]> git.cameronkatri.com Git - mandoc.git/blob - preconv.c
Delete five standards that are:
[mandoc.git] / preconv.c
1 /* $Id: preconv.c,v 1.12 2014/11/14 04:24:04 schwarze Exp $ */
2 /*
3 * Copyright (c) 2011 Kristaps Dzonsons <kristaps@bsd.lv>
4 * Copyright (c) 2014 Ingo Schwarze <schwarze@openbsd.org>
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 "config.h"
19
20 #include <sys/types.h>
21
22 #include <stdio.h>
23 #include <string.h>
24 #include "mandoc.h"
25 #include "libmandoc.h"
26
27 int
28 preconv_encode(struct buf *ib, size_t *ii, struct buf *ob, size_t *oi,
29 int *filenc)
30 {
31 size_t i;
32 int state;
33 unsigned int accum;
34 unsigned char cu;
35
36 if ( ! (*filenc & MPARSE_UTF8))
37 goto latin;
38
39 state = 0;
40 accum = 0U;
41
42 for (i = *ii; i < ib->sz; i++) {
43 cu = ib->buf[i];
44 if (state) {
45 if ( ! (cu & 128) || (cu & 64)) {
46 /* Bad sequence header. */
47 break;
48 }
49
50 /* Accept only legitimate bit patterns. */
51
52 if (cu > 191 || cu < 128) {
53 /* Bad in-sequence bits. */
54 break;
55 }
56
57 accum |= (cu & 63) << --state * 6;
58
59 if (state)
60 continue;
61
62 if (accum < 0x80)
63 ob->buf[(*oi)++] = accum;
64 else
65 *oi += snprintf(ob->buf + *oi,
66 11, "\\[u%.4X]", accum);
67 *ii = i + 1;
68 *filenc &= ~MPARSE_LATIN1;
69 return(1);
70 } else {
71 /*
72 * Entering a UTF-8 state: if we encounter a
73 * UTF-8 bitmask, calculate the expected UTF-8
74 * state from it.
75 */
76 for (state = 0; state < 7; state++)
77 if ( ! (cu & (1 << (7 - state))))
78 break;
79
80 /* Accept only legitimate bit patterns. */
81
82 switch (state--) {
83 case (4):
84 if (cu <= 244 && cu >= 240) {
85 accum = (cu & 7) << 18;
86 continue;
87 }
88 /* Bad 4-sequence start bits. */
89 break;
90 case (3):
91 if (cu <= 239 && cu >= 224) {
92 accum = (cu & 15) << 12;
93 continue;
94 }
95 /* Bad 3-sequence start bits. */
96 break;
97 case (2):
98 if (cu <= 223 && cu >= 194) {
99 accum = (cu & 31) << 6;
100 continue;
101 }
102 /* Bad 2-sequence start bits. */
103 break;
104 default:
105 /* Bad sequence bit mask. */
106 break;
107 }
108 break;
109 }
110 }
111
112 /* FALLTHROUGH: Invalid or incomplete UTF-8 sequence. */
113
114 latin:
115 if ( ! (*filenc & MPARSE_LATIN1))
116 return(0);
117
118 *oi += snprintf(ob->buf + *oi, 11,
119 "\\[u%.4X]", (unsigned char)ib->buf[(*ii)++]);
120
121 *filenc &= ~MPARSE_UTF8;
122 return(1);
123 }
124
125 int
126 preconv_cue(const struct buf *b, size_t offset)
127 {
128 const char *ln, *eoln, *eoph;
129 size_t sz, phsz;
130
131 ln = b->buf + offset;
132 sz = b->sz - offset;
133
134 /* Look for the end-of-line. */
135
136 if (NULL == (eoln = memchr(ln, '\n', sz)))
137 eoln = ln + sz;
138
139 /* Check if we have the correct header/trailer. */
140
141 if ((sz = (size_t)(eoln - ln)) < 10 ||
142 memcmp(ln, ".\\\" -*-", 7) || memcmp(eoln - 3, "-*-", 3))
143 return(MPARSE_UTF8 | MPARSE_LATIN1);
144
145 /* Move after the header and adjust for the trailer. */
146
147 ln += 7;
148 sz -= 10;
149
150 while (sz > 0) {
151 while (sz > 0 && ' ' == *ln) {
152 ln++;
153 sz--;
154 }
155 if (0 == sz)
156 break;
157
158 /* Find the end-of-phrase marker (or eoln). */
159
160 if (NULL == (eoph = memchr(ln, ';', sz)))
161 eoph = eoln - 3;
162 else
163 eoph++;
164
165 /* Only account for the "coding" phrase. */
166
167 if ((phsz = eoph - ln) < 7 ||
168 strncasecmp(ln, "coding:", 7)) {
169 sz -= phsz;
170 ln += phsz;
171 continue;
172 }
173
174 sz -= 7;
175 ln += 7;
176
177 while (sz > 0 && ' ' == *ln) {
178 ln++;
179 sz--;
180 }
181 if (0 == sz)
182 return(0);
183
184 /* Check us against known encodings. */
185
186 if (phsz > 4 && !strncasecmp(ln, "utf-8", 5))
187 return(MPARSE_UTF8);
188 if (phsz > 10 && !strncasecmp(ln, "iso-latin-1", 11))
189 return(MPARSE_LATIN1);
190 return(0);
191 }
192 return(MPARSE_UTF8 | MPARSE_LATIN1);
193 }