1 /* Declarations having to do with GNU Emacs syntax tables.
2
3 Copyright (C) 1985, 1993-1994, 1997-1998, 2001-2023 Free Software
4 Foundation, Inc.
5
6 This file is part of GNU Emacs.
7
8 GNU Emacs is free software: you can redistribute it and/or modify
9 it under the terms of the GNU General Public License as published by
10 the Free Software Foundation, either version 3 of the License, or (at
11 your option) any later version.
12
13 GNU Emacs is distributed in the hope that it will be useful,
14 but WITHOUT ANY WARRANTY; without even the implied warranty of
15 MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the
16 GNU General Public License for more details.
17
18 You should have received a copy of the GNU General Public License
19 along with GNU Emacs. If not, see <https://www.gnu.org/licenses/>. */
20
21 #ifndef EMACS_SYNTAX_H
22 #define EMACS_SYNTAX_H
23
24 #include "buffer.h"
25 #include "lisp.h"
26
27 INLINE_HEADER_BEGIN
28
29 extern void update_syntax_table (ptrdiff_t, EMACS_INT, bool, Lisp_Object);
30 extern void update_syntax_table_forward (ptrdiff_t, bool, Lisp_Object);
31
32 /* The standard syntax table is stored where it will automatically
33 be used in all new buffers. */
34 #define Vstandard_syntax_table BVAR (&buffer_defaults, syntax_table)
35
36 /* A syntax table is a chartable whose elements are cons cells
37 (CODE+FLAGS . MATCHING-CHAR). MATCHING-CHAR can be nil if the char
38 is not a kind of parenthesis.
39
40 The low 8 bits of CODE+FLAGS is a code, as follows: */
41
42 enum syntaxcode
43 {
44 Swhitespace, /* for a whitespace character */
45 Spunct, /* for random punctuation characters */
46 Sword, /* for a word constituent */
47 Ssymbol, /* symbol constituent but not word constituent */
48 Sopen, /* for a beginning delimiter */
49 Sclose, /* for an ending delimiter */
50 Squote, /* for a prefix character like Lisp ' */
51 Sstring, /* for a string-grouping character like Lisp " */
52 Smath, /* for delimiters like $ in Tex. */
53 Sescape, /* for a character that begins a C-style escape */
54 Scharquote, /* for a character that quotes the following character */
55 Scomment, /* for a comment-starting character */
56 Sendcomment, /* for a comment-ending character */
57 Sinherit, /* use the standard syntax table for this character */
58 Scomment_fence, /* Starts/ends comment which is delimited on the
59 other side by any char with the same syntaxcode. */
60 Sstring_fence, /* Starts/ends string which is delimited on the
61 other side by any char with the same syntaxcode. */
62 Smax /* Upper bound on codes that are meaningful. */
63 };
64
65
66 struct gl_state_s
67 {
68 Lisp_Object object; /* The object we are scanning. */
69 ptrdiff_t start; /* Where to stop. */
70 ptrdiff_t stop; /* Where to stop. */
71 bool use_global; /* Whether to use global_code
72 or c_s_t. */
73 Lisp_Object global_code; /* Syntax code of current char. */
74 Lisp_Object current_syntax_table; /* Syntax table for current pos. */
75 Lisp_Object old_prop; /* Syntax-table prop at prev pos. */
76 ptrdiff_t b_property; /* First index where c_s_t is valid. */
77 ptrdiff_t e_property; /* First index where c_s_t is
78 not valid. */
79 bool e_property_truncated; /* true if e_property if was truncated
80 by parse_sexp_propertize_done. */
81 INTERVAL forward_i; /* Where to start lookup on forward. */
82 INTERVAL backward_i; /* or backward movement. The
83 data in c_s_t is valid
84 between these intervals,
85 and possibly at the
86 intervals too, depending
87 on: */
88 /* Offset for positions specified to UPDATE_SYNTAX_TABLE. */
89 ptrdiff_t offset;
90 };
91
92 extern struct gl_state_s gl_state;
93
94 /* Fetch the information from the entry for character C
95 in the current buffer's syntax table,
96 or (if VIA_PROPERTY) from globally kept data (gl_state).
97 Does inheritance. */
98
99 INLINE Lisp_Object
100 syntax_property_entry (int c, bool via_property)
101 {
102 if (via_property)
103 return (gl_state.use_global
104 ? gl_state.global_code
105 : CHAR_TABLE_REF (gl_state.current_syntax_table, c));
106 return CHAR_TABLE_REF (BVAR (current_buffer, syntax_table), c);
107 }
108 INLINE Lisp_Object
109 SYNTAX_ENTRY (int c)
110 {
111 return syntax_property_entry (c, false);
112 }
113
114 /* Extract the information from the entry for character C
115 in the current syntax table. */
116
117 INLINE int
118 syntax_property_with_flags (int c, bool via_property)
119 {
120 Lisp_Object ent = syntax_property_entry (c, via_property);
121 return CONSP (ent) ? XFIXNUM (XCAR (ent)) : Swhitespace;
122 }
123 INLINE int
124 SYNTAX_WITH_FLAGS (int c)
125 {
126 return syntax_property_with_flags (c, false);
127 }
128
129 INLINE enum syntaxcode
130 syntax_property (int c, bool via_property)
131 {
132 return syntax_property_with_flags (c, via_property) & 0xff;
133 }
134 INLINE enum syntaxcode
135 SYNTAX (int c)
136 {
137 return syntax_property (c, false);
138 }
139
140
141 /* Whether the syntax of the character C has the prefix flag set. */
142 extern bool syntax_prefix_flag_p (int c);
143
144 /* This array, indexed by a character less than 256, contains the
145 syntax code which that character signifies (as an unsigned char).
146 For example, syntax_spec_code['w'] == Sword. */
147
148 extern unsigned char const syntax_spec_code[0400];
149
150 /* Convert the byte offset BYTEPOS into a character position,
151 for the object recorded in gl_state with SETUP_SYNTAX_TABLE_FOR_OBJECT.
152
153 The value is meant for use in code that does nothing when
154 parse_sexp_lookup_properties is false, so return 0 in that case,
155 for speed. */
156
157 INLINE ptrdiff_t
158 SYNTAX_TABLE_BYTE_TO_CHAR (ptrdiff_t bytepos)
159 {
160 return (! parse_sexp_lookup_properties
161 ? 0
162 : STRINGP (gl_state.object)
163 ? string_byte_to_char (gl_state.object, bytepos)
164 : BUFFERP (gl_state.object)
165 ? ((buf_bytepos_to_charpos
166 (XBUFFER (gl_state.object),
167 (bytepos + BUF_BEGV_BYTE (XBUFFER (gl_state.object)) - 1)))
168 - BUF_BEGV (XBUFFER (gl_state.object)) + 1)
169 : NILP (gl_state.object)
170 ? BYTE_TO_CHAR (bytepos + BEGV_BYTE - 1) - BEGV + 1
171 : bytepos);
172 }
173
174 /* Make syntax table state (gl_state) good for CHARPOS, assuming it is
175 currently good for a position before CHARPOS. */
176
177 INLINE void
178 UPDATE_SYNTAX_TABLE_FORWARD (ptrdiff_t charpos)
179 { /* Performs just-in-time syntax-propertization. */
180 if (parse_sexp_lookup_properties && charpos >= gl_state.e_property)
181 update_syntax_table_forward (charpos + gl_state.offset,
182 false, gl_state.object);
183 }
184
185 /* Make syntax table state (gl_state) good for CHARPOS, assuming it is
186 currently good for a position after CHARPOS. */
187
188 INLINE void
189 UPDATE_SYNTAX_TABLE_BACKWARD (ptrdiff_t charpos)
190 {
191 if (parse_sexp_lookup_properties && charpos < gl_state.b_property)
192 update_syntax_table (charpos + gl_state.offset, -1, false, gl_state.object);
193 }
194
195 /* Make syntax table good for CHARPOS. */
196
197 INLINE void
198 UPDATE_SYNTAX_TABLE (ptrdiff_t charpos)
199 {
200 UPDATE_SYNTAX_TABLE_BACKWARD (charpos);
201 UPDATE_SYNTAX_TABLE_FORWARD (charpos);
202 }
203
204 /* Set up the buffer-global syntax table. */
205
206 INLINE void
207 SETUP_BUFFER_SYNTAX_TABLE (void)
208 {
209 gl_state.use_global = false;
210 gl_state.e_property_truncated = false;
211 gl_state.current_syntax_table = BVAR (current_buffer, syntax_table);
212 }
213
214 extern ptrdiff_t scan_words (ptrdiff_t, EMACS_INT);
215 extern void SETUP_SYNTAX_TABLE_FOR_OBJECT (Lisp_Object, ptrdiff_t, ptrdiff_t);
216
217 INLINE_HEADER_END
218
219 #endif /* EMACS_SYNTAX_H */