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 "compat.h"
18 
19 #include <stdlib.h>
20 #include <string.h>
21 
22 #include "parser.h"
23 #include "telescope.h"
24 
25 void
26 parser_init(struct tab *tab, parserfn fn)
27 {
28 	erase_buffer(&tab->buffer);
29 	fn(&tab->buffer.page);
30 	tab->buffer.page.init = fn;
31 }
32 
33 int
34 parser_parse(struct tab *tab, const char *chunk, size_t len)
35 {
36 	return tab->buffer.page.parse(&tab->buffer.page, chunk, len);
37 }
38 
39 int
40 parser_free(struct tab *tab)
41 {
42 	int	 r;
43 	char	*tilde, *slash;
44 
45 	r = tab->buffer.page.free(&tab->buffer.page);
46 
47 	if (*tab->buffer.page.title != '\0')
48 		return r;
49 
50 	/*
51 	 * heuristic: see if there is a "tilde user" and use that as
52 	 * page title, using the full domain name as fallback.
53 	 */
54 	if ((tilde = strstr(tab->hist_cur->h, "/~")) != NULL) {
55 		strlcpy(tab->buffer.page.title, tilde+1,
56 		    sizeof(tab->buffer.page.title));
57 
58 		if ((slash = strchr(tab->buffer.page.title, '/')) != NULL)
59 			*slash = '\0';
60 	} else
61 		strlcpy(tab->buffer.page.title, tab->uri.host,
62 		    sizeof(tab->buffer.page.title));
63 
64 	return r;
65 }
66 
67 int
68 parser_serialize(struct tab *tab, struct evbuffer *evb)
69 {
70 	struct line	*line;
71 	const char	*text;
72 	int		 r;
73 
74 	if (tab->buffer.page.serialize != NULL)
75 		return tab->buffer.page.serialize(&tab->buffer.page, evb);
76 
77 	/* a default implementation good enough for plain text */
78 	TAILQ_FOREACH(line, &tab->buffer.page.head, lines) {
79 		if ((text = line->line) == NULL)
80 			text = "";
81 
82 		r = evbuffer_add_printf(evb, "%s\n", text);
83 		if (r == -1)
84 			return 0;
85 	}
86 
87 	return 1;
88 }
89 
90 int
91 parser_append(struct parser *p, const char *buf, size_t len)
92 {
93 	size_t newlen;
94 	char *t;
95 
96 	newlen = len + p->len;
97 	if ((t = calloc(1, newlen)) == NULL)
98 		return 0;
99 	memcpy(t, p->buf, p->len);
100 	memcpy(t + p->len, buf, len);
101 	free(p->buf);
102 	p->buf = t;
103 	p->len = newlen;
104 	return 1;
105 }
106 
107 int
108 parser_set_buf(struct parser *p, const char *buf, size_t len)
109 {
110 	char *tmp;
111 
112 	if (len == 0) {
113 		p->len = 0;
114 		free(p->buf);
115 		p->buf = NULL;
116 		return 1;
117 	}
118 
119 	/*
120 	 * p->buf and buf can (and probably almost always will)
121 	 * overlap!
122 	 */
123 
124 	if ((tmp = calloc(1, len)) == NULL)
125 		return 0;
126 	memcpy(tmp, buf, len);
127 	free(p->buf);
128 	p->buf = tmp;
129 	p->len = len;
130 	return 1;
131 }
132 
133 int
134 parser_foreach_line(struct parser *p, const char *buf, size_t size,
135     parsechunkfn fn)
136 {
137 	char		*b, *e;
138 	unsigned int	 ch;
139 	size_t		 i, l, len;
140 
141 	if (!parser_append(p, buf, size))
142 		return 0;
143 	b = p->buf;
144 	len = p->len;
145 
146 	if (!(p->flags & PARSER_IN_BODY) && len < 3)
147 		return 1;
148 
149 	if (!(p->flags & PARSER_IN_BODY)) {
150 		p->flags |= PARSER_IN_BODY;
151 
152 		/*
153 		 * drop the BOM: only UTF-8 is supported, and there
154 		 * it's useless; some editors may still add one
155 		 * though.
156 		 */
157 		if (memmem(b, len, "\xEF\xBB\xBF", 3) == b) {
158 			b += 3;
159 			len -= 3;
160 		}
161 	}
162 
163 	/* drop every "funny" ASCII character */
164 	for (i = 0; i < len; ) {
165 		ch = b[i];
166 		if ((ch >= ' ' || ch == '\n' || ch == '\t')
167 		    && ch != 127) { /* del */
168 			++i;
169 			continue;
170 		}
171 		memmove(&b[i], &b[i+1], len - i - 1);
172 		len--;
173 	}
174 
175 	while (len > 0) {
176 		if ((e = memmem((char*)b, len, "\n", 1)) == NULL)
177 			break;
178 		l = e - b;
179 
180 		if (!fn(p, b, l))
181 			return 0;
182 
183 		len -= l;
184 		b += l;
185 
186 		if (len > 0) {
187 			/* skip \n */
188 			len--;
189 			b++;
190 		}
191 	}
192 
193 	return parser_set_buf(p, b, len);
194 }