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 #include "ui.h"
25 
26 /*
27  * Load a text/gemini page given the string page.  Always returns 0.
28  */
29 int
30 load_page_from_str(struct tab *tab, const char *page)
31 {
32 	erase_buffer(&tab->buffer);
33 	parser_init(tab, gemtext_initparser);
34 	if (!tab->buffer.page.parse(&tab->buffer.page, page, strlen(page)))
35 		abort();
36 	if (!tab->buffer.page.free(&tab->buffer.page))
37 		abort();
38 	ui_on_tab_refresh(tab);
39 	ui_on_tab_loaded(tab);
40 	return 0;
41 }
42 
43 void
44 parser_init(struct tab *tab, parserfn fn)
45 {
46 	fn(&tab->buffer.page);
47 }
48 
49 int
50 parser_parse(struct tab *tab, const char *chunk, size_t len)
51 {
52 	return tab->buffer.page.parse(&tab->buffer.page, chunk, len);
53 }
54 
55 int
56 parser_free(struct tab *tab)
57 {
58 	int	 r;
59 	char	*tilde, *slash;
60 
61 	r = tab->buffer.page.free(&tab->buffer.page);
62 
63 	/* fallback to the host as title if nothing else */
64 	if (*tab->buffer.page.title != '\0')
65 		return r;
66 
67 	if ((tilde = strstr(tab->hist_cur->h, "/~")) != NULL) {
68 		strlcpy(tab->buffer.page.title, tilde+1,
69 		    sizeof(tab->buffer.page.title));
70 
71 		if ((slash = strchr(tab->buffer.page.title, '/')) != NULL)
72 			*slash = '\0';
73 	} else
74 		strlcpy(tab->buffer.page.title, tab->uri.host,
75 		    sizeof(tab->buffer.page.title));
76 
77 	return r;
78 }
79 
80 int
81 parser_append(struct parser *p, const char *buf, size_t len)
82 {
83 	size_t newlen;
84 	char *t;
85 
86 	newlen = len + p->len;
87 	if ((t = calloc(1, newlen)) == NULL)
88 		return 0;
89 	memcpy(t, p->buf, p->len);
90 	memcpy(t + p->len, buf, len);
91 	free(p->buf);
92 	p->buf = t;
93 	p->len = newlen;
94 	return 1;
95 }
96 
97 int
98 parser_set_buf(struct parser *p, const char *buf, size_t len)
99 {
100 	char *tmp;
101 
102 	if (len == 0) {
103 		p->len = 0;
104 		free(p->buf);
105 		p->buf = NULL;
106 		return 1;
107 	}
108 
109 	/*
110 	 * p->buf and buf can (and probably almost always will)
111 	 * overlap!
112 	 */
113 
114 	if ((tmp = calloc(1, len)) == NULL)
115 		return 0;
116 	memcpy(tmp, buf, len);
117 	free(p->buf);
118 	p->buf = tmp;
119 	p->len = len;
120 	return 1;
121 }
122 
123 int
124 parser_foreach_line(struct parser *p, const char *buf, size_t size,
125     parsechunkfn fn)
126 {
127 	char		*b, *e;
128 	unsigned int	 ch;
129 	size_t		 i, l, len;
130 
131 	if (!parser_append(p, buf, size))
132 		return 0;
133 	b = p->buf;
134 	len = p->len;
135 
136 	if (!(p->flags & PARSER_IN_BODY) && len < 3)
137 		return 1;
138 
139 	if (!(p->flags & PARSER_IN_BODY)) {
140 		p->flags |= PARSER_IN_BODY;
141 
142 		/*
143 		 * drop the BOM: only UTF-8 is supported, and there
144 		 * it's useless; some editors may still add one
145 		 * though.
146 		 */
147 		if (memmem(b, len, "\xEF\xBB\xBF", 3) == b) {
148 			b += 3;
149 			len -= 3;
150 		}
151 	}
152 
153 	/* drop every "funny" ASCII character */
154 	for (i = 0; i < len; ) {
155 		ch = b[i];
156 		if ((ch >= ' ' || ch == '\n' || ch == '\t')
157 		    && ch != 127) { /* del */
158 			++i;
159 			continue;
160 		}
161 		memmove(&b[i], &b[i+1], len - i - 1);
162 		len--;
163 	}
164 
165 	while (len > 0) {
166 		if ((e = memmem((char*)b, len, "\n", 1)) == NULL)
167 			break;
168 		l = e - b;
169 
170 		if (!fn(p, b, l))
171 			return 0;
172 
173 		len -= l;
174 		b += l;
175 
176 		if (len > 0) {
177 			/* skip \n */
178 			len--;
179 			b++;
180 		}
181 	}
182 
183 	return parser_set_buf(p, b, len);
184 }