Annotation of mandoc/chars.c, Revision 1.36
1.36 ! kristaps 1: /* $Id: chars.c,v 1.35 2011/04/20 22:50:22 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 {
! 58: enum mcharst type;
1.1 kristaps 59: struct ln **htab;
60: };
61:
62: static inline int match(const struct ln *,
63: const char *, size_t, int);
1.36 ! kristaps 64: static const struct ln *find(struct mchars *, const char *, size_t, int);
1.1 kristaps 65:
66: void
1.36 ! kristaps 67: mchars_free(struct mchars *arg)
1.1 kristaps 68: {
69:
1.36 ! kristaps 70: free(arg->htab);
! 71: free(arg);
1.1 kristaps 72: }
73:
1.36 ! kristaps 74: struct mchars *
! 75: mchars_init(enum mcharst type)
1.1 kristaps 76: {
1.36 ! kristaps 77: struct mchars *tab;
1.1 kristaps 78: struct ln **htab;
79: struct ln *pp;
80: int i, hash;
81:
82: /*
83: * Constructs a very basic chaining hashtable. The hash routine
84: * is simply the integral value of the first character.
85: * Subsequent entries are chained in the order they're processed
86: * (they're in-line re-ordered during lookup).
87: */
88:
1.36 ! kristaps 89: tab = mandoc_malloc(sizeof(struct mchars));
1.33 kristaps 90: htab = mandoc_calloc(PRINT_HI - PRINT_LO + 1, sizeof(struct ln **));
1.1 kristaps 91:
92: for (i = 0; i < LINES_MAX; i++) {
1.2 kristaps 93: hash = (int)lines[i].code[0] - PRINT_LO;
1.1 kristaps 94:
95: if (NULL == (pp = htab[hash])) {
96: htab[hash] = &lines[i];
97: continue;
98: }
99:
100: for ( ; pp->next; pp = pp->next)
101: /* Scan ahead. */ ;
102: pp->next = &lines[i];
103: }
104:
105: tab->htab = htab;
1.10 kristaps 106: tab->type = type;
1.1 kristaps 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:
141: /*
142: * Numbered character to literal character,
143: * represented as a null-terminated string for additional safety.
144: */
145: const char *
1.36 ! kristaps 146: mchars_num2char(const char *p, size_t sz)
1.32 schwarze 147: {
148: int i;
149: static char c[2];
150:
151: if (sz > 3)
152: return(NULL);
153: i = atoi(p);
154: if (i < 0 || i > 255)
155: return(NULL);
156: c[0] = (char)i;
157: c[1] = '\0';
158: return(c);
1.21 kristaps 159: }
160:
161:
162: /*
163: * Special character to string array.
164: */
1.1 kristaps 165: const char *
1.36 ! kristaps 166: mchars_spec2str(struct mchars *arg, const char *p, size_t sz, size_t *rsz)
1.1 kristaps 167: {
1.21 kristaps 168: const struct ln *ln;
169:
1.36 ! kristaps 170: ln = find(arg, p, sz, CHARS_CHAR);
1.21 kristaps 171: if (NULL == ln)
172: return(NULL);
1.1 kristaps 173:
1.24 kristaps 174: *rsz = strlen(ln->ascii);
1.21 kristaps 175: return(ln->ascii);
1.1 kristaps 176: }
177:
178:
1.21 kristaps 179: /*
180: * Reserved word to string array.
181: */
1.1 kristaps 182: const char *
1.36 ! kristaps 183: mchars_res2str(struct mchars *arg, const char *p, size_t sz, size_t *rsz)
1.1 kristaps 184: {
1.21 kristaps 185: const struct ln *ln;
1.1 kristaps 186:
1.36 ! kristaps 187: ln = find(arg, p, sz, CHARS_STRING);
1.21 kristaps 188: if (NULL == ln)
189: return(NULL);
190:
1.24 kristaps 191: *rsz = strlen(ln->ascii);
1.21 kristaps 192: return(ln->ascii);
1.1 kristaps 193: }
194:
1.21 kristaps 195: static const struct ln *
1.36 ! kristaps 196: find(struct mchars *tab, const char *p, size_t sz, int type)
1.1 kristaps 197: {
198: struct ln *pp, *prev;
199: struct ln **htab;
200: int hash;
201:
202: assert(p);
1.23 kristaps 203: if (0 == sz)
204: return(NULL);
1.1 kristaps 205:
1.2 kristaps 206: if (p[0] < PRINT_LO || p[0] > PRINT_HI)
1.1 kristaps 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:
1.2 kristaps 215: hash = (int)p[0] - PRINT_LO;
1.1 kristaps 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, type)) {
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:
1.21 kristaps 233: return(pp);
1.1 kristaps 234: }
235:
236: return(NULL);
237: }
238:
239: static inline int
240: match(const struct ln *ln, const char *p, size_t sz, int type)
241: {
242:
243: if ( ! (ln->type & type))
244: return(0);
1.22 kristaps 245: if (strncmp(ln->code, p, sz))
1.1 kristaps 246: return(0);
1.22 kristaps 247: return('\0' == ln->code[(int)sz]);
1.1 kristaps 248: }
CVSweb