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  */
16 
17 #include "parser.h"
18 #include "telescope.h"
19 
20 #include <stdlib.h>
21 #include <string.h>
22 
23 void
24 parser_init(struct tab *tab, parserfn fn)
25 {
26 	fn(&tab->buffer.page);
27 }
28 
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 }
34 
35 int
36 parser_free(struct tab *tab)
37 {
38 	int	 r;
39 	char	*tilde, *slash;
40 
41 	r = tab->buffer.page.free(&tab->buffer.page);
42 
43 	/* fallback to the host as title if nothing else */
44 	if (*tab->buffer.page.title != '\0')
45 		return r;
46 
47 	if ((tilde = strstr(tab->hist_cur->h, "/~")) != NULL) {
48 		strlcpy(tab->buffer.page.title, tilde+1,
49 		    sizeof(tab->buffer.page.title));
50 
51 		if ((slash = strchr(tab->buffer.page.title, '/')) != NULL)
52 			*slash = '\0';
53 	} else
54 		strlcpy(tab->buffer.page.title, tab->uri.host,
55 		    sizeof(tab->buffer.page.title));
56 
57 	return r;
58 }
59 
60 int
61 parser_append(struct parser *p, const char *buf, size_t len)
62 {
63 	size_t newlen;
64 	char *t;
65 
66 	newlen = len + p->len;
67 	if ((t = calloc(1, newlen)) == NULL)
68 		return 0;
69 	memcpy(t, p->buf, p->len);
70 	memcpy(t + p->len, buf, len);
71 	free(p->buf);
72 	p->buf = t;
73 	p->len = newlen;
74 	return 1;
75 }
76 
77 int
78 parser_set_buf(struct parser *p, const char *buf, size_t len)
79 {
80 	char *tmp;
81 
82 	if (len == 0) {
83 		p->len = 0;
84 		free(p->buf);
85 		p->buf = NULL;
86 		return 1;
87 	}
88 
89 	/*
90 	 * p->buf and buf can (and probably almost always will)
91 	 * overlap!
92 	 */
93 
94 	if ((tmp = calloc(1, len)) == NULL)
95 		return 0;
96 	memcpy(tmp, buf, len);
97 	free(p->buf);
98 	p->buf = tmp;
99 	p->len = len;
100 	return 1;
101 }
102 
103 int
104 parser_foreach_line(struct parser *p, const char *buf, size_t size,
105     parsechunkfn fn)
106 {
107 	char		*b, *e;
108 	unsigned int	 ch;
109 	size_t		 i, l, len;
110 
111 	if (!parser_append(p, buf, size))
112 		return 0;
113 	b = p->buf;
114 	len = p->len;
115 
116 	if (!(p->flags & PARSER_IN_BODY) && len < 3)
117 		return 1;
118 
119 	if (!(p->flags & PARSER_IN_BODY)) {
120                 p->flags |= PARSER_IN_BODY;
121 
122 		/*
123 		 * drop the BOM: only UTF-8 is supported, and there
124 		 * it's useless; some editors may still add one
125 		 * though.
126 		 */
127 		if (memmem(b, len, "\xEF\xBB\xBF", 3) == b) {
128 			b += 3;
129 			len -= 3;
130 		}
131 	}
132 
133 	/* drop every "funny" ASCII character */
134 	for (i = 0; i < len; ) {
135 		ch = b[i];
136 		if ((ch >= ' ' || ch == '\n' || ch == '\t')
137 		    && ch != 127) { /* del */
138 			++i;
139 			continue;
140 		}
141 		memmove(&b[i], &b[i+1], len - i - 1);
142 		len--;
143 	}
144 
145 	while (len > 0) {
146 		if ((e = memmem((char*)b, len, "\n", 1)) == NULL)
147 			break;
148 		l = e - b;
149 
150 		if (!fn(p, b, l))
151 			return 0;
152 
153 		len -= l;
154 		b += l;
155 
156 		if (len > 0) {
157 			/* skip \n */
158 			len--;
159 			b++;
160 		}
161 	}
162 
163 	return parser_set_buf(p, b, len);
164 }