src/syntax.h

   1 /* Declarations having to do with GNU Emacs syntax tables.
   2    Copyright (C) 1985, 1993, 1994, 1997, 1998, 2001, 2002, 2003, 2004,
   3                  2005, 2006, 2007  Free Software Foundation, Inc.
   4
   5 This file is part of GNU Emacs.
   6
   7 GNU Emacs is free software; you can redistribute it and/or modify
   8 it under the terms of the GNU General Public License as published by
   9 the Free Software Foundation; either version 3, or (at your option)
  10 any later version.
  11
  12 GNU Emacs is distributed in the hope that it will be useful,
  13 but WITHOUT ANY WARRANTY; without even the implied warranty of
  14 MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the
  15 GNU General Public License for more details.
  16
  17 You should have received a copy of the GNU General Public License
  18 along with GNU Emacs; see the file COPYING.  If not, write to
  19 the Free Software Foundation, Inc., 51 Franklin Street, Fifth Floor,
  20 Boston, MA 02110-1301, USA.  */
  21
  22
  23 extern Lisp_Object Qsyntax_table_p;
  24 extern void update_syntax_table P_ ((int, int, int, Lisp_Object));
  25
  26 /* The standard syntax table is stored where it will automatically
  27    be used in all new buffers.  */
  28 #define Vstandard_syntax_table buffer_defaults.syntax_table
  29
  30 /* A syntax table is a chartable whose elements are cons cells
  31    (CODE+FLAGS . MATCHING-CHAR).  MATCHING-CHAR can be nil if the char
  32    is not a kind of parenthesis.
  33
  34    The low 8 bits of CODE+FLAGS is a code, as follows:  */
  35
  36 enum syntaxcode
  37   {
  38     Swhitespace, /* for a whitespace character */
  39     Spunct,      /* for random punctuation characters */
  40     Sword,       /* for a word constituent */
  41     Ssymbol,     /* symbol constituent but not word constituent */
  42     Sopen,       /* for a beginning delimiter */
  43     Sclose,      /* for an ending delimiter */
  44     Squote,      /* for a prefix character like Lisp ' */
  45     Sstring,     /* for a string-grouping character like Lisp " */
  46     Smath,       /* for delimiters like $ in Tex.  */
  47     Sescape,     /* for a character that begins a C-style escape */
  48     Scharquote,  /* for a character that quotes the following character */
  49     Scomment,    /* for a comment-starting character */
  50     Sendcomment, /* for a comment-ending character */
  51     Sinherit,    /* use the standard syntax table for this character */
  52     Scomment_fence, /* Starts/ends comment which is delimited on the
  53                        other side by any char with the same syntaxcode.  */
  54     Sstring_fence,  /* Starts/ends string which is delimited on the
  55                        other side by any char with the same syntaxcode.  */
  56     Smax         /* Upper bound on codes that are meaningful */
  57   };
  58
  59 /* Set the syntax entry VAL for char C in table TABLE.  */
  60
  61 #define SET_RAW_SYNTAX_ENTRY(table, c, val)                             \
  62   ((((c) & 0xFF) == (c))                                                \
  63    ? (XCHAR_TABLE (table)->contents[(unsigned char) (c)] = (val))       \
  64    : Faset ((table), make_number (c), (val)))
  65
  66 /* Fetch the syntax entry for char C in syntax table TABLE.
  67    This macro is called only when C is less than CHAR_TABLE_ORDINARY_SLOTS.
  68    Do inheritance.  */
  69
  70 #ifdef __GNUC__
  71 #define SYNTAX_ENTRY_FOLLOW_PARENT(table, c)                    \
  72   ({ Lisp_Object _syntax_tbl = (table);                         \
  73      Lisp_Object _syntax_temp = XCHAR_TABLE (_syntax_tbl)->contents[(c)]; \
  74      while (NILP (_syntax_temp))                                \
  75        {                                                        \
  76          _syntax_tbl = XCHAR_TABLE (_syntax_tbl)->parent;       \
  77          if (NILP (_syntax_tbl))                                \
  78            break;                                               \
  79          _syntax_temp = XCHAR_TABLE (_syntax_tbl)->contents[(c)]; \
  80        }                                                        \
  81      _syntax_temp; })
  82 #else
  83 extern Lisp_Object syntax_temp;
  84 extern Lisp_Object syntax_parent_lookup P_ ((Lisp_Object, int));
  85
  86 #define SYNTAX_ENTRY_FOLLOW_PARENT(table, c)            \
  87   (syntax_temp = XCHAR_TABLE (table)->contents[(c)],    \
  88    (NILP (syntax_temp)                                  \
  89     ? syntax_parent_lookup (table, (c))                 \
  90     : syntax_temp))
  91 #endif
  92
  93 /* SYNTAX_ENTRY fetches the information from the entry for character C
  94    in syntax table TABLE, or from globally kept data (gl_state).
  95    Does inheritance.  */
  96 /* CURRENT_SYNTAX_TABLE gives the syntax table valid for current
  97    position, it is either the buffer's syntax table, or syntax table
  98    found in text properties.  */
  99
 100 #ifdef SYNTAX_ENTRY_VIA_PROPERTY
 101 #  define SYNTAX_ENTRY(c)                                             \
 102     (gl_state.use_global ? gl_state.global_code : SYNTAX_ENTRY_INT (c))
 103 #  define CURRENT_SYNTAX_TABLE gl_state.current_syntax_table
 104 #else
 105 #  define SYNTAX_ENTRY SYNTAX_ENTRY_INT
 106 #  define CURRENT_SYNTAX_TABLE current_buffer->syntax_table
 107 #endif
 108
 109 #define SYNTAX_ENTRY_INT(c)                             \
 110   ((((c) & 0xFF) == (c))                                \
 111    ? SYNTAX_ENTRY_FOLLOW_PARENT (CURRENT_SYNTAX_TABLE,  \
 112                                  (unsigned char) (c))   \
 113    : Faref (CURRENT_SYNTAX_TABLE,                       \
 114             make_number (c)))
 115
 116 /* Extract the information from the entry for character C
 117    in the current syntax table.  */
 118
 119 #ifdef __GNUC__
 120 #define SYNTAX(c)                                                       \
 121   ({ Lisp_Object _syntax_temp;                                          \
 122      _syntax_temp = SYNTAX_ENTRY (c);                                   \
 123      (CONSP (_syntax_temp)                                              \
 124       ? (enum syntaxcode) (XINT (XCAR (_syntax_temp)) & 0xff)           \
 125       : Swhitespace); })
 126
 127 #define SYNTAX_WITH_FLAGS(c)                                            \
 128   ({ Lisp_Object _syntax_temp;                                          \
 129      _syntax_temp = SYNTAX_ENTRY (c);                                   \
 130      (CONSP (_syntax_temp)                                              \
 131       ? XINT (XCAR (_syntax_temp))                                      \
 132       : (int) Swhitespace); })
 133
 134 #define SYNTAX_MATCH(c)                                                 \
 135   ({ Lisp_Object _syntax_temp;                                          \
 136      _syntax_temp = SYNTAX_ENTRY (c);                                   \
 137      (CONSP (_syntax_temp)                                              \
 138       ? XCDR (_syntax_temp)                                             \
 139       : Qnil); })
 140 #else
 141 #define SYNTAX(c)                                                       \
 142   (syntax_temp = SYNTAX_ENTRY ((c)),                                    \
 143    (CONSP (syntax_temp)                                                 \
 144     ? (enum syntaxcode) (XINT (XCAR (syntax_temp)) & 0xff)      \
 145     : Swhitespace))
 146
 147 #define SYNTAX_WITH_FLAGS(c)                                            \
 148   (syntax_temp = SYNTAX_ENTRY ((c)),                                    \
 149    (CONSP (syntax_temp)                                                 \
 150     ? XINT (XCAR (syntax_temp))                                 \
 151     : (int) Swhitespace))
 152
 153 #define SYNTAX_MATCH(c)                                                 \
 154   (syntax_temp = SYNTAX_ENTRY ((c)),                                    \
 155    (CONSP (syntax_temp)                                                 \
 156     ? XCDR (syntax_temp)                                                \
 157     : Qnil))
 158 #endif
 159
 160 /* Then there are seven single-bit flags that have the following meanings:
 161   1. This character is the first of a two-character comment-start sequence.
 162   2. This character is the second of a two-character comment-start sequence.
 163   3. This character is the first of a two-character comment-end sequence.
 164   4. This character is the second of a two-character comment-end sequence.
 165   5. This character is a prefix, for backward-prefix-chars.
 166   6. see below
 167   7. This character is part of a nestable comment sequence.
 168   Note that any two-character sequence whose first character has flag 1
 169   and whose second character has flag 2 will be interpreted as a comment start.
 170
 171   bit 6 is used to discriminate between two different comment styles.
 172   Languages such as C++ allow two orthogonal syntax start/end pairs
 173   and bit 6 is used to determine whether a comment-end or Scommentend
 174   ends style a or b.  Comment start sequences can start style a or b.
 175   Style a is always the default.
 176   */
 177
 178 /* These macros extract a particular flag for a given character.  */
 179
 180 #define SYNTAX_COMSTART_FIRST(c) ((SYNTAX_WITH_FLAGS (c) >> 16) & 1)
 181
 182 #define SYNTAX_COMSTART_SECOND(c) ((SYNTAX_WITH_FLAGS (c) >> 17) & 1)
 183
 184 #define SYNTAX_COMEND_FIRST(c) ((SYNTAX_WITH_FLAGS (c) >> 18) & 1)
 185
 186 #define SYNTAX_COMEND_SECOND(c) ((SYNTAX_WITH_FLAGS (c) >> 19) & 1)
 187
 188 #define SYNTAX_PREFIX(c) ((SYNTAX_WITH_FLAGS (c) >> 20) & 1)
 189
 190 #define SYNTAX_COMMENT_STYLE(c) ((SYNTAX_WITH_FLAGS (c) >> 21) & 1)
 191
 192 #define SYNTAX_COMMENT_NESTED(c) ((SYNTAX_WITH_FLAGS (c) >> 22) & 1)
 193
 194 /* These macros extract specific flags from an integer
 195    that holds the syntax code and the flags.  */
 196
 197 #define SYNTAX_FLAGS_COMSTART_FIRST(flags) (((flags) >> 16) & 1)
 198
 199 #define SYNTAX_FLAGS_COMSTART_SECOND(flags) (((flags) >> 17) & 1)
 200
 201 #define SYNTAX_FLAGS_COMEND_FIRST(flags) (((flags) >> 18) & 1)
 202
 203 #define SYNTAX_FLAGS_COMEND_SECOND(flags) (((flags) >> 19) & 1)
 204
 205 #define SYNTAX_FLAGS_PREFIX(flags) (((flags) >> 20) & 1)
 206
 207 #define SYNTAX_FLAGS_COMMENT_STYLE(flags) (((flags) >> 21) & 1)
 208
 209 #define SYNTAX_FLAGS_COMMENT_NESTED(flags) (((flags) >> 22) & 1)
 210
 211 /* This array, indexed by a character, contains the syntax code which that
 212  character signifies (as a char).  For example,
 213  (enum syntaxcode) syntax_spec_code['w'] is Sword.  */
 214
 215 extern unsigned char syntax_spec_code[0400];
 216
 217 /* Indexed by syntax code, give the letter that describes it.  */
 218
 219 extern char syntax_code_spec[16];
 220
 221 /* Convert the byte offset BYTEPOS into a character position,
 222    for the object recorded in gl_state with SETUP_SYNTAX_TABLE_FOR_OBJECT.
 223
 224    The value is meant for use in the UPDATE_SYNTAX_TABLE... macros.
 225    These macros do nothing when parse_sexp_lookup_properties is 0,
 226    so we return 0 in that case, for speed.  */
 227
 228 #define SYNTAX_TABLE_BYTE_TO_CHAR(bytepos)                              \
 229   (! parse_sexp_lookup_properties                                       \
 230    ? 0                                                                  \
 231    : STRINGP (gl_state.object)                                          \
 232    ? string_byte_to_char (gl_state.object, (bytepos))                   \
 233    : BUFFERP (gl_state.object)                                          \
 234    ? buf_bytepos_to_charpos (XBUFFER (gl_state.object),                 \
 235                              (bytepos) + BUF_BEGV_BYTE (XBUFFER (gl_state.object)) - 1) - BUF_BEGV (XBUFFER (gl_state.object)) + 1      \
 236    : NILP (gl_state.object)                                             \
 237    ? BYTE_TO_CHAR ((bytepos) + BEGV_BYTE - 1) - BEGV + 1                \
 238    : (bytepos))
 239
 240 /* Make syntax table state (gl_state) good for CHARPOS, assuming it is
 241    currently good for a position before CHARPOS.  */
 242
 243 #define UPDATE_SYNTAX_TABLE_FORWARD(charpos)                    \
 244   (parse_sexp_lookup_properties                                 \
 245    && (charpos) >= gl_state.e_property                          \
 246    ? (update_syntax_table ((charpos) + gl_state.offset, 1, 0,   \
 247                            gl_state.object),                    \
 248       1)                                                        \
 249    : 0)
 250
 251 /* Make syntax table state (gl_state) good for CHARPOS, assuming it is
 252    currently good for a position after CHARPOS.  */
 253
 254 #define UPDATE_SYNTAX_TABLE_BACKWARD(charpos)                   \
 255   (parse_sexp_lookup_properties                                 \
 256    && (charpos) < gl_state.b_property                           \
 257    ? (update_syntax_table ((charpos) + gl_state.offset, -1, 0,  \
 258                            gl_state.object),                    \
 259       1)                                                        \
 260    : 0)
 261
 262 /* Make syntax table good for CHARPOS.  */
 263
 264 #define UPDATE_SYNTAX_TABLE(charpos)                            \
 265   (parse_sexp_lookup_properties                                 \
 266    && (charpos) < gl_state.b_property                           \
 267    ? (update_syntax_table ((charpos) + gl_state.offset, -1, 0,  \
 268                            gl_state.object),                    \
 269       1)                                                        \
 270    : (parse_sexp_lookup_properties                              \
 271       && (charpos) >= gl_state.e_property                       \
 272       ? (update_syntax_table ((charpos) + gl_state.offset, 1, 0,\
 273                               gl_state.object),                 \
 274          1)                                                     \
 275       : 0))
 276
 277 /* This macro should be called with FROM at the start of forward
 278    search, or after the last position of the backward search.  It
 279    makes sure that the first char is picked up with correct table, so
 280    one does not need to call UPDATE_SYNTAX_TABLE immediately after the
 281    call.
 282    Sign of COUNT gives the direction of the search.
 283  */
 284
 285 #define SETUP_SYNTAX_TABLE(FROM, COUNT)                                 \
 286 if (1)                                                                  \
 287   {                                                                     \
 288     gl_state.b_property = BEGV;                                         \
 289     gl_state.e_property = ZV + 1;                                       \
 290     gl_state.object = Qnil;                                             \
 291     gl_state.use_global = 0;                                            \
 292     gl_state.offset = 0;                                                \
 293     gl_state.current_syntax_table = current_buffer->syntax_table;       \
 294     if (parse_sexp_lookup_properties)                                   \
 295       if ((COUNT) > 0 || (FROM) > BEGV)                                 \
 296         update_syntax_table ((COUNT) > 0 ? (FROM) : (FROM) - 1, (COUNT),\
 297                              1, Qnil);                                  \
 298   }                                                                     \
 299 else
 300
 301 /* Same as above, but in OBJECT.  If OBJECT is nil, use current buffer.
 302    If it is t, ignore properties altogether.
 303
 304    This is meant for regex.c to use.  For buffers, regex.c passes arguments
 305    to the UPDATE_SYNTAX_TABLE macros which are relative to BEGV.
 306    So if it is a buffer, we set the offset field to BEGV.  */
 307
 308 #define SETUP_SYNTAX_TABLE_FOR_OBJECT(OBJECT, FROM, COUNT)              \
 309 if (1)                                                                  \
 310   {                                                                     \
 311     gl_state.object = (OBJECT);                                         \
 312     if (BUFFERP (gl_state.object))                                      \
 313       {                                                                 \
 314         struct buffer *buf = XBUFFER (gl_state.object);                 \
 315         gl_state.b_property = 1;                                        \
 316         gl_state.e_property = BUF_ZV (buf) - BUF_BEGV (buf) + 1;        \
 317         gl_state.offset = BUF_BEGV (buf) - 1;                           \
 318       }                                                                 \
 319     else if (NILP (gl_state.object))                                    \
 320       {                                                                 \
 321         gl_state.b_property = 1;                                        \
 322         gl_state.e_property = ZV - BEGV + 1;                            \
 323         gl_state.offset = BEGV - 1;                                     \
 324       }                                                                 \
 325     else if (EQ (gl_state.object, Qt))                                  \
 326       {                                                                 \
 327         gl_state.b_property = 0;                                        \
 328         gl_state.e_property = 1500000000;                               \
 329         gl_state.offset = 0;                                            \
 330       }                                                                 \
 331     else                                                                \
 332       {                                                                 \
 333         gl_state.b_property = 0;                                        \
 334         gl_state.e_property = 1 + SCHARS (gl_state.object);             \
 335         gl_state.offset = 0;                                            \
 336       }                                                                 \
 337     gl_state.use_global = 0;                                            \
 338     gl_state.current_syntax_table = current_buffer->syntax_table;       \
 339     if (parse_sexp_lookup_properties)                                   \
 340       update_syntax_table (((FROM) + gl_state.offset                    \
 341                             + (COUNT > 0 ? 0 :  -1)),                   \
 342                            COUNT, 1, gl_state.object);                  \
 343   }                                                                     \
 344 else
 345
 346 struct gl_state_s
 347 {
 348   Lisp_Object object;                   /* The object we are scanning. */
 349   int start;                            /* Where to stop. */
 350   int stop;                             /* Where to stop. */
 351   int use_global;                       /* Whether to use global_code
 352                                            or c_s_t. */
 353   Lisp_Object global_code;              /* Syntax code of current char. */
 354   Lisp_Object current_syntax_table;     /* Syntax table for current pos. */
 355   Lisp_Object old_prop;                 /* Syntax-table prop at prev pos. */
 356   int b_property;                       /* First index where c_s_t is valid. */
 357   int e_property;                       /* First index where c_s_t is
 358                                            not valid. */
 359   INTERVAL forward_i;                   /* Where to start lookup on forward */
 360   INTERVAL backward_i;                  /* or backward movement.  The
 361                                            data in c_s_t is valid
 362                                            between these intervals,
 363                                            and possibly at the
 364                                            intervals too, depending
 365                                            on: */
 366   /* Offset for positions specified to UPDATE_SYNTAX_TABLE.  */
 367   int offset;
 368 };
 369
 370 extern struct gl_state_s gl_state;
 371 extern int parse_sexp_lookup_properties;
 372 extern INTERVAL interval_of P_ ((int, Lisp_Object));
 373
 374 extern int scan_words P_ ((int, int));
 375
 376 /* arch-tag: 28833cca-cd73-4741-8c85-a3111166a0e0
 377    (do not change this comment) */