Annotation of mandoc/chars.c, Revision 1.37
1.37 ! kristaps 1: /* $Id: chars.c,v 1.36 2011/04/29 22:18:12 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>
1.10 kristaps 23: #include <stdio.h>
1.1 kristaps 24: #include <stdlib.h>
25: #include <string.h>
26:
1.18 kristaps 27: #include "mandoc.h"
1.1 kristaps 28:
1.2 kristaps 29: #define PRINT_HI 126
30: #define PRINT_LO 32
1.1 kristaps 31:
32: struct ln {
33: struct ln *next;
34: const char *code;
1.2 kristaps 35: const char *ascii;
1.21 kristaps 36: int unicode;
1.1 kristaps 37: int type;
38: #define CHARS_CHAR (1 << 0)
39: #define CHARS_STRING (1 << 1)
1.12 kristaps 40: #define CHARS_BOTH (CHARS_CHAR | CHARS_STRING)
1.1 kristaps 41: };
42:
1.35 kristaps 43: #define LINES_MAX 353
1.1 kristaps 44:
1.24 kristaps 45: #define CHAR(in, ch, code) \
46: { NULL, (in), (ch), (code), CHARS_CHAR },
47: #define STRING(in, ch, code) \
48: { NULL, (in), (ch), (code), CHARS_STRING },
49: #define BOTH(in, ch, code) \
50: { NULL, (in), (ch), (code), CHARS_BOTH },
1.1 kristaps 51:
1.13 kristaps 52: #define CHAR_TBL_START static struct ln lines[LINES_MAX] = {
53: #define CHAR_TBL_END };
54:
1.1 kristaps 55: #include "chars.in"
56:
1.36 kristaps 57: struct mchars {
1.1 kristaps 58: struct ln **htab;
59: };
60:
61: static inline int match(const struct ln *,
62: const char *, size_t, int);
1.36 kristaps 63: static const struct ln *find(struct mchars *, const char *, size_t, int);
1.1 kristaps 64:
65: void
1.36 kristaps 66: mchars_free(struct mchars *arg)
1.1 kristaps 67: {
68:
1.36 kristaps 69: free(arg->htab);
70: free(arg);
1.1 kristaps 71: }
72:
1.36 kristaps 73: struct mchars *
1.37 ! kristaps 74: mchars_init(void)
1.1 kristaps 75: {
1.36 kristaps 76: struct mchars *tab;
1.1 kristaps 77: struct ln **htab;
78: struct ln *pp;
79: int i, hash;
80:
81: /*
82: * Constructs a very basic chaining hashtable. The hash routine
83: * is simply the integral value of the first character.
84: * Subsequent entries are chained in the order they're processed
85: * (they're in-line re-ordered during lookup).
86: */
87:
1.36 kristaps 88: tab = mandoc_malloc(sizeof(struct mchars));
1.33 kristaps 89: htab = mandoc_calloc(PRINT_HI - PRINT_LO + 1, sizeof(struct ln **));
1.1 kristaps 90:
91: for (i = 0; i < LINES_MAX; i++) {
1.2 kristaps 92: hash = (int)lines[i].code[0] - PRINT_LO;
1.1 kristaps 93:
94: if (NULL == (pp = htab[hash])) {
95: htab[hash] = &lines[i];
96: continue;
97: }
98:
99: for ( ; pp->next; pp = pp->next)
100: /* Scan ahead. */ ;
101: pp->next = &lines[i];
102: }
103:
104: tab->htab = htab;
105: return(tab);
106: }
107:
108:
1.21 kristaps 109: /*
110: * Special character to Unicode codepoint.
111: */
112: int
1.36 kristaps 113: mchars_spec2cp(struct mchars *arg, const char *p, size_t sz)
1.21 kristaps 114: {
115: const struct ln *ln;
116:
1.36 kristaps 117: ln = find(arg, p, sz, CHARS_CHAR);
1.21 kristaps 118: if (NULL == ln)
119: return(-1);
120: return(ln->unicode);
121: }
122:
123:
124: /*
125: * Reserved word to Unicode codepoint.
126: */
127: int
1.36 kristaps 128: mchars_res2cp(struct mchars *arg, const char *p, size_t sz)
1.21 kristaps 129: {
130: const struct ln *ln;
131:
1.36 kristaps 132: ln = find(arg, p, sz, CHARS_STRING);
1.21 kristaps 133: if (NULL == ln)
134: return(-1);
135: return(ln->unicode);
1.32 schwarze 136: }
137:
138:
139: /*
140: * Numbered character to literal character,
141: * represented as a null-terminated string for additional safety.
142: */
143: const char *
1.36 kristaps 144: mchars_num2char(const char *p, size_t sz)
1.32 schwarze 145: {
146: int i;
147: static char c[2];
148:
149: if (sz > 3)
150: return(NULL);
151: i = atoi(p);
152: if (i < 0 || i > 255)
153: return(NULL);
154: c[0] = (char)i;
155: c[1] = '\0';
156: return(c);
1.21 kristaps 157: }
158:
159:
160: /*
161: * Special character to string array.
162: */
1.1 kristaps 163: const char *
1.36 kristaps 164: mchars_spec2str(struct mchars *arg, const char *p, size_t sz, size_t *rsz)
1.1 kristaps 165: {
1.21 kristaps 166: const struct ln *ln;
167:
1.36 kristaps 168: ln = find(arg, p, sz, CHARS_CHAR);
1.21 kristaps 169: if (NULL == ln)
170: return(NULL);
1.1 kristaps 171:
1.24 kristaps 172: *rsz = strlen(ln->ascii);
1.21 kristaps 173: return(ln->ascii);
1.1 kristaps 174: }
175:
176:
1.21 kristaps 177: /*
178: * Reserved word to string array.
179: */
1.1 kristaps 180: const char *
1.36 kristaps 181: mchars_res2str(struct mchars *arg, const char *p, size_t sz, size_t *rsz)
1.1 kristaps 182: {
1.21 kristaps 183: const struct ln *ln;
1.1 kristaps 184:
1.36 kristaps 185: ln = find(arg, p, sz, CHARS_STRING);
1.21 kristaps 186: if (NULL == ln)
187: return(NULL);
188:
1.24 kristaps 189: *rsz = strlen(ln->ascii);
1.21 kristaps 190: return(ln->ascii);
1.1 kristaps 191: }
192:
1.21 kristaps 193: static const struct ln *
1.36 kristaps 194: find(struct mchars *tab, const char *p, size_t sz, int type)
1.1 kristaps 195: {
196: struct ln *pp, *prev;
197: struct ln **htab;
198: int hash;
199:
200: assert(p);
1.23 kristaps 201: if (0 == sz)
202: return(NULL);
1.1 kristaps 203:
1.2 kristaps 204: if (p[0] < PRINT_LO || p[0] > PRINT_HI)
1.1 kristaps 205: return(NULL);
206:
207: /*
208: * Lookup the symbol in the symbol hash. See ascii2htab for the
209: * hashtable specs. This dynamically re-orders the hash chain
210: * to optimise for repeat hits.
211: */
212:
1.2 kristaps 213: hash = (int)p[0] - PRINT_LO;
1.1 kristaps 214: htab = tab->htab;
215:
216: if (NULL == (pp = htab[hash]))
217: return(NULL);
218:
219: for (prev = NULL; pp; pp = pp->next) {
220: if ( ! match(pp, p, sz, type)) {
221: prev = pp;
222: continue;
223: }
224:
225: if (prev) {
226: prev->next = pp->next;
227: pp->next = htab[hash];
228: htab[hash] = pp;
229: }
230:
1.21 kristaps 231: return(pp);
1.1 kristaps 232: }
233:
234: return(NULL);
235: }
236:
237: static inline int
238: match(const struct ln *ln, const char *p, size_t sz, int type)
239: {
240:
241: if ( ! (ln->type & type))
242: return(0);
1.22 kristaps 243: if (strncmp(ln->code, p, sz))
1.1 kristaps 244: return(0);
1.22 kristaps 245: return('\0' == ln->code[(int)sz]);
1.1 kristaps 246: }
CVSweb