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 (c)))
 114
 115 /* Extract the information from the entry for character C
 116    in the current syntax table.  */
 117
 118 #ifdef __GNUC__
 119 #define SYNTAX(c)                                                       \
 120   ({ Lisp_Object temp;                                                  \
 121      temp = SYNTAX_ENTRY (c);                                           \
 122      (CONSP (temp)                                                      \
 123       ? (enum syntaxcode) (XINT (XCAR (temp)) & 0xff)           \
 124       : Swhitespace); })
 125
 126 #define SYNTAX_WITH_FLAGS(c)                                            \
 127   ({ Lisp_Object temp;                                                  \
 128      temp = SYNTAX_ENTRY (c);                                           \
 129      (CONSP (temp)                                                      \
 130       ? XINT (XCAR (temp))                                      \
 131       : (int) Swhitespace); })
 132
 133 #define SYNTAX_MATCH(c)                                                 \
 134   ({ Lisp_Object temp;                                                  \
 135      temp = SYNTAX_ENTRY (c);                                           \
 136      (CONSP (temp)                                                      \
 137       ? XCDR (temp)                                             \
 138       : Qnil); })
 139 #else
 140 #define SYNTAX(c)                                                       \
 141   (syntax_temp = SYNTAX_ENTRY ((c)),                                    \
 142    (CONSP (syntax_temp)                                                 \
 143     ? (enum syntaxcode) (XINT (XCAR (syntax_temp)) & 0xff)      \
 144     : Swhitespace))
 145
 146 #define SYNTAX_WITH_FLAGS(c)                                            \
 147   (syntax_temp = SYNTAX_ENTRY ((c)),                                    \
 148    (CONSP (syntax_temp)                                                 \
 149     ? XINT (XCAR (syntax_temp))                                 \
 150     : (int) Swhitespace))
 151
 152 #define SYNTAX_MATCH(c)                                                 \
 153   (syntax_temp = SYNTAX_ENTRY ((c)),                                    \
 154    (CONSP (syntax_temp)                                                 \
 155     ? XCDR (syntax_temp)                                                \
 156     : Qnil))
 157 #endif
 158
 159 /* Then there are seven single-bit flags that have the following meanings:
 160   1. This character is the first of a two-character comment-start sequence.
 161   2. This character is the second of a two-character comment-start sequence.
 162   3. This character is the first of a two-character comment-end sequence.
 163   4. This character is the second of a two-character comment-end sequence.
 164   5. This character is a prefix, for backward-prefix-chars.
 165   6. see below
 166   7. This character is part of a nestable comment sequence.
 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 #define SYNTAX_COMMENT_NESTED(c) ((SYNTAX_WITH_FLAGS (c) >> 22) & 1)
 192
 193 /* These macros extract specific flags from an integer
 194    that holds the syntax code and the flags.  */
 195
 196 #define SYNTAX_FLAGS_COMSTART_FIRST(flags) (((flags) >> 16) & 1)
 197
 198 #define SYNTAX_FLAGS_COMSTART_SECOND(flags) (((flags) >> 17) & 1)
 199
 200 #define SYNTAX_FLAGS_COMEND_FIRST(flags) (((flags) >> 18) & 1)
 201
 202 #define SYNTAX_FLAGS_COMEND_SECOND(flags) (((flags) >> 19) & 1)
 203
 204 #define SYNTAX_FLAGS_PREFIX(flags) (((flags) >> 20) & 1)
 205
 206 #define SYNTAX_FLAGS_COMMENT_STYLE(flags) (((flags) >> 21) & 1)
 207
 208 #define SYNTAX_FLAGS_COMMENT_NESTED(flags) (((flags) >> 22) & 1)
 209
 210 /* This array, indexed by a character, contains the syntax code which that
 211  character signifies (as a char).  For example,
 212  (enum syntaxcode) syntax_spec_code['w'] is Sword.  */
 213
 214 extern unsigned char syntax_spec_code[0400];
 215
 216 /* Indexed by syntax code, give the letter that describes it.  */
 217
 218 extern char syntax_code_spec[16];
 219
 220 /* Convert the byte offset BYTEPOS into a character position,
 221    for the object recorded in gl_state with SETUP_SYNTAX_TABLE_FOR_OBJECT.
 222
 223    The value is meant for use in the UPDATE_SYNTAX_TABLE... macros.
 224    These macros do nothing when parse_sexp_lookup_properties is 0,
 225    so we return 0 in that case, for speed.  */
 226
 227 #define SYNTAX_TABLE_BYTE_TO_CHAR(bytepos)                              \
 228   (! parse_sexp_lookup_properties                                       \
 229    ? 0                                                                  \
 230    : STRINGP (gl_state.object)                                          \
 231    ? string_byte_to_char (gl_state.object, (bytepos))                   \
 232    : BUFFERP (gl_state.object)                                          \
 233    ? buf_bytepos_to_charpos (XBUFFER (gl_state.object),                 \
 234                              (bytepos) + BUF_BEGV_BYTE (XBUFFER (gl_state.object)) - 1) - BUF_BEGV (XBUFFER (gl_state.object)) + 1      \
 235    : NILP (gl_state.object)                                             \
 236    ? BYTE_TO_CHAR ((bytepos) + BEGV_BYTE - 1) - BEGV + 1                \
 237    : (bytepos))
 238
 239 /* Make syntax table state (gl_state) good for CHARPOS, assuming it is
 240    currently good for a position before CHARPOS.  */
 241
 242 #define UPDATE_SYNTAX_TABLE_FORWARD(charpos)                    \
 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 /* Make syntax table state (gl_state) good for CHARPOS, assuming it is
 251    currently good for a position after CHARPOS.  */
 252
 253 #define UPDATE_SYNTAX_TABLE_BACKWARD(charpos)                   \
 254   (parse_sexp_lookup_properties                                 \
 255    && (charpos) <= gl_state.b_property                          \
 256    ? (update_syntax_table ((charpos) + gl_state.offset, -1, 0,  \
 257                            gl_state.object),                    \
 258       1)                                                        \
 259    : 0)
 260
 261 /* Make syntax table good for CHARPOS.  */
 262
 263 #define UPDATE_SYNTAX_TABLE(charpos)                            \
 264   (parse_sexp_lookup_properties                                 \
 265    && (charpos) <= gl_state.b_property                          \
 266    ? (update_syntax_table ((charpos) + gl_state.offset, -1, 0,  \
 267                            gl_state.object),                    \
 268       1)                                                        \
 269    : (parse_sexp_lookup_properties                              \
 270       && (charpos) >= gl_state.e_property                       \
 271       ? (update_syntax_table ((charpos) + gl_state.offset, 1, 0,\
 272                               gl_state.object),                 \
 273          1)                                                     \
 274       : 0))
 275
 276 /* This macro should be called with FROM at the start of forward
 277    search, or after the last position of the backward search.  It
 278    makes sure that the first char is picked up with correct table, so
 279    one does not need to call UPDATE_SYNTAX_TABLE immediately after the
 280    call.
 281    Sign of COUNT gives the direction of the search.
 282  */
 283
 284 #define SETUP_SYNTAX_TABLE(FROM, COUNT)                                 \
 285 if (1)                                                                  \
 286   {                                                                     \
 287     gl_state.b_property = BEGV - 1;                                     \
 288     gl_state.e_property = ZV + 1;                                       \
 289     gl_state.object = Qnil;                                             \
 290     gl_state.use_global = 0;                                            \
 291     gl_state.offset = 0;                                                \
 292     gl_state.current_syntax_table = current_buffer->syntax_table;       \
 293     if (parse_sexp_lookup_properties)                                   \
 294       if ((COUNT) > 0 || (FROM) > BEGV)                                 \
 295         update_syntax_table ((COUNT) > 0 ? (FROM) : (FROM) - 1, (COUNT),\
 296                              1, Qnil);                                  \
 297   }                                                                     \
 298 else
 299
 300 /* Same as above, but in OBJECT.  If OBJECT is nil, use current buffer.
 301    If it is t, ignore properties altogether.
 302
 303    This is meant for regex.c to use.  For buffers, regex.c passes arguments
 304    to the UPDATE_SYNTAX_TABLE macros which are relative to BEGV.
 305    So if it is a buffer, we set the offset field to BEGV.  */
 306
 307 #define SETUP_SYNTAX_TABLE_FOR_OBJECT(OBJECT, FROM, COUNT)              \
 308 if (1)                                                                  \
 309   {                                                                     \
 310     gl_state.object = (OBJECT);                                         \
 311     if (BUFFERP (gl_state.object))                                      \
 312       {                                                                 \
 313         struct buffer *buf = XBUFFER (gl_state.object);                 \
 314         gl_state.b_property = 0;                                        \
 315         gl_state.e_property = BUF_ZV (buf) - BUF_BEGV (buf) + 1;        \
 316         gl_state.offset = BUF_BEGV (buf) - 1;                           \
 317       }                                                                 \
 318     else if (NILP (gl_state.object))                                    \
 319       {                                                                 \
 320         gl_state.b_property = 0;                                        \
 321         gl_state.e_property = ZV - BEGV + 1;                            \
 322         gl_state.offset = BEGV - 1;                                     \
 323       }                                                                 \
 324     else if (EQ (gl_state.object, Qt))                                  \
 325       {                                                                 \
 326         gl_state.b_property = - 1;                                      \
 327         gl_state.e_property = 1500000000;                               \
 328         gl_state.offset = 0;                                            \
 329       }                                                                 \
 330     else                                                                \
 331       {                                                                 \
 332         gl_state.b_property = -1;                                       \
 333         gl_state.e_property = 1 + XSTRING (gl_state.object)->size;      \
 334         gl_state.offset = 0;                                            \
 335       }                                                                 \
 336     gl_state.use_global = 0;                                            \
 337     gl_state.current_syntax_table = current_buffer->syntax_table;       \
 338     if (parse_sexp_lookup_properties)                                   \
 339       update_syntax_table (((FROM) + gl_state.offset                    \
 340                             + (COUNT > 0 ? 0 :  -1)),                   \
 341                            COUNT, 1, gl_state.object);                  \
 342   }                                                                     \
 343 else
 344
 345 struct gl_state_s
 346 {
 347   Lisp_Object object;                   /* The object we are scanning. */
 348   int start;                            /* Where to stop. */
 349   int stop;                             /* Where to stop. */
 350   int use_global;                       /* Whether to use global_code
 351                                            or c_s_t. */
 352   Lisp_Object global_code;              /* Syntax code of current char. */
 353   Lisp_Object current_syntax_table;     /* Syntax table for current pos. */
 354   Lisp_Object old_prop;                 /* Syntax-table prop at prev pos. */
 355   int b_property;                       /* Last index where c_s_t is
 356                                            not 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   char left_ok;
 369   char right_ok;
 370 };
 371
 372 extern struct gl_state_s gl_state;
 373 extern int parse_sexp_lookup_properties;
 374 extern INTERVAL interval_of P_ ((int, Lisp_Object));
 375
 376 extern int scan_words P_ ((int, int));