1 /* $Id: chars.c,v 1.46 2011/05/24 21:31:23 kristaps Exp $ */
3 * Copyright (c) 2009, 2010 Kristaps Dzonsons <kristaps@bsd.lv>
4 * Copyright (c) 2011 Ingo Schwarze <schwarze@openbsd.org>
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.
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.
29 #include "libmandoc.h"
43 #define CHAR(in, ch, code) \
44 { NULL, (in), (ch), (code) },
46 #define CHAR_TBL_START static struct ln lines[LINES_MAX] = {
47 #define CHAR_TBL_END };
55 static inline int match(const struct ln *, const char *, size_t);
56 static const struct ln *find(struct mchars *, const char *, size_t);
59 mchars_free(struct mchars *arg)
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).
81 tab = mandoc_malloc(sizeof(struct mchars));
82 htab = mandoc_calloc(PRINT_HI - PRINT_LO + 1, sizeof(struct ln **));
84 for (i = 0; i < LINES_MAX; i++) {
85 hash = (int)lines[i].code[0] - PRINT_LO;
87 if (NULL == (pp = htab[hash])) {
88 htab[hash] = &lines[i];
92 for ( ; pp->next; pp = pp->next)
103 * Special character to Unicode codepoint.
106 mchars_spec2cp(struct mchars *arg, const char *p, size_t sz)
110 ln = find(arg, p, sz);
117 * Numbered character string to ASCII codepoint.
118 * This can only be a printable character (i.e., alnum, punct, space) so
119 * prevent the character from ruining our state (backspace, newline, and
121 * If the character is illegal, returns '\0'.
124 mchars_num2char(const char *p, size_t sz)
128 if ((i = mandoc_strntou(p, sz, 10)) < 0)
130 return(isprint(i) ? i : '\0');
134 * Hex character string to Unicode codepoint.
135 * If the character is illegal, returns '\0'.
138 mchars_num2uc(const char *p, size_t sz)
142 if ((i = mandoc_strntou(p, sz, 16)) < 0)
144 /* FIXME: make sure we're not in a bogus range. */
145 return(i > 0x80 && i <= 0x10FFFF ? i : '\0');
149 * Special character to string array.
152 mchars_spec2str(struct mchars *arg, const char *p, size_t sz, size_t *rsz)
156 ln = find(arg, p, sz);
160 *rsz = strlen(ln->ascii);
164 static const struct ln *
165 find(struct mchars *tab, const char *p, size_t sz)
167 struct ln *pp, *prev;
175 if (p[0] < PRINT_LO || p[0] > PRINT_HI)
179 * Lookup the symbol in the symbol hash. See ascii2htab for the
180 * hashtable specs. This dynamically re-orders the hash chain
181 * to optimise for repeat hits.
184 hash = (int)p[0] - PRINT_LO;
187 if (NULL == (pp = htab[hash]))
190 for (prev = NULL; pp; pp = pp->next) {
191 if ( ! match(pp, p, sz)) {
197 prev->next = pp->next;
198 pp->next = htab[hash];
209 match(const struct ln *ln, const char *p, size_t sz)
212 if (strncmp(ln->code, p, sz))
214 return('\0' == ln->code[(int)sz]);