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 "hist.h"
23 #include "parser.h"
24 #include "telescope.h"
25 
26 void
27 parser_init(struct tab *tab, parserfn fn)
28 {
29 	erase_buffer(&tab->buffer);
30 	fn(&tab->buffer.page);
31 	tab->buffer.page.init = fn;
32 }
33 
34 int
35 parser_parse(struct tab *tab, const char *chunk, size_t len)
36 {
37 	return tab->buffer.page.parse(&tab->buffer.page, chunk, len);
38 }
39 
40 int
41 parser_parsef(struct tab *tab, const char *fmt, ...)
42 {
43 	char *s;
44 	va_list ap;
45 	int r;
46 
47 	va_start(ap, fmt);
48 	r = vasprintf(&s, fmt, ap);
49 	va_end(ap);
50 
51 	if (r == -1)
52 		return 0;
53 
54 	r = parser_parse(tab, s, strlen(s));
55 	free(s);
56 	return r;
57 }
58 
59 int
60 parser_free(struct tab *tab)
61 {
62 	int	 r;
63 	char	*tilde, *slash;
64 
65 	r = tab->buffer.page.free(&tab->buffer.page);
66 
67 	if (*tab->buffer.page.title != '\0')
68 		return r;
69 
70 	/*
71 	 * heuristic: see if there is a "tilde user" and use that as
72 	 * page title, using the full domain name as fallback.
73 	 */
74 	if ((tilde = strstr(hist_cur(tab->hist), "/~")) != NULL) {
75 		strlcpy(tab->buffer.page.title, tilde+1,
76 		    sizeof(tab->buffer.page.title));
77 
78 		if ((slash = strchr(tab->buffer.page.title, '/')) != NULL)
79 			*slash = '\0';
80 	} else
81 		strlcpy(tab->buffer.page.title, tab->iri.iri_host,
82 		    sizeof(tab->buffer.page.title));
83 
84 	return r;
85 }
86 
87 int
88 parser_serialize(struct tab *tab, FILE *fp)
89 {
90 	struct line	*line;
91 	const char	*text;
92 	int		 r;
93 
94 	if (tab->buffer.page.serialize != NULL)
95 		return tab->buffer.page.serialize(&tab->buffer.page, fp);
96 
97 	/* a default implementation good enough for plain text */
98 	TAILQ_FOREACH(line, &tab->buffer.page.head, lines) {
99 		if ((text = line->line) == NULL)
100 			text = "";
101 
102 		r = fprintf(fp, "%s\n", text);
103 		if (r == -1)
104 			return 0;
105 	}
106 
107 	return 1;
108 }
109 
110 int
111 parser_append(struct parser *p, const char *buf, size_t len)
112 {
113 	size_t newlen;
114 	char *t;
115 
116 	newlen = len + p->len;
117 	if ((t = calloc(1, newlen)) == NULL)
118 		return 0;
119 	memcpy(t, p->buf, p->len);
120 	memcpy(t + p->len, buf, len);
121 	free(p->buf);
122 	p->buf = t;
123 	p->len = newlen;
124 	return 1;
125 }
126 
127 int
128 parser_set_buf(struct parser *p, const char *buf, size_t len)
129 {
130 	char *tmp;
131 
132 	if (len == 0) {
133 		p->len = 0;
134 		free(p->buf);
135 		p->buf = NULL;
136 		return 1;
137 	}
138 
139 	/*
140 	 * p->buf and buf can (and probably almost always will)
141 	 * overlap!
142 	 */
143 
144 	if ((tmp = calloc(1, len)) == NULL)
145 		return 0;
146 	memcpy(tmp, buf, len);
147 	free(p->buf);
148 	p->buf = tmp;
149 	p->len = len;
150 	return 1;
151 }
152 
153 int
154 parser_foreach_line(struct parser *p, const char *buf, size_t size,
155     parsechunkfn fn)
156 {
157 	char		*b, *e;
158 	unsigned int	 ch;
159 	size_t		 i, l, len;
160 
161 	if (!parser_append(p, buf, size))
162 		return 0;
163 	b = p->buf;
164 	len = p->len;
165 
166 	if (!(p->flags & PARSER_IN_BODY) && len < 3)
167 		return 1;
168 
169 	if (!(p->flags & PARSER_IN_BODY)) {
170 		p->flags |= PARSER_IN_BODY;
171 
172 		/*
173 		 * drop the BOM: only UTF-8 is supported, and there
174 		 * it's useless; some editors may still add one
175 		 * though.
176 		 */
177 		if (memmem(b, len, "\xEF\xBB\xBF", 3) == b) {
178 			b += 3;
179 			len -= 3;
180 		}
181 	}
182 
183 	/* drop every "funny" ASCII character */
184 	for (i = 0; i < len; ) {
185 		ch = b[i];
186 		if ((ch >= ' ' || ch == '\n' || ch == '\t')
187 		    && ch != 127) { /* del */
188 			++i;
189 			continue;
190 		}
191 		memmove(&b[i], &b[i+1], len - i - 1);
192 		len--;
193 	}
194 
195 	while (len > 0) {
196 		if ((e = memmem((char*)b, len, "\n", 1)) == NULL)
197 			break;
198 		l = e - b;
199 
200 		if (!fn(p, b, l))
201 			return 0;
202 
203 		len -= l;
204 		b += l;
205 
206 		if (len > 0) {
207 			/* skip \n */
208 			len--;
209 			b++;
210 		}
211 	}
212 
213 	return parser_set_buf(p, b, len);
214 }