src/syntax.h

   1 /* Declarations having to do with GNU Emacs syntax tables.
   2    Copyright (C) 1985, 1993, 1994, 1997 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 Lisp_Object Fsyntax_table_p (), Fsyntax_table (), Fset_syntax_table ();
  24 extern void update_syntax_table ();
  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 a char with the same syntaxcode.  */
  54     Sstring_fence,  /* Starts/ends string which is delimited on the
  55                        other side by a 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   ((c) < CHAR_TABLE_SINGLE_BYTE_SLOTS                                   \
  63    ? (XCHAR_TABLE (table)->contents[(unsigned char) (c)] = (val))       \
  64    : Faset ((table), make_number (c), (val)))
  65
  66 /* Fetch the syntax entry for char C in syntax table TABLE.
  67    This macro is called only when C is less than CHAR_TABLE_ORDINARY_SLOTS.
  68    Do inheritance.  */
  69
  70 #ifdef __GNUC__
  71 #define SYNTAX_ENTRY_FOLLOW_PARENT(table, c)                    \
  72   ({ Lisp_Object tbl = table;                                   \
  73      Lisp_Object temp = XCHAR_TABLE (tbl)->contents[(c)];       \
  74      while (NILP (temp))                                        \
  75        {                                                        \
  76          tbl = XCHAR_TABLE (tbl)->parent;                       \
  77          if (NILP (tbl))                                        \
  78            break;                                               \
  79          temp = XCHAR_TABLE (tbl)->contents[(c)];               \
  80        }                                                        \
  81      temp; })
  82 #else
  83 extern Lisp_Object syntax_temp;
  84 extern Lisp_Object syntax_parent_lookup ();
  85
  86 #define SYNTAX_ENTRY_FOLLOW_PARENT(table, c)            \
  87   (syntax_temp = XCHAR_TABLE (table)->contents[(c)],    \
  88    (NILP (syntax_temp)                                  \
  89     ? syntax_parent_lookup (table, (c))                 \
  90     : syntax_temp))
  91 #endif
  92
  93 /* SYNTAX_ENTRY fetches the information from the entry for character C
  94    in syntax table TABLE, or from globally kept data (gl_state).
  95    Does inheritance.  */
  96 /* CURRENT_SYNTAX_TABLE gives the syntax table valid for current
  97    position, it is either the buffer's syntax table, or syntax table
  98    found in text properties.  */
  99
 100 #ifdef SYNTAX_ENTRY_VIA_PROPERTY
 101 #  define SYNTAX_ENTRY(c)                                             \
 102     (gl_state.use_global ? gl_state.global_code : SYNTAX_ENTRY_INT (c))
 103 #  define CURRENT_SYNTAX_TABLE gl_state.current_syntax_table
 104 #else
 105 #  define SYNTAX_ENTRY SYNTAX_ENTRY_INT
 106 #  define CURRENT_SYNTAX_TABLE current_buffer->syntax_table
 107 #endif
 108
 109 #define SYNTAX_ENTRY_INT(c)                                             \
 110   ((c) < CHAR_TABLE_SINGLE_BYTE_SLOTS                           \
 111    ? SYNTAX_ENTRY_FOLLOW_PARENT (CURRENT_SYNTAX_TABLE,  \
 112                                  (unsigned char) (c))           \
 113    : Faref (CURRENT_SYNTAX_TABLE, 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 (XCONS (temp)->car) & 0xff)             \
 124       : Swhitespace); })
 125
 126 #define SYNTAX_WITH_FLAGS(c)                                            \
 127   ({ Lisp_Object temp;                                                  \
 128      temp = SYNTAX_ENTRY (c);                                           \
 129      (CONSP (temp)                                                      \
 130       ? XINT (XCONS (temp)->car)                                        \
 131       : (int) Swhitespace); })
 132
 133 #define SYNTAX_MATCH(c)                                                 \
 134   ({ Lisp_Object temp;                                                  \
 135      temp = SYNTAX_ENTRY (c);                                           \
 136      (CONSP (temp)                                                      \
 137       ? XCONS (temp)->cdr                                               \
 138       : Qnil); })
 139 #else
 140 #define SYNTAX(c)                                                       \
 141   (syntax_temp = SYNTAX_ENTRY ((c)),                                    \
 142    (CONSP (syntax_temp)                                                 \
 143     ? (enum syntaxcode) (XINT (XCONS (syntax_temp)->car) & 0xff)        \
 144     : Swhitespace))
 145
 146 #define SYNTAX_WITH_FLAGS(c)                                            \
 147   (syntax_temp = SYNTAX_ENTRY ((c)),                                    \
 148    (CONSP (syntax_temp)                                                 \
 149     ? XINT (XCONS (syntax_temp)->car)                                   \
 150     : (int) Swhitespace))
 151
 152 #define SYNTAX_MATCH(c)                                                 \
 153   (syntax_temp = SYNTAX_ENTRY ((c)),                                    \
 154    (CONSP (syntax_temp)                                                 \
 155     ? XCONS (syntax_temp)->cdr                                          \
 156     : Qnil))
 157 #endif
 158
 159 /* Then there are six 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   Note that any two-character sequence whose first character has flag 1
 166   and whose second character has flag 2 will be interpreted as a comment start.
 167
 168   bit 6 is used to discriminate between two different comment styles.
 169   Languages such as C++ allow two orthogonal syntax start/end pairs
 170   and bit 6 is used to determine whether a comment-end or Scommentend
 171   ends style a or b.  Comment start sequences can start style a or b.
 172   Style a is always the default.
 173   */
 174
 175 #define SYNTAX_COMSTART_FIRST(c) ((SYNTAX_WITH_FLAGS (c) >> 16) & 1)
 176
 177 #define SYNTAX_COMSTART_SECOND(c) ((SYNTAX_WITH_FLAGS (c) >> 17) & 1)
 178
 179 #define SYNTAX_COMEND_FIRST(c) ((SYNTAX_WITH_FLAGS (c) >> 18) & 1)
 180
 181 #define SYNTAX_COMEND_SECOND(c) ((SYNTAX_WITH_FLAGS (c) >> 19) & 1)
 182
 183 #define SYNTAX_PREFIX(c) ((SYNTAX_WITH_FLAGS (c) >> 20) & 1)
 184
 185 /* extract the comment style bit from the syntax table entry */
 186 #define SYNTAX_COMMENT_STYLE(c) ((SYNTAX_WITH_FLAGS (c) >> 21) & 1)
 187
 188 /* This array, indexed by a character, contains the syntax code which that
 189  character signifies (as a char).  For example,
 190  (enum syntaxcode) syntax_spec_code['w'] is Sword.  */
 191
 192 extern unsigned char syntax_spec_code[0400];
 193
 194 /* Indexed by syntax code, give the letter that describes it.  */
 195
 196 extern char syntax_code_spec[16];
 197
 198 /* Make syntax table state (gl_state) good for POS, assuming it is
 199    currently good for a position before POS.  */
 200
 201 #define UPDATE_SYNTAX_TABLE_FORWARD(pos)                \
 202   ((pos) >= gl_state.e_property - gl_state.offset       \
 203    ? (update_syntax_table ((pos) + gl_state.offset, 1, 0), 1) : 0)
 204
 205 /* Make syntax table state (gl_state) good for POS, assuming it is
 206    currently good for a position after POS.  */
 207
 208 #define UPDATE_SYNTAX_TABLE_BACKWARD(pos)               \
 209   ((pos) <= gl_state.b_property - gl_state.offset       \
 210    ? (update_syntax_table ((pos) + gl_state.offset, -1, 0), 1) : 0)
 211
 212 /* Make syntax table good for POS.  */
 213
 214 #define UPDATE_SYNTAX_TABLE(pos)                                        \
 215   ((pos) <= gl_state.b_property - gl_state.offset                       \
 216    ? (update_syntax_table ((pos) + gl_state.offset, -1, 0), 1)          \
 217    : ((pos) >= gl_state.e_property - gl_state.offset                    \
 218       ? (update_syntax_table ((pos) + gl_state.offset, 1, 0), 1) : 0))
 219
 220 /* This macro should be called with FROM at the start of forward
 221    search, or after the last position of the backward search.  It
 222    makes sure that the first char is picked up with correct table, so
 223    one does not need to call UPDATE_SYNTAX_TABLE immediately after the
 224    call.
 225    Sign of COUNT gives the direction of the search.
 226  */
 227
 228 #define SETUP_SYNTAX_TABLE(from,count)                                  \
 229   gl_state.b_property = BEGV - 1;                                       \
 230   gl_state.e_property = ZV + 1;                                         \
 231   gl_state.use_global = 0;                                              \
 232   gl_state.offset = 0;                                                  \
 233   gl_state.current_syntax_table = current_buffer->syntax_table;         \
 234   if (parse_sexp_lookup_properties)                                     \
 235     update_syntax_table ((count) > 0 ? (from) : (from) - 1, (count),    \
 236                          1, Qnil);
 237
 238 /* Same as above, but in OBJECT.  If OBJECT is nil, use current buffer.
 239    If it is t, ignore properties altogether.
 240
 241    This is meant for regex.c to use.  For buffers, regex.c passes arguments
 242    to the UPDATE_SYNTAX_TABLE macros which are relative to BEGV.
 243    So if it is a buffer,a we set the offset field to BEGV.  */
 244
 245 #define SETUP_SYNTAX_TABLE_FOR_OBJECT(object, from, count)              \
 246   if (BUFFERP (object) || NILP (object))                                \
 247     {                                                                   \
 248       gl_state.b_property = BEGV - 1;                                   \
 249       gl_state.e_property = ZV;                                         \
 250       gl_state.offset = BEGV;                                           \
 251     }                                                                   \
 252   else if (EQ (object, Qt))                                             \
 253     {                                                                   \
 254       gl_state.b_property = - 1;                                        \
 255       gl_state.e_property = 1500000000;                                 \
 256       gl_state.offset = 0;                                              \
 257     }                                                                   \
 258   else                                                                  \
 259     {                                                                   \
 260       gl_state.b_property = -1;                                         \
 261       gl_state.e_property = 1 + XSTRING (object)->size;                 \
 262       gl_state.offset = 0;                                              \
 263     }                                                                   \
 264   gl_state.use_global = 0;                                              \
 265   gl_state.current_syntax_table = current_buffer->syntax_table;         \
 266   if (parse_sexp_lookup_properties)                                     \
 267       update_syntax_table (count > 0 ? (from) : (from) - 1, count, 1, object);
 268
 269 struct gl_state_s
 270 {
 271   int start;                            /* Where to stop. */
 272   int stop;                             /* Where to stop. */
 273   int use_global;                       /* Whether to use global_code
 274                                            or c_s_t. */
 275   Lisp_Object global_code;              /* Syntax code of current char. */
 276   Lisp_Object current_syntax_table;     /* Syntax table for current pos. */
 277   Lisp_Object old_prop;                 /* Syntax-table prop at prev pos. */
 278   int b_property;                       /* Last index where c_s_t is
 279                                            not valid. */
 280   int e_property;                       /* First index where c_s_t is
 281                                            not valid. */
 282   INTERVAL forward_i;                   /* Where to start lookup on forward */
 283   INTERVAL backward_i;                  /* or backward movement.  The
 284                                            data in c_s_t is valid
 285                                            between these intervals,
 286                                            and possibly at the
 287                                            intervals too, depending
 288                                            on: */
 289   /* Offset for positions specified to UPDATE_SYNTAX_TABLE.  */
 290   int offset;
 291   char left_ok;
 292   char right_ok;
 293 };
 294
 295 extern struct gl_state_s gl_state;
 296 extern int parse_sexp_lookup_properties;
 297 extern INTERVAL interval_of ();