1 /*-
2 * Copyright (c) 2014 Sebastian Freundt
3 * All rights reserved.
4 *
5 * Redistribution and use in source and binary forms, with or without
6 * modification, are permitted provided that the following conditions
7 * are met:
8 * 1. Redistributions of source code must retain the above copyright
9 * notice, this list of conditions and the following disclaimer
10 * in this position and unchanged.
11 * 2. Redistributions in binary form must reproduce the above copyright
12 * notice, this list of conditions and the following disclaimer in the
13 * documentation and/or other materials provided with the distribution.
14 *
15 * THIS SOFTWARE IS PROVIDED BY THE AUTHOR(S) ``AS IS'' AND ANY EXPRESS OR
16 * IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE IMPLIED WARRANTIES
17 * OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE ARE DISCLAIMED.
18 * IN NO EVENT SHALL THE AUTHOR(S) BE LIABLE FOR ANY DIRECT, INDIRECT,
19 * INCIDENTAL, SPECIAL, EXEMPLARY, OR CONSEQUENTIAL DAMAGES (INCLUDING, BUT
20 * NOT LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS OR SERVICES; LOSS OF USE,
21 * DATA, OR PROFITS; OR BUSINESS INTERRUPTION) HOWEVER CAUSED AND ON ANY
22 * THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT LIABILITY, OR TORT
23 * (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY OUT OF THE USE OF
24 * THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF SUCH DAMAGE.
25 */
26
27 #include "test.h"
28
DEFINE_TEST(test_read_format_warc)29 DEFINE_TEST(test_read_format_warc)
30 {
31 char buff[256U];
32 const char reffile[] = "test_read_format_warc.warc";
33 struct archive_entry *ae;
34 struct archive *a;
35
36 extract_reference_file(reffile);
37 assert((a = archive_read_new()) != NULL);
38 assertEqualIntA(a, ARCHIVE_OK, archive_read_support_filter_all(a));
39 assertEqualIntA(a, ARCHIVE_OK, archive_read_support_format_all(a));
40 assertEqualIntA(a, ARCHIVE_OK,
41 archive_read_open_filename(a, reffile, 10240));
42
43 /* First Entry */
44 assertEqualIntA(a, ARCHIVE_OK, archive_read_next_header(a, &ae));
45 assertEqualString("sometest.txt", archive_entry_pathname(ae));
46 assertEqualInt(1402399833, archive_entry_mtime(ae));
47 assertEqualInt(0, archive_entry_uid(ae));
48 assertEqualInt(0, archive_entry_gid(ae));
49 assertEqualInt(AE_IFREG, archive_entry_filetype(ae));
50 assertEqualInt(65, archive_entry_size(ae));
51 assertEqualInt(65, archive_read_data(a, buff, sizeof(buff)));
52 assertEqualMem(buff, "This is a sample text file for libarchive's WARC reader/writer.\n\n", 65);
53 assertEqualInt(archive_entry_is_encrypted(ae), 0);
54 assertEqualIntA(a, archive_read_has_encrypted_entries(a), ARCHIVE_READ_FORMAT_ENCRYPTION_UNSUPPORTED);
55
56 /* Second Entry */
57 assertEqualIntA(a, ARCHIVE_OK, archive_read_next_header(a, &ae));
58 assertEqualString("moretest.txt", archive_entry_pathname(ae));
59 assertEqualInt(1402399884, archive_entry_mtime(ae));
60 assertEqualInt(0, archive_entry_uid(ae));
61 assertEqualInt(0, archive_entry_gid(ae));
62 assertEqualInt(AE_IFREG, archive_entry_filetype(ae));
63 assertEqualInt(76, archive_entry_size(ae));
64 assertA(76 == archive_read_data(a, buff, sizeof(buff)));
65 assertEqualMem(buff, "The beauty is that WARC remains ASCII only iff all contents are ASCII only.\n", 76);
66 assertEqualInt(archive_entry_is_encrypted(ae), 0);
67 assertEqualIntA(a, archive_read_has_encrypted_entries(a), ARCHIVE_READ_FORMAT_ENCRYPTION_UNSUPPORTED);
68
69 /* Test EOF */
70 assertEqualIntA(a, ARCHIVE_EOF, archive_read_next_header(a, &ae));
71 assertEqualInt(2, archive_file_count(a));
72
73 /* Verify archive format. */
74 assertEqualIntA(a, ARCHIVE_FILTER_NONE, archive_filter_code(a, 0));
75 assertEqualIntA(a, ARCHIVE_FORMAT_WARC, archive_format(a));
76
77 /* Verify closing and resource freeing */
78 assertEqualIntA(a, ARCHIVE_OK, archive_read_close(a));
79 assertEqualInt(ARCHIVE_OK, archive_read_free(a));
80 }
81
DEFINE_TEST(test_read_format_warc_incomplete)82 DEFINE_TEST(test_read_format_warc_incomplete)
83 {
84 const char reffile[] = "test_read_format_warc_incomplete.warc";
85 struct archive_entry *ae;
86 struct archive *a;
87
88 extract_reference_file(reffile);
89 assert((a = archive_read_new()) != NULL);
90 assertEqualIntA(a, ARCHIVE_OK, archive_read_support_filter_all(a));
91 assertEqualIntA(a, ARCHIVE_OK, archive_read_support_format_all(a));
92 assertEqualIntA(a, ARCHIVE_OK,
93 archive_read_open_filename(a, reffile, 10240));
94
95 /* Entry cannot be parsed */
96 assertEqualIntA(a, ARCHIVE_FATAL, archive_read_next_header(a, &ae));
97
98 /* Verify archive format. */
99 assertEqualIntA(a, ARCHIVE_FILTER_NONE, archive_filter_code(a, 0));
100
101 /* Verify closing and resource freeing */
102 assertEqualIntA(a, ARCHIVE_OK, archive_read_close(a));
103 assertEqualInt(ARCHIVE_OK, archive_read_free(a));
104 }
105
DEFINE_TEST(test_read_format_warc_skip_after_extract)106 DEFINE_TEST(test_read_format_warc_skip_after_extract)
107 {
108 #define WARC_APPEND(s) do { \
109 memcpy(p, s, sizeof(s) - 1); \
110 p += sizeof(s) - 1; \
111 } while (0)
112 const size_t body_size = 200000;
113 char *warc;
114 char *p;
115 char second[6];
116 size_t used;
117 int fd;
118 struct archive_entry *ae;
119 struct archive *a;
120
121 warc = malloc(body_size + 1024);
122 assert(warc != NULL);
123 p = warc;
124
125 WARC_APPEND("WARC/1.0\r\n"
126 "WARC-Type: resource\r\n"
127 "WARC-Date: 2014-06-10T10:10:10Z\r\n"
128 "WARC-Target-URI: file://first.txt\r\n"
129 "Content-Length: 200000\r\n"
130 "\r\n");
131 memset(p, 'A', body_size);
132 p += body_size;
133 WARC_APPEND("\r\n\r\n"
134 "WARC/1.0\r\n"
135 "WARC-Type: resource\r\n"
136 "WARC-Date: 2014-06-10T10:10:11Z\r\n"
137 "WARC-Target-URI: file://second.txt\r\n"
138 "Content-Length: 6\r\n"
139 "\r\n"
140 "SECOND"
141 "\r\n\r\n");
142 used = (size_t)(p - warc);
143
144 assert((a = archive_read_new()) != NULL);
145 assertEqualIntA(a, ARCHIVE_OK, archive_read_support_filter_all(a));
146 assertEqualIntA(a, ARCHIVE_OK, archive_read_support_format_all(a));
147 assertEqualIntA(a, ARCHIVE_OK,
148 archive_read_open_memory2(a, warc, used, 10240));
149
150 assertEqualIntA(a, ARCHIVE_OK, archive_read_next_header(a, &ae));
151 assertEqualString("first.txt", archive_entry_pathname(ae));
152
153 fd = open("warc.out", O_WRONLY | O_CREAT | O_TRUNC | O_BINARY, 0644);
154 assert(fd >= 0);
155 assertEqualIntA(a, ARCHIVE_OK, archive_read_data_into_fd(a, fd));
156 assertEqualInt(0, close(fd));
157
158 assertEqualIntA(a, ARCHIVE_OK, archive_read_next_header(a, &ae));
159 assertEqualString("second.txt", archive_entry_pathname(ae));
160 assertEqualInt(6, archive_entry_size(ae));
161 assertEqualInt(6, archive_read_data(a, second, sizeof(second)));
162 assertEqualMem(second, "SECOND", 6);
163
164 assertEqualIntA(a, ARCHIVE_OK, archive_read_close(a));
165 assertEqualInt(ARCHIVE_OK, archive_read_free(a));
166 free(warc);
167 #undef WARC_APPEND
168 }
169
DEFINE_TEST(test_read_format_warc_truncated_body)170 DEFINE_TEST(test_read_format_warc_truncated_body)
171 {
172 static const char warc[] =
173 "WARC/1.0\r\n"
174 "WARC-Type: resource\r\n"
175 "WARC-Date: 2014-06-10T10:10:10Z\r\n"
176 "WARC-Target-URI: file://test.txt\r\n"
177 "Content-Length: 10\r\n"
178 "\r\n"
179 "ABC";
180 char body[3];
181 struct archive_entry *ae;
182 struct archive *a;
183
184 assert((a = archive_read_new()) != NULL);
185 assertEqualIntA(a, ARCHIVE_OK,
186 archive_read_support_filter_all(a));
187 assertEqualIntA(a, ARCHIVE_OK,
188 archive_read_support_format_all(a));
189 assertEqualIntA(a, ARCHIVE_OK,
190 archive_read_open_memory(a, warc, sizeof(warc) - 1U));
191
192 assertEqualIntA(a, ARCHIVE_OK,
193 archive_read_next_header(a, &ae));
194 assertEqualInt(3,
195 archive_read_data(a, body, sizeof(body)));
196 assertEqualMem(body, "ABC", sizeof(body));
197 assertEqualIntA(a, ARCHIVE_FATAL,
198 archive_read_data(a, body, sizeof(body)));
199 assertEqualString("Truncated WARC file data",
200 archive_error_string(a));
201
202 assertEqualIntA(a, ARCHIVE_OK, archive_read_close(a));
203 assertEqualInt(ARCHIVE_OK, archive_read_free(a));
204 }
205