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, 2008  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   CHAR_TABLE_SET ((table), c, (val))
  63
  64 /* Set the syntax entry VAL for char-range RANGE in table TABLE.
  65    RANGE is a cons (FROM . TO) specifying the range of characters.  */
  66
  67 #define SET_RAW_SYNTAX_ENTRY_RANGE(table, range, val)   \
  68   Fset_char_table_range ((table), (range), (val))
  69
  70 /* SYNTAX_ENTRY fetches the information from the entry for character C
  71    in syntax table TABLE, or from globally kept data (gl_state).
  72    Does inheritance.  */
  73 /* CURRENT_SYNTAX_TABLE gives the syntax table valid for current
  74    position, it is either the buffer's syntax table, or syntax table
  75    found in text properties.  */
  76
  77 #ifdef SYNTAX_ENTRY_VIA_PROPERTY
  78 #  define SYNTAX_ENTRY(c)                                             \
  79     (gl_state.use_global ? gl_state.global_code : SYNTAX_ENTRY_INT (c))
  80 #  define CURRENT_SYNTAX_TABLE gl_state.current_syntax_table
  81 #else
  82 #  define SYNTAX_ENTRY SYNTAX_ENTRY_INT
  83 #  define CURRENT_SYNTAX_TABLE current_buffer->syntax_table
  84 #endif
  85
  86 #define SYNTAX_ENTRY_INT(c) CHAR_TABLE_REF (CURRENT_SYNTAX_TABLE, (c))
  87
  88 /* Extract the information from the entry for character C
  89    in the current syntax table.  */
  90
  91 #ifdef __GNUC__
  92 #define SYNTAX(c)                                                       \
  93   ({ Lisp_Object _syntax_temp;                                          \
  94      _syntax_temp = SYNTAX_ENTRY (c);                                   \
  95      (CONSP (_syntax_temp)                                              \
  96       ? (enum syntaxcode) (XINT (XCAR (_syntax_temp)) & 0xff)           \
  97       : Swhitespace); })
  98
  99 #define SYNTAX_WITH_FLAGS(c)                                            \
 100   ({ Lisp_Object _syntax_temp;                                          \
 101      _syntax_temp = SYNTAX_ENTRY (c);                                   \
 102      (CONSP (_syntax_temp)                                              \
 103       ? XINT (XCAR (_syntax_temp))                                      \
 104       : (int) Swhitespace); })
 105
 106 #define SYNTAX_MATCH(c)                                                 \
 107   ({ Lisp_Object _syntax_temp;                                          \
 108      _syntax_temp = SYNTAX_ENTRY (c);                                   \
 109      (CONSP (_syntax_temp)                                              \
 110       ? XCDR (_syntax_temp)                                             \
 111       : Qnil); })
 112 #else
 113 extern Lisp_Object syntax_temp;
 114 #define SYNTAX(c)                                                       \
 115   (syntax_temp = SYNTAX_ENTRY ((c)),                                    \
 116    (CONSP (syntax_temp)                                                 \
 117     ? (enum syntaxcode) (XINT (XCAR (syntax_temp)) & 0xff)      \
 118     : Swhitespace))
 119
 120 #define SYNTAX_WITH_FLAGS(c)                                            \
 121   (syntax_temp = SYNTAX_ENTRY ((c)),                                    \
 122    (CONSP (syntax_temp)                                                 \
 123     ? XINT (XCAR (syntax_temp))                                 \
 124     : (int) Swhitespace))
 125
 126 #define SYNTAX_MATCH(c)                                                 \
 127   (syntax_temp = SYNTAX_ENTRY ((c)),                                    \
 128    (CONSP (syntax_temp)                                                 \
 129     ? XCDR (syntax_temp)                                                \
 130     : Qnil))
 131 #endif
 132
 133 /* Then there are seven single-bit flags that have the following meanings:
 134   1. This character is the first of a two-character comment-start sequence.
 135   2. This character is the second of a two-character comment-start sequence.
 136   3. This character is the first of a two-character comment-end sequence.
 137   4. This character is the second of a two-character comment-end sequence.
 138   5. This character is a prefix, for backward-prefix-chars.
 139   6. see below
 140   7. This character is part of a nestable comment sequence.
 141   Note that any two-character sequence whose first character has flag 1
 142   and whose second character has flag 2 will be interpreted as a comment start.
 143
 144   bit 6 is used to discriminate between two different comment styles.
 145   Languages such as C++ allow two orthogonal syntax start/end pairs
 146   and bit 6 is used to determine whether a comment-end or Scommentend
 147   ends style a or b.  Comment start sequences can start style a or b.
 148   Style a is always the default.
 149   */
 150
 151 /* These macros extract a particular flag for a given character.  */
 152
 153 #define SYNTAX_COMSTART_FIRST(c) ((SYNTAX_WITH_FLAGS (c) >> 16) & 1)
 154
 155 #define SYNTAX_COMSTART_SECOND(c) ((SYNTAX_WITH_FLAGS (c) >> 17) & 1)
 156
 157 #define SYNTAX_COMEND_FIRST(c) ((SYNTAX_WITH_FLAGS (c) >> 18) & 1)
 158
 159 #define SYNTAX_COMEND_SECOND(c) ((SYNTAX_WITH_FLAGS (c) >> 19) & 1)
 160
 161 #define SYNTAX_PREFIX(c) ((SYNTAX_WITH_FLAGS (c) >> 20) & 1)
 162
 163 #define SYNTAX_COMMENT_STYLE(c) ((SYNTAX_WITH_FLAGS (c) >> 21) & 1)
 164
 165 #define SYNTAX_COMMENT_NESTED(c) ((SYNTAX_WITH_FLAGS (c) >> 22) & 1)
 166
 167 /* These macros extract specific flags from an integer
 168    that holds the syntax code and the flags.  */
 169
 170 #define SYNTAX_FLAGS_COMSTART_FIRST(flags) (((flags) >> 16) & 1)
 171
 172 #define SYNTAX_FLAGS_COMSTART_SECOND(flags) (((flags) >> 17) & 1)
 173
 174 #define SYNTAX_FLAGS_COMEND_FIRST(flags) (((flags) >> 18) & 1)
 175
 176 #define SYNTAX_FLAGS_COMEND_SECOND(flags) (((flags) >> 19) & 1)
 177
 178 #define SYNTAX_FLAGS_PREFIX(flags) (((flags) >> 20) & 1)
 179
 180 #define SYNTAX_FLAGS_COMMENT_STYLE(flags) (((flags) >> 21) & 1)
 181
 182 #define SYNTAX_FLAGS_COMMENT_NESTED(flags) (((flags) >> 22) & 1)
 183
 184 /* This array, indexed by a character, contains the syntax code which that
 185  character signifies (as a char).  For example,
 186  (enum syntaxcode) syntax_spec_code['w'] is Sword.  */
 187
 188 extern unsigned char syntax_spec_code[0400];
 189
 190 /* Indexed by syntax code, give the letter that describes it.  */
 191
 192 extern char syntax_code_spec[16];
 193
 194 /* Convert the byte offset BYTEPOS into a character position,
 195    for the object recorded in gl_state with SETUP_SYNTAX_TABLE_FOR_OBJECT.
 196
 197    The value is meant for use in the UPDATE_SYNTAX_TABLE... macros.
 198    These macros do nothing when parse_sexp_lookup_properties is 0,
 199    so we return 0 in that case, for speed.  */
 200
 201 #define SYNTAX_TABLE_BYTE_TO_CHAR(bytepos)                              \
 202   (! parse_sexp_lookup_properties                                       \
 203    ? 0                                                                  \
 204    : STRINGP (gl_state.object)                                          \
 205    ? string_byte_to_char (gl_state.object, (bytepos))                   \
 206    : BUFFERP (gl_state.object)                                          \
 207    ? buf_bytepos_to_charpos (XBUFFER (gl_state.object),                 \
 208                              (bytepos) + BUF_BEGV_BYTE (XBUFFER (gl_state.object)) - 1) - BUF_BEGV (XBUFFER (gl_state.object)) + 1      \
 209    : NILP (gl_state.object)                                             \
 210    ? BYTE_TO_CHAR ((bytepos) + BEGV_BYTE - 1) - BEGV + 1                \
 211    : (bytepos))
 212
 213 /* Make syntax table state (gl_state) good for CHARPOS, assuming it is
 214    currently good for a position before CHARPOS.  */
 215
 216 #define UPDATE_SYNTAX_TABLE_FORWARD(charpos)                    \
 217   (parse_sexp_lookup_properties                                 \
 218    && (charpos) >= gl_state.e_property                          \
 219    ? (update_syntax_table ((charpos) + gl_state.offset, 1, 0,   \
 220                            gl_state.object),                    \
 221       1)                                                        \
 222    : 0)
 223
 224 /* Make syntax table state (gl_state) good for CHARPOS, assuming it is
 225    currently good for a position after CHARPOS.  */
 226
 227 #define UPDATE_SYNTAX_TABLE_BACKWARD(charpos)                   \
 228   (parse_sexp_lookup_properties                                 \
 229    && (charpos) < gl_state.b_property                           \
 230    ? (update_syntax_table ((charpos) + gl_state.offset, -1, 0,  \
 231                            gl_state.object),                    \
 232       1)                                                        \
 233    : 0)
 234
 235 /* Make syntax table good for CHARPOS.  */
 236
 237 #define UPDATE_SYNTAX_TABLE(charpos)                            \
 238   (parse_sexp_lookup_properties                                 \
 239    && (charpos) < gl_state.b_property                           \
 240    ? (update_syntax_table ((charpos) + gl_state.offset, -1, 0,  \
 241                            gl_state.object),                    \
 242       1)                                                        \
 243    : (parse_sexp_lookup_properties                              \
 244       && (charpos) >= gl_state.e_property                       \
 245       ? (update_syntax_table ((charpos) + gl_state.offset, 1, 0,\
 246                               gl_state.object),                 \
 247          1)                                                     \
 248       : 0))
 249
 250 /* This macro should be called with FROM at the start of forward
 251    search, or after the last position of the backward search.  It
 252    makes sure that the first char is picked up with correct table, so
 253    one does not need to call UPDATE_SYNTAX_TABLE immediately after the
 254    call.
 255    Sign of COUNT gives the direction of the search.
 256  */
 257
 258 #define SETUP_SYNTAX_TABLE(FROM, COUNT)                                 \
 259 do                                                                      \
 260   {                                                                     \
 261     gl_state.b_property = BEGV;                                         \
 262     gl_state.e_property = ZV + 1;                                       \
 263     gl_state.object = Qnil;                                             \
 264     gl_state.use_global = 0;                                            \
 265     gl_state.offset = 0;                                                \
 266     gl_state.current_syntax_table = current_buffer->syntax_table;       \
 267     if (parse_sexp_lookup_properties)                                   \
 268       if ((COUNT) > 0 || (FROM) > BEGV)                                 \
 269         update_syntax_table ((COUNT) > 0 ? (FROM) : (FROM) - 1, (COUNT),\
 270                              1, Qnil);                                  \
 271   }                                                                     \
 272 while (0)
 273
 274 /* Same as above, but in OBJECT.  If OBJECT is nil, use current buffer.
 275    If it is t, ignore properties altogether.
 276
 277    This is meant for regex.c to use.  For buffers, regex.c passes arguments
 278    to the UPDATE_SYNTAX_TABLE macros which are relative to BEGV.
 279    So if it is a buffer, we set the offset field to BEGV.  */
 280
 281 #define SETUP_SYNTAX_TABLE_FOR_OBJECT(OBJECT, FROM, COUNT)              \
 282 do                                                                      \
 283   {                                                                     \
 284     gl_state.object = (OBJECT);                                         \
 285     if (BUFFERP (gl_state.object))                                      \
 286       {                                                                 \
 287         struct buffer *buf = XBUFFER (gl_state.object);                 \
 288         gl_state.b_property = 1;                                        \
 289         gl_state.e_property = BUF_ZV (buf) - BUF_BEGV (buf) + 1;        \
 290         gl_state.offset = BUF_BEGV (buf) - 1;                           \
 291       }                                                                 \
 292     else if (NILP (gl_state.object))                                    \
 293       {                                                                 \
 294         gl_state.b_property = 1;                                        \
 295         gl_state.e_property = ZV - BEGV + 1;                            \
 296         gl_state.offset = BEGV - 1;                                     \
 297       }                                                                 \
 298     else if (EQ (gl_state.object, Qt))                                  \
 299       {                                                                 \
 300         gl_state.b_property = 0;                                        \
 301         gl_state.e_property = 1500000000;                               \
 302         gl_state.offset = 0;                                            \
 303       }                                                                 \
 304     else                                                                \
 305       {                                                                 \
 306         gl_state.b_property = 0;                                        \
 307         gl_state.e_property = 1 + SCHARS (gl_state.object);             \
 308         gl_state.offset = 0;                                            \
 309       }                                                                 \
 310     gl_state.use_global = 0;                                            \
 311     gl_state.current_syntax_table = current_buffer->syntax_table;       \
 312     if (parse_sexp_lookup_properties)                                   \
 313       update_syntax_table (((FROM) + gl_state.offset                    \
 314                             + (COUNT > 0 ? 0 :  -1)),                   \
 315                            COUNT, 1, gl_state.object);                  \
 316   }                                                                     \
 317 while (0)
 318
 319 struct gl_state_s
 320 {
 321   Lisp_Object object;                   /* The object we are scanning. */
 322   int start;                            /* Where to stop. */
 323   int stop;                             /* Where to stop. */
 324   int use_global;                       /* Whether to use global_code
 325                                            or c_s_t. */
 326   Lisp_Object global_code;              /* Syntax code of current char. */
 327   Lisp_Object current_syntax_table;     /* Syntax table for current pos. */
 328   Lisp_Object old_prop;                 /* Syntax-table prop at prev pos. */
 329   int b_property;                       /* First index where c_s_t is valid. */
 330   int e_property;                       /* First index where c_s_t is
 331                                            not valid. */
 332   INTERVAL forward_i;                   /* Where to start lookup on forward */
 333   INTERVAL backward_i;                  /* or backward movement.  The
 334                                            data in c_s_t is valid
 335                                            between these intervals,
 336                                            and possibly at the
 337                                            intervals too, depending
 338                                            on: */
 339   /* Offset for positions specified to UPDATE_SYNTAX_TABLE.  */
 340   int offset;
 341 };
 342
 343 extern struct gl_state_s gl_state;
 344 extern int parse_sexp_lookup_properties;
 345 extern INTERVAL interval_of P_ ((int, Lisp_Object));
 346
 347 extern int scan_words P_ ((int, int));
 348
 349 /* arch-tag: 28833cca-cd73-4741-8c85-a3111166a0e0
 350    (do not change this comment) */