syntax.c 106 KB
Newer Older
Richard M. Stallman's avatar
Richard M. Stallman committed
1
/* GNU Emacs routines to deal with syntax tables; also word and list parsing.
Paul Eggert's avatar
Paul Eggert committed
2
   Copyright (C) 1985, 1987, 1993-1995, 1997-1999, 2001-2015 Free
3
   Software Foundation, Inc.
Richard M. Stallman's avatar
Richard M. Stallman committed
4 5 6

This file is part of GNU Emacs.

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

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

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


21
#include <config.h>
22 23

#include <sys/types.h>
24

Richard M. Stallman's avatar
Richard M. Stallman committed
25
#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 84 85 86 87 88 89 90 91 92 93 94 95 96 97 98 99 100 101 102 103 104 105 106 107
static bool
SYNTAX_FLAGS_COMSTART_FIRST (int flags)
{
  return (flags >> 16) & 1;
}
static bool
SYNTAX_FLAGS_COMSTART_SECOND (int flags)
{
  return (flags >> 17) & 1;
}
static bool
SYNTAX_FLAGS_COMEND_FIRST (int flags)
{
  return (flags >> 18) & 1;
}
static bool
SYNTAX_FLAGS_COMEND_SECOND (int flags)
{
  return (flags >> 19) & 1;
}
static bool
SYNTAX_FLAGS_PREFIX (int flags)
{
  return (flags >> 20) & 1;
}
static bool
SYNTAX_FLAGS_COMMENT_STYLEB (int flags)
{
  return (flags >> 21) & 1;
}
static bool
SYNTAX_FLAGS_COMMENT_STYLEC (int flags)
{
  return (flags >> 23) & 1;
}
static int
SYNTAX_FLAGS_COMMENT_STYLEC2 (int flags)
{
  return (flags >> 22) & 2; /* SYNTAX_FLAGS_COMMENT_STYLEC (flags) * 2 */
}
static bool
SYNTAX_FLAGS_COMMENT_NESTED (int flags)
{
  return (flags >> 22) & 1;
}
108 109 110

/* FLAGS should be the flags of the main char of the comment marker, e.g.
   the second for comstart and the first for comend.  */
111 112 113 114 115 116 117
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));
}
118

119
/* Extract a particular flag for a given character.  */
120

121 122 123 124 125
static bool
SYNTAX_COMEND_FIRST (int c)
{
  return SYNTAX_FLAGS_COMEND_FIRST (SYNTAX_WITH_FLAGS (c));
}
126 127

/* We use these constants in place for comment-style and
128
   string-ender-char to distinguish comments/strings started by
129 130
   comment_fence and string_fence codes.  */

131 132 133 134 135
enum
  {
    ST_COMMENT_STYLE = 256 + 1,
    ST_STRING_STYLE = 256 + 2
  };
Richard M. Stallman's avatar
Richard M. Stallman committed
136

137 138 139 140
/* This is the internal form of the parse state used in parse-partial-sexp.  */

struct lisp_parse_state
  {
141
    EMACS_INT depth;	/* Depth at end of parsing.  */
142
    int instring;  /* -1 if not within string, else desired terminator.  */
143
    EMACS_INT incomment; /* -1 if in unnestable comment else comment nesting */
144
    int comstyle;  /* comment style a=0, or b=1, or ST_COMMENT_STYLE.  */
Paul Eggert's avatar
Paul Eggert committed
145
    bool quoted;   /* True if just after an escape char at end of parsing.  */
146
    EMACS_INT mindepth;	/* Minimum depth seen while scanning.  */
147
    /* Char number of most recent start-of-expression at current level */
148
    ptrdiff_t thislevelstart;
149
    /* Char number of start of containing expression */
150 151
    ptrdiff_t prevlevelstart;
    ptrdiff_t location;	     /* Char number at which parsing stopped.  */
152
    ptrdiff_t location_byte; /* Corresponding byte position.  */
153
    ptrdiff_t comstr_start;  /* Position of last comment/string starter.  */
154 155
    Lisp_Object levelstarts; /* Char numbers of starts-of-expression
				of levels (starting from outermost).  */
156 157
  };

158 159 160
/* 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.
161
   find_start_value_byte is the corresponding byte position.
162 163 164 165
   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.  */

166 167 168
static ptrdiff_t find_start_pos;
static ptrdiff_t find_start_value;
static ptrdiff_t find_start_value_byte;
169
static struct buffer *find_start_buffer;
170 171
static ptrdiff_t find_start_begv;
static EMACS_INT find_start_modiff;
172 173


Paul Eggert's avatar
Paul Eggert committed
174 175 176
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);
177
static void scan_sexps_forward (struct lisp_parse_state *,
178
                                ptrdiff_t, ptrdiff_t, ptrdiff_t, EMACS_INT,
Paul Eggert's avatar
Paul Eggert committed
179 180
                                bool, Lisp_Object, int);
static bool in_classes (int, Lisp_Object);
181
static void parse_sexp_propertize (ptrdiff_t charpos);
Paul Eggert's avatar
Paul Eggert committed
182 183

/* This setter is used only in this file, so it can be private.  */
184
static void
Paul Eggert's avatar
Paul Eggert committed
185 186
bset_syntax_table (struct buffer *b, Lisp_Object val)
{
187
  b->syntax_table_ = val;
Paul Eggert's avatar
Paul Eggert committed
188
}
189

190
/* Whether the syntax of the character C has the prefix flag set.  */
Paul Eggert's avatar
Paul Eggert committed
191 192
bool
syntax_prefix_flag_p (int c)
193
{
194
  return SYNTAX_FLAGS_PREFIX (SYNTAX_WITH_FLAGS (c));
195
}
196 197 198

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

199
enum { INTERVALS_AT_ONCE = 10 };	/* 1 + max-number of intervals
200 201
					   to scan to property-change.  */

202 203 204 205 206 207 208 209 210 211 212 213 214 215 216 217 218 219 220 221 222 223 224 225 226 227 228 229 230 231 232 233 234 235 236 237 238 239 240 241 242 243 244 245 246
/* 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)
247
    {
248
      if (count > 0)
249
	update_syntax_table_forward (from, true, true, Qnil);
250
      else if (from > BEGV)
251
	{
252 253
	  update_syntax_table (from - 1, count, true, Qnil);
	  parse_sexp_propertize (from - 1);
254 255
	}
    }
256 257 258 259 260 261 262 263 264 265 266 267 268 269 270 271 272 273 274 275 276 277 278 279 280 281 282 283 284 285 286 287 288 289 290 291 292 293 294 295 296 297 298 299 300 301
}

/* 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);
}

302 303
/* Update gl_state to an appropriate interval which contains CHARPOS.  The
   sign of COUNT give the relative position of CHARPOS wrt the previously
304
   valid interval.  If INIT, only [be]_property fields of gl_state are
305
   valid at start, the rest is filled basing on OBJECT.
306

307
   `gl_state.*_i' are the intervals, and CHARPOS is further in the search
308 309
   direction than the intervals - or in an interval.  We update the
   current syntax-table basing on the property of this interval, and
310
   update the interval to start further than CHARPOS - or be
Dmitry Antipov's avatar
Dmitry Antipov committed
311
   NULL.  We also update lim_property to be the next value of
312
   charpos to call this subroutine again - or be before/after the
313 314 315
   start/end of OBJECT.  */

void
Paul Eggert's avatar
Paul Eggert committed
316
update_syntax_table (ptrdiff_t charpos, EMACS_INT count, bool init,
317
		     Lisp_Object object)
318 319
{
  Lisp_Object tmp_table;
Paul Eggert's avatar
Paul Eggert committed
320
  int cnt = 0;
321
  bool invalidate = true;
322
  INTERVAL i;
323 324 325

  if (init)
    {
326
      gl_state.old_prop = Qnil;
327 328
      gl_state.start = gl_state.b_property;
      gl_state.stop = gl_state.e_property;
329 330
      i = interval_of (charpos, object);
      gl_state.backward_i = gl_state.forward_i = i;
331
      invalidate = false;
Dmitry Antipov's avatar
Dmitry Antipov committed
332
      if (!i)
333
	return;
334
      /* interval_of updates only ->position of the return value, so
335
	 update the parents manually to speed up update_interval.  */
336
      while (!NULL_PARENT (i))
337 338
	{
	  if (AM_RIGHT_CHILD (i))
339
	    INTERVAL_PARENT (i)->position = i->position
340
	      - LEFT_TOTAL_LENGTH (i) + TOTAL_LENGTH (i) /* right end */
341 342
	      - TOTAL_LENGTH (INTERVAL_PARENT (i))
	      + LEFT_TOTAL_LENGTH (INTERVAL_PARENT (i));
343
	  else
344
	    INTERVAL_PARENT (i)->position = i->position - LEFT_TOTAL_LENGTH (i)
345
	      + TOTAL_LENGTH (i);
346
	  i = INTERVAL_PARENT (i);
347 348
	}
      i = gl_state.forward_i;
349
      gl_state.b_property = i->position - gl_state.offset;
350
      gl_state.e_property = INTERVAL_LAST_POS (i) - gl_state.offset;
351 352
      goto update;
    }
353
  i = count > 0 ? gl_state.forward_i : gl_state.backward_i;
354

355
  /* We are guaranteed to be called with CHARPOS either in i,
356
     or further off.  */
Dmitry Antipov's avatar
Dmitry Antipov committed
357
  if (!i)
358
    error ("Error in syntax_table logic for to-the-end intervals");
359
  else if (charpos < i->position)		/* Move left.  */
360 361
    {
      if (count > 0)
362
	error ("Error in syntax_table logic for intervals <-");
363
      /* Update the interval.  */
364
      i = update_interval (i, charpos);
365
      if (INTERVAL_LAST_POS (i) != gl_state.b_property)
366
	{
367
	  invalidate = false;
368
	  gl_state.forward_i = i;
369
	  gl_state.e_property = INTERVAL_LAST_POS (i) - gl_state.offset;
370
	}
371
    }
372
  else if (charpos >= INTERVAL_LAST_POS (i)) /* Move right.  */
373 374
    {
      if (count < 0)
375
	error ("Error in syntax_table logic for intervals ->");
376
      /* Update the interval.  */
377
      i = update_interval (i, charpos);
378
      if (i->position != gl_state.e_property)
379
	{
380
	  invalidate = false;
381
	  gl_state.backward_i = i;
382
	  gl_state.b_property = i->position - gl_state.offset;
383 384 385 386 387 388 389 390
	}
    }

  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
391

392 393 394
  if (invalidate)		/* Did not get to adjacent interval.  */
    {				/* with the same table => */
				/* invalidate the old range.  */
395 396 397
      if (count > 0)
	{
	  gl_state.backward_i = i;
398 399 400
	  gl_state.b_property = i->position - gl_state.offset;
	}
      else
401
	{
402
	  gl_state.forward_i = i;
403
	  gl_state.e_property = INTERVAL_LAST_POS (i) - gl_state.offset;
404 405
	}
    }
406

407
  if (!EQ (tmp_table, gl_state.old_prop))
408
    {
409 410 411 412 413
      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
414
	}
415 416 417 418 419
      else if (CONSP (tmp_table))
	{
	  gl_state.use_global = 1;
	  gl_state.global_code = tmp_table;
	}
Juanma Barranquero's avatar
Juanma Barranquero committed
420
      else
421 422
	{
	  gl_state.use_global = 0;
Tom Tromey's avatar
Tom Tromey committed
423
	  gl_state.current_syntax_table = BVAR (current_buffer, syntax_table);
424
	}
425 426
    }

Dmitry Antipov's avatar
Dmitry Antipov committed
427
  while (i)
428 429 430 431
    {
      if (cnt && !EQ (tmp_table, textget (i->plist, Qsyntax_table)))
	{
	  if (count > 0)
432 433 434 435 436 437
	    {
	      gl_state.e_property = i->position - gl_state.offset;
	      gl_state.forward_i = i;
	    }
	  else
	    {
438 439
	      gl_state.b_property
		= i->position + LENGTH (i) - gl_state.offset;
440 441 442
	      gl_state.backward_i = i;
	    }
	  return;
443
	}
Juanma Barranquero's avatar
Juanma Barranquero committed
444
      else if (cnt == INTERVALS_AT_ONCE)
445 446
	{
	  if (count > 0)
447
	    {
448 449 450 451 452
	      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
453
		+ (next_interval (i) ? 0 : 1);
454 455 456 457 458 459 460 461
	      gl_state.forward_i = i;
	    }
	  else
	    {
	      gl_state.b_property = i->position - gl_state.offset;
	      gl_state.backward_i = i;
	    }
	  return;
462 463 464 465
	}
      cnt++;
      i = count > 0 ? next_interval (i) : previous_interval (i);
    }
Dmitry Antipov's avatar
Dmitry Antipov committed
466
  eassert (i == NULL); /* This property goes to the end.  */
467
  if (count > 0)
468 469 470 471
    {
      gl_state.e_property = gl_state.stop;
      gl_state.forward_i = i;
    }
472 473
  else
    gl_state.b_property = gl_state.start;
474
}
475 476 477 478

static void
parse_sexp_propertize (ptrdiff_t charpos)
{
479 480 481 482 483 484 485 486 487 488 489 490 491 492 493 494 495 496 497 498
  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
499 500
  else if (gl_state.e_property_truncated
	   && gl_state.e_property < syntax_propertize__done)
501 502 503 504
    { /* 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;
505
      update_syntax_table_forward (charpos, false, true, Qnil);
506
    }
507 508 509
}

void
510
update_syntax_table_forward (ptrdiff_t charpos, bool init, bool propertize,
511
			     Lisp_Object object)
512
{
513
  if (gl_state.e_property_truncated)
514
    {
515 516 517 518 519 520
      eassert (NILP (object));
      eassert (charpos >= gl_state.e_property);
    }
  else
    {
      update_syntax_table (charpos, 1, init, object);
521 522
      propertize &= (NILP (object)
		     && gl_state.e_property > syntax_propertize__done);
523
    }
524 525 526

  if (propertize)
    parse_sexp_propertize (charpos);
527
}
528

Paul Eggert's avatar
Paul Eggert committed
529
/* Returns true if char at CHARPOS is quoted.
530
   Global syntax-table data should be set up already to be good at CHARPOS
531
   or after.  On return global syntax data is good for lookup at CHARPOS.  */
532

Paul Eggert's avatar
Paul Eggert committed
533
static bool
534
char_quoted (ptrdiff_t charpos, ptrdiff_t bytepos)
535
{
Paul Eggert's avatar
Paul Eggert committed
536 537 538
  enum syntaxcode code;
  ptrdiff_t beg = BEGV;
  bool quoted = 0;
539
  ptrdiff_t orig = charpos;
540

541
  while (charpos > beg)
542
    {
543
      int c;
544
      DEC_BOTH (charpos, bytepos);
545

546
      UPDATE_SYNTAX_TABLE_BACKWARD (charpos);
547
      c = FETCH_CHAR_AS_MULTIBYTE (bytepos);
548
      code = SYNTAX (c);
549 550 551 552
      if (! (code == Scharquote || code == Sescape))
	break;

      quoted = !quoted;
553
    }
554 555

  UPDATE_SYNTAX_TABLE (orig);
556 557
  return quoted;
}
558 559 560 561

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

562
static ptrdiff_t
563
dec_bytepos (ptrdiff_t bytepos)
564
{
Tom Tromey's avatar
Tom Tromey committed
565
  if (NILP (BVAR (current_buffer, enable_multibyte_characters)))
566 567
    return bytepos - 1;

568 569 570
  DEC_POS (bytepos);
  return bytepos;
}
571

Juanma Barranquero's avatar
Juanma Barranquero committed
572
/* Return a defun-start position before POS and not too far before.
Richard M. Stallman's avatar
Richard M. Stallman committed
573 574 575
   It should be the last one before POS, or nearly the last.

   When open_paren_in_column_0_is_defun_start is nonzero,
576
   only the beginning of the buffer is treated as a defun-start.
Richard M. Stallman's avatar
Richard M. Stallman committed
577 578 579 580

   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.
581 582 583 584

   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.  */
585

586 587
static ptrdiff_t
find_defun_start (ptrdiff_t pos, ptrdiff_t pos_byte)
588
{
589
  ptrdiff_t opoint = PT, opoint_byte = PT_BYTE;
590 591 592 593 594 595 596 597 598 599 600 601

  /* 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;

602 603 604 605 606 607 608
  if (!open_paren_in_column_0_is_defun_start)
    {
      find_start_value = BEGV;
      find_start_value_byte = BEGV_BYTE;
      goto found;
    }

609
  /* Back up to start of line.  */
610
  scan_newline (pos, pos_byte, BEGV, BEGV_BYTE, -1, 1);
611

612 613 614
  /* We optimize syntax-table lookup for rare updates.  Thus we accept
     only those `^\s(' which are good in global _and_ text-property
     syntax-tables.  */
615
  SETUP_BUFFER_SYNTAX_TABLE ();
616
  while (PT > BEGV)
617
    {
618 619
      int c;

620 621
      /* Open-paren at start of line means we may have found our
	 defun-start.  */
622
      c = FETCH_CHAR_AS_MULTIBYTE (PT_BYTE);
623
      if (SYNTAX (c) == Sopen)
624
	{
625
	  SETUP_SYNTAX_TABLE (PT + 1, -1);	/* Try again... */
626
	  c = FETCH_CHAR_AS_MULTIBYTE (PT_BYTE);
627
	  if (SYNTAX (c) == Sopen)
628 629
	    break;
	  /* Now fallback to the default value.  */
630
	  SETUP_BUFFER_SYNTAX_TABLE ();
631
	}
632 633
      /* Move to beg of previous line.  */
      scan_newline (PT, PT_BYTE, BEGV, BEGV_BYTE, -2, 1);
634 635 636
    }

  /* Record what we found, for the next try.  */
637 638
  find_start_value = PT;
  find_start_value_byte = PT_BYTE;
639 640 641
  TEMP_SET_PT_BOTH (opoint, opoint_byte);

 found:
642 643 644 645 646 647 648 649
  find_start_buffer = current_buffer;
  find_start_modiff = MODIFF;
  find_start_begv = BEGV;
  find_start_pos = pos;

  return find_start_value;
}

650 651
/* Return the SYNTAX_COMEND_FIRST of the character before POS, POS_BYTE.  */

Paul Eggert's avatar
Paul Eggert committed
652
static bool
653
prev_char_comend_first (ptrdiff_t pos, ptrdiff_t pos_byte)
654
{
Paul Eggert's avatar
Paul Eggert committed
655 656
  int c;
  bool val;
657 658 659 660 661 662 663 664 665

  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
666
/* Check whether charpos FROM is at the end of a comment.
667 668 669
   FROM_BYTE is the bytepos corresponding to FROM.
   Do not move back before STOP.

Paul Eggert's avatar
Paul Eggert committed
670
   Return true if we find a comment ending at FROM/FROM_BYTE.
671 672 673

   If successful, store the charpos of the comment's beginning
   into *CHARPOS_PTR, and the bytepos into *BYTEPOS_PTR.
674 675 676

   Global syntax data remains valid for backward search starting at
   the returned value (or at FROM, if the search was not successful).  */
677

Paul Eggert's avatar
Paul Eggert committed
678 679 680 681
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)
682 683 684 685 686 687 688 689 690 691
{
  /* 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.  */
692
  int string_style = -1;	/* Presumed outside of any string.  */
Paul Eggert's avatar
Paul Eggert committed
693
  bool string_lossage = 0;
694
  /* Not a real lossage: indicates that we have passed a matching comment
Juanma Barranquero's avatar
Juanma Barranquero committed
695
     starter plus a non-matching comment-ender, meaning that any matching
696 697 698
     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
699
  bool comment_lossage = 0;
700 701 702 703
  ptrdiff_t comment_end = from;
  ptrdiff_t comment_end_byte = from_byte;
  ptrdiff_t comstart_pos = 0;
  ptrdiff_t comstart_byte IF_LINT (= 0);
704 705
  /* Place where the containing defun starts,
     or 0 if we didn't come across it yet.  */
706 707
  ptrdiff_t defun_start = 0;
  ptrdiff_t defun_start_byte = 0;
Paul Eggert's avatar
Paul Eggert committed
708
  enum syntaxcode code;
709
  ptrdiff_t nesting = 1;		/* Current comment nesting.  */
710
  int c;
711 712 713 714 715
  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.  */
716 717 718 719 720

  /* At beginning of range to scan, we're outside of strings;
     that determines quote parity to the comment-end.  */
  while (from != stop)
    {
721
      ptrdiff_t temp_byte;
Paul Eggert's avatar
Paul Eggert committed
722 723
      int prev_syntax;
      bool com2start, com2end, comstart;
724

725
      /* Move back and examine a character.  */
726
      DEC_BOTH (from, from_byte);
727 728
      UPDATE_SYNTAX_TABLE_BACKWARD (from);

729
      prev_syntax = syntax;
730
      c = FETCH_CHAR_AS_MULTIBYTE (from_byte);
731
      syntax = SYNTAX_WITH_FLAGS (c);
732 733
      code = SYNTAX (c);

734 735 736
      /* Check for 2-char comment markers.  */
      com2start = (SYNTAX_FLAGS_COMSTART_FIRST (syntax)
		   && SYNTAX_FLAGS_COMSTART_SECOND (prev_syntax)
737 738
		   && (comstyle
		       == SYNTAX_FLAGS_COMMENT_STYLE (prev_syntax, syntax))
739 740 741 742
		   && (SYNTAX_FLAGS_COMMENT_NESTED (prev_syntax)
		       || SYNTAX_FLAGS_COMMENT_NESTED (syntax)) == comnested);
      com2end = (SYNTAX_FLAGS_COMEND_FIRST (syntax)
		 && SYNTAX_FLAGS_COMEND_SECOND (prev_syntax));
743
      comstart = (com2start || code == Scomment);
744

745 746 747 748 749 750 751 752 753 754
      /* 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,
755 756 757
	 we don't try to be clever.  E.g. |*| in C, or }% in modes that
	 have %..\n and %{..}%.  */
      if (from > stop && (com2end || comstart))
758
	{
759
	  ptrdiff_t next = from, next_byte = from_byte;
760
	  int next_c, next_syntax;
761 762
	  DEC_BOTH (next, next_byte);
	  UPDATE_SYNTAX_TABLE_BACKWARD (next);
763
	  next_c = FETCH_CHAR_AS_MULTIBYTE (next_byte);
764
	  next_syntax = SYNTAX_WITH_FLAGS (next_c);
765
	  if (((comstart || comnested)
766 767 768 769
	       && SYNTAX_FLAGS_COMEND_SECOND (syntax)
	       && SYNTAX_FLAGS_COMEND_FIRST (next_syntax))
	      || ((com2end || comnested)
		  && SYNTAX_FLAGS_COMSTART_SECOND (syntax)
770 771
		  && (comstyle
		      == SYNTAX_FLAGS_COMMENT_STYLE (syntax, prev_syntax))
772 773 774
		  && SYNTAX_FLAGS_COMSTART_FIRST (next_syntax)))
	    goto lossage;
	  /* UPDATE_SYNTAX_TABLE_FORWARD (next + 1); */
775
	}
776 777 778 779 780 781 782 783 784 785 786 787 788 789 790

      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
791
	       && (comstyle != SYNTAX_FLAGS_COMMENT_STYLE (syntax, 0)
792
		   || SYNTAX_FLAGS_COMMENT_NESTED (syntax) != comnested))
793
	continue;
794

795 796
      /* Ignore escaped characters, except comment-enders.  */
      if (code != Sendcomment && char_quoted (from, from_byte))
797 798
	continue;

799
      switch (code)
800
	{
801 802 803 804 805 806 807 808 809 810 811 812 813 814
	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.  */
815
	    string_lossage = 1;
816
	  break;
Juanma Barranquero's avatar
Juanma Barranquero committed
817

818 819
	case Scomment:
	  /* We've already checked that it is the relevant comstyle.  */
820
	  if (string_style != -1 || comment_lossage || string_lossage)
821 822 823 824
	    /* There are odd string quotes involved, so let's be careful.
	       Test case in Pascal: " { " a { " } */
	    goto lossage;

825 826 827 828 829 830 831
	  if (!comnested)
	    {
	      /* Record best comment-starter so far.  */
	      comstart_pos = from;
	      comstart_byte = from_byte;
	    }
	  else if (--nesting <= 0)
832 833 834 835 836 837
	    /* 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;
838 839
	  break;

840
	case Sendcomment:
841
	  if (SYNTAX_FLAGS_COMMENT_STYLE (syntax, 0) == comstyle
842
	      && ((com2end && SYNTAX_FLAGS_COMMENT_NESTED (prev_syntax))
843
		  || SYNTAX_FLAGS_COMMENT_NESTED (syntax)) == comnested)
844 845 846 847 848 849 850 851 852
	    /* 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.  */
	    }
853 854 855 856 857 858 859 860 861
	  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;
862
	  break;
863

864 865 866 867 868 869 870 871 872 873 874
	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.  */
	    }
875
	  break;
876 877

	default:
878
	  break;
879
	}
880 881 882 883 884
    }

  if (comstart_pos == 0)
    {
      from = comment_end;
885
      from_byte = comment_end_byte;
886
      UPDATE_SYNTAX_TABLE_FORWARD (comment_end);
887
    }
888 889 890
  /* 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 */
891 892
    {
      from = comstart_pos;
893
      from_byte = comstart_byte;
894
      UPDATE_SYNTAX_TABLE_FORWARD (from - 1);
895
    }
896
  else lossage:
897
    {
898
      struct lisp_parse_state state;
899
      bool adjusted = true;
900 901
      /* We had two kinds of string delimiters mixed up
	 together.  Decode this going forwards.
902
	 Scan fwd from a known safe place (beginning-of-defun)
903 904
	 to the one in question; this records where we
	 last passed a comment starter.  */
905 906 907 908 909
      /* 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;
910
	  adjusted = (defun_start > BEGV);
911
	}
912
      do
913
	{
914 915
	  scan_sexps_forward (&state,
			      defun_start, defun_start_byte,
916 917
			      comment_end, TYPE_MINIMUM (EMACS_INT),
			      0, Qnil, 0);
918
	  defun_start = comment_end;
919 920 921 922 923 924 925 926 927 928
	  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);
	    }

929 930 931 932 933 934 935 936 937 938 939
	  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 (* " *)  */
		{
940
		  /* FIXME: We should advance by one or two chars.  */
941 942 943 944 945 946
		  defun_start = state.comstr_start + 2;
		  defun_start_byte = CHAR_TO_BYTE (defun_start);
		}
	    }
	} while (defun_start < comment_end);

947
      from_byte = CHAR_TO_BYTE (from);
948 949
      UPDATE_SYNTAX_TABLE_FORWARD (from - 1);
    }
Juanma Barranquero's avatar
Juanma Barranquero committed
950

951
 done:
952 953 954
  *charpos_ptr = from;
  *bytepos_ptr = from_byte;

Paul Eggert's avatar
Paul Eggert committed
955
  return from != comment_end;
956 957
}

Richard M. Stallman's avatar
Richard M. Stallman committed
958
DEFUN ("syntax-table-p", Fsyntax_table_p, Ssyntax_table_p, 1, 1, 0,
959 960
       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
961
  (Lisp_Object object)
Richard M. Stallman's avatar
Richard M. Stallman committed
962
{
963
  if (CHAR_TABLE_P (object)
964
      && EQ (XCHAR_TABLE (object)->purpose, Qsyntax_table))
Richard M. Stallman's avatar
Richard M. Stallman committed
965 966 967 968
    return Qt;
  return Qnil;
}

969
static void
970
check_syntax_table (Lisp_Object obj)
Richard M. Stallman's avatar
Richard M. Stallman committed
971
{
972 973
  CHECK_TYPE (CHAR_TABLE_P (obj) && EQ (XCHAR_TABLE (obj)->purpose, Qsyntax_table),
	      Qsyntax_table_p, obj);
Juanma Barranquero's avatar
Juanma Barranquero committed
974
}
Richard M. Stallman's avatar
Richard M. Stallman committed
975 976

DEFUN ("syntax-table", Fsyntax_table, Ssyntax_table, 0, 0, 0,
977 978
       doc: /* Return the current syntax table.
This is the one specified by the current buffer.  */)
Dan Nicolaescu's avatar
Dan Nicolaescu committed
979
  (void)
Richard M. Stallman's avatar
Richard M. Stallman committed
980
{
Tom Tromey's avatar
Tom Tromey committed
981
  return BVAR (current_buffer,