Annotation of mandoc/chars.c, Revision 1.43
1.43 ! kristaps 1: /* $Id: chars.c,v 1.42 2011/05/15 15:47:46 kristaps Exp $ */
1.1 kristaps 2: /*
1.25 schwarze 3: * Copyright (c) 2009, 2010 Kristaps Dzonsons <kristaps@bsd.lv>
1.32 schwarze 4: * Copyright (c) 2011 Ingo Schwarze <schwarze@openbsd.org>
1.1 kristaps 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: */
1.14 kristaps 18: #ifdef HAVE_CONFIG_H
19: #include "config.h"
20: #endif
21:
1.1 kristaps 22: #include <assert.h>
1.40 kristaps 23: #include <ctype.h>
1.10 kristaps 24: #include <stdio.h>
1.1 kristaps 25: #include <stdlib.h>
26: #include <string.h>
27:
1.18 kristaps 28: #include "mandoc.h"
1.43 ! kristaps 29: #include "libmandoc.h"
1.1 kristaps 30:
1.2 kristaps 31: #define PRINT_HI 126
32: #define PRINT_LO 32
1.1 kristaps 33:
34: struct ln {
35: struct ln *next;
36: const char *code;
1.2 kristaps 37: const char *ascii;
1.21 kristaps 38: int unicode;
1.1 kristaps 39: int type;
40: #define CHARS_CHAR (1 << 0)
41: #define CHARS_STRING (1 << 1)
1.12 kristaps 42: #define CHARS_BOTH (CHARS_CHAR | CHARS_STRING)
1.1 kristaps 43: };
44:
1.35 kristaps 45: #define LINES_MAX 353
1.1 kristaps 46:
1.24 kristaps 47: #define CHAR(in, ch, code) \
48: { NULL, (in), (ch), (code), CHARS_CHAR },
49: #define STRING(in, ch, code) \
50: { NULL, (in), (ch), (code), CHARS_STRING },
51: #define BOTH(in, ch, code) \
52: { NULL, (in), (ch), (code), CHARS_BOTH },
1.1 kristaps 53:
1.13 kristaps 54: #define CHAR_TBL_START static struct ln lines[LINES_MAX] = {
55: #define CHAR_TBL_END };
56:
1.1 kristaps 57: #include "chars.in"
58:
1.36 kristaps 59: struct mchars {
1.1 kristaps 60: struct ln **htab;
61: };
62:
63: static inline int match(const struct ln *,
64: const char *, size_t, int);
1.36 kristaps 65: static const struct ln *find(struct mchars *, const char *, size_t, int);
1.1 kristaps 66:
67: void
1.36 kristaps 68: mchars_free(struct mchars *arg)
1.1 kristaps 69: {
70:
1.36 kristaps 71: free(arg->htab);
72: free(arg);
1.1 kristaps 73: }
74:
1.36 kristaps 75: struct mchars *
1.38 kristaps 76: mchars_alloc(void)
1.1 kristaps 77: {
1.36 kristaps 78: struct mchars *tab;
1.1 kristaps 79: struct ln **htab;
80: struct ln *pp;
81: int i, hash;
82:
83: /*
84: * Constructs a very basic chaining hashtable. The hash routine
85: * is simply the integral value of the first character.
86: * Subsequent entries are chained in the order they're processed
87: * (they're in-line re-ordered during lookup).
88: */
89:
1.36 kristaps 90: tab = mandoc_malloc(sizeof(struct mchars));
1.33 kristaps 91: htab = mandoc_calloc(PRINT_HI - PRINT_LO + 1, sizeof(struct ln **));
1.1 kristaps 92:
93: for (i = 0; i < LINES_MAX; i++) {
1.2 kristaps 94: hash = (int)lines[i].code[0] - PRINT_LO;
1.1 kristaps 95:
96: if (NULL == (pp = htab[hash])) {
97: htab[hash] = &lines[i];
98: continue;
99: }
100:
101: for ( ; pp->next; pp = pp->next)
102: /* Scan ahead. */ ;
103: pp->next = &lines[i];
104: }
105:
106: tab->htab = htab;
107: return(tab);
108: }
109:
110:
1.21 kristaps 111: /*
112: * Special character to Unicode codepoint.
113: */
114: int
1.36 kristaps 115: mchars_spec2cp(struct mchars *arg, const char *p, size_t sz)
1.21 kristaps 116: {
117: const struct ln *ln;
118:
1.36 kristaps 119: ln = find(arg, p, sz, CHARS_CHAR);
1.21 kristaps 120: if (NULL == ln)
121: return(-1);
122: return(ln->unicode);
123: }
124:
125:
126: /*
127: * Reserved word to Unicode codepoint.
128: */
129: int
1.36 kristaps 130: mchars_res2cp(struct mchars *arg, const char *p, size_t sz)
1.21 kristaps 131: {
132: const struct ln *ln;
133:
1.36 kristaps 134: ln = find(arg, p, sz, CHARS_STRING);
1.21 kristaps 135: if (NULL == ln)
136: return(-1);
137: return(ln->unicode);
1.32 schwarze 138: }
139:
140: /*
1.39 kristaps 141: * Numbered character to literal character.
1.40 kristaps 142: * This can only be a printable character (i.e., alnum, punct, space) so
143: * prevent the character from ruining our state (backspace, newline, and
144: * so on).
1.42 kristaps 145: * If the character is illegal, returns '\0'.
1.32 schwarze 146: */
1.39 kristaps 147: char
1.36 kristaps 148: mchars_num2char(const char *p, size_t sz)
1.32 schwarze 149: {
150: int i;
151:
1.43 ! kristaps 152: if ((i = mandoc_strntou(p, sz, 10)) < 0)
1.39 kristaps 153: return('\0');
154:
1.41 kristaps 155: return(isprint(i) ? i : '\0');
1.21 kristaps 156: }
157:
158: /*
159: * Special character to string array.
160: */
1.1 kristaps 161: const char *
1.36 kristaps 162: mchars_spec2str(struct mchars *arg, const char *p, size_t sz, size_t *rsz)
1.1 kristaps 163: {
1.21 kristaps 164: const struct ln *ln;
165:
1.36 kristaps 166: ln = find(arg, p, sz, CHARS_CHAR);
1.21 kristaps 167: if (NULL == ln)
168: return(NULL);
1.1 kristaps 169:
1.24 kristaps 170: *rsz = strlen(ln->ascii);
1.21 kristaps 171: return(ln->ascii);
1.1 kristaps 172: }
173:
1.21 kristaps 174: /*
175: * Reserved word to string array.
176: */
1.1 kristaps 177: const char *
1.36 kristaps 178: mchars_res2str(struct mchars *arg, const char *p, size_t sz, size_t *rsz)
1.1 kristaps 179: {
1.21 kristaps 180: const struct ln *ln;
1.1 kristaps 181:
1.36 kristaps 182: ln = find(arg, p, sz, CHARS_STRING);
1.21 kristaps 183: if (NULL == ln)
184: return(NULL);
185:
1.24 kristaps 186: *rsz = strlen(ln->ascii);
1.21 kristaps 187: return(ln->ascii);
1.1 kristaps 188: }
189:
1.21 kristaps 190: static const struct ln *
1.36 kristaps 191: find(struct mchars *tab, const char *p, size_t sz, int type)
1.1 kristaps 192: {
193: struct ln *pp, *prev;
194: struct ln **htab;
195: int hash;
196:
197: assert(p);
1.23 kristaps 198: if (0 == sz)
199: return(NULL);
1.1 kristaps 200:
1.2 kristaps 201: if (p[0] < PRINT_LO || p[0] > PRINT_HI)
1.1 kristaps 202: return(NULL);
203:
204: /*
205: * Lookup the symbol in the symbol hash. See ascii2htab for the
206: * hashtable specs. This dynamically re-orders the hash chain
207: * to optimise for repeat hits.
208: */
209:
1.2 kristaps 210: hash = (int)p[0] - PRINT_LO;
1.1 kristaps 211: htab = tab->htab;
212:
213: if (NULL == (pp = htab[hash]))
214: return(NULL);
215:
216: for (prev = NULL; pp; pp = pp->next) {
217: if ( ! match(pp, p, sz, type)) {
218: prev = pp;
219: continue;
220: }
221:
222: if (prev) {
223: prev->next = pp->next;
224: pp->next = htab[hash];
225: htab[hash] = pp;
226: }
227:
1.21 kristaps 228: return(pp);
1.1 kristaps 229: }
230:
231: return(NULL);
232: }
233:
234: static inline int
235: match(const struct ln *ln, const char *p, size_t sz, int type)
236: {
237:
238: if ( ! (ln->type & type))
239: return(0);
1.22 kristaps 240: if (strncmp(ln->code, p, sz))
1.1 kristaps 241: return(0);
1.22 kristaps 242: return('\0' == ln->code[(int)sz]);
1.1 kristaps 243: }
CVSweb