syntax.c 109 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, 2001-2017 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

Richard M. Stallman's avatar
Richard M. Stallman committed
23
#include "lisp.h"
24
#include "character.h"
25
#include "buffer.h"
Richard M. Stallman's avatar
Richard M. Stallman committed
26
#include "regex.h"
Richard M. Stallman's avatar
Richard M. Stallman committed
27
#include "syntax.h"
28
#include "intervals.h"
29 30
#include "category.h"

31 32 33 34 35
/* 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
36
/* Eight single-bit flags have the following meanings:
37 38 39 40 41 42 43 44 45 46 47
  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
48
  Bits 6 and 8 discriminate among different comment styles.
49
  Languages such as C++ allow two orthogonal syntax start/end pairs
Paul Eggert's avatar
Paul Eggert committed
50
  and bit 6 determines whether a comment-end or Scommentend
51 52
  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
53
  For 2-char comment markers, the style b flag is looked up only on the second
54
  char of the comment marker and on the first char of the comment ender.
Paul Eggert's avatar
Paul Eggert committed
55 56
  For style c (like the nested flag), the flag can be placed on any of
  the chars.  */
57

58
/* These functions extract specific flags from an integer
59 60
   that holds the syntax code and the flags.  */

61 62 63 64 65 66 67 68 69 70 71 72 73 74 75 76 77 78 79 80 81
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
82 83 84 85 86
SYNTAX_FLAGS_COMSTARTEND_FIRST (int flags)
{
  return (flags & 0x50000) != 0;
}
static bool
87 88 89 90 91 92 93 94 95 96 97 98 99 100 101 102 103 104 105 106 107 108 109 110
SYNTAX_FLAGS_PREFIX (int flags)
{
  return (flags >> 20) & 1;
}
static bool
SYNTAX_FLAGS_COMMENT_STYLEB (int flags)
{
  return (flags >> 21) & 1;
}
static bool
SYNTAX_FLAGS_COMMENT_STYLEC (int flags)
{
  return (flags >> 23) & 1;
}
static int
SYNTAX_FLAGS_COMMENT_STYLEC2 (int flags)
{
  return (flags >> 22) & 2; /* SYNTAX_FLAGS_COMMENT_STYLEC (flags) * 2 */
}
static bool
SYNTAX_FLAGS_COMMENT_NESTED (int flags)
{
  return (flags >> 22) & 1;
}
111 112 113

/* FLAGS should be the flags of the main char of the comment marker, e.g.
   the second for comstart and the first for comend.  */
114 115 116 117 118 119 120
static int
SYNTAX_FLAGS_COMMENT_STYLE (int flags, int other_flags)
{
  return (SYNTAX_FLAGS_COMMENT_STYLEB (flags)
	  | SYNTAX_FLAGS_COMMENT_STYLEC2 (flags)
	  | SYNTAX_FLAGS_COMMENT_STYLEC2 (other_flags));
}
121

122
/* Extract a particular flag for a given character.  */
123

124 125 126 127 128
static bool
SYNTAX_COMEND_FIRST (int c)
{
  return SYNTAX_FLAGS_COMEND_FIRST (SYNTAX_WITH_FLAGS (c));
}
129 130

/* We use these constants in place for comment-style and
131
   string-ender-char to distinguish comments/strings started by
132 133
   comment_fence and string_fence codes.  */

134 135 136 137 138
enum
  {
    ST_COMMENT_STYLE = 256 + 1,
    ST_STRING_STYLE = 256 + 2
  };
Richard M. Stallman's avatar
Richard M. Stallman committed
139

140 141 142 143
/* This is the internal form of the parse state used in parse-partial-sexp.  */

struct lisp_parse_state
  {
144
    EMACS_INT depth;	/* Depth at end of parsing.  */
145
    int instring;  /* -1 if not within string, else desired terminator.  */
146
    EMACS_INT incomment; /* -1 if in unnestable comment else comment nesting */
147
    int comstyle;  /* comment style a=0, or b=1, or ST_COMMENT_STYLE.  */
Paul Eggert's avatar
Paul Eggert committed
148
    bool quoted;   /* True if just after an escape char at end of parsing.  */
149
    EMACS_INT mindepth;	/* Minimum depth seen while scanning.  */
150
    /* Char number of most recent start-of-expression at current level */
151
    ptrdiff_t thislevelstart;
152
    /* Char number of start of containing expression */
153 154
    ptrdiff_t prevlevelstart;
    ptrdiff_t location;	     /* Char number at which parsing stopped.  */
155
    ptrdiff_t location_byte; /* Corresponding byte position.  */
156
    ptrdiff_t comstr_start;  /* Position of last comment/string starter.  */
157 158
    Lisp_Object levelstarts; /* Char numbers of starts-of-expression
				of levels (starting from outermost).  */
159 160 161 162
    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. */
163 164
  };

165 166 167
/* 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.
168
   find_start_value_byte is the corresponding byte position.
169 170 171 172
   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.  */

173 174 175
static ptrdiff_t find_start_pos;
static ptrdiff_t find_start_value;
static ptrdiff_t find_start_value_byte;
176
static struct buffer *find_start_buffer;
177 178
static ptrdiff_t find_start_begv;
static EMACS_INT find_start_modiff;
179 180


Paul Eggert's avatar
Paul Eggert committed
181 182 183
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);
184
static void scan_sexps_forward (struct lisp_parse_state *,
185
                                ptrdiff_t, ptrdiff_t, ptrdiff_t, EMACS_INT,
186 187
                                bool, int);
static void internalize_parse_state (Lisp_Object, struct lisp_parse_state *);
Paul Eggert's avatar
Paul Eggert committed
188
static bool in_classes (int, Lisp_Object);
189
static void parse_sexp_propertize (ptrdiff_t charpos);
Paul Eggert's avatar
Paul Eggert committed
190 191

/* This setter is used only in this file, so it can be private.  */
192
static void
Paul Eggert's avatar
Paul Eggert committed
193 194
bset_syntax_table (struct buffer *b, Lisp_Object val)
{
195
  b->syntax_table_ = val;
Paul Eggert's avatar
Paul Eggert committed
196
}
197

198
/* Whether the syntax of the character C has the prefix flag set.  */
Paul Eggert's avatar
Paul Eggert committed
199 200
bool
syntax_prefix_flag_p (int c)
201
{
202
  return SYNTAX_FLAGS_PREFIX (SYNTAX_WITH_FLAGS (c));
203
}
204 205 206

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

207
enum { INTERVALS_AT_ONCE = 10 };	/* 1 + max-number of intervals
208 209
					   to scan to property-change.  */

210 211 212 213 214 215 216 217 218 219 220 221 222 223 224 225 226 227 228 229 230 231 232 233 234 235 236 237 238 239 240 241 242 243 244 245 246 247 248 249 250 251 252 253 254
/* 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)
255
    {
256
      if (count > 0)
257
	update_syntax_table_forward (from, true, Qnil);
258
      else if (from > BEGV)
259
	{
260 261
	  update_syntax_table (from - 1, count, true, Qnil);
	  parse_sexp_propertize (from - 1);
262 263
	}
    }
264 265 266 267 268 269 270 271 272 273 274 275 276 277 278 279 280 281 282 283 284 285 286 287 288 289 290 291 292 293 294 295 296 297 298 299 300 301 302 303 304 305 306 307 308 309
}

/* 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);
}

310 311
/* Update gl_state to an appropriate interval which contains CHARPOS.  The
   sign of COUNT give the relative position of CHARPOS wrt the previously
312
   valid interval.  If INIT, only [be]_property fields of gl_state are
313
   valid at start, the rest is filled basing on OBJECT.
314

315
   `gl_state.*_i' are the intervals, and CHARPOS is further in the search
316 317
   direction than the intervals - or in an interval.  We update the
   current syntax-table basing on the property of this interval, and
318
   update the interval to start further than CHARPOS - or be
Dmitry Antipov's avatar
Dmitry Antipov committed
319
   NULL.  We also update lim_property to be the next value of
320
   charpos to call this subroutine again - or be before/after the
321 322 323
   start/end of OBJECT.  */

void
Paul Eggert's avatar
Paul Eggert committed
324
update_syntax_table (ptrdiff_t charpos, EMACS_INT count, bool init,
325
		     Lisp_Object object)
326 327
{
  Lisp_Object tmp_table;
Paul Eggert's avatar
Paul Eggert committed
328
  int cnt = 0;
329
  bool invalidate = true;
330
  INTERVAL i;
331 332 333

  if (init)
    {
334
      gl_state.old_prop = Qnil;
335 336
      gl_state.start = gl_state.b_property;
      gl_state.stop = gl_state.e_property;
337 338
      i = interval_of (charpos, object);
      gl_state.backward_i = gl_state.forward_i = i;
339
      invalidate = false;
Dmitry Antipov's avatar
Dmitry Antipov committed
340
      if (!i)
341
	return;
342
      /* interval_of updates only ->position of the return value, so
343
	 update the parents manually to speed up update_interval.  */
344
      while (!NULL_PARENT (i))
345 346
	{
	  if (AM_RIGHT_CHILD (i))
347
	    INTERVAL_PARENT (i)->position = i->position
348
	      - LEFT_TOTAL_LENGTH (i) + TOTAL_LENGTH (i) /* right end */
349 350
	      - TOTAL_LENGTH (INTERVAL_PARENT (i))
	      + LEFT_TOTAL_LENGTH (INTERVAL_PARENT (i));
351
	  else
352
	    INTERVAL_PARENT (i)->position = i->position - LEFT_TOTAL_LENGTH (i)
353
	      + TOTAL_LENGTH (i);
354
	  i = INTERVAL_PARENT (i);
355 356
	}
      i = gl_state.forward_i;
357
      gl_state.b_property = i->position - gl_state.offset;
358
      gl_state.e_property = INTERVAL_LAST_POS (i) - gl_state.offset;
359 360
      goto update;
    }
361
  i = count > 0 ? gl_state.forward_i : gl_state.backward_i;
362

363
  /* We are guaranteed to be called with CHARPOS either in i,
364
     or further off.  */
Dmitry Antipov's avatar
Dmitry Antipov committed
365
  if (!i)
366
    error ("Error in syntax_table logic for to-the-end intervals");
367
  else if (charpos < i->position)		/* Move left.  */
368 369
    {
      if (count > 0)
370
	error ("Error in syntax_table logic for intervals <-");
371
      /* Update the interval.  */
372
      i = update_interval (i, charpos);
373
      if (INTERVAL_LAST_POS (i) != gl_state.b_property)
374
	{
375
	  invalidate = false;
376
	  gl_state.forward_i = i;
377
	  gl_state.e_property = INTERVAL_LAST_POS (i) - gl_state.offset;
378
	}
379
    }
380
  else if (charpos >= INTERVAL_LAST_POS (i)) /* Move right.  */
381 382
    {
      if (count < 0)
383
	error ("Error in syntax_table logic for intervals ->");
384
      /* Update the interval.  */
385
      i = update_interval (i, charpos);
386
      if (i->position != gl_state.e_property)
387
	{
388
	  invalidate = false;
389
	  gl_state.backward_i = i;
390
	  gl_state.b_property = i->position - gl_state.offset;
391 392 393 394 395 396 397 398
	}
    }

  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
399

400 401 402
  if (invalidate)		/* Did not get to adjacent interval.  */
    {				/* with the same table => */
				/* invalidate the old range.  */
403 404 405
      if (count > 0)
	{
	  gl_state.backward_i = i;
406 407 408
	  gl_state.b_property = i->position - gl_state.offset;
	}
      else
409
	{
410
	  gl_state.forward_i = i;
411
	  gl_state.e_property = INTERVAL_LAST_POS (i) - gl_state.offset;
412 413
	}
    }
414

415
  if (!EQ (tmp_table, gl_state.old_prop))
416
    {
417 418 419 420 421
      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
422
	}
423 424 425 426 427
      else if (CONSP (tmp_table))
	{
	  gl_state.use_global = 1;
	  gl_state.global_code = tmp_table;
	}
Juanma Barranquero's avatar
Juanma Barranquero committed
428
      else
429 430
	{
	  gl_state.use_global = 0;
Tom Tromey's avatar
Tom Tromey committed
431
	  gl_state.current_syntax_table = BVAR (current_buffer, syntax_table);
432
	}
433 434
    }

Dmitry Antipov's avatar
Dmitry Antipov committed
435
  while (i)
436 437 438 439
    {
      if (cnt && !EQ (tmp_table, textget (i->plist, Qsyntax_table)))
	{
	  if (count > 0)
440 441 442 443 444 445
	    {
	      gl_state.e_property = i->position - gl_state.offset;
	      gl_state.forward_i = i;
	    }
	  else
	    {
446 447
	      gl_state.b_property
		= i->position + LENGTH (i) - gl_state.offset;
448 449 450
	      gl_state.backward_i = i;
	    }
	  return;
451
	}
Juanma Barranquero's avatar
Juanma Barranquero committed
452
      else if (cnt == INTERVALS_AT_ONCE)
453 454
	{
	  if (count > 0)
455
	    {
456 457 458 459 460
	      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
461
		+ (next_interval (i) ? 0 : 1);
462 463 464 465 466 467 468 469
	      gl_state.forward_i = i;
	    }
	  else
	    {
	      gl_state.b_property = i->position - gl_state.offset;
	      gl_state.backward_i = i;
	    }
	  return;
470 471 472 473
	}
      cnt++;
      i = count > 0 ? next_interval (i) : previous_interval (i);
    }
Dmitry Antipov's avatar
Dmitry Antipov committed
474
  eassert (i == NULL); /* This property goes to the end.  */
475
  if (count > 0)
476 477 478 479
    {
      gl_state.e_property = gl_state.stop;
      gl_state.forward_i = i;
    }
480 481
  else
    gl_state.b_property = gl_state.start;
482
}
483 484 485 486

static void
parse_sexp_propertize (ptrdiff_t charpos)
{
487 488 489 490 491 492 493 494 495 496 497 498 499 500 501 502 503 504 505 506
  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
507 508
  else if (gl_state.e_property_truncated
	   && gl_state.e_property < syntax_propertize__done)
509 510 511 512
    { /* 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;
513
      update_syntax_table_forward (charpos, false, Qnil);
514
    }
515 516 517
}

void
518
update_syntax_table_forward (ptrdiff_t charpos, bool init,
519
			     Lisp_Object object)
520
{
521
  if (gl_state.e_property_truncated)
522
    {
523 524
      eassert (NILP (object));
      eassert (charpos >= gl_state.e_property);
525
      parse_sexp_propertize (charpos);
526 527 528 529
    }
  else
    {
      update_syntax_table (charpos, 1, init, object);
530 531
      if (NILP (object) && gl_state.e_property > syntax_propertize__done)
	parse_sexp_propertize (charpos);
532 533
    }
}
534

Paul Eggert's avatar
Paul Eggert committed
535
/* Returns true if char at CHARPOS is quoted.
536
   Global syntax-table data should be set up already to be good at CHARPOS
537
   or after.  On return global syntax data is good for lookup at CHARPOS.  */
538

Paul Eggert's avatar
Paul Eggert committed
539
static bool
540
char_quoted (ptrdiff_t charpos, ptrdiff_t bytepos)
541
{
Paul Eggert's avatar
Paul Eggert committed
542 543 544
  enum syntaxcode code;
  ptrdiff_t beg = BEGV;
  bool quoted = 0;
545
  ptrdiff_t orig = charpos;
546

547
  while (charpos > beg)
548
    {
549
      int c;
550
      DEC_BOTH (charpos, bytepos);
551

552
      UPDATE_SYNTAX_TABLE_BACKWARD (charpos);
553
      c = FETCH_CHAR_AS_MULTIBYTE (bytepos);
554
      code = SYNTAX (c);
555 556 557 558
      if (! (code == Scharquote || code == Sescape))
	break;

      quoted = !quoted;
559
    }
560 561

  UPDATE_SYNTAX_TABLE (orig);
562 563
  return quoted;
}
564 565 566 567

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

568
static ptrdiff_t
569
dec_bytepos (ptrdiff_t bytepos)
570
{
Tom Tromey's avatar
Tom Tromey committed
571
  if (NILP (BVAR (current_buffer, enable_multibyte_characters)))
572 573
    return bytepos - 1;

574 575 576
  DEC_POS (bytepos);
  return bytepos;
}
577

Juanma Barranquero's avatar
Juanma Barranquero committed
578
/* Return a defun-start position before POS and not too far before.
Richard M. Stallman's avatar
Richard M. Stallman committed
579 580 581
   It should be the last one before POS, or nearly the last.

   When open_paren_in_column_0_is_defun_start is nonzero,
582
   only the beginning of the buffer is treated as a defun-start.
Richard M. Stallman's avatar
Richard M. Stallman committed
583 584 585 586

   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.
587 588 589 590

   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.  */
591

592 593
static ptrdiff_t
find_defun_start (ptrdiff_t pos, ptrdiff_t pos_byte)
594
{
595
  ptrdiff_t opoint = PT, opoint_byte = PT_BYTE;
596 597 598 599 600 601 602 603 604 605 606 607

  /* 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;

608 609 610 611 612 613 614
  if (!open_paren_in_column_0_is_defun_start)
    {
      find_start_value = BEGV;
      find_start_value_byte = BEGV_BYTE;
      goto found;
    }

615
  /* Back up to start of line.  */
616
  scan_newline (pos, pos_byte, BEGV, BEGV_BYTE, -1, 1);
617

618 619 620
  /* We optimize syntax-table lookup for rare updates.  Thus we accept
     only those `^\s(' which are good in global _and_ text-property
     syntax-tables.  */
621
  SETUP_BUFFER_SYNTAX_TABLE ();
622
  while (PT > BEGV)
623
    {
624 625
      int c;

626 627
      /* Open-paren at start of line means we may have found our
	 defun-start.  */
628
      c = FETCH_CHAR_AS_MULTIBYTE (PT_BYTE);
629
      if (SYNTAX (c) == Sopen)
630
	{
631
	  SETUP_SYNTAX_TABLE (PT + 1, -1);	/* Try again... */
632
	  c = FETCH_CHAR_AS_MULTIBYTE (PT_BYTE);
633
	  if (SYNTAX (c) == Sopen)
634 635
	    break;
	  /* Now fallback to the default value.  */
636
	  SETUP_BUFFER_SYNTAX_TABLE ();
637
	}
638 639
      /* Move to beg of previous line.  */
      scan_newline (PT, PT_BYTE, BEGV, BEGV_BYTE, -2, 1);
640 641 642
    }

  /* Record what we found, for the next try.  */
643 644
  find_start_value = PT;
  find_start_value_byte = PT_BYTE;
645 646 647
  TEMP_SET_PT_BOTH (opoint, opoint_byte);

 found:
648 649 650 651 652 653 654 655
  find_start_buffer = current_buffer;
  find_start_modiff = MODIFF;
  find_start_begv = BEGV;
  find_start_pos = pos;

  return find_start_value;
}

656 657
/* Return the SYNTAX_COMEND_FIRST of the character before POS, POS_BYTE.  */

Paul Eggert's avatar
Paul Eggert committed
658
static bool
659
prev_char_comend_first (ptrdiff_t pos, ptrdiff_t pos_byte)
660
{
Paul Eggert's avatar
Paul Eggert committed
661 662
  int c;
  bool val;
663 664 665 666 667 668 669 670 671

  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
672
/* Check whether charpos FROM is at the end of a comment.
673 674 675
   FROM_BYTE is the bytepos corresponding to FROM.
   Do not move back before STOP.

Paul Eggert's avatar
Paul Eggert committed
676
   Return true if we find a comment ending at FROM/FROM_BYTE.
677 678 679

   If successful, store the charpos of the comment's beginning
   into *CHARPOS_PTR, and the bytepos into *BYTEPOS_PTR.
680 681 682

   Global syntax data remains valid for backward search starting at
   the returned value (or at FROM, if the search was not successful).  */
683

Paul Eggert's avatar
Paul Eggert committed
684 685 686 687
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)
688 689 690 691 692 693 694 695 696 697
{
  /* 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.  */
698
  int string_style = -1;	/* Presumed outside of any string.  */
Paul Eggert's avatar
Paul Eggert committed
699
  bool string_lossage = 0;
700
  /* Not a real lossage: indicates that we have passed a matching comment
Juanma Barranquero's avatar
Juanma Barranquero committed
701
     starter plus a non-matching comment-ender, meaning that any matching
702 703 704
     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
705
  bool comment_lossage = 0;
706 707 708
  ptrdiff_t comment_end = from;
  ptrdiff_t comment_end_byte = from_byte;
  ptrdiff_t comstart_pos = 0;
709
  ptrdiff_t comstart_byte;
710 711
  /* Place where the containing defun starts,
     or 0 if we didn't come across it yet.  */
712 713
  ptrdiff_t defun_start = 0;
  ptrdiff_t defun_start_byte = 0;
Paul Eggert's avatar
Paul Eggert committed
714
  enum syntaxcode code;
715
  ptrdiff_t nesting = 1;		/* Current comment nesting.  */
716
  int c;
717 718 719 720 721
  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.  */
722 723 724 725 726

  /* At beginning of range to scan, we're outside of strings;
     that determines quote parity to the comment-end.  */
  while (from != stop)
    {
727
      ptrdiff_t temp_byte;
Paul Eggert's avatar
Paul Eggert committed
728 729
      int prev_syntax;
      bool com2start, com2end, comstart;
730

731
      /* Move back and examine a character.  */
732
      DEC_BOTH (from, from_byte);
733 734
      UPDATE_SYNTAX_TABLE_BACKWARD (from);

735
      prev_syntax = syntax;
736
      c = FETCH_CHAR_AS_MULTIBYTE (from_byte);
737
      syntax = SYNTAX_WITH_FLAGS (c);
738 739
      code = SYNTAX (c);

740 741 742
      /* Check for 2-char comment markers.  */
      com2start = (SYNTAX_FLAGS_COMSTART_FIRST (syntax)
		   && SYNTAX_FLAGS_COMSTART_SECOND (prev_syntax)
743 744
		   && (comstyle
		       == SYNTAX_FLAGS_COMMENT_STYLE (prev_syntax, syntax))
745 746 747 748
		   && (SYNTAX_FLAGS_COMMENT_NESTED (prev_syntax)
		       || SYNTAX_FLAGS_COMMENT_NESTED (syntax)) == comnested);
      com2end = (SYNTAX_FLAGS_COMEND_FIRST (syntax)
		 && SYNTAX_FLAGS_COMEND_SECOND (prev_syntax));
749
      comstart = (com2start || code == Scomment);
750

751 752 753 754 755 756 757 758 759 760
      /* 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,
761 762 763
	 we don't try to be clever.  E.g. |*| in C, or }% in modes that
	 have %..\n and %{..}%.  */
      if (from > stop && (com2end || comstart))
764
	{
765
	  ptrdiff_t next = from, next_byte = from_byte;
766
	  int next_c, next_syntax;
767 768
	  DEC_BOTH (next, next_byte);
	  UPDATE_SYNTAX_TABLE_BACKWARD (next);
769
	  next_c = FETCH_CHAR_AS_MULTIBYTE (next_byte);
770
	  next_syntax = SYNTAX_WITH_FLAGS (next_c);
771
	  if (((comstart || comnested)
772 773 774 775
	       && SYNTAX_FLAGS_COMEND_SECOND (syntax)
	       && SYNTAX_FLAGS_COMEND_FIRST (next_syntax))
	      || ((com2end || comnested)
		  && SYNTAX_FLAGS_COMSTART_SECOND (syntax)
776 777
		  && (comstyle
		      == SYNTAX_FLAGS_COMMENT_STYLE (syntax, prev_syntax))
778 779 780
		  && SYNTAX_FLAGS_COMSTART_FIRST (next_syntax)))
	    goto lossage;
	  /* UPDATE_SYNTAX_TABLE_FORWARD (next + 1); */
781
	}
782 783 784 785 786 787 788 789 790 791 792 793 794 795 796

      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
797
	       && (comstyle != SYNTAX_FLAGS_COMMENT_STYLE (syntax, 0)
798
		   || SYNTAX_FLAGS_COMMENT_NESTED (syntax) != comnested))
799
	continue;
800

801 802 803 804
      /* Ignore escaped characters, except comment-enders which cannot
         be escaped.  */
      if ((Vcomment_end_can_be_escaped || code != Sendcomment)
          && char_quoted (from, from_byte))
805 806
	continue;

807
      switch (code)
808
	{
809 810 811 812 813 814 815 816 817 818 819 820 821 822
	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.  */
823
	    string_lossage = 1;
824
	  break;
Juanma Barranquero's avatar
Juanma Barranquero committed
825

826 827
	case Scomment:
	  /* We've already checked that it is the relevant comstyle.  */
828
	  if (string_style != -1 || comment_lossage || string_lossage)
829 830 831 832
	    /* There are odd string quotes involved, so let's be careful.
	       Test case in Pascal: " { " a { " } */
	    goto lossage;

833 834 835 836 837 838 839
	  if (!comnested)
	    {
	      /* Record best comment-starter so far.  */
	      comstart_pos = from;
	      comstart_byte = from_byte;
	    }
	  else if (--nesting <= 0)
840 841 842 843 844 845
	    /* 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;
846 847
	  break;

848
	case Sendcomment:
849
	  if (SYNTAX_FLAGS_COMMENT_STYLE (syntax, 0) == comstyle
850
	      && ((com2end && SYNTAX_FLAGS_COMMENT_NESTED (prev_syntax))
851
		  || SYNTAX_FLAGS_COMMENT_NESTED (syntax)) == comnested)
852 853 854 855 856 857 858 859 860
	    /* 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.  */
	    }
861 862 863 864 865 866 867 868 869
	  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;
870
	  break;
871

872 873 874 875 876 877 878 879 880 881 882
	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.  */
	    }
883
	  break;
884 885

	default:
886
	  break;
887
	}
888 889 890 891 892
    }

  if (comstart_pos == 0)
    {
      from = comment_end;
893
      from_byte = comment_end_byte;
894
      UPDATE_SYNTAX_TABLE_FORWARD (comment_end);
895
    }
896 897 898
  /* 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 */
899 900
    {
      from = comstart_pos;
901
      from_byte = comstart_byte;
902
      UPDATE_SYNTAX_TABLE_FORWARD (from - 1);
903
    }
904
  else lossage:
905
    {
906
      struct lisp_parse_state state;
907
      bool adjusted = true;
908 909
      /* We had two kinds of string delimiters mixed up
	 together.  Decode this going forwards.
910
	 Scan fwd from a known safe place (beginning-of-defun)
911 912
	 to the one in question; this records where we
	 last passed a comment starter.  */
913 914 915 916 917
      /* 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;
918
	  adjusted = (defun_start > BEGV);
919
	}
920
      do
921
	{
922
          internalize_parse_state (Qnil, &state);
923 924
	  scan_sexps_forward (&state,
			      defun_start, defun_start_byte,
925
			      comment_end, TYPE_MINIMUM (EMACS_INT),
926
			      0, 0);
927
	  defun_start = comment_end;
928 929 930 931 932 933 934 935 936 937
	  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);
	    }

938 939 940 941 942 943 944 945 946 947 948
	  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 (* " *)  */
		{
949
		  /* FIXME: We should advance by one or two chars.  */
950 951 952 953 954 955
		  defun_start = state.comstr_start + 2;
		  defun_start_byte = CHAR_TO_BYTE (defun_start);
		}
	    }
	} while (defun_start < comment_end);

956
      from_byte = CHAR_TO_BYTE (from);
957 958
      UPDATE_SYNTAX_TABLE_FORWARD (from - 1);
    }
Juanma Barranquero's avatar
Juanma Barranquero committed
959

960
 done:
961 962 963
  *charpos_ptr = from;
  *bytepos_ptr = from_byte;

Paul Eggert's avatar
Paul Eggert committed
964
  return from != comment_end;
965 966
}

Richard M. Stallman's avatar
Richard M. Stallman committed
967
DEFUN ("syntax-table-p", Fsyntax_table_p, Ssyntax_table_p, 1, 1, 0,
968 969
       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
970
  (Lisp_Object object)
Richard M. Stallman's avatar
Richard M. Stallman committed
971
{
972
  if (CHAR_TABLE_P (object)
973
      && EQ (XCHAR_TABLE (object)->purpose, Qsyntax_table))
Richard M. Stallman's avatar
Richard M. Stallman committed
974 975 976 977
    return Qt;
  return Qnil;
}

978
static void
979
check_syntax_table (Lisp_Object obj)
Richard M. Stallman's avatar
Richard M. Stallman committed
980
{
981