Blob


1 /*
2 * Copyright (c) 2021 Omar Polo <op@omarpolo.com>
3 *
4 * Permission to use, copy, modify, and distribute this software for any
5 * purpose with or without fee is hereby granted, provided that the above
6 * copyright notice and this permission notice appear in all copies.
7 *
8 * THE SOFTWARE IS PROVIDED "AS IS" AND THE AUTHOR DISCLAIMS ALL WARRANTIES
9 * WITH REGARD TO THIS SOFTWARE INCLUDING ALL IMPLIED WARRANTIES OF
10 * MERCHANTABILITY AND FITNESS. IN NO EVENT SHALL THE AUTHOR BE LIABLE FOR
11 * ANY SPECIAL, DIRECT, INDIRECT, OR CONSEQUENTIAL DAMAGES OR ANY DAMAGES
12 * WHATSOEVER RESULTING FROM LOSS OF USE, DATA OR PROFITS, WHETHER IN AN
13 * ACTION OF CONTRACT, NEGLIGENCE OR OTHER TORTIOUS ACTION, ARISING OUT OF
14 * OR IN CONNECTION WITH THE USE OR PERFORMANCE OF THIS SOFTWARE.
15 */
17 #include "parser.h"
18 #include "telescope.h"
20 #include <stdlib.h>
21 #include <string.h>
23 void
24 parser_init(struct tab *tab, parserfn fn)
25 {
26 fn(&tab->buffer.page);
27 }
29 int
30 parser_parse(struct tab *tab, const char *chunk, size_t len)
31 {
32 return tab->buffer.page.parse(&tab->buffer.page, chunk, len);
33 }
35 int
36 parser_free(struct tab *tab)
37 {
38 int r;
40 r = tab->buffer.page.free(&tab->buffer.page);
42 /* fallback to the host as title if nothing else */
43 if (*tab->buffer.page.title == '\0')
44 strlcpy(tab->buffer.page.title, tab->uri.host,
45 sizeof(tab->buffer.page.title));
47 return r;
48 }
50 int
51 parser_append(struct parser *p, const char *buf, size_t len)
52 {
53 size_t newlen;
54 char *t;
56 newlen = len + p->len;
57 if ((t = calloc(1, newlen)) == NULL)
58 return 0;
59 memcpy(t, p->buf, p->len);
60 memcpy(t + p->len, buf, len);
61 free(p->buf);
62 p->buf = t;
63 p->len = newlen;
64 return 1;
65 }
67 int
68 parser_set_buf(struct parser *p, const char *buf, size_t len)
69 {
70 char *tmp;
72 if (len == 0) {
73 p->len = 0;
74 free(p->buf);
75 p->buf = NULL;
76 return 1;
77 }
79 /*
80 * p->buf and buf can (and probably almost always will)
81 * overlap!
82 */
84 if ((tmp = calloc(1, len)) == NULL)
85 return 0;
86 memcpy(tmp, buf, len);
87 free(p->buf);
88 p->buf = tmp;
89 p->len = len;
90 return 1;
91 }
93 int
94 parser_foreach_line(struct parser *p, const char *buf, size_t size,
95 parsechunkfn fn)
96 {
97 char *b, *e;
98 unsigned int ch;
99 size_t i, l, len;
101 if (!parser_append(p, buf, size))
102 return 0;
103 b = p->buf;
104 len = p->len;
106 if (!(p->flags & PARSER_IN_BODY) && len < 3)
107 return 1;
109 if (!(p->flags & PARSER_IN_BODY)) {
110 p->flags |= PARSER_IN_BODY;
112 /*
113 * drop the BOM: only UTF-8 is supported, and there
114 * it's useless; some editors may still add one
115 * though.
116 */
117 if (memmem(b, len, "\xEF\xBB\xBF", 3) == b) {
118 b += 3;
119 len -= 3;
123 /* drop every "funny" ASCII character */
124 for (i = 0; i < len; ) {
125 ch = b[i];
126 if ((ch >= ' ' || ch == '\n' || ch == '\t')
127 && ch != 127) { /* del */
128 ++i;
129 continue;
131 memmove(&b[i], &b[i+1], len - i - 1);
132 len--;
135 while (len > 0) {
136 if ((e = memmem((char*)b, len, "\n", 1)) == NULL)
137 break;
138 l = e - b;
140 if (!fn(p, b, l))
141 return 0;
143 len -= l;
144 b += l;
146 if (len > 0) {
147 /* skip \n */
148 len--;
149 b++;
153 return parser_set_buf(p, b, len);