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_parsef(struct tab *tab, const char *fmt, ...)
41 {
42 	char *s;
43 	va_list ap;
44 	int r;
45 
46 	va_start(ap, fmt);
47 	r = vasprintf(&s, fmt, ap);
48 	va_end(ap);
49 
50 	if (r == -1)
51 		return 0;
52 
53 	r = parser_parse(tab, s, strlen(s));
54 	free(s);
55 	return r;
56 }
57 
58 int
59 parser_free(struct tab *tab)
60 {
61 	int	 r;
62 	char	*tilde, *slash;
63 
64 	r = tab->buffer.page.free(&tab->buffer.page);
65 
66 	if (*tab->buffer.page.title != '\0')
67 		return r;
68 
69 	/*
70 	 * heuristic: see if there is a "tilde user" and use that as
71 	 * page title, using the full domain name as fallback.
72 	 */
73 	if ((tilde = strstr(tab->hist_cur->h, "/~")) != NULL) {
74 		strlcpy(tab->buffer.page.title, tilde+1,
75 		    sizeof(tab->buffer.page.title));
76 
77 		if ((slash = strchr(tab->buffer.page.title, '/')) != NULL)
78 			*slash = '\0';
79 	} else
80 		strlcpy(tab->buffer.page.title, tab->uri.host,
81 		    sizeof(tab->buffer.page.title));
82 
83 	return r;
84 }
85 
86 int
87 parser_serialize(struct tab *tab, FILE *fp)
88 {
89 	struct line	*line;
90 	const char	*text;
91 	int		 r;
92 
93 	if (tab->buffer.page.serialize != NULL)
94 		return tab->buffer.page.serialize(&tab->buffer.page, fp);
95 
96 	/* a default implementation good enough for plain text */
97 	TAILQ_FOREACH(line, &tab->buffer.page.head, lines) {
98 		if ((text = line->line) == NULL)
99 			text = "";
100 
101 		r = fprintf(fp, "%s\n", text);
102 		if (r == -1)
103 			return 0;
104 	}
105 
106 	return 1;
107 }
108 
109 int
110 parser_append(struct parser *p, const char *buf, size_t len)
111 {
112 	size_t newlen;
113 	char *t;
114 
115 	newlen = len + p->len;
116 	if ((t = calloc(1, newlen)) == NULL)
117 		return 0;
118 	memcpy(t, p->buf, p->len);
119 	memcpy(t + p->len, buf, len);
120 	free(p->buf);
121 	p->buf = t;
122 	p->len = newlen;
123 	return 1;
124 }
125 
126 int
127 parser_set_buf(struct parser *p, const char *buf, size_t len)
128 {
129 	char *tmp;
130 
131 	if (len == 0) {
132 		p->len = 0;
133 		free(p->buf);
134 		p->buf = NULL;
135 		return 1;
136 	}
137 
138 	/*
139 	 * p->buf and buf can (and probably almost always will)
140 	 * overlap!
141 	 */
142 
143 	if ((tmp = calloc(1, len)) == NULL)
144 		return 0;
145 	memcpy(tmp, buf, len);
146 	free(p->buf);
147 	p->buf = tmp;
148 	p->len = len;
149 	return 1;
150 }
151 
152 int
153 parser_foreach_line(struct parser *p, const char *buf, size_t size,
154     parsechunkfn fn)
155 {
156 	char		*b, *e;
157 	unsigned int	 ch;
158 	size_t		 i, l, len;
159 
160 	if (!parser_append(p, buf, size))
161 		return 0;
162 	b = p->buf;
163 	len = p->len;
164 
165 	if (!(p->flags & PARSER_IN_BODY) && len < 3)
166 		return 1;
167 
168 	if (!(p->flags & PARSER_IN_BODY)) {
169 		p->flags |= PARSER_IN_BODY;
170 
171 		/*
172 		 * drop the BOM: only UTF-8 is supported, and there
173 		 * it's useless; some editors may still add one
174 		 * though.
175 		 */
176 		if (memmem(b, len, "\xEF\xBB\xBF", 3) == b) {
177 			b += 3;
178 			len -= 3;
179 		}
180 	}
181 
182 	/* drop every "funny" ASCII character */
183 	for (i = 0; i < len; ) {
184 		ch = b[i];
185 		if ((ch >= ' ' || ch == '\n' || ch == '\t')
186 		    && ch != 127) { /* del */
187 			++i;
188 			continue;
189 		}
190 		memmove(&b[i], &b[i+1], len - i - 1);
191 		len--;
192 	}
193 
194 	while (len > 0) {
195 		if ((e = memmem((char*)b, len, "\n", 1)) == NULL)
196 			break;
197 		l = e - b;
198 
199 		if (!fn(p, b, l))
200 			return 0;
201 
202 		len -= l;
203 		b += l;
204 
205 		if (len > 0) {
206 			/* skip \n */
207 			len--;
208 			b++;
209 		}
210 	}
211 
212 	return parser_set_buf(p, b, len);
213 }