Commit | Line | Data |
---|---|---|
9889c728 | 1 | /* Declarations having to do with GNU Emacs syntax tables. |
5eea1c5a | 2 | Copyright (C) 1985, 1993, 1994, 1997 Free Software Foundation, Inc. |
9889c728 JB |
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 | |
7c938215 | 8 | the Free Software Foundation; either version 2, or (at your option) |
9889c728 JB |
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 | |
3b7ad313 EN |
18 | the Free Software Foundation, Inc., 59 Temple Place - Suite 330, |
19 | Boston, MA 02111-1307, USA. */ | |
9889c728 JB |
20 | |
21 | ||
22 | extern Lisp_Object Qsyntax_table_p; | |
23 | extern Lisp_Object Fsyntax_table_p (), Fsyntax_table (), Fset_syntax_table (); | |
5eea1c5a | 24 | extern void update_syntax_table (); |
9889c728 JB |
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 | ||
e46c910e RS |
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. | |
9889c728 | 33 | |
e46c910e | 34 | The low 8 bits of CODE+FLAGS is a code, as follows: */ |
9889c728 JB |
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 " */ | |
5eea1c5a | 46 | Smath, /* for delimiters like $ in Tex. */ |
9889c728 JB |
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 */ | |
c8cdcb16 | 51 | Sinherit, /* use the standard syntax table for this character */ |
5eea1c5a RS |
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. */ | |
9889c728 JB |
56 | Smax /* Upper bound on codes that are meaningful */ |
57 | }; | |
58 | ||
e0b8ff93 | 59 | /* Set the syntax entry VAL for char C in table TABLE. */ |
e46c910e | 60 | |
0e35bfd8 | 61 | #define SET_RAW_SYNTAX_ENTRY(table, c, val) \ |
9ca6ab7d | 62 | ((c) < CHAR_TABLE_SINGLE_BYTE_SLOTS \ |
0e35bfd8 KH |
63 | ? (XCHAR_TABLE (table)->contents[(unsigned char) (c)] = (val)) \ |
64 | : Faset ((table), make_number (c), (val))) | |
e0b8ff93 KH |
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; }) | |
c8cdcb16 | 82 | #else |
e0b8ff93 KH |
83 | extern Lisp_Object syntax_temp; |
84 | extern Lisp_Object syntax_parent_lookup (); | |
e46c910e | 85 | |
e0b8ff93 KH |
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)) | |
c8cdcb16 | 91 | #endif |
9889c728 | 92 | |
5eea1c5a RS |
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 | |
e0b8ff93 | 108 | |
5eea1c5a | 109 | #define SYNTAX_ENTRY_INT(c) \ |
9ca6ab7d | 110 | ((c) < CHAR_TABLE_SINGLE_BYTE_SLOTS \ |
5eea1c5a | 111 | ? SYNTAX_ENTRY_FOLLOW_PARENT (CURRENT_SYNTAX_TABLE, \ |
0e35bfd8 | 112 | (unsigned char) (c)) \ |
5eea1c5a | 113 | : Faref (CURRENT_SYNTAX_TABLE, make_number ((c)))) |
e0b8ff93 | 114 | |
e46c910e | 115 | /* Extract the information from the entry for character C |
e0b8ff93 | 116 | in the current syntax table. */ |
c8cdcb16 RS |
117 | |
118 | #ifdef __GNUC__ | |
e46c910e RS |
119 | #define SYNTAX(c) \ |
120 | ({ Lisp_Object temp; \ | |
121 | temp = SYNTAX_ENTRY (c); \ | |
122 | (CONSP (temp) \ | |
123 | ? (enum syntaxcode) (XINT (XCONS (temp)->car) & 0xff) \ | |
e0b8ff93 | 124 | : Swhitespace); }) |
e46c910e RS |
125 | |
126 | #define SYNTAX_WITH_FLAGS(c) \ | |
127 | ({ Lisp_Object temp; \ | |
128 | temp = SYNTAX_ENTRY (c); \ | |
129 | (CONSP (temp) \ | |
130 | ? XINT (XCONS (temp)->car) \ | |
e0b8ff93 | 131 | : (int) Swhitespace); }) |
e46c910e RS |
132 | |
133 | #define SYNTAX_MATCH(c) \ | |
134 | ({ Lisp_Object temp; \ | |
135 | temp = SYNTAX_ENTRY (c); \ | |
136 | (CONSP (temp) \ | |
137 | ? XINT (XCONS (temp)->cdr) \ | |
e0b8ff93 | 138 | : Qnil); }) |
c8cdcb16 | 139 | #else |
e46c910e | 140 | #define SYNTAX(c) \ |
9d40ebd2 | 141 | (syntax_temp = SYNTAX_ENTRY ((c)), \ |
e46c910e RS |
142 | (CONSP (syntax_temp) \ |
143 | ? (enum syntaxcode) (XINT (XCONS (syntax_temp)->car) & 0xff) \ | |
e0b8ff93 | 144 | : Swhitespace)) |
e46c910e RS |
145 | |
146 | #define SYNTAX_WITH_FLAGS(c) \ | |
9d40ebd2 | 147 | (syntax_temp = SYNTAX_ENTRY ((c)), \ |
e46c910e RS |
148 | (CONSP (syntax_temp) \ |
149 | ? XINT (XCONS (syntax_temp)->car) \ | |
e0b8ff93 | 150 | : (int) Swhitespace)) |
e46c910e RS |
151 | |
152 | #define SYNTAX_MATCH(c) \ | |
9d40ebd2 | 153 | (syntax_temp = SYNTAX_ENTRY ((c)), \ |
e46c910e RS |
154 | (CONSP (syntax_temp) \ |
155 | ? XINT (XCONS (syntax_temp)->cdr) \ | |
e0b8ff93 | 156 | : Qnil)) |
c8cdcb16 | 157 | #endif |
9889c728 | 158 | |
a306d6f1 | 159 | /* Then there are six single-bit flags that have the following meanings: |
9889c728 JB |
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. | |
a306d6f1 RS |
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 | |
5eea1c5a | 171 | ends style a or b. Comment start sequences can start style a or b. |
a306d6f1 RS |
172 | Style a is always the default. |
173 | */ | |
9889c728 | 174 | |
e46c910e | 175 | #define SYNTAX_COMSTART_FIRST(c) ((SYNTAX_WITH_FLAGS (c) >> 16) & 1) |
c8cdcb16 | 176 | |
e46c910e | 177 | #define SYNTAX_COMSTART_SECOND(c) ((SYNTAX_WITH_FLAGS (c) >> 17) & 1) |
c8cdcb16 | 178 | |
e46c910e | 179 | #define SYNTAX_COMEND_FIRST(c) ((SYNTAX_WITH_FLAGS (c) >> 18) & 1) |
c8cdcb16 | 180 | |
e46c910e | 181 | #define SYNTAX_COMEND_SECOND(c) ((SYNTAX_WITH_FLAGS (c) >> 19) & 1) |
c8cdcb16 | 182 | |
e46c910e | 183 | #define SYNTAX_PREFIX(c) ((SYNTAX_WITH_FLAGS (c) >> 20) & 1) |
9889c728 | 184 | |
a306d6f1 | 185 | /* extract the comment style bit from the syntax table entry */ |
e46c910e | 186 | #define SYNTAX_COMMENT_STYLE(c) ((SYNTAX_WITH_FLAGS (c) >> 21) & 1) |
a306d6f1 | 187 | |
9889c728 JB |
188 | /* This array, indexed by a character, contains the syntax code which that |
189 | character signifies (as a char). For example, | |
5eea1c5a | 190 | (enum syntaxcode) syntax_spec_code['w'] is Sword. */ |
9889c728 JB |
191 | |
192 | extern unsigned char syntax_spec_code[0400]; | |
193 | ||
5eea1c5a RS |
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 | #define UPDATE_SYNTAX_TABLE_FORWARD(pos) \ | |
201 | ((pos) >= gl_state.e_property ? \ | |
202 | ( update_syntax_table ((pos), 1, 0), 1 ) : 0) | |
203 | ||
204 | ||
205 | /* Make syntax table state (gl_state) good for POS, assuming it is | |
206 | currently good for a position after POS. */ | |
207 | #define UPDATE_SYNTAX_TABLE_BACKWARD(pos) \ | |
208 | ((pos) <= gl_state.b_property ? \ | |
209 | ( update_syntax_table ((pos), -1, 0), 1 ) : 0) | |
210 | ||
211 | /* Make syntax table good for POS. */ | |
212 | #define UPDATE_SYNTAX_TABLE(pos) \ | |
213 | ((pos) <= gl_state.b_property ? \ | |
214 | ( update_syntax_table ((pos), -1, 0), 1 ) : \ | |
215 | ( (pos) >= gl_state.e_property ? \ | |
216 | ( update_syntax_table ((pos), 1, 0), 1 ) : 0)) | |
217 | ||
218 | /* This macro should be called with FROM at the start of forward | |
219 | search, or after the last position of the backward search. It | |
220 | makes sure that the first char is picked up with correct table, so | |
221 | one does not need to call UPDATE_SYNTAX_TABLE immediately after the | |
222 | call. | |
223 | Sign of COUNT gives the direction of the search. | |
224 | */ | |
225 | ||
226 | #define SETUP_SYNTAX_TABLE(from,count) \ | |
227 | gl_state.b_property = BEGV - 1; \ | |
228 | gl_state.e_property = ZV + 1; \ | |
229 | gl_state.use_global = 0; \ | |
230 | gl_state.current_syntax_table = current_buffer->syntax_table; \ | |
231 | if (parse_sexp_lookup_properties) \ | |
232 | update_syntax_table ((count) > 0 ? (from) : (from) - 1, (count), 1, Qnil); | |
233 | ||
234 | /* Same as above, but in OBJECT. If OBJECT is nil, use current buffer. | |
235 | If it is t, ignore properties altogether. */ | |
236 | ||
237 | #define SETUP_SYNTAX_TABLE_FOR_OBJECT(object, from, count) \ | |
238 | if (BUFFERP (object)) \ | |
239 | { \ | |
240 | gl_state.b_property = BEGV - 1; \ | |
241 | gl_state.e_property = ZV; \ | |
242 | } \ | |
243 | else if (EQ (object, Qt)) \ | |
244 | { \ | |
245 | gl_state.b_property = - 1; \ | |
246 | gl_state.e_property = 1500000000; \ | |
247 | } \ | |
248 | else \ | |
249 | { \ | |
250 | gl_state.b_property = -1; \ | |
251 | gl_state.e_property = 1 + XSTRING (object)->size; \ | |
252 | } \ | |
253 | gl_state.use_global = 0; \ | |
254 | gl_state.current_syntax_table = current_buffer->syntax_table; \ | |
255 | if (parse_sexp_lookup_properties) \ | |
256 | update_syntax_table (count > 0 ? (from) : (from) - 1, count, 1, object); | |
257 | ||
258 | struct gl_state_s | |
259 | { | |
260 | int start; /* Where to stop. */ | |
261 | int stop; /* Where to stop. */ | |
262 | int use_global; /* Whether to use global_code | |
263 | or c_s_t. */ | |
264 | Lisp_Object global_code; /* Syntax code of current char. */ | |
265 | Lisp_Object current_syntax_table; /* Syntax table for current pos. */ | |
266 | Lisp_Object old_prop; /* Syntax-table prop at prev pos. */ | |
267 | int b_property; /* Last index where c_s_t is | |
268 | not valid. */ | |
269 | int e_property; /* First index where c_s_t is | |
270 | not valid. */ | |
271 | INTERVAL forward_i; /* Where to start lookup on forward */ | |
272 | INTERVAL backward_i; /* or backward movement. The | |
273 | data in c_s_t is valid | |
274 | between these intervals, | |
275 | and possibly at the | |
276 | intervals too, depending | |
277 | on: */ | |
278 | char left_ok; | |
279 | char right_ok; | |
280 | }; | |
281 | ||
282 | extern struct gl_state_s gl_state; | |
283 | extern int parse_sexp_lookup_properties; | |
284 | extern INTERVAL interval_of(); |