Annotation of src/usr.bin/mandoc/chars.c, Revision 1.16
1.16 ! schwarze 1: /* $Id: chars.c,v 1.15 2011/01/04 22:28:17 schwarze Exp $ */
1.1 schwarze 2: /*
1.9 schwarze 3: * Copyright (c) 2009, 2010 Kristaps Dzonsons <kristaps@bsd.lv>
1.16 ! schwarze 4: * Copyright (c) 2011 Ingo Schwarze <schwarze@openbsd.org>
1.1 schwarze 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: #include <assert.h>
1.3 schwarze 19: #include <stdio.h>
1.1 schwarze 20: #include <stdlib.h>
21: #include <string.h>
22:
1.7 schwarze 23: #include "mandoc.h"
1.1 schwarze 24: #include "chars.h"
25:
26: #define PRINT_HI 126
27: #define PRINT_LO 32
28:
29: struct ln {
30: struct ln *next;
31: const char *code;
32: const char *ascii;
1.9 schwarze 33: int unicode;
1.1 schwarze 34: int type;
35: #define CHARS_CHAR (1 << 0)
36: #define CHARS_STRING (1 << 1)
1.4 schwarze 37: #define CHARS_BOTH (CHARS_CHAR | CHARS_STRING)
1.1 schwarze 38: };
39:
1.14 schwarze 40: #define LINES_MAX 351
1.1 schwarze 41:
1.10 schwarze 42: #define CHAR(in, ch, code) \
43: { NULL, (in), (ch), (code), CHARS_CHAR },
44: #define STRING(in, ch, code) \
45: { NULL, (in), (ch), (code), CHARS_STRING },
46: #define BOTH(in, ch, code) \
47: { NULL, (in), (ch), (code), CHARS_BOTH },
1.1 schwarze 48:
1.4 schwarze 49: #define CHAR_TBL_START static struct ln lines[LINES_MAX] = {
50: #define CHAR_TBL_END };
51:
1.1 schwarze 52: #include "chars.in"
53:
1.15 schwarze 54: struct ctab {
1.1 schwarze 55: enum chars type;
56: struct ln **htab;
57: };
58:
59: static inline int match(const struct ln *,
60: const char *, size_t, int);
1.15 schwarze 61: static const struct ln *find(struct ctab *, const char *, size_t, int);
1.1 schwarze 62:
63:
64: void
65: chars_free(void *arg)
66: {
1.15 schwarze 67: struct ctab *tab;
1.1 schwarze 68:
1.15 schwarze 69: tab = (struct ctab *)arg;
1.1 schwarze 70:
71: free(tab->htab);
72: free(tab);
73: }
74:
75:
76: void *
77: chars_init(enum chars type)
78: {
1.15 schwarze 79: struct ctab *tab;
1.1 schwarze 80: struct ln **htab;
81: struct ln *pp;
82: int i, hash;
83:
84: /*
85: * Constructs a very basic chaining hashtable. The hash routine
86: * is simply the integral value of the first character.
87: * Subsequent entries are chained in the order they're processed
88: * (they're in-line re-ordered during lookup).
89: */
90:
1.15 schwarze 91: tab = malloc(sizeof(struct ctab));
1.3 schwarze 92: if (NULL == tab) {
93: perror(NULL);
1.14 schwarze 94: exit((int)MANDOCLEVEL_SYSERR);
1.3 schwarze 95: }
1.1 schwarze 96:
97: htab = calloc(PRINT_HI - PRINT_LO + 1, sizeof(struct ln **));
1.3 schwarze 98: if (NULL == htab) {
99: perror(NULL);
1.14 schwarze 100: exit((int)MANDOCLEVEL_SYSERR);
1.3 schwarze 101: }
1.1 schwarze 102:
103: for (i = 0; i < LINES_MAX; i++) {
104: hash = (int)lines[i].code[0] - PRINT_LO;
105:
106: if (NULL == (pp = htab[hash])) {
107: htab[hash] = &lines[i];
108: continue;
109: }
110:
111: for ( ; pp->next; pp = pp->next)
112: /* Scan ahead. */ ;
113: pp->next = &lines[i];
114: }
115:
116: tab->htab = htab;
1.3 schwarze 117: tab->type = type;
1.1 schwarze 118: return(tab);
119: }
120:
121:
1.9 schwarze 122: /*
123: * Special character to Unicode codepoint.
124: */
125: int
126: chars_spec2cp(void *arg, const char *p, size_t sz)
127: {
128: const struct ln *ln;
129:
1.15 schwarze 130: ln = find((struct ctab *)arg, p, sz, CHARS_CHAR);
1.9 schwarze 131: if (NULL == ln)
132: return(-1);
133: return(ln->unicode);
134: }
135:
136:
137: /*
138: * Reserved word to Unicode codepoint.
139: */
140: int
141: chars_res2cp(void *arg, const char *p, size_t sz)
142: {
143: const struct ln *ln;
144:
1.15 schwarze 145: ln = find((struct ctab *)arg, p, sz, CHARS_STRING);
1.9 schwarze 146: if (NULL == ln)
147: return(-1);
148: return(ln->unicode);
1.16 ! schwarze 149: }
! 150:
! 151:
! 152: /*
! 153: * Numbered character to literal character,
! 154: * represented as a null-terminated string for additional safety.
! 155: */
! 156: const char *
! 157: chars_num2char(const char *p, size_t sz)
! 158: {
! 159: int i;
! 160: static char c[2];
! 161:
! 162: if (sz > 3)
! 163: return(NULL);
! 164: i = atoi(p);
! 165: if (i < 0 || i > 255)
! 166: return(NULL);
! 167: c[0] = (char)i;
! 168: c[1] = '\0';
! 169: return(c);
1.9 schwarze 170: }
171:
172:
173: /*
174: * Special character to string array.
175: */
1.1 schwarze 176: const char *
1.9 schwarze 177: chars_spec2str(void *arg, const char *p, size_t sz, size_t *rsz)
1.1 schwarze 178: {
1.9 schwarze 179: const struct ln *ln;
180:
1.15 schwarze 181: ln = find((struct ctab *)arg, p, sz, CHARS_CHAR);
1.9 schwarze 182: if (NULL == ln)
183: return(NULL);
1.1 schwarze 184:
1.10 schwarze 185: *rsz = strlen(ln->ascii);
1.9 schwarze 186: return(ln->ascii);
1.1 schwarze 187: }
188:
189:
1.9 schwarze 190: /*
191: * Reserved word to string array.
192: */
1.1 schwarze 193: const char *
1.9 schwarze 194: chars_res2str(void *arg, const char *p, size_t sz, size_t *rsz)
1.1 schwarze 195: {
1.9 schwarze 196: const struct ln *ln;
197:
1.15 schwarze 198: ln = find((struct ctab *)arg, p, sz, CHARS_STRING);
1.9 schwarze 199: if (NULL == ln)
200: return(NULL);
1.1 schwarze 201:
1.10 schwarze 202: *rsz = strlen(ln->ascii);
1.9 schwarze 203: return(ln->ascii);
1.1 schwarze 204: }
205:
206:
1.9 schwarze 207: static const struct ln *
1.15 schwarze 208: find(struct ctab *tab, const char *p, size_t sz, int type)
1.1 schwarze 209: {
210: struct ln *pp, *prev;
211: struct ln **htab;
212: int hash;
213:
214: assert(p);
1.9 schwarze 215: if (0 == sz)
216: return(NULL);
1.1 schwarze 217:
218: if (p[0] < PRINT_LO || p[0] > PRINT_HI)
219: return(NULL);
220:
221: /*
222: * Lookup the symbol in the symbol hash. See ascii2htab for the
223: * hashtable specs. This dynamically re-orders the hash chain
224: * to optimise for repeat hits.
225: */
226:
227: hash = (int)p[0] - PRINT_LO;
228: htab = tab->htab;
229:
230: if (NULL == (pp = htab[hash]))
231: return(NULL);
232:
233: for (prev = NULL; pp; pp = pp->next) {
234: if ( ! match(pp, p, sz, type)) {
235: prev = pp;
236: continue;
237: }
238:
239: if (prev) {
240: prev->next = pp->next;
241: pp->next = htab[hash];
242: htab[hash] = pp;
243: }
244:
1.9 schwarze 245: return(pp);
1.1 schwarze 246: }
247:
248: return(NULL);
249: }
250:
251:
252: static inline int
253: match(const struct ln *ln, const char *p, size_t sz, int type)
254: {
255:
256: if ( ! (ln->type & type))
257: return(0);
1.9 schwarze 258: if (strncmp(ln->code, p, sz))
1.1 schwarze 259: return(0);
1.9 schwarze 260: return('\0' == ln->code[(int)sz]);
1.1 schwarze 261: }
![[BACK]](/icons/back.gif){kind=icon}
Return to chars.c CVS log ![[TXT]](/icons/text.gif){kind=icon}
![[DIR]](/icons/dir.gif){kind=icon}
Up to [local] / src / usr.bin / mandoc