| 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 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 P_ ((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 | /* Then there are seven single-bit flags that have the following meanings: |
| 132 | 1. This character is the first of a two-character comment-start sequence. |
| 133 | 2. This character is the second of a two-character comment-start sequence. |
| 134 | 3. This character is the first of a two-character comment-end sequence. |
| 135 | 4. This character is the second of a two-character comment-end sequence. |
| 136 | 5. This character is a prefix, for backward-prefix-chars. |
| 137 | 6. see below |
| 138 | 7. This character is part of a nestable comment sequence. |
| 139 | Note that any two-character sequence whose first character has flag 1 |
| 140 | and whose second character has flag 2 will be interpreted as a comment start. |
| 141 | |
| 142 | bit 6 is used to discriminate between two different comment styles. |
| 143 | Languages such as C++ allow two orthogonal syntax start/end pairs |
| 144 | and bit 6 is used to determine whether a comment-end or Scommentend |
| 145 | ends style a or b. Comment start sequences can start style a or b. |
| 146 | Style a is always the default. |
| 147 | */ |
| 148 | |
| 149 | /* These macros extract a particular flag for a given character. */ |
| 150 | |
| 151 | #define SYNTAX_COMSTART_FIRST(c) ((SYNTAX_WITH_FLAGS (c) >> 16) & 1) |
| 152 | |
| 153 | #define SYNTAX_COMSTART_SECOND(c) ((SYNTAX_WITH_FLAGS (c) >> 17) & 1) |
| 154 | |
| 155 | #define SYNTAX_COMEND_FIRST(c) ((SYNTAX_WITH_FLAGS (c) >> 18) & 1) |
| 156 | |
| 157 | #define SYNTAX_COMEND_SECOND(c) ((SYNTAX_WITH_FLAGS (c) >> 19) & 1) |
| 158 | |
| 159 | #define SYNTAX_PREFIX(c) ((SYNTAX_WITH_FLAGS (c) >> 20) & 1) |
| 160 | |
| 161 | #define SYNTAX_COMMENT_STYLE(c) ((SYNTAX_WITH_FLAGS (c) >> 21) & 1) |
| 162 | |
| 163 | #define SYNTAX_COMMENT_NESTED(c) ((SYNTAX_WITH_FLAGS (c) >> 22) & 1) |
| 164 | |
| 165 | /* These macros extract specific flags from an integer |
| 166 | that holds the syntax code and the flags. */ |
| 167 | |
| 168 | #define SYNTAX_FLAGS_COMSTART_FIRST(flags) (((flags) >> 16) & 1) |
| 169 | |
| 170 | #define SYNTAX_FLAGS_COMSTART_SECOND(flags) (((flags) >> 17) & 1) |
| 171 | |
| 172 | #define SYNTAX_FLAGS_COMEND_FIRST(flags) (((flags) >> 18) & 1) |
| 173 | |
| 174 | #define SYNTAX_FLAGS_COMEND_SECOND(flags) (((flags) >> 19) & 1) |
| 175 | |
| 176 | #define SYNTAX_FLAGS_PREFIX(flags) (((flags) >> 20) & 1) |
| 177 | |
| 178 | #define SYNTAX_FLAGS_COMMENT_STYLE(flags) (((flags) >> 21) & 1) |
| 179 | |
| 180 | #define SYNTAX_FLAGS_COMMENT_NESTED(flags) (((flags) >> 22) & 1) |
| 181 | |
| 182 | /* This array, indexed by a character, contains the syntax code which that |
| 183 | character signifies (as a char). For example, |
| 184 | (enum syntaxcode) syntax_spec_code['w'] is Sword. */ |
| 185 | |
| 186 | extern unsigned char syntax_spec_code[0400]; |
| 187 | |
| 188 | /* Indexed by syntax code, give the letter that describes it. */ |
| 189 | |
| 190 | extern char syntax_code_spec[16]; |
| 191 | |
| 192 | /* Convert the byte offset BYTEPOS into a character position, |
| 193 | for the object recorded in gl_state with SETUP_SYNTAX_TABLE_FOR_OBJECT. |
| 194 | |
| 195 | The value is meant for use in the UPDATE_SYNTAX_TABLE... macros. |
| 196 | These macros do nothing when parse_sexp_lookup_properties is 0, |
| 197 | so we return 0 in that case, for speed. */ |
| 198 | |
| 199 | #define SYNTAX_TABLE_BYTE_TO_CHAR(bytepos) \ |
| 200 | (! parse_sexp_lookup_properties \ |
| 201 | ? 0 \ |
| 202 | : STRINGP (gl_state.object) \ |
| 203 | ? string_byte_to_char (gl_state.object, (bytepos)) \ |
| 204 | : BUFFERP (gl_state.object) \ |
| 205 | ? buf_bytepos_to_charpos (XBUFFER (gl_state.object), \ |
| 206 | (bytepos) + BUF_BEGV_BYTE (XBUFFER (gl_state.object)) - 1) - BUF_BEGV (XBUFFER (gl_state.object)) + 1 \ |
| 207 | : NILP (gl_state.object) \ |
| 208 | ? BYTE_TO_CHAR ((bytepos) + BEGV_BYTE - 1) - BEGV + 1 \ |
| 209 | : (bytepos)) |
| 210 | |
| 211 | /* Make syntax table state (gl_state) good for CHARPOS, assuming it is |
| 212 | currently good for a position before CHARPOS. */ |
| 213 | |
| 214 | #define UPDATE_SYNTAX_TABLE_FORWARD(charpos) \ |
| 215 | (parse_sexp_lookup_properties \ |
| 216 | && (charpos) >= gl_state.e_property \ |
| 217 | ? (update_syntax_table ((charpos) + gl_state.offset, 1, 0, \ |
| 218 | gl_state.object), \ |
| 219 | 1) \ |
| 220 | : 0) |
| 221 | |
| 222 | /* Make syntax table state (gl_state) good for CHARPOS, assuming it is |
| 223 | currently good for a position after CHARPOS. */ |
| 224 | |
| 225 | #define UPDATE_SYNTAX_TABLE_BACKWARD(charpos) \ |
| 226 | (parse_sexp_lookup_properties \ |
| 227 | && (charpos) < gl_state.b_property \ |
| 228 | ? (update_syntax_table ((charpos) + gl_state.offset, -1, 0, \ |
| 229 | gl_state.object), \ |
| 230 | 1) \ |
| 231 | : 0) |
| 232 | |
| 233 | /* Make syntax table good for CHARPOS. */ |
| 234 | |
| 235 | #define UPDATE_SYNTAX_TABLE(charpos) \ |
| 236 | (parse_sexp_lookup_properties \ |
| 237 | && (charpos) < gl_state.b_property \ |
| 238 | ? (update_syntax_table ((charpos) + gl_state.offset, -1, 0, \ |
| 239 | gl_state.object), \ |
| 240 | 1) \ |
| 241 | : (parse_sexp_lookup_properties \ |
| 242 | && (charpos) >= gl_state.e_property \ |
| 243 | ? (update_syntax_table ((charpos) + gl_state.offset, 1, 0,\ |
| 244 | gl_state.object), \ |
| 245 | 1) \ |
| 246 | : 0)) |
| 247 | |
| 248 | /* This macro should be called with FROM at the start of forward |
| 249 | search, or after the last position of the backward search. It |
| 250 | makes sure that the first char is picked up with correct table, so |
| 251 | one does not need to call UPDATE_SYNTAX_TABLE immediately after the |
| 252 | call. |
| 253 | Sign of COUNT gives the direction of the search. |
| 254 | */ |
| 255 | |
| 256 | #define SETUP_SYNTAX_TABLE(FROM, COUNT) \ |
| 257 | do \ |
| 258 | { \ |
| 259 | gl_state.b_property = BEGV; \ |
| 260 | gl_state.e_property = ZV + 1; \ |
| 261 | gl_state.object = Qnil; \ |
| 262 | gl_state.use_global = 0; \ |
| 263 | gl_state.offset = 0; \ |
| 264 | gl_state.current_syntax_table = current_buffer->syntax_table; \ |
| 265 | if (parse_sexp_lookup_properties) \ |
| 266 | if ((COUNT) > 0 || (FROM) > BEGV) \ |
| 267 | update_syntax_table ((COUNT) > 0 ? (FROM) : (FROM) - 1, (COUNT),\ |
| 268 | 1, Qnil); \ |
| 269 | } \ |
| 270 | while (0) |
| 271 | |
| 272 | /* Same as above, but in OBJECT. If OBJECT is nil, use current buffer. |
| 273 | If it is t, ignore properties altogether. |
| 274 | |
| 275 | This is meant for regex.c to use. For buffers, regex.c passes arguments |
| 276 | to the UPDATE_SYNTAX_TABLE macros which are relative to BEGV. |
| 277 | So if it is a buffer, we set the offset field to BEGV. */ |
| 278 | |
| 279 | #define SETUP_SYNTAX_TABLE_FOR_OBJECT(OBJECT, FROM, COUNT) \ |
| 280 | do \ |
| 281 | { \ |
| 282 | gl_state.object = (OBJECT); \ |
| 283 | if (BUFFERP (gl_state.object)) \ |
| 284 | { \ |
| 285 | struct buffer *buf = XBUFFER (gl_state.object); \ |
| 286 | gl_state.b_property = 1; \ |
| 287 | gl_state.e_property = BUF_ZV (buf) - BUF_BEGV (buf) + 1; \ |
| 288 | gl_state.offset = BUF_BEGV (buf) - 1; \ |
| 289 | } \ |
| 290 | else if (NILP (gl_state.object)) \ |
| 291 | { \ |
| 292 | gl_state.b_property = 1; \ |
| 293 | gl_state.e_property = ZV - BEGV + 1; \ |
| 294 | gl_state.offset = BEGV - 1; \ |
| 295 | } \ |
| 296 | else if (EQ (gl_state.object, Qt)) \ |
| 297 | { \ |
| 298 | gl_state.b_property = 0; \ |
| 299 | gl_state.e_property = 1500000000; \ |
| 300 | gl_state.offset = 0; \ |
| 301 | } \ |
| 302 | else \ |
| 303 | { \ |
| 304 | gl_state.b_property = 0; \ |
| 305 | gl_state.e_property = 1 + SCHARS (gl_state.object); \ |
| 306 | gl_state.offset = 0; \ |
| 307 | } \ |
| 308 | gl_state.use_global = 0; \ |
| 309 | gl_state.current_syntax_table = current_buffer->syntax_table; \ |
| 310 | if (parse_sexp_lookup_properties) \ |
| 311 | update_syntax_table (((FROM) + gl_state.offset \ |
| 312 | + (COUNT > 0 ? 0 : -1)), \ |
| 313 | COUNT, 1, gl_state.object); \ |
| 314 | } \ |
| 315 | while (0) |
| 316 | |
| 317 | struct gl_state_s |
| 318 | { |
| 319 | Lisp_Object object; /* The object we are scanning. */ |
| 320 | int start; /* Where to stop. */ |
| 321 | int stop; /* Where to stop. */ |
| 322 | int use_global; /* Whether to use global_code |
| 323 | or c_s_t. */ |
| 324 | Lisp_Object global_code; /* Syntax code of current char. */ |
| 325 | Lisp_Object current_syntax_table; /* Syntax table for current pos. */ |
| 326 | Lisp_Object old_prop; /* Syntax-table prop at prev pos. */ |
| 327 | int b_property; /* First index where c_s_t is valid. */ |
| 328 | int e_property; /* First index where c_s_t is |
| 329 | not valid. */ |
| 330 | INTERVAL forward_i; /* Where to start lookup on forward */ |
| 331 | INTERVAL backward_i; /* or backward movement. The |
| 332 | data in c_s_t is valid |
| 333 | between these intervals, |
| 334 | and possibly at the |
| 335 | intervals too, depending |
| 336 | on: */ |
| 337 | /* Offset for positions specified to UPDATE_SYNTAX_TABLE. */ |
| 338 | int offset; |
| 339 | }; |
| 340 | |
| 341 | extern struct gl_state_s gl_state; |
| 342 | extern int parse_sexp_lookup_properties; |
| 343 | extern INTERVAL interval_of P_ ((int, Lisp_Object)); |
| 344 | |
| 345 | extern int scan_words P_ ((int, int)); |
| 346 | |
| 347 | /* arch-tag: 28833cca-cd73-4741-8c85-a3111166a0e0 |
| 348 | (do not change this comment) */ |