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, make_number ((c))))
 113
 114 /* Extract the information from the entry for character C
 115    in the current syntax table.  */
 116
 117 #ifdef __GNUC__
 118 #define SYNTAX(c)                                                       \
 119   ({ Lisp_Object temp;                                                  \
 120      temp = SYNTAX_ENTRY (c);                                           \
 121      (CONSP (temp)                                                      \
 122       ? (enum syntaxcode) (XINT (XCONS (temp)->car) & 0xff)             \
 123       : Swhitespace); })
 124
 125 #define SYNTAX_WITH_FLAGS(c)                                            \
 126   ({ Lisp_Object temp;                                                  \
 127      temp = SYNTAX_ENTRY (c);                                           \
 128      (CONSP (temp)                                                      \
 129       ? XINT (XCONS (temp)->car)                                        \
 130       : (int) Swhitespace); })
 131
 132 #define SYNTAX_MATCH(c)                                                 \
 133   ({ Lisp_Object temp;                                                  \
 134      temp = SYNTAX_ENTRY (c);                                           \
 135      (CONSP (temp)                                                      \
 136       ? XCONS (temp)->cdr                                               \
 137       : Qnil); })
 138 #else
 139 #define SYNTAX(c)                                                       \
 140   (syntax_temp = SYNTAX_ENTRY ((c)),                                    \
 141    (CONSP (syntax_temp)                                                 \
 142     ? (enum syntaxcode) (XINT (XCONS (syntax_temp)->car) & 0xff)        \
 143     : Swhitespace))
 144
 145 #define SYNTAX_WITH_FLAGS(c)                                            \
 146   (syntax_temp = SYNTAX_ENTRY ((c)),                                    \
 147    (CONSP (syntax_temp)                                                 \
 148     ? XINT (XCONS (syntax_temp)->car)                                   \
 149     : (int) Swhitespace))
 150
 151 #define SYNTAX_MATCH(c)                                                 \
 152   (syntax_temp = SYNTAX_ENTRY ((c)),                                    \
 153    (CONSP (syntax_temp)                                                 \
 154     ? XCONS (syntax_temp)->cdr                                          \
 155     : Qnil))
 156 #endif
 157
 158 /* Then there are six single-bit flags that have the following meanings:
 159   1. This character is the first of a two-character comment-start sequence.
 160   2. This character is the second of a two-character comment-start sequence.
 161   3. This character is the first of a two-character comment-end sequence.
 162   4. This character is the second of a two-character comment-end sequence.
 163   5. This character is a prefix, for backward-prefix-chars.
 164   Note that any two-character sequence whose first character has flag 1
 165   and whose second character has flag 2 will be interpreted as a comment start.
 166
 167   bit 6 is used to discriminate between two different comment styles.
 168   Languages such as C++ allow two orthogonal syntax start/end pairs
 169   and bit 6 is used to determine whether a comment-end or Scommentend
 170   ends style a or b.  Comment start sequences can start style a or b.
 171   Style a is always the default.
 172   */
 173
 174 #define SYNTAX_COMSTART_FIRST(c) ((SYNTAX_WITH_FLAGS (c) >> 16) & 1)
 175
 176 #define SYNTAX_COMSTART_SECOND(c) ((SYNTAX_WITH_FLAGS (c) >> 17) & 1)
 177
 178 #define SYNTAX_COMEND_FIRST(c) ((SYNTAX_WITH_FLAGS (c) >> 18) & 1)
 179
 180 #define SYNTAX_COMEND_SECOND(c) ((SYNTAX_WITH_FLAGS (c) >> 19) & 1)
 181
 182 #define SYNTAX_PREFIX(c) ((SYNTAX_WITH_FLAGS (c) >> 20) & 1)
 183
 184 /* extract the comment style bit from the syntax table entry */
 185 #define SYNTAX_COMMENT_STYLE(c) ((SYNTAX_WITH_FLAGS (c) >> 21) & 1)
 186
 187 /* This array, indexed by a character, contains the syntax code which that
 188  character signifies (as a char).  For example,
 189  (enum syntaxcode) syntax_spec_code['w'] is Sword.  */
 190
 191 extern unsigned char syntax_spec_code[0400];
 192
 193 /* Indexed by syntax code, give the letter that describes it.  */
 194
 195 extern char syntax_code_spec[16];
 196
 197 /* Convert the byte offset BYTEPOS into a character position,
 198    for the object recorded in gl_state with SETUP_SYNTAX_TABLE_FOR_OBJECT.  */
 199
 200 #define SYNTAX_TABLE_BYTE_TO_CHAR(bytepos)                              \
 201   (STRINGP (gl_state.object)                                            \
 202    ? string_byte_to_char (gl_state.object, (bytepos))                   \
 203    : BUFFERP (gl_state.object)                                          \
 204    ? buf_bytepos_to_charpos (XBUFFER (gl_state.object), (bytepos))      \
 205    : NILP (gl_state.object)                                             \
 206    ? BYTE_TO_CHAR ((bytepos))                                           \
 207    : (bytepos))
 208
 209 /* Make syntax table state (gl_state) good for POS, assuming it is
 210    currently good for a position before POS.  */
 211
 212 #define UPDATE_SYNTAX_TABLE_FORWARD(pos)                        \
 213   ((pos) >= gl_state.e_property                                 \
 214    ? (update_syntax_table ((pos) + gl_state.offset, 1, 0,       \
 215                            gl_state.object),                    \
 216       1)                                                        \
 217    : 0)
 218
 219 /* Make syntax table state (gl_state) good for POS, assuming it is
 220    currently good for a position after POS.  */
 221
 222 #define UPDATE_SYNTAX_TABLE_BACKWARD(pos)                       \
 223   ((pos) <= gl_state.b_property                                 \
 224    ? (update_syntax_table ((pos) + gl_state.offset, -1, 0,      \
 225                            gl_state.object),                    \
 226       1)                                                        \
 227    : 0)
 228
 229 /* Make syntax table good for POS.  */
 230
 231 #define UPDATE_SYNTAX_TABLE(pos)                                \
 232   ((pos) <= gl_state.b_property                                 \
 233    ? (update_syntax_table ((pos) + gl_state.offset, -1, 0,      \
 234                            gl_state.object),                    \
 235       1)                                                        \
 236    : ((pos) >= gl_state.e_property                              \
 237       ? (update_syntax_table ((pos) + gl_state.offset, 1, 0,    \
 238                               gl_state.object),                 \
 239          1)                                                     \
 240       : 0))
 241
 242 /* This macro should be called with FROM at the start of forward
 243    search, or after the last position of the backward search.  It
 244    makes sure that the first char is picked up with correct table, so
 245    one does not need to call UPDATE_SYNTAX_TABLE immediately after the
 246    call.
 247    Sign of COUNT gives the direction of the search.
 248  */
 249
 250 #define SETUP_SYNTAX_TABLE(FROM, COUNT)                                 \
 251   gl_state.b_property = BEGV - 1;                                       \
 252   gl_state.e_property = ZV + 1;                                         \
 253   gl_state.object = Qnil;                                               \
 254   gl_state.use_global = 0;                                              \
 255   gl_state.offset = 0;                                                  \
 256   gl_state.current_syntax_table = current_buffer->syntax_table;         \
 257   if (parse_sexp_lookup_properties)                                     \
 258     update_syntax_table ((COUNT) > 0 ? (FROM) : (FROM) - 1, (COUNT),    \
 259                          1, Qnil);
 260
 261 /* Same as above, but in OBJECT.  If OBJECT is nil, use current buffer.
 262    If it is t, ignore properties altogether.
 263
 264    This is meant for regex.c to use.  For buffers, regex.c passes arguments
 265    to the UPDATE_SYNTAX_TABLE macros which are relative to BEGV.
 266    So if it is a buffer, we set the offset field to BEGV.  */
 267
 268 #define SETUP_SYNTAX_TABLE_FOR_OBJECT(OBJECT, FROM, COUNT)              \
 269 if (1)                                                                  \
 270   {                                                                     \
 271     gl_state.object = (OBJECT);                                         \
 272     if (BUFFERP (gl_state.object))                                      \
 273       {                                                                 \
 274         struct buffer *buf = XBUFFER (gl_state.object);                 \
 275         gl_state.b_property = 0;                                        \
 276         gl_state.e_property = BUF_ZV (buf) - BUF_BEGV (buf) + 1;        \
 277         gl_state.offset = BUF_BEGV (buf) - 1;                           \
 278       }                                                                 \
 279     else if (NILP (gl_state.object))                                    \
 280       {                                                                 \
 281         gl_state.b_property = 0;                                        \
 282         gl_state.e_property = ZV - BEGV + 1;                            \
 283         gl_state.offset = BEGV - 1;                                     \
 284       }                                                                 \
 285     else if (EQ (gl_state.object, Qt))                                  \
 286       {                                                                 \
 287         gl_state.b_property = - 1;                                      \
 288         gl_state.e_property = 1500000000;                               \
 289         gl_state.offset = 0;                                            \
 290       }                                                                 \
 291     else                                                                \
 292       {                                                                 \
 293         gl_state.b_property = -1;                                       \
 294         gl_state.e_property = 1 + XSTRING (gl_state.object)->size;      \
 295         gl_state.offset = 0;                                            \
 296       }                                                                 \
 297     gl_state.use_global = 0;                                            \
 298     gl_state.current_syntax_table = current_buffer->syntax_table;       \
 299     if (parse_sexp_lookup_properties)                                   \
 300       update_syntax_table ((BYTE_TO_CHAR ((FROM) + gl_state.offset)     \
 301                             + (COUNT > 0 ? 0 :  -1)),                   \
 302                            COUNT, 1, gl_state.object);                  \
 303   }                                                                     \
 304 else
 305
 306 struct gl_state_s
 307 {
 308   Lisp_Object object;                   /* The object we are scanning. */
 309   int start;                            /* Where to stop. */
 310   int stop;                             /* Where to stop. */
 311   int use_global;                       /* Whether to use global_code
 312                                            or c_s_t. */
 313   Lisp_Object global_code;              /* Syntax code of current char. */
 314   Lisp_Object current_syntax_table;     /* Syntax table for current pos. */
 315   Lisp_Object old_prop;                 /* Syntax-table prop at prev pos. */
 316   int b_property;                       /* Last index where c_s_t is
 317                                            not valid. */
 318   int e_property;                       /* First index where c_s_t is
 319                                            not valid. */
 320   INTERVAL forward_i;                   /* Where to start lookup on forward */
 321   INTERVAL backward_i;                  /* or backward movement.  The
 322                                            data in c_s_t is valid
 323                                            between these intervals,
 324                                            and possibly at the
 325                                            intervals too, depending
 326                                            on: */
 327   /* Offset for positions specified to UPDATE_SYNTAX_TABLE.  */
 328   int offset;
 329   char left_ok;
 330   char right_ok;
 331 };
 332
 333 extern struct gl_state_s gl_state;
 334 extern int parse_sexp_lookup_properties;
 335 extern INTERVAL interval_of P_ ((int, Lisp_Object));
 336
 337 extern int scan_words P_ ((int, int));