src/syntax.h

   1 /* Declarations having to do with GNU Emacs syntax tables.
   2    Copyright (C) 1985, 93, 94, 97, 1998 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   ((c) < CHAR_TABLE_SINGLE_BYTE_SLOTS                                   \
  62    ? (XCHAR_TABLE (table)->contents[(unsigned char) (c)] = (val))       \
  63    : Faset ((table), make_number (c), (val)))
  64
  65 /* Fetch the syntax entry for char C in syntax table TABLE.
  66    This macro is called only when C is less than CHAR_TABLE_ORDINARY_SLOTS.
  67    Do inheritance.  */
  68
  69 #ifdef __GNUC__
  70 #define SYNTAX_ENTRY_FOLLOW_PARENT(table, c)                    \
  71   ({ Lisp_Object tbl = table;                                   \
  72      Lisp_Object temp = XCHAR_TABLE (tbl)->contents[(c)];       \
  73      while (NILP (temp))                                        \
  74        {                                                        \
  75          tbl = XCHAR_TABLE (tbl)->parent;                       \
  76          if (NILP (tbl))                                        \
  77            break;                                               \
  78          temp = XCHAR_TABLE (tbl)->contents[(c)];               \
  79        }                                                        \
  80      temp; })
  81 #else
  82 extern Lisp_Object syntax_temp;
  83 extern Lisp_Object syntax_parent_lookup P_ ((Lisp_Object, int));
  84
  85 #define SYNTAX_ENTRY_FOLLOW_PARENT(table, c)            \
  86   (syntax_temp = XCHAR_TABLE (table)->contents[(c)],    \
  87    (NILP (syntax_temp)                                  \
  88     ? syntax_parent_lookup (table, (c))                 \
  89     : syntax_temp))
  90 #endif
  91
  92 /* SYNTAX_ENTRY fetches the information from the entry for character C
  93    in syntax table TABLE, or from globally kept data (gl_state).
  94    Does inheritance.  */
  95 /* CURRENT_SYNTAX_TABLE gives the syntax table valid for current
  96    position, it is either the buffer's syntax table, or syntax table
  97    found in text properties.  */
  98
  99 #ifdef SYNTAX_ENTRY_VIA_PROPERTY
 100 #  define SYNTAX_ENTRY(c)                                             \
 101     (gl_state.use_global ? gl_state.global_code : SYNTAX_ENTRY_INT (c))
 102 #  define CURRENT_SYNTAX_TABLE gl_state.current_syntax_table
 103 #else
 104 #  define SYNTAX_ENTRY SYNTAX_ENTRY_INT
 105 #  define CURRENT_SYNTAX_TABLE current_buffer->syntax_table
 106 #endif
 107
 108 #define SYNTAX_ENTRY_INT(c)                                     \
 109   ((c) < CHAR_TABLE_SINGLE_BYTE_SLOTS                           \
 110    ? SYNTAX_ENTRY_FOLLOW_PARENT (CURRENT_SYNTAX_TABLE,          \
 111                                  (unsigned char) (c))           \
 112    : Faref (CURRENT_SYNTAX_TABLE,                               \
 113             make_number (COMPOSITE_CHAR_P (c)                   \
 114                          ? cmpchar_component ((c), 0, 1)        \
 115                          : (c))))
 116
 117 /* Extract the information from the entry for character C
 118    in the current syntax table.  */
 119
 120 #ifdef __GNUC__
 121 #define SYNTAX(c)                                                       \
 122   ({ Lisp_Object temp;                                                  \
 123      temp = SYNTAX_ENTRY (c);                                           \
 124      (CONSP (temp)                                                      \
 125       ? (enum syntaxcode) (XINT (XCONS (temp)->car) & 0xff)             \
 126       : Swhitespace); })
 127
 128 #define SYNTAX_WITH_FLAGS(c)                                            \
 129   ({ Lisp_Object temp;                                                  \
 130      temp = SYNTAX_ENTRY (c);                                           \
 131      (CONSP (temp)                                                      \
 132       ? XINT (XCONS (temp)->car)                                        \
 133       : (int) Swhitespace); })
 134
 135 #define SYNTAX_MATCH(c)                                                 \
 136   ({ Lisp_Object temp;                                                  \
 137      temp = SYNTAX_ENTRY (c);                                           \
 138      (CONSP (temp)                                                      \
 139       ? XCONS (temp)->cdr                                               \
 140       : Qnil); })
 141 #else
 142 #define SYNTAX(c)                                                       \
 143   (syntax_temp = SYNTAX_ENTRY ((c)),                                    \
 144    (CONSP (syntax_temp)                                                 \
 145     ? (enum syntaxcode) (XINT (XCONS (syntax_temp)->car) & 0xff)        \
 146     : Swhitespace))
 147
 148 #define SYNTAX_WITH_FLAGS(c)                                            \
 149   (syntax_temp = SYNTAX_ENTRY ((c)),                                    \
 150    (CONSP (syntax_temp)                                                 \
 151     ? XINT (XCONS (syntax_temp)->car)                                   \
 152     : (int) Swhitespace))
 153
 154 #define SYNTAX_MATCH(c)                                                 \
 155   (syntax_temp = SYNTAX_ENTRY ((c)),                                    \
 156    (CONSP (syntax_temp)                                                 \
 157     ? XCONS (syntax_temp)->cdr                                          \
 158     : Qnil))
 159 #endif
 160
 161 /* Then there are six single-bit flags that have the following meanings:
 162   1. This character is the first of a two-character comment-start sequence.
 163   2. This character is the second of a two-character comment-start sequence.
 164   3. This character is the first of a two-character comment-end sequence.
 165   4. This character is the second of a two-character comment-end sequence.
 166   5. This character is a prefix, for backward-prefix-chars.
 167   Note that any two-character sequence whose first character has flag 1
 168   and whose second character has flag 2 will be interpreted as a comment start.
 169
 170   bit 6 is used to discriminate between two different comment styles.
 171   Languages such as C++ allow two orthogonal syntax start/end pairs
 172   and bit 6 is used to determine whether a comment-end or Scommentend
 173   ends style a or b.  Comment start sequences can start style a or b.
 174   Style a is always the default.
 175   */
 176
 177 /* These macros extract a particular flag for a given character.  */
 178
 179 #define SYNTAX_COMSTART_FIRST(c) ((SYNTAX_WITH_FLAGS (c) >> 16) & 1)
 180
 181 #define SYNTAX_COMSTART_SECOND(c) ((SYNTAX_WITH_FLAGS (c) >> 17) & 1)
 182
 183 #define SYNTAX_COMEND_FIRST(c) ((SYNTAX_WITH_FLAGS (c) >> 18) & 1)
 184
 185 #define SYNTAX_COMEND_SECOND(c) ((SYNTAX_WITH_FLAGS (c) >> 19) & 1)
 186
 187 #define SYNTAX_PREFIX(c) ((SYNTAX_WITH_FLAGS (c) >> 20) & 1)
 188
 189 #define SYNTAX_COMMENT_STYLE(c) ((SYNTAX_WITH_FLAGS (c) >> 21) & 1)
 190
 191 /* These macros extract specific flags from an integer
 192    that holds the syntax code and the flags.  */
 193
 194 #define SYNTAX_FLAGS_COMSTART_FIRST(flags) (((flags) >> 16) & 1)
 195
 196 #define SYNTAX_FLAGS_COMSTART_SECOND(flags) (((flags) >> 17) & 1)
 197
 198 #define SYNTAX_FLAGS_COMEND_FIRST(flags) (((flags) >> 18) & 1)
 199
 200 #define SYNTAX_FLAGS_COMEND_SECOND(flags) (((flags) >> 19) & 1)
 201
 202 #define SYNTAX_FLAGS_PREFIX(flags) (((flags) >> 20) & 1)
 203
 204 #define SYNTAX_FLAGS_COMMENT_STYLE(flags) (((flags) >> 21) & 1)
 205
 206 /* This array, indexed by a character, contains the syntax code which that
 207  character signifies (as a char).  For example,
 208  (enum syntaxcode) syntax_spec_code['w'] is Sword.  */
 209
 210 extern unsigned char syntax_spec_code[0400];
 211
 212 /* Indexed by syntax code, give the letter that describes it.  */
 213
 214 extern char syntax_code_spec[16];
 215
 216 /* Convert the byte offset BYTEPOS into a character position,
 217    for the object recorded in gl_state with SETUP_SYNTAX_TABLE_FOR_OBJECT.
 218
 219    The value is meant for use in the UPDATE_SYNTAX_TABLE... macros.
 220    These macros do nothing when parse_sexp_lookup_properties is 0,
 221    so we return 0 in that case, for speed.  */
 222
 223 #define SYNTAX_TABLE_BYTE_TO_CHAR(bytepos)                              \
 224   (! parse_sexp_lookup_properties                                       \
 225    ? 0                                                                  \
 226    : STRINGP (gl_state.object)                                          \
 227    ? string_byte_to_char (gl_state.object, (bytepos))                   \
 228    : BUFFERP (gl_state.object)                                          \
 229    ? buf_bytepos_to_charpos (XBUFFER (gl_state.object),                 \
 230                              (bytepos) + BUF_BEGV_BYTE (XBUFFER (gl_state.object)) - 1) - BUF_BEGV (XBUFFER (gl_state.object)) + 1      \
 231    : NILP (gl_state.object)                                             \
 232    ? BYTE_TO_CHAR ((bytepos) + BEGV_BYTE - 1) - BEGV + 1                \
 233    : (bytepos))
 234
 235 /* Make syntax table state (gl_state) good for CHARPOS, assuming it is
 236    currently good for a position before CHARPOS.  */
 237
 238 #define UPDATE_SYNTAX_TABLE_FORWARD(charpos)                    \
 239   (parse_sexp_lookup_properties                                 \
 240    && (charpos) >= gl_state.e_property                          \
 241    ? (update_syntax_table ((charpos) + gl_state.offset, 1, 0,   \
 242                            gl_state.object),                    \
 243       1)                                                        \
 244    : 0)
 245
 246 /* Make syntax table state (gl_state) good for CHARPOS, assuming it is
 247    currently good for a position after CHARPOS.  */
 248
 249 #define UPDATE_SYNTAX_TABLE_BACKWARD(charpos)                   \
 250   (parse_sexp_lookup_properties                                 \
 251    && (charpos) <= gl_state.b_property                          \
 252    ? (update_syntax_table ((charpos) + gl_state.offset, -1, 0,  \
 253                            gl_state.object),                    \
 254       1)                                                        \
 255    : 0)
 256
 257 /* Make syntax table good for CHARPOS.  */
 258
 259 #define UPDATE_SYNTAX_TABLE(charpos)                            \
 260   (parse_sexp_lookup_properties                                 \
 261    && (charpos) <= gl_state.b_property                          \
 262    ? (update_syntax_table ((charpos) + gl_state.offset, -1, 0,  \
 263                            gl_state.object),                    \
 264       1)                                                        \
 265    : (parse_sexp_lookup_properties                              \
 266       && (charpos) >= gl_state.e_property                       \
 267       ? (update_syntax_table ((charpos) + gl_state.offset, 1, 0,\
 268                               gl_state.object),                 \
 269          1)                                                     \
 270       : 0))
 271
 272 /* This macro should be called with FROM at the start of forward
 273    search, or after the last position of the backward search.  It
 274    makes sure that the first char is picked up with correct table, so
 275    one does not need to call UPDATE_SYNTAX_TABLE immediately after the
 276    call.
 277    Sign of COUNT gives the direction of the search.
 278  */
 279
 280 #define SETUP_SYNTAX_TABLE(FROM, COUNT)                                 \
 281 if (1)                                                                  \
 282   {                                                                     \
 283     gl_state.b_property = BEGV - 1;                                     \
 284     gl_state.e_property = ZV + 1;                                       \
 285     gl_state.object = Qnil;                                             \
 286     gl_state.use_global = 0;                                            \
 287     gl_state.offset = 0;                                                \
 288     gl_state.current_syntax_table = current_buffer->syntax_table;       \
 289     if (parse_sexp_lookup_properties)                                   \
 290       if ((COUNT) > 0 || (FROM) > BEGV)                                 \
 291         update_syntax_table ((COUNT) > 0 ? (FROM) : (FROM) - 1, (COUNT),\
 292                              1, Qnil);                                  \
 293   }                                                                     \
 294 else
 295
 296 /* Same as above, but in OBJECT.  If OBJECT is nil, use current buffer.
 297    If it is t, ignore properties altogether.
 298
 299    This is meant for regex.c to use.  For buffers, regex.c passes arguments
 300    to the UPDATE_SYNTAX_TABLE macros which are relative to BEGV.
 301    So if it is a buffer, we set the offset field to BEGV.  */
 302
 303 #define SETUP_SYNTAX_TABLE_FOR_OBJECT(OBJECT, FROM, COUNT)              \
 304 if (1)                                                                  \
 305   {                                                                     \
 306     gl_state.object = (OBJECT);                                         \
 307     if (BUFFERP (gl_state.object))                                      \
 308       {                                                                 \
 309         struct buffer *buf = XBUFFER (gl_state.object);                 \
 310         gl_state.b_property = 0;                                        \
 311         gl_state.e_property = BUF_ZV (buf) - BUF_BEGV (buf) + 1;        \
 312         gl_state.offset = BUF_BEGV (buf) - 1;                           \
 313       }                                                                 \
 314     else if (NILP (gl_state.object))                                    \
 315       {                                                                 \
 316         gl_state.b_property = 0;                                        \
 317         gl_state.e_property = ZV - BEGV + 1;                            \
 318         gl_state.offset = BEGV - 1;                                     \
 319       }                                                                 \
 320     else if (EQ (gl_state.object, Qt))                                  \
 321       {                                                                 \
 322         gl_state.b_property = - 1;                                      \
 323         gl_state.e_property = 1500000000;                               \
 324         gl_state.offset = 0;                                            \
 325       }                                                                 \
 326     else                                                                \
 327       {                                                                 \
 328         gl_state.b_property = -1;                                       \
 329         gl_state.e_property = 1 + XSTRING (gl_state.object)->size;      \
 330         gl_state.offset = 0;                                            \
 331       }                                                                 \
 332     gl_state.use_global = 0;                                            \
 333     gl_state.current_syntax_table = current_buffer->syntax_table;       \
 334     if (parse_sexp_lookup_properties)                                   \
 335       update_syntax_table (((FROM) + gl_state.offset                    \
 336                             + (COUNT > 0 ? 0 :  -1)),                   \
 337                            COUNT, 1, gl_state.object);                  \
 338   }                                                                     \
 339 else
 340
 341 struct gl_state_s
 342 {
 343   Lisp_Object object;                   /* The object we are scanning. */
 344   int start;                            /* Where to stop. */
 345   int stop;                             /* Where to stop. */
 346   int use_global;                       /* Whether to use global_code
 347                                            or c_s_t. */
 348   Lisp_Object global_code;              /* Syntax code of current char. */
 349   Lisp_Object current_syntax_table;     /* Syntax table for current pos. */
 350   Lisp_Object old_prop;                 /* Syntax-table prop at prev pos. */
 351   int b_property;                       /* Last index where c_s_t is
 352                                            not valid. */
 353   int e_property;                       /* First index where c_s_t is
 354                                            not valid. */
 355   INTERVAL forward_i;                   /* Where to start lookup on forward */
 356   INTERVAL backward_i;                  /* or backward movement.  The
 357                                            data in c_s_t is valid
 358                                            between these intervals,
 359                                            and possibly at the
 360                                            intervals too, depending
 361                                            on: */
 362   /* Offset for positions specified to UPDATE_SYNTAX_TABLE.  */
 363   int offset;
 364   char left_ok;
 365   char right_ok;
 366 };
 367
 368 extern struct gl_state_s gl_state;
 369 extern int parse_sexp_lookup_properties;
 370 extern INTERVAL interval_of P_ ((int, Lisp_Object));
 371
 372 extern int scan_words P_ ((int, int));