Scharquote || code == Sescape))
	break;

      quoted = !quoted;
561
    }
562 563

  UPDATE_SYNTAX_TABLE (orig);
564 565
  return quoted;
}
566 567 568 569

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

570
static ptrdiff_t
571
dec_bytepos (ptrdiff_t bytepos)
572
{
Tom Tromey's avatar
Tom Tromey committed
573
  if (NILP (BVAR (current_buffer, enable_multibyte_characters)))
574 575
    return bytepos - 1;

576 577 578
  DEC_POS (bytepos);
  return bytepos;
}
579

Juanma Barranquero's avatar
Juanma Barranquero committed
580
/* Return a defun-start position before POS and not too far before.
Richard M. Stallman's avatar
Richard M. Stallman committed
581 582 583
   It should be the last one before POS, or nearly the last.

   When open_paren_in_column_0_is_defun_start is nonzero,
584
   only the beginning of the buffer is treated as a defun-start.
Richard M. Stallman's avatar
Richard M. Stallman committed
585 586 587 588

   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.
589 590 591 592

   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.  */
593

594 595
static ptrdiff_t
find_defun_start (ptrdiff_t pos, ptrdiff_t pos_byte)
596
{
597
  ptrdiff_t opoint = PT, opoint_byte = PT_BYTE;
598 599 600 601 602 603 604 605 606 607 608 609

  /* 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;

610 611 612 613 614 615 616
  if (!open_paren_in_column_0_is_defun_start)
    {
      find_start_value = BEGV;
      find_start_value_byte = BEGV_BYTE;
      goto found;
    }

617
  /* Back up to start of line.  */
618
  scan_newline (pos, pos_byte, BEGV, BEGV_BYTE, -1, 1);
619

620 621 622
  /* We optimize syntax-table lookup for rare updates.  Thus we accept
     only those `^\s(' which are good in global _and_ text-property
     syntax-tables.  */
623
  SETUP_BUFFER_SYNTAX_TABLE ();
624
  while (PT > BEGV)
625
    {
626 627
      int c;

628 629
      /* Open-paren at start of line means we may have found our
	 defun-start.  */
630
      c = FETCH_CHAR_AS_MULTIBYTE (PT_BYTE);
631
      if (SYNTAX (c) == Sopen)
632
	{
633
	  SETUP_SYNTAX_TABLE (PT + 1, -1);	/* Try again... */
634
	  c = FETCH_CHAR_AS_MULTIBYTE (PT_BYTE);
635
	  if (SYNTAX (c) == Sopen)
636 637
	    break;
	  /* Now fallback to the default value.  */
638
	  SETUP_BUFFER_SYNTAX_TABLE ();
639
	}
640 641
      /* Move to beg of previous line.  */
      scan_newline (PT, PT_BYTE, BEGV, BEGV_BYTE, -2, 1);
642 643 644
    }

  /* Record what we found, for the next try.  */
645 646
  find_start_value = PT;
  find_start_value_byte = PT_BYTE;
647 648 649
  TEMP_SET_PT_BOTH (opoint, opoint_byte);

 found:
650 651 652 653 654 655 656 657
  find_start_buffer = current_buffer;
  find_start_modiff = MODIFF;
  find_start_begv = BEGV;
  find_start_pos = pos;

  return find_start_value;
}

658 659
/* Return the SYNTAX_COMEND_FIRST of the character before POS, POS_BYTE.  */

Paul Eggert's avatar
Paul Eggert committed
660
static bool
661
prev_char_comend_first (ptrdiff_t pos, ptrdiff_t pos_byte)
662
{
Paul Eggert's avatar
Paul Eggert committed
663 664
  int c;
  bool val;
665 666 667 668 669 670 671 672 673

  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
674
/* Check whether charpos FROM is at the end of a comment.
675 676 677
   FROM_BYTE is the bytepos corresponding to FROM.
   Do not move back before STOP.

Paul Eggert's avatar
Paul Eggert committed
678
   Return true if we find a comment ending at FROM/FROM_BYTE.
679 680 681

   If successful, store the charpos of the comment's beginning
   into *CHARPOS_PTR, and the bytepos into *BYTEPOS_PTR.
682 683 684

   Global syntax data remains valid for backward search starting at
   the returned value (or at FROM, if the search was not successful).  */
685

Paul Eggert's avatar
Paul Eggert committed
686 687 688 689
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)
690 691 692 693 694 695 696 697 698 699
{
  /* 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.  */
700
  int string_style = -1;	/* Presumed outside of any string.  */
Paul Eggert's avatar
Paul Eggert committed
701
  bool string_lossage = 0;
702
  /* Not a real lossage: indicates that we have passed a matching comment
Juanma Barranquero's avatar
Juanma Barranquero committed
703
     starter plus a non-matching comment-ender, meaning that any matching
704 705 706
     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
707
  bool comment_lossage = 0;
708 709 710 711
  ptrdiff_t comment_end = from;
  ptrdiff_t comment_end_byte = from_byte;
  ptrdiff_t comstart_pos = 0;
  ptrdiff_t comstart_byte IF_LINT (= 0);
712 713
  /* Place where the containing defun starts,
     or 0 if we didn't come across it yet.  */
714 715
  ptrdiff_t defun_start = 0;
  ptrdiff_t defun_start_byte = 0;
Paul Eggert's avatar
Paul Eggert committed
716
  enum syntaxcode code;
717
  ptrdiff_t nesting = 1;		/* Current comment nesting.  */
718
  int c;
719 720 721 722 723
  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.  */
724 725 726 727 728

  /* At beginning of range to scan, we're outside of strings;
     that determines quote parity to the comment-end.  */
  while (from != stop)
    {
729
      ptrdiff_t temp_byte;
Paul Eggert's avatar
Paul Eggert committed
730 731
      int prev_syntax;
      bool com2start, com2end, comstart;
732

733
      /* Move back and examine a character.  */
734
      DEC_BOTH (from, from_byte);
735 736
      UPDATE_SYNTAX_TABLE_BACKWARD (from);

737
      prev_syntax = syntax;
738
      c = FETCH_CHAR_AS_MULTIBYTE (from_byte);
739
      syntax = SYNTAX_WITH_FLAGS (c);
740 741
      code = SYNTAX (c);

742 743 744
      /* Check for 2-char comment markers.  */
      com2start = (SYNTAX_FLAGS_COMSTART_FIRST (syntax)
		   && SYNTAX_FLAGS_COMSTART_SECOND (prev_syntax)
745 746
		   && (comstyle
		       == SYNTAX_FLAGS_COMMENT_STYLE (prev_syntax, syntax))
747 748 749 750
		   && (SYNTAX_FLAGS_COMMENT_NESTED (prev_syntax)
		       || SYNTAX_FLAGS_COMMENT_NESTED (syntax)) == comnested);
      com2end = (SYNTAX_FLAGS_COMEND_FIRST (syntax)
		 && SYNTAX_FLAGS_COMEND_SECOND (prev_syntax));
751
      comstart = (com2start || code == Scomment);
752

753 754 755 756 757 758 759 760 761 762
      /* 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,
763 764 765
	 we don't try to be clever.  E.g. |*| in C, or }% in modes that
	 have %..\n and %{..}%.  */
      if (from > stop && (com2end || comstart))
766
	{
767
	  ptrdiff_t next = from, next_byte = from_byte;
768
	  int next_c, next_syntax;
769 770
	  DEC_BOTH (next, next_byte);
	  UPDATE_SYNTAX_TABLE_BACKWARD (next);
771
	  next_c = FETCH_CHAR_AS_MULTIBYTE (next_byte);
772
	  next_syntax = SYNTAX_WITH_FLAGS (next_c);
773
	  if (((comstart || comnested)
774 775 776 777
	       && SYNTAX_FLAGS_COMEND_SECOND (syntax)
	       && SYNTAX_FLAGS_COMEND_FIRST (next_syntax))
	      || ((com2end || comnested)
		  && SYNTAX_FLAGS_COMSTART_SECOND (syntax)
778 779
		  && (comstyle
		      == SYNTAX_FLAGS_COMMENT_STYLE (syntax, prev_syntax))
780 781 782
		  && SYNTAX_FLAGS_COMSTART_FIRST (next_syntax)))
	    goto lossage;
	  /* UPDATE_SYNTAX_TABLE_FORWARD (next + 1); */
783
	}
784 785 786 787 788 789 790 791 792 793 794 795 796 797 798

      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
799
	       && (comstyle != SYNTAX_FLAGS_COMMENT_STYLE (syntax, 0)
800
		   || SYNTAX_FLAGS_COMMENT_NESTED (syntax) != comnested))
801
	continue;
802

803 804 805 806
      /* Ignore escaped characters, except comment-enders which cannot
         be escaped.  */
      if ((Vcomment_end_can_be_escaped || code != Sendcomment)
          && char_quoted (from, from_byte))
807 808
	continue;

809
      switch (code)
810
	{
811 812 813 814 815 816 817 818 819 820 821 822 823 824
	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.  */
825
	    string_lossage = 1;
826
	  break;
Juanma Barranquero's avatar
Juanma Barranquero committed
827

828 829
	case Scomment:
	  /* We've already checked that it is the relevant comstyle.  */
830
	  if (string_style != -1 || comment_lossage || string_lossage)
831 832 833 834
	    /* There are odd string quotes involved, so let's be careful.
	       Test case in Pascal: " { " a { " } */
	    goto lossage;

835 836 837 838 839 840 841
	  if (!comnested)
	    {
	      /* Record best comment-starter so far.  */
	      comstart_pos = from;
	      comstart_byte = from_byte;
	    }
	  else if (--nesting <= 0)
842 843 844 845 846 847
	    /* 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;
848 849
	  break;

850
	case Sendcomment:
851
	  if (SYNTAX_FLAGS_COMMENT_STYLE (syntax, 0) == comstyle
852
	      && ((com2end && SYNTAX_FLAGS_COMMENT_NESTED (prev_syntax))
853
		  || SYNTAX_FLAGS_COMMENT_NESTED (syntax)) == comnested)
854 855 856 857 858 859 860 861 862
	    /* 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.  */
	    }
863 864 865 866 867 868 869 870 871
	  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;
872
	  break;
873

874 875 876 877 878 879 880 881 882 883 884
	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.  */
	    }
885
	  break;
886 887

	default:
888
	  break;
889
	}
890 891 892 893 894
    }

  if (comstart_pos == 0)
    {
      from = comment_end;
895
      from_byte = comment_end_byte;
896
      UPDATE_SYNTAX_TABLE_FORWARD (comment_end);
897
    }
898 899 900
  /* 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 */
901 902
    {
      from = comstart_pos;
903
      from_byte = comstart_byte;
904
      UPDATE_SYNTAX_TABLE_FORWARD (from - 1);
905
    }
906
  else lossage:
907
    {
908
      struct lisp_parse_state state;
909
      bool adjusted = true;
910 911
      /* We had two kinds of string delimiters mixed up
	 together.  Decode this going forwards.
912
	 Scan fwd from a known safe place (beginning-of-defun)
913 914
	 to the one in question; this records where we
	 last passed a comment starter.  */
915 916 917 918 919
      /* 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;
920
	  adjusted = (defun_start > BEGV);
921
	}
922
      do
923
	{
924
          internalize_parse_state (Qnil, &state);
925 926
	  scan_sexps_forward (&state,
			      defun_start, defun_start_byte,
927
			      comment_end, TYPE_MINIMUM (EMACS_INT),
928
			      0, 0);
929
	  defun_start = comment_end;
930 931 932 933 934 935 936 937 938 939
	  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);
	    }

940 941 942 943 944 945 946 947 948 949 950
	  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 (* " *)  */
		{
951
		  /* FIXME: We should advance by one or two chars.  */
952 953 954 955 956 957
		  defun_start = state.comstr_start + 2;
		  defun_start_byte = CHAR_TO_BYTE (defun_start);
		}
	    }
	} while (defun_start < comment_end);

958
      from_byte = CHAR_TO_BYTE (from);
959 960
      UPDATE_SYNTAX_TABLE_FORWARD (from - 1);
    }
Juanma Barranquero's avatar
Juanma Barranquero committed
961

962
 done:
963 964 965
  *charpos_ptr = from;
  *bytepos_ptr = from_byte;

Paul Eggert's avatar
Paul Eggert committed
966
  return from != comment_end;
967 968
}

Richard M. Stallman's avatar
Richard M. Stallman committed
969
DEFUN ("syntax-table-p", Fsyntax_table_p, Ssyntax_table_p, 1, 1, 0,
970 971
       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
972
  (Lisp_Object object)
Richard M. Stallman's avatar
Richard M. Stallman committed
973
{
974
  if (CHAR_TABLE_P (object)
975
      && EQ (XCHAR_TABLE (object)->purpose, Qsyntax_table))
Richard M. Stallman's avatar
Richard M. Stallman committed
976 977 978 979
    return Qt;
  return Qnil;
}

980
static void
981
check_syntax_table (Lisp_Object obj)
Richard M. Stallman's avatar
Richard M. Stallman committed
982
{
983 984
  CHECK_TYPE (CHAR_TABLE_P (obj) && EQ (XCHAR_TABLE (obj)->purpose, Qsyntax_table),
	      Qsyntax_table_p, obj);
Juanma Barranquero's avatar
Juanma Barranquero committed
985
}
Richard M. Stallman's avatar
Richard M. Stallman committed
986 987

DEFUN ("syntax-table", Fsyntax_table, Ssyntax_table, 0, 0, 0,
988 989
       doc: /* Return the current syntax table.
This is the one specified by the current buffer.  */)
Dan Nicolaescu's avatar
Dan Nicolaescu committed
990
  (void)
Richard M. Stallman's avatar
Richard M. Stallman committed
991
{
Tom Tromey's avatar
Tom Tromey committed
992
  return BVAR (current_buffer, syntax_table);
Richard M. Stallman's avatar
Richard M. Stallman committed
993 994 995 996
}

DEFUN ("standard-syntax-table", Fstandard_syntax_table,
   Sstandard_syntax_table, 0, 0, 0,
997 998
       doc: /* Return the standard syntax table.
This is the one used for new buffers.  */)
Dan Nicolaescu's avatar
Dan Nicolaescu committed
999
  (void)
Richard M. Stallman's avatar
Richard M. Stallman committed
1000 1001 1002 1003 1004
{
  return Vstandard_syntax_table;
}

DEFUN ("copy-syntax-table", Fcopy_syntax_table, Scopy_syntax_table, 0, 1, 0,
1005 1006
       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
1007
  (Lisp_Object table)
Richard M. Stallman's avatar
Richard M. Stallman committed
1008
{
1009 1010
  Lisp_Object copy;

Jim Blandy's avatar
Jim Blandy committed
1011
  if (!NILP (table))
1012 1013 1014 1015 1016
    check_syntax_table (table);
  else
    table = Vstandard_syntax_table;

  copy = Fcopy_sequence (table);
1017 1018 1019

  /* 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
1020
  set_char_table_defalt (copy, Qnil);
1021 1022 1023 1024 1025 1026

  /* 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);
1027
  return copy;
Richard M. Stallman's avatar
Richard M. Stallman committed
1028 1029 1030
}

DEFUN ("set-syntax-table", Fset_syntax_table, Sset_syntax_table, 1, 1, 0,
1031 1032
       doc: /* Select a new syntax table for the current buffer.
One argument, a syntax table.  */)
Dan Nicolaescu's avatar
Dan Nicolaescu committed
1033
  (Lisp_Object table)
Richard M. Stallman's avatar
Richard M. Stallman committed
1034
{
1035
  int idx;
1036
  check_syntax_table (table);
Paul Eggert's avatar
Paul Eggert committed
1037
  bset_syntax_table (current_buffer, table);
Richard M. Stallman's avatar
Richard M. Stallman committed
1038
  /* Indicate that this buffer now has a specified syntax table.  */
Gerd Moellmann's avatar
Gerd Moellmann committed
1039 1040
  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
1041 1042 1043 1044 1045
  return table;
}

/* Convert a letter which signifies a syntax code
 into the code it signifies.
1046
 This is used by modify-syntax-entry, and other things.  */
Richard M. Stallman's avatar
Richard M. Stallman committed
1047

Paul Eggert's avatar
Paul Eggert committed
1048
unsigned char const syntax_spec_code[0400] =
Richard M. Stallman's avatar
Richard M. Stallman committed
1049 1050 1051 1052
  { 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
1053 1054
    Swhitespace, Scomment_fence, Sstring, 0377, Smath, 0377, 0377, Squote,
    Sopen, Sclose, 0377, 0377, 0377, Swhitespace, Spunct, Scharquote,