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 
20 #include <stdlib.h>
21 #include <string.h>
22 
23 int
24 parser_append(struct parser *p, const char *buf, size_t len)
25 {
26 	size_t newlen;
27 	char *t;
28 
29 	newlen = len + p->len;
30 	if ((t = calloc(1, newlen)) == NULL)
31 		return 0;
32 	memcpy(t, p->buf, p->len);
33 	memcpy(t + p->len, buf, len);
34 	free(p->buf);
35 	p->buf = t;
36 	p->len = newlen;
37 	return 1;
38 }
39 
40 int
41 parser_set_buf(struct parser *p, const char *buf, size_t len)
42 {
43 	char *tmp;
44 
45 	if (len == 0) {
46 		p->len = 0;
47 		free(p->buf);
48 		p->buf = NULL;
49 		return 1;
50 	}
51 
52 	/*
53 	 * p->buf and buf can (and probably almost always will)
54 	 * overlap!
55 	 */
56 
57 	if ((tmp = calloc(1, len)) == NULL)
58 		return 0;
59 	memcpy(tmp, buf, len);
60 	free(p->buf);
61 	p->buf = tmp;
62 	p->len = len;
63 	return 1;
64 }
65 
66 int
67 parser_foreach_line(struct parser *p, const char *buf, size_t size,
68     parsechunkfn fn)
69 {
70 	char		*b, *e;
71 	unsigned int	 ch;
72 	size_t		 i, l, len;
73 
74 	if (!parser_append(p, buf, size))
75 		return 0;
76 	b = p->buf;
77 	len = p->len;
78 
79 	if (!(p->flags & PARSER_IN_BODY) && len < 3)
80 		return 1;
81 
82 	if (!(p->flags & PARSER_IN_BODY)) {
83                 p->flags |= PARSER_IN_BODY;
84 
85 		/*
86 		 * drop the BOM: only UTF-8 is supported, and there
87 		 * it's useless; some editors may still add one
88 		 * though.
89 		 */
90 		if (memmem(b, len, "\xEF\xBB\xBF", 3) == b) {
91 			b += 3;
92 			len -= 3;
93 		}
94 	}
95 
96 	/* drop every "funny" ASCII character */
97 	for (i = 0; i < len; ) {
98 		ch = b[i];
99 		if ((ch >= ' ' || ch == '\n' || ch == '\t')
100 		    && ch != 127) { /* del */
101 			++i;
102 			continue;
103 		}
104 		memmove(&b[i], &b[i+1], len - i);
105 		len--;
106 	}
107 
108 	while (len > 0) {
109 		if ((e = memmem((char*)b, len, "\n", 1)) == NULL)
110 			break;
111 		l = e - b;
112 
113 		if (!fn(p, b, l))
114 			return 0;
115 
116 		len -= l;
117 		b += l;
118 
119 		if (len > 0) {
120 			/* skip \n */
121 			len--;
122 			b++;
123 		}
124 	}
125 
126 	return parser_set_buf(p, b, len);
127 }