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 #include "ui.h"
20 
21 #include <stdlib.h>
22 #include <string.h>
23 
24 /*
25  * Load a text/gemini page given the string page.  Always returns 0.
26  */
27 int
28 load_page_from_str(struct tab *tab, const char *page)
29 {
30 	erase_buffer(&tab->buffer);
31 	parser_init(tab, gemtext_initparser);
32 	if (!tab->buffer.page.parse(&tab->buffer.page, page, strlen(page)))
33 		abort();
34 	if (!tab->buffer.page.free(&tab->buffer.page))
35 		abort();
36 	ui_on_tab_refresh(tab);
37 	ui_on_tab_loaded(tab);
38 	return 0;
39 }
40 
41 void
42 parser_init(struct tab *tab, parserfn fn)
43 {
44 	fn(&tab->buffer.page);
45 }
46 
47 int
48 parser_parse(struct tab *tab, const char *chunk, size_t len)
49 {
50 	return tab->buffer.page.parse(&tab->buffer.page, chunk, len);
51 }
52 
53 int
54 parser_free(struct tab *tab)
55 {
56 	int	 r;
57 	char	*tilde, *slash;
58 
59 	r = tab->buffer.page.free(&tab->buffer.page);
60 
61 	/* fallback to the host as title if nothing else */
62 	if (*tab->buffer.page.title != '\0')
63 		return r;
64 
65 	if ((tilde = strstr(tab->hist_cur->h, "/~")) != NULL) {
66 		strlcpy(tab->buffer.page.title, tilde+1,
67 		    sizeof(tab->buffer.page.title));
68 
69 		if ((slash = strchr(tab->buffer.page.title, '/')) != NULL)
70 			*slash = '\0';
71 	} else
72 		strlcpy(tab->buffer.page.title, tab->uri.host,
73 		    sizeof(tab->buffer.page.title));
74 
75 	return r;
76 }
77 
78 int
79 parser_append(struct parser *p, const char *buf, size_t len)
80 {
81 	size_t newlen;
82 	char *t;
83 
84 	newlen = len + p->len;
85 	if ((t = calloc(1, newlen)) == NULL)
86 		return 0;
87 	memcpy(t, p->buf, p->len);
88 	memcpy(t + p->len, buf, len);
89 	free(p->buf);
90 	p->buf = t;
91 	p->len = newlen;
92 	return 1;
93 }
94 
95 int
96 parser_set_buf(struct parser *p, const char *buf, size_t len)
97 {
98 	char *tmp;
99 
100 	if (len == 0) {
101 		p->len = 0;
102 		free(p->buf);
103 		p->buf = NULL;
104 		return 1;
105 	}
106 
107 	/*
108 	 * p->buf and buf can (and probably almost always will)
109 	 * overlap!
110 	 */
111 
112 	if ((tmp = calloc(1, len)) == NULL)
113 		return 0;
114 	memcpy(tmp, buf, len);
115 	free(p->buf);
116 	p->buf = tmp;
117 	p->len = len;
118 	return 1;
119 }
120 
121 int
122 parser_foreach_line(struct parser *p, const char *buf, size_t size,
123     parsechunkfn fn)
124 {
125 	char		*b, *e;
126 	unsigned int	 ch;
127 	size_t		 i, l, len;
128 
129 	if (!parser_append(p, buf, size))
130 		return 0;
131 	b = p->buf;
132 	len = p->len;
133 
134 	if (!(p->flags & PARSER_IN_BODY) && len < 3)
135 		return 1;
136 
137 	if (!(p->flags & PARSER_IN_BODY)) {
138                 p->flags |= PARSER_IN_BODY;
139 
140 		/*
141 		 * drop the BOM: only UTF-8 is supported, and there
142 		 * it's useless; some editors may still add one
143 		 * though.
144 		 */
145 		if (memmem(b, len, "\xEF\xBB\xBF", 3) == b) {
146 			b += 3;
147 			len -= 3;
148 		}
149 	}
150 
151 	/* drop every "funny" ASCII character */
152 	for (i = 0; i < len; ) {
153 		ch = b[i];
154 		if ((ch >= ' ' || ch == '\n' || ch == '\t')
155 		    && ch != 127) { /* del */
156 			++i;
157 			continue;
158 		}
159 		memmove(&b[i], &b[i+1], len - i - 1);
160 		len--;
161 	}
162 
163 	while (len > 0) {
164 		if ((e = memmem((char*)b, len, "\n", 1)) == NULL)
165 			break;
166 		l = e - b;
167 
168 		if (!fn(p, b, l))
169 			return 0;
170 
171 		len -= l;
172 		b += l;
173 
174 		if (len > 0) {
175 			/* skip \n */
176 			len--;
177 			b++;
178 		}
179 	}
180 
181 	return parser_set_buf(p, b, len);
182 }