]> git.cameronkatri.com Git - mandoc.git/blob - chars.c
b32c5bbaa261861de2a1550a33c5eddd1c89eac4
[mandoc.git] / chars.c
1 /* $Id: chars.c,v 1.45 2011/05/24 21:24:16 kristaps Exp $ */
2 /*
3 * Copyright (c) 2009, 2010 Kristaps Dzonsons <kristaps@bsd.lv>
4 * Copyright (c) 2011 Ingo Schwarze <schwarze@openbsd.org>
5 *
6 * Permission to use, copy, modify, and distribute this software for any
7 * purpose with or without fee is hereby granted, provided that the above
8 * copyright notice and this permission notice appear in all copies.
9 *
10 * THE SOFTWARE IS PROVIDED "AS IS" AND THE AUTHOR DISCLAIMS ALL WARRANTIES
11 * WITH REGARD TO THIS SOFTWARE INCLUDING ALL IMPLIED WARRANTIES OF
12 * MERCHANTABILITY AND FITNESS. IN NO EVENT SHALL THE AUTHOR BE LIABLE FOR
13 * ANY SPECIAL, DIRECT, INDIRECT, OR CONSEQUENTIAL DAMAGES OR ANY DAMAGES
14 * WHATSOEVER RESULTING FROM LOSS OF USE, DATA OR PROFITS, WHETHER IN AN
15 * ACTION OF CONTRACT, NEGLIGENCE OR OTHER TORTIOUS ACTION, ARISING OUT OF
16 * OR IN CONNECTION WITH THE USE OR PERFORMANCE OF THIS SOFTWARE.
17 */
18 #ifdef HAVE_CONFIG_H
19 #include "config.h"
20 #endif
21
22 #include <assert.h>
23 #include <ctype.h>
24 #include <stdio.h>
25 #include <stdlib.h>
26 #include <string.h>
27
28 #include "mandoc.h"
29 #include "libmandoc.h"
30
31 #define PRINT_HI 126
32 #define PRINT_LO 32
33
34 struct ln {
35 struct ln *next;
36 const char *code;
37 const char *ascii;
38 int unicode;
39 };
40
41 #define LINES_MAX 325
42
43 #define CHAR(in, ch, code) \
44 { NULL, (in), (ch), (code) },
45
46 #define CHAR_TBL_START static struct ln lines[LINES_MAX] = {
47 #define CHAR_TBL_END };
48
49 #include "chars.in"
50
51 struct mchars {
52 struct ln **htab;
53 };
54
55 static inline int match(const struct ln *, const char *, size_t);
56 static const struct ln *find(struct mchars *, const char *, size_t);
57
58 void
59 mchars_free(struct mchars *arg)
60 {
61
62 free(arg->htab);
63 free(arg);
64 }
65
66 struct mchars *
67 mchars_alloc(void)
68 {
69 struct mchars *tab;
70 struct ln **htab;
71 struct ln *pp;
72 int i, hash;
73
74 /*
75 * Constructs a very basic chaining hashtable. The hash routine
76 * is simply the integral value of the first character.
77 * Subsequent entries are chained in the order they're processed
78 * (they're in-line re-ordered during lookup).
79 */
80
81 tab = mandoc_malloc(sizeof(struct mchars));
82 htab = mandoc_calloc(PRINT_HI - PRINT_LO + 1, sizeof(struct ln **));
83
84 for (i = 0; i < LINES_MAX; i++) {
85 hash = (int)lines[i].code[0] - PRINT_LO;
86
87 if (NULL == (pp = htab[hash])) {
88 htab[hash] = &lines[i];
89 continue;
90 }
91
92 for ( ; pp->next; pp = pp->next)
93 /* Scan ahead. */ ;
94 pp->next = &lines[i];
95 }
96
97 tab->htab = htab;
98 return(tab);
99 }
100
101
102 /*
103 * Special character to Unicode codepoint.
104 */
105 int
106 mchars_spec2cp(struct mchars *arg, const char *p, size_t sz)
107 {
108 const struct ln *ln;
109
110 ln = find(arg, p, sz);
111 if (NULL == ln)
112 return(-1);
113 return(ln->unicode);
114 }
115
116
117 /*
118 * Reserved word to Unicode codepoint.
119 */
120 int
121 mchars_res2cp(struct mchars *arg, const char *p, size_t sz)
122 {
123 const struct ln *ln;
124
125 ln = find(arg, p, sz);
126 if (NULL == ln)
127 return(-1);
128 return(ln->unicode);
129 }
130
131 /*
132 * Numbered character string to ASCII codepoint.
133 * This can only be a printable character (i.e., alnum, punct, space) so
134 * prevent the character from ruining our state (backspace, newline, and
135 * so on).
136 * If the character is illegal, returns '\0'.
137 */
138 char
139 mchars_num2char(const char *p, size_t sz)
140 {
141 int i;
142
143 if ((i = mandoc_strntou(p, sz, 10)) < 0)
144 return('\0');
145 return(isprint(i) ? i : '\0');
146 }
147
148 /*
149 * Hex character string to Unicode codepoint.
150 * If the character is illegal, returns '\0'.
151 */
152 int
153 mchars_num2uc(const char *p, size_t sz)
154 {
155 int i;
156
157 if ((i = mandoc_strntou(p, sz, 16)) < 0)
158 return('\0');
159 /* FIXME: make sure we're not in a bogus range. */
160 return(i > 0x80 && i <= 0x10FFFF ? i : '\0');
161 }
162
163 /*
164 * Special character to string array.
165 */
166 const char *
167 mchars_spec2str(struct mchars *arg, const char *p, size_t sz, size_t *rsz)
168 {
169 const struct ln *ln;
170
171 ln = find(arg, p, sz);
172 if (NULL == ln)
173 return(NULL);
174
175 *rsz = strlen(ln->ascii);
176 return(ln->ascii);
177 }
178
179 /*
180 * Reserved word to string array.
181 */
182 const char *
183 mchars_res2str(struct mchars *arg, const char *p, size_t sz, size_t *rsz)
184 {
185 const struct ln *ln;
186
187 ln = find(arg, p, sz);
188 if (NULL == ln)
189 return(NULL);
190
191 *rsz = strlen(ln->ascii);
192 return(ln->ascii);
193 }
194
195 static const struct ln *
196 find(struct mchars *tab, const char *p, size_t sz)
197 {
198 struct ln *pp, *prev;
199 struct ln **htab;
200 int hash;
201
202 assert(p);
203 if (0 == sz)
204 return(NULL);
205
206 if (p[0] < PRINT_LO || p[0] > PRINT_HI)
207 return(NULL);
208
209 /*
210 * Lookup the symbol in the symbol hash. See ascii2htab for the
211 * hashtable specs. This dynamically re-orders the hash chain
212 * to optimise for repeat hits.
213 */
214
215 hash = (int)p[0] - PRINT_LO;
216 htab = tab->htab;
217
218 if (NULL == (pp = htab[hash]))
219 return(NULL);
220
221 for (prev = NULL; pp; pp = pp->next) {
222 if ( ! match(pp, p, sz)) {
223 prev = pp;
224 continue;
225 }
226
227 if (prev) {
228 prev->next = pp->next;
229 pp->next = htab[hash];
230 htab[hash] = pp;
231 }
232
233 return(pp);
234 }
235
236 return(NULL);
237 }
238
239 static inline int
240 match(const struct ln *ln, const char *p, size_t sz)
241 {
242
243 if (strncmp(ln->code, p, sz))
244 return(0);
245 return('\0' == ln->code[(int)sz]);
246 }