src/syntax.h

   1 /* Declarations having to do with GNU Emacs syntax tables.
   2    Copyright (C) 1985, 1993, 1994, 1997, 1998, 2001, 2002, 2003, 2004,
   3                  2005, 2006, 2007, 2008, 2009, 2010, 2011  Free Software Foundation, Inc.
   4
   5 This file is part of GNU Emacs.
   6
   7 GNU Emacs is free software: you can redistribute it and/or modify
   8 it under the terms of the GNU General Public License as published by
   9 the Free Software Foundation, either version 3 of the License, or
  10 (at your option) any later version.
  11
  12 GNU Emacs is distributed in the hope that it will be useful,
  13 but WITHOUT ANY WARRANTY; without even the implied warranty of
  14 MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the
  15 GNU General Public License for more details.
  16
  17 You should have received a copy of the GNU General Public License
  18 along with GNU Emacs.  If not, see <http://www.gnu.org/licenses/>.  */
  19
  20
  21 extern Lisp_Object Qsyntax_table_p;
  22 extern void update_syntax_table (EMACS_INT, int, int, Lisp_Object);
  23
  24 /* The standard syntax table is stored where it will automatically
  25    be used in all new buffers.  */
  26 #define Vstandard_syntax_table buffer_defaults.syntax_table
  27
  28 /* A syntax table is a chartable whose elements are cons cells
  29    (CODE+FLAGS . MATCHING-CHAR).  MATCHING-CHAR can be nil if the char
  30    is not a kind of parenthesis.
  31
  32    The low 8 bits of CODE+FLAGS is a code, as follows:  */
  33
  34 enum syntaxcode
  35   {
  36     Swhitespace, /* for a whitespace character */
  37     Spunct,      /* for random punctuation characters */
  38     Sword,       /* for a word constituent */
  39     Ssymbol,     /* symbol constituent but not word constituent */
  40     Sopen,       /* for a beginning delimiter */
  41     Sclose,      /* for an ending delimiter */
  42     Squote,      /* for a prefix character like Lisp ' */
  43     Sstring,     /* for a string-grouping character like Lisp " */
  44     Smath,       /* for delimiters like $ in Tex.  */
  45     Sescape,     /* for a character that begins a C-style escape */
  46     Scharquote,  /* for a character that quotes the following character */
  47     Scomment,    /* for a comment-starting character */
  48     Sendcomment, /* for a comment-ending character */
  49     Sinherit,    /* use the standard syntax table for this character */
  50     Scomment_fence, /* Starts/ends comment which is delimited on the
  51                        other side by any char with the same syntaxcode.  */
  52     Sstring_fence,  /* Starts/ends string which is delimited on the
  53                        other side by any char with the same syntaxcode.  */
  54     Smax         /* Upper bound on codes that are meaningful */
  55   };
  56
  57 /* Set the syntax entry VAL for char C in table TABLE.  */
  58
  59 #define SET_RAW_SYNTAX_ENTRY(table, c, val)     \
  60   CHAR_TABLE_SET ((table), c, (val))
  61
  62 /* Set the syntax entry VAL for char-range RANGE in table TABLE.
  63    RANGE is a cons (FROM . TO) specifying the range of characters.  */
  64
  65 #define SET_RAW_SYNTAX_ENTRY_RANGE(table, range, val)   \
  66   Fset_char_table_range ((table), (range), (val))
  67
  68 /* SYNTAX_ENTRY fetches the information from the entry for character C
  69    in syntax table TABLE, or from globally kept data (gl_state).
  70    Does inheritance.  */
  71 /* CURRENT_SYNTAX_TABLE gives the syntax table valid for current
  72    position, it is either the buffer's syntax table, or syntax table
  73    found in text properties.  */
  74
  75 #ifdef SYNTAX_ENTRY_VIA_PROPERTY
  76 #  define SYNTAX_ENTRY(c)                                             \
  77     (gl_state.use_global ? gl_state.global_code : SYNTAX_ENTRY_INT (c))
  78 #  define CURRENT_SYNTAX_TABLE gl_state.current_syntax_table
  79 #else
  80 #  define SYNTAX_ENTRY SYNTAX_ENTRY_INT
  81 #  define CURRENT_SYNTAX_TABLE current_buffer->syntax_table
  82 #endif
  83
  84 #define SYNTAX_ENTRY_INT(c) CHAR_TABLE_REF (CURRENT_SYNTAX_TABLE, (c))
  85
  86 /* Extract the information from the entry for character C
  87    in the current syntax table.  */
  88
  89 #ifdef __GNUC__
  90 #define SYNTAX(c)                                                       \
  91   ({ Lisp_Object _syntax_temp;                                          \
  92      _syntax_temp = SYNTAX_ENTRY (c);                                   \
  93      (CONSP (_syntax_temp)                                              \
  94       ? (enum syntaxcode) (XINT (XCAR (_syntax_temp)) & 0xff)           \
  95       : Swhitespace); })
  96
  97 #define SYNTAX_WITH_FLAGS(c)                                            \
  98   ({ Lisp_Object _syntax_temp;                                          \
  99      _syntax_temp = SYNTAX_ENTRY (c);                                   \
 100      (CONSP (_syntax_temp)                                              \
 101       ? XINT (XCAR (_syntax_temp))                                      \
 102       : (int) Swhitespace); })
 103
 104 #define SYNTAX_MATCH(c)                                                 \
 105   ({ Lisp_Object _syntax_temp;                                          \
 106      _syntax_temp = SYNTAX_ENTRY (c);                                   \
 107      (CONSP (_syntax_temp)                                              \
 108       ? XCDR (_syntax_temp)                                             \
 109       : Qnil); })
 110 #else
 111 extern Lisp_Object syntax_temp;
 112 #define SYNTAX(c)                                                       \
 113   (syntax_temp = SYNTAX_ENTRY ((c)),                                    \
 114    (CONSP (syntax_temp)                                                 \
 115     ? (enum syntaxcode) (XINT (XCAR (syntax_temp)) & 0xff)      \
 116     : Swhitespace))
 117
 118 #define SYNTAX_WITH_FLAGS(c)                                            \
 119   (syntax_temp = SYNTAX_ENTRY ((c)),                                    \
 120    (CONSP (syntax_temp)                                                 \
 121     ? XINT (XCAR (syntax_temp))                                 \
 122     : (int) Swhitespace))
 123
 124 #define SYNTAX_MATCH(c)                                                 \
 125   (syntax_temp = SYNTAX_ENTRY ((c)),                                    \
 126    (CONSP (syntax_temp)                                                 \
 127     ? XCDR (syntax_temp)                                                \
 128     : Qnil))
 129 #endif
 130
 131
 132 /* Whether the syntax of the character C has the prefix flag set.  */
 133 extern int syntax_prefix_flag_p (int c);
 134
 135 /* This array, indexed by a character, contains the syntax code which that
 136  character signifies (as a char).  For example,
 137  (enum syntaxcode) syntax_spec_code['w'] is Sword.  */
 138
 139 extern unsigned char syntax_spec_code[0400];
 140
 141 /* Indexed by syntax code, give the letter that describes it.  */
 142
 143 extern char syntax_code_spec[16];
 144
 145 /* Convert the byte offset BYTEPOS into a character position,
 146    for the object recorded in gl_state with SETUP_SYNTAX_TABLE_FOR_OBJECT.
 147
 148    The value is meant for use in the UPDATE_SYNTAX_TABLE... macros.
 149    These macros do nothing when parse_sexp_lookup_properties is 0,
 150    so we return 0 in that case, for speed.  */
 151
 152 #define SYNTAX_TABLE_BYTE_TO_CHAR(bytepos)                              \
 153   (! parse_sexp_lookup_properties                                       \
 154    ? 0                                                                  \
 155    : STRINGP (gl_state.object)                                          \
 156    ? string_byte_to_char (gl_state.object, (bytepos))                   \
 157    : BUFFERP (gl_state.object)                                          \
 158    ? buf_bytepos_to_charpos (XBUFFER (gl_state.object),                 \
 159                              (bytepos) + BUF_BEGV_BYTE (XBUFFER (gl_state.object)) - 1) - BUF_BEGV (XBUFFER (gl_state.object)) + 1      \
 160    : NILP (gl_state.object)                                             \
 161    ? BYTE_TO_CHAR ((bytepos) + BEGV_BYTE - 1) - BEGV + 1                \
 162    : (bytepos))
 163
 164 /* Make syntax table state (gl_state) good for CHARPOS, assuming it is
 165    currently good for a position before CHARPOS.  */
 166
 167 #define UPDATE_SYNTAX_TABLE_FORWARD(charpos)                    \
 168   (parse_sexp_lookup_properties                                 \
 169    && (charpos) >= gl_state.e_property                          \
 170    ? (update_syntax_table ((charpos) + gl_state.offset, 1, 0,   \
 171                            gl_state.object),                    \
 172       1)                                                        \
 173    : 0)
 174
 175 /* Make syntax table state (gl_state) good for CHARPOS, assuming it is
 176    currently good for a position after CHARPOS.  */
 177
 178 #define UPDATE_SYNTAX_TABLE_BACKWARD(charpos)                   \
 179   (parse_sexp_lookup_properties                                 \
 180    && (charpos) < gl_state.b_property                           \
 181    ? (update_syntax_table ((charpos) + gl_state.offset, -1, 0,  \
 182                            gl_state.object),                    \
 183       1)                                                        \
 184    : 0)
 185
 186 /* Make syntax table good for CHARPOS.  */
 187
 188 #define UPDATE_SYNTAX_TABLE(charpos)                            \
 189   (parse_sexp_lookup_properties                                 \
 190    && (charpos) < gl_state.b_property                           \
 191    ? (update_syntax_table ((charpos) + gl_state.offset, -1, 0,  \
 192                            gl_state.object),                    \
 193       1)                                                        \
 194    : (parse_sexp_lookup_properties                              \
 195       && (charpos) >= gl_state.e_property                       \
 196       ? (update_syntax_table ((charpos) + gl_state.offset, 1, 0,\
 197                               gl_state.object),                 \
 198          1)                                                     \
 199       : 0))
 200
 201 /* This macro sets up the buffer-global syntax table.  */
 202 #define SETUP_BUFFER_SYNTAX_TABLE()                                     \
 203 do                                                                      \
 204   {                                                                     \
 205     gl_state.use_global = 0;                                            \
 206     gl_state.current_syntax_table = current_buffer->syntax_table;       \
 207   } while (0)
 208
 209 /* This macro should be called with FROM at the start of forward
 210    search, or after the last position of the backward search.  It
 211    makes sure that the first char is picked up with correct table, so
 212    one does not need to call UPDATE_SYNTAX_TABLE immediately after the
 213    call.
 214    Sign of COUNT gives the direction of the search.
 215  */
 216
 217 #define SETUP_SYNTAX_TABLE(FROM, COUNT)                                 \
 218 do                                                                      \
 219   {                                                                     \
 220     SETUP_BUFFER_SYNTAX_TABLE ();                                       \
 221     gl_state.b_property = BEGV;                                         \
 222     gl_state.e_property = ZV + 1;                                       \
 223     gl_state.object = Qnil;                                             \
 224     gl_state.offset = 0;                                                \
 225     if (parse_sexp_lookup_properties)                                   \
 226       if ((COUNT) > 0 || (FROM) > BEGV)                                 \
 227         update_syntax_table ((COUNT) > 0 ? (FROM) : (FROM) - 1, (COUNT),\
 228                              1, Qnil);                                  \
 229   }                                                                     \
 230 while (0)
 231
 232 /* Same as above, but in OBJECT.  If OBJECT is nil, use current buffer.
 233    If it is t, ignore properties altogether.
 234
 235    This is meant for regex.c to use.  For buffers, regex.c passes arguments
 236    to the UPDATE_SYNTAX_TABLE macros which are relative to BEGV.
 237    So if it is a buffer, we set the offset field to BEGV.  */
 238
 239 #define SETUP_SYNTAX_TABLE_FOR_OBJECT(OBJECT, FROM, COUNT)              \
 240 do                                                                      \
 241   {                                                                     \
 242     SETUP_BUFFER_SYNTAX_TABLE ();                                       \
 243     gl_state.object = (OBJECT);                                         \
 244     if (BUFFERP (gl_state.object))                                      \
 245       {                                                                 \
 246         struct buffer *buf = XBUFFER (gl_state.object);                 \
 247         gl_state.b_property = 1;                                        \
 248         gl_state.e_property = BUF_ZV (buf) - BUF_BEGV (buf) + 1;        \
 249         gl_state.offset = BUF_BEGV (buf) - 1;                           \
 250       }                                                                 \
 251     else if (NILP (gl_state.object))                                    \
 252       {                                                                 \
 253         gl_state.b_property = 1;                                        \
 254         gl_state.e_property = ZV - BEGV + 1;                            \
 255         gl_state.offset = BEGV - 1;                                     \
 256       }                                                                 \
 257     else if (EQ (gl_state.object, Qt))                                  \
 258       {                                                                 \
 259         gl_state.b_property = 0;                                        \
 260         gl_state.e_property = 1500000000;                               \
 261         gl_state.offset = 0;                                            \
 262       }                                                                 \
 263     else                                                                \
 264       {                                                                 \
 265         gl_state.b_property = 0;                                        \
 266         gl_state.e_property = 1 + SCHARS (gl_state.object);             \
 267         gl_state.offset = 0;                                            \
 268       }                                                                 \
 269     if (parse_sexp_lookup_properties)                                   \
 270       update_syntax_table (((FROM) + gl_state.offset                    \
 271                             + (COUNT > 0 ? 0 :  -1)),                   \
 272                            COUNT, 1, gl_state.object);                  \
 273   }                                                                     \
 274 while (0)
 275
 276 struct gl_state_s
 277 {
 278   Lisp_Object object;                   /* The object we are scanning. */
 279   int start;                            /* Where to stop. */
 280   int stop;                             /* Where to stop. */
 281   int use_global;                       /* Whether to use global_code
 282                                            or c_s_t. */
 283   Lisp_Object global_code;              /* Syntax code of current char. */
 284   Lisp_Object current_syntax_table;     /* Syntax table for current pos. */
 285   Lisp_Object old_prop;                 /* Syntax-table prop at prev pos. */
 286   int b_property;                       /* First index where c_s_t is valid. */
 287   int e_property;                       /* First index where c_s_t is
 288                                            not valid. */
 289   INTERVAL forward_i;                   /* Where to start lookup on forward */
 290   INTERVAL backward_i;                  /* or backward movement.  The
 291                                            data in c_s_t is valid
 292                                            between these intervals,
 293                                            and possibly at the
 294                                            intervals too, depending
 295                                            on: */
 296   /* Offset for positions specified to UPDATE_SYNTAX_TABLE.  */
 297   int offset;
 298 };
 299
 300 extern struct gl_state_s gl_state;
 301 extern int parse_sexp_lookup_properties;
 302
 303 extern EMACS_INT scan_words (EMACS_INT, EMACS_INT);
 304