src/syntax.h

   1 /* Declarations having to do with GNU Emacs syntax tables.
   2    Copyright (C) 1985, 93, 94, 97, 1998, 2002 Free Software Foundation, Inc.
   3
   4 This file is part of GNU Emacs.
   5
   6 GNU Emacs is free software; you can redistribute it and/or modify
   7 it under the terms of the GNU General Public License as published by
   8 the Free Software Foundation; either version 2, or (at your option)
   9 any later version.
  10
  11 GNU Emacs is distributed in the hope that it will be useful,
  12 but WITHOUT ANY WARRANTY; without even the implied warranty of
  13 MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the
  14 GNU General Public License for more details.
  15
  16 You should have received a copy of the GNU General Public License
  17 along with GNU Emacs; see the file COPYING.  If not, write to
  18 the Free Software Foundation, Inc., 59 Temple Place - Suite 330,
  19 Boston, MA 02111-1307, USA.  */
  20
  21
  22 extern Lisp_Object Qsyntax_table_p;
  23 extern void update_syntax_table P_ ((int, int, int, Lisp_Object));
  24
  25 /* The standard syntax table is stored where it will automatically
  26    be used in all new buffers.  */
  27 #define Vstandard_syntax_table buffer_defaults.syntax_table
  28
  29 /* A syntax table is a chartable whose elements are cons cells
  30    (CODE+FLAGS . MATCHING-CHAR).  MATCHING-CHAR can be nil if the char
  31    is not a kind of parenthesis.
  32
  33    The low 8 bits of CODE+FLAGS is a code, as follows:  */
  34
  35 enum syntaxcode
  36   {
  37     Swhitespace, /* for a whitespace character */
  38     Spunct,      /* for random punctuation characters */
  39     Sword,       /* for a word constituent */
  40     Ssymbol,     /* symbol constituent but not word constituent */
  41     Sopen,       /* for a beginning delimiter */
  42     Sclose,      /* for an ending delimiter */
  43     Squote,      /* for a prefix character like Lisp ' */
  44     Sstring,     /* for a string-grouping character like Lisp " */
  45     Smath,       /* for delimiters like $ in Tex.  */
  46     Sescape,     /* for a character that begins a C-style escape */
  47     Scharquote,  /* for a character that quotes the following character */
  48     Scomment,    /* for a comment-starting character */
  49     Sendcomment, /* for a comment-ending character */
  50     Sinherit,    /* use the standard syntax table for this character */
  51     Scomment_fence, /* Starts/ends comment which is delimited on the
  52                        other side by any char with the same syntaxcode.  */
  53     Sstring_fence,  /* Starts/ends string which is delimited on the
  54                        other side by any char with the same syntaxcode.  */
  55     Smax         /* Upper bound on codes that are meaningful */
  56   };
  57
  58 /* Set the syntax entry VAL for char C in table TABLE.  */
  59
  60 #define SET_RAW_SYNTAX_ENTRY(table, c, val)     \
  61   CHAR_TABLE_SET ((table), c, (val))
  62
  63 /* Set the syntax entry VAL for char-range RANGE in table TABLE.
  64    RANGE is a cons (FROM . TO) specifying the range of characters.  */
  65
  66 #define SET_RAW_SYNTAX_ENTRY_RANGE(table, range, val)   \
  67   Fset_char_table_range ((table), (range), (val))
  68
  69 /* SYNTAX_ENTRY fetches the information from the entry for character C
  70    in syntax table TABLE, or from globally kept data (gl_state).
  71    Does inheritance.  */
  72 /* CURRENT_SYNTAX_TABLE gives the syntax table valid for current
  73    position, it is either the buffer's syntax table, or syntax table
  74    found in text properties.  */
  75
  76 #ifdef SYNTAX_ENTRY_VIA_PROPERTY
  77 #  define SYNTAX_ENTRY(c)                                             \
  78     (gl_state.use_global ? gl_state.global_code : SYNTAX_ENTRY_INT (c))
  79 #  define CURRENT_SYNTAX_TABLE gl_state.current_syntax_table
  80 #else
  81 #  define SYNTAX_ENTRY SYNTAX_ENTRY_INT
  82 #  define CURRENT_SYNTAX_TABLE current_buffer->syntax_table
  83 #endif
  84
  85 #define SYNTAX_ENTRY_INT(c) CHAR_TABLE_REF (CURRENT_SYNTAX_TABLE, (c))
  86
  87 /* Extract the information from the entry for character C
  88    in the current syntax table.  */
  89
  90 #ifdef __GNUC__
  91 #define SYNTAX(c)                                                       \
  92   ({ Lisp_Object temp;                                                  \
  93      temp = SYNTAX_ENTRY (c);                                           \
  94      (CONSP (temp)                                                      \
  95       ? (enum syntaxcode) (XINT (XCAR (temp)) & 0xff)           \
  96       : Swhitespace); })
  97
  98 #define SYNTAX_WITH_FLAGS(c)                                            \
  99   ({ Lisp_Object temp;                                                  \
 100      temp = SYNTAX_ENTRY (c);                                           \
 101      (CONSP (temp)                                                      \
 102       ? XINT (XCAR (temp))                                      \
 103       : (int) Swhitespace); })
 104
 105 #define SYNTAX_MATCH(c)                                                 \
 106   ({ Lisp_Object temp;                                                  \
 107      temp = SYNTAX_ENTRY (c);                                           \
 108      (CONSP (temp)                                                      \
 109       ? XCDR (temp)                                             \
 110       : Qnil); })
 111 #else
 112 extern Lisp_Object syntax_temp;
 113 #define SYNTAX(c)                                                       \
 114   (syntax_temp = SYNTAX_ENTRY ((c)),                                    \
 115    (CONSP (syntax_temp)                                                 \
 116     ? (enum syntaxcode) (XINT (XCAR (syntax_temp)) & 0xff)      \
 117     : Swhitespace))
 118
 119 #define SYNTAX_WITH_FLAGS(c)                                            \
 120   (syntax_temp = SYNTAX_ENTRY ((c)),                                    \
 121    (CONSP (syntax_temp)                                                 \
 122     ? XINT (XCAR (syntax_temp))                                 \
 123     : (int) Swhitespace))
 124
 125 #define SYNTAX_MATCH(c)                                                 \
 126   (syntax_temp = SYNTAX_ENTRY ((c)),                                    \
 127    (CONSP (syntax_temp)                                                 \
 128     ? XCDR (syntax_temp)                                                \
 129     : Qnil))
 130 #endif
 131
 132 /* Then there are seven single-bit flags that have the following meanings:
 133   1. This character is the first of a two-character comment-start sequence.
 134   2. This character is the second of a two-character comment-start sequence.
 135   3. This character is the first of a two-character comment-end sequence.
 136   4. This character is the second of a two-character comment-end sequence.
 137   5. This character is a prefix, for backward-prefix-chars.
 138   6. see below
 139   7. This character is part of a nestable comment sequence.
 140   Note that any two-character sequence whose first character has flag 1
 141   and whose second character has flag 2 will be interpreted as a comment start.
 142
 143   bit 6 is used to discriminate between two different comment styles.
 144   Languages such as C++ allow two orthogonal syntax start/end pairs
 145   and bit 6 is used to determine whether a comment-end or Scommentend
 146   ends style a or b.  Comment start sequences can start style a or b.
 147   Style a is always the default.
 148   */
 149
 150 /* These macros extract a particular flag for a given character.  */
 151
 152 #define SYNTAX_COMSTART_FIRST(c) ((SYNTAX_WITH_FLAGS (c) >> 16) & 1)
 153
 154 #define SYNTAX_COMSTART_SECOND(c) ((SYNTAX_WITH_FLAGS (c) >> 17) & 1)
 155
 156 #define SYNTAX_COMEND_FIRST(c) ((SYNTAX_WITH_FLAGS (c) >> 18) & 1)
 157
 158 #define SYNTAX_COMEND_SECOND(c) ((SYNTAX_WITH_FLAGS (c) >> 19) & 1)
 159
 160 #define SYNTAX_PREFIX(c) ((SYNTAX_WITH_FLAGS (c) >> 20) & 1)
 161
 162 #define SYNTAX_COMMENT_STYLE(c) ((SYNTAX_WITH_FLAGS (c) >> 21) & 1)
 163
 164 #define SYNTAX_COMMENT_NESTED(c) ((SYNTAX_WITH_FLAGS (c) >> 22) & 1)
 165
 166 /* These macros extract specific flags from an integer
 167    that holds the syntax code and the flags.  */
 168
 169 #define SYNTAX_FLAGS_COMSTART_FIRST(flags) (((flags) >> 16) & 1)
 170
 171 #define SYNTAX_FLAGS_COMSTART_SECOND(flags) (((flags) >> 17) & 1)
 172
 173 #define SYNTAX_FLAGS_COMEND_FIRST(flags) (((flags) >> 18) & 1)
 174
 175 #define SYNTAX_FLAGS_COMEND_SECOND(flags) (((flags) >> 19) & 1)
 176
 177 #define SYNTAX_FLAGS_PREFIX(flags) (((flags) >> 20) & 1)
 178
 179 #define SYNTAX_FLAGS_COMMENT_STYLE(flags) (((flags) >> 21) & 1)
 180
 181 #define SYNTAX_FLAGS_COMMENT_NESTED(flags) (((flags) >> 22) & 1)
 182
 183 /* This array, indexed by a character, contains the syntax code which that
 184  character signifies (as a char).  For example,
 185  (enum syntaxcode) syntax_spec_code['w'] is Sword.  */
 186
 187 extern unsigned char syntax_spec_code[0400];
 188
 189 /* Indexed by syntax code, give the letter that describes it.  */
 190
 191 extern char syntax_code_spec[16];
 192
 193 /* Convert the byte offset BYTEPOS into a character position,
 194    for the object recorded in gl_state with SETUP_SYNTAX_TABLE_FOR_OBJECT.
 195
 196    The value is meant for use in the UPDATE_SYNTAX_TABLE... macros.
 197    These macros do nothing when parse_sexp_lookup_properties is 0,
 198    so we return 0 in that case, for speed.  */
 199
 200 #define SYNTAX_TABLE_BYTE_TO_CHAR(bytepos)                              \
 201   (! parse_sexp_lookup_properties                                       \
 202    ? 0                                                                  \
 203    : STRINGP (gl_state.object)                                          \
 204    ? string_byte_to_char (gl_state.object, (bytepos))                   \
 205    : BUFFERP (gl_state.object)                                          \
 206    ? buf_bytepos_to_charpos (XBUFFER (gl_state.object),                 \
 207                              (bytepos) + BUF_BEGV_BYTE (XBUFFER (gl_state.object)) - 1) - BUF_BEGV (XBUFFER (gl_state.object)) + 1      \
 208    : NILP (gl_state.object)                                             \
 209    ? BYTE_TO_CHAR ((bytepos) + BEGV_BYTE - 1) - BEGV + 1                \
 210    : (bytepos))
 211
 212 /* Make syntax table state (gl_state) good for CHARPOS, assuming it is
 213    currently good for a position before CHARPOS.  */
 214
 215 #define UPDATE_SYNTAX_TABLE_FORWARD(charpos)                    \
 216   (parse_sexp_lookup_properties                                 \
 217    && (charpos) >= gl_state.e_property                          \
 218    ? (update_syntax_table ((charpos) + gl_state.offset, 1, 0,   \
 219                            gl_state.object),                    \
 220       1)                                                        \
 221    : 0)
 222
 223 /* Make syntax table state (gl_state) good for CHARPOS, assuming it is
 224    currently good for a position after CHARPOS.  */
 225
 226 #define UPDATE_SYNTAX_TABLE_BACKWARD(charpos)                   \
 227   (parse_sexp_lookup_properties                                 \
 228    && (charpos) < gl_state.b_property                           \
 229    ? (update_syntax_table ((charpos) + gl_state.offset, -1, 0,  \
 230                            gl_state.object),                    \
 231       1)                                                        \
 232    : 0)
 233
 234 /* Make syntax table good for CHARPOS.  */
 235
 236 #define UPDATE_SYNTAX_TABLE(charpos)                            \
 237   (parse_sexp_lookup_properties                                 \
 238    && (charpos) < gl_state.b_property                           \
 239    ? (update_syntax_table ((charpos) + gl_state.offset, -1, 0,  \
 240                            gl_state.object),                    \
 241       1)                                                        \
 242    : (parse_sexp_lookup_properties                              \
 243       && (charpos) >= gl_state.e_property                       \
 244       ? (update_syntax_table ((charpos) + gl_state.offset, 1, 0,\
 245                               gl_state.object),                 \
 246          1)                                                     \
 247       : 0))
 248
 249 /* This macro should be called with FROM at the start of forward
 250    search, or after the last position of the backward search.  It
 251    makes sure that the first char is picked up with correct table, so
 252    one does not need to call UPDATE_SYNTAX_TABLE immediately after the
 253    call.
 254    Sign of COUNT gives the direction of the search.
 255  */
 256
 257 #define SETUP_SYNTAX_TABLE(FROM, COUNT)                                 \
 258 if (1)                                                                  \
 259   {                                                                     \
 260     gl_state.b_property = BEGV;                                         \
 261     gl_state.e_property = ZV + 1;                                       \
 262     gl_state.object = Qnil;                                             \
 263     gl_state.use_global = 0;                                            \
 264     gl_state.offset = 0;                                                \
 265     gl_state.current_syntax_table = current_buffer->syntax_table;       \
 266     if (parse_sexp_lookup_properties)                                   \
 267       if ((COUNT) > 0 || (FROM) > BEGV)                                 \
 268         update_syntax_table ((COUNT) > 0 ? (FROM) : (FROM) - 1, (COUNT),\
 269                              1, Qnil);                                  \
 270   }                                                                     \
 271 else
 272
 273 /* Same as above, but in OBJECT.  If OBJECT is nil, use current buffer.
 274    If it is t, ignore properties altogether.
 275
 276    This is meant for regex.c to use.  For buffers, regex.c passes arguments
 277    to the UPDATE_SYNTAX_TABLE macros which are relative to BEGV.
 278    So if it is a buffer, we set the offset field to BEGV.  */
 279
 280 #define SETUP_SYNTAX_TABLE_FOR_OBJECT(OBJECT, FROM, COUNT)              \
 281 if (1)                                                                  \
 282   {                                                                     \
 283     gl_state.object = (OBJECT);                                         \
 284     if (BUFFERP (gl_state.object))                                      \
 285       {                                                                 \
 286         struct buffer *buf = XBUFFER (gl_state.object);                 \
 287         gl_state.b_property = 1;                                        \
 288         gl_state.e_property = BUF_ZV (buf) - BUF_BEGV (buf) + 1;        \
 289         gl_state.offset = BUF_BEGV (buf) - 1;                           \
 290       }                                                                 \
 291     else if (NILP (gl_state.object))                                    \
 292       {                                                                 \
 293         gl_state.b_property = 1;                                        \
 294         gl_state.e_property = ZV - BEGV + 1;                            \
 295         gl_state.offset = BEGV - 1;                                     \
 296       }                                                                 \
 297     else if (EQ (gl_state.object, Qt))                                  \
 298       {                                                                 \
 299         gl_state.b_property = 0;                                        \
 300         gl_state.e_property = 1500000000;                               \
 301         gl_state.offset = 0;                                            \
 302       }                                                                 \
 303     else                                                                \
 304       {                                                                 \
 305         gl_state.b_property = 0;                                        \
 306         gl_state.e_property = 1 + XSTRING (gl_state.object)->size;      \
 307         gl_state.offset = 0;                                            \
 308       }                                                                 \
 309     gl_state.use_global = 0;                                            \
 310     gl_state.current_syntax_table = current_buffer->syntax_table;       \
 311     if (parse_sexp_lookup_properties)                                   \
 312       update_syntax_table (((FROM) + gl_state.offset                    \
 313                             + (COUNT > 0 ? 0 :  -1)),                   \
 314                            COUNT, 1, gl_state.object);                  \
 315   }                                                                     \
 316 else
 317
 318 struct gl_state_s
 319 {
 320   Lisp_Object object;                   /* The object we are scanning. */
 321   int start;                            /* Where to stop. */
 322   int stop;                             /* Where to stop. */
 323   int use_global;                       /* Whether to use global_code
 324                                            or c_s_t. */
 325   Lisp_Object global_code;              /* Syntax code of current char. */
 326   Lisp_Object current_syntax_table;     /* Syntax table for current pos. */
 327   Lisp_Object old_prop;                 /* Syntax-table prop at prev pos. */
 328   int b_property;                       /* First index where c_s_t is valid. */
 329   int e_property;                       /* First index where c_s_t is
 330                                            not valid. */
 331   INTERVAL forward_i;                   /* Where to start lookup on forward */
 332   INTERVAL backward_i;                  /* or backward movement.  The
 333                                            data in c_s_t is valid
 334                                            between these intervals,
 335                                            and possibly at the
 336                                            intervals too, depending
 337                                            on: */
 338   /* Offset for positions specified to UPDATE_SYNTAX_TABLE.  */
 339   int offset;
 340 };
 341
 342 extern struct gl_state_s gl_state;
 343 extern int parse_sexp_lookup_properties;
 344 extern INTERVAL interval_of P_ ((int, Lisp_Object));
 345
 346 extern int scan_words P_ ((int, int));