1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
91
92
93
94
95
96
97
98
99
100
101
102
103
104
105
106
107
108
109
110
111
112
113
114
115
116
117
118
119
120
121
122
123
124
125
126
127
128
129
130
131
132
133
134
135
136
137
138
139
140
141
142
143
144
145
146
147
148
149
150
151
152
153
154
155
156
157
158
159
160
161
162
163
164
165
166
167
168
169
170
171
172
173
174
175
176
177
178
179
180
181
182
183
184
185
186
187
188
189
190
191
192
193
194
195
196
197
198
199
200
201
202
203
204
205
206
207
208
209
210
211
212
213
214
215
216
217
218
219
220
221
222
223
224
225
226
227
228
229
230
231
232
233
234
235
236
237
238
239
240
241
242
243
244
245
246
247
248
249
250
251
252
253
254
255
256
257
258
259
260
261
262
263
264
265
266
267
268
269
270
271
272
273
274
275
276
277
278
279
280
281
282
283
284
285
286
287
288
289
290
291
292
293
294
295
296
297
298
299
300
301
302
303
304
305
306
307
308
309
310
311
312
313
314
315
316
317
318
319
320
321
322
323
324
325
326
327
328
329
330
331
332
333
334
335
336
337
338
339
340
341
342
343
344
345
346
347
348
349
350
351
352
353
354
355
356
357
358
359
360
361
362
363
364
365
366
367
368
369
370
371
372
373
374
375
376
377
378
379
380
381
382
383
384
385
386
387
388
389
390
391
392
393
394
395
396
397
398
399
400
401
402
403
404
405
406
407
408
409
410
411
412
413
414
415
416
417
418
419
420
421
422
423
|
// Copyright (c) 2015 The Chromium Authors. All rights reserved.
// Use of this source code is governed by a BSD-style license that can be
// found in the LICENSE file.
#include "components/safe_browsing_db/util.h"
#include <stddef.h>
#include "base/macros.h"
#include "base/strings/string_util.h"
#include "base/trace_event/trace_event.h"
#include "crypto/sha2.h"
#include "net/base/escape.h"
#include "url/gurl.h"
#include "url/url_util.h"
namespace safe_browsing {
// Utility functions -----------------------------------------------------------
namespace {
bool IsKnownList(const std::string& name) {
for (size_t i = 0; i < arraysize(kAllLists); ++i) {
if (!strcmp(kAllLists[i], name.c_str())) {
return true;
}
}
return false;
}
} // namespace
// SBCachedFullHashResult ------------------------------------------------------
SBCachedFullHashResult::SBCachedFullHashResult() {}
SBCachedFullHashResult::SBCachedFullHashResult(
const base::Time& in_expire_after)
: expire_after(in_expire_after) {}
SBCachedFullHashResult::SBCachedFullHashResult(
const SBCachedFullHashResult& other) = default;
SBCachedFullHashResult::~SBCachedFullHashResult() {}
// Listnames that browser can process.
const char kMalwareList[] = "goog-malware-shavar";
const char kPhishingList[] = "goog-phish-shavar";
const char kBinUrlList[] = "goog-badbinurl-shavar";
const char kCsdWhiteList[] = "goog-csdwhite-sha256";
const char kDownloadWhiteList[] = "goog-downloadwhite-digest256";
const char kExtensionBlacklist[] = "goog-badcrxids-digestvar";
const char kIPBlacklist[] = "goog-badip-digest256";
const char kUnwantedUrlList[] = "goog-unwanted-shavar";
const char kInclusionWhitelist[] = "goog-csdinclusionwhite-sha256";
const char kModuleWhitelist[] = "goog-whitemodule-digest256";
const char kResourceBlacklist[] = "goog-badresource-shavar";
const char* kAllLists[11] = {
kMalwareList, kPhishingList, kBinUrlList, kCsdWhiteList,
kDownloadWhiteList, kExtensionBlacklist, kIPBlacklist, kUnwantedUrlList,
kInclusionWhitelist, kModuleWhitelist, kResourceBlacklist,
};
ListType GetListId(const base::StringPiece& name) {
ListType id;
if (name == kMalwareList) {
id = MALWARE;
} else if (name == kPhishingList) {
id = PHISH;
} else if (name == kBinUrlList) {
id = BINURL;
} else if (name == kCsdWhiteList) {
id = CSDWHITELIST;
} else if (name == kDownloadWhiteList) {
id = DOWNLOADWHITELIST;
} else if (name == kExtensionBlacklist) {
id = EXTENSIONBLACKLIST;
} else if (name == kIPBlacklist) {
id = IPBLACKLIST;
} else if (name == kUnwantedUrlList) {
id = UNWANTEDURL;
} else if (name == kInclusionWhitelist) {
id = INCLUSIONWHITELIST;
} else if (name == kModuleWhitelist) {
id = MODULEWHITELIST;
} else if (name == kResourceBlacklist) {
id = RESOURCEBLACKLIST;
} else {
id = INVALID;
}
return id;
}
bool GetListName(ListType list_id, std::string* list) {
switch (list_id) {
case MALWARE:
*list = kMalwareList;
break;
case PHISH:
*list = kPhishingList;
break;
case BINURL:
*list = kBinUrlList;
break;
case CSDWHITELIST:
*list = kCsdWhiteList;
break;
case DOWNLOADWHITELIST:
*list = kDownloadWhiteList;
break;
case EXTENSIONBLACKLIST:
*list = kExtensionBlacklist;
break;
case IPBLACKLIST:
*list = kIPBlacklist;
break;
case UNWANTEDURL:
*list = kUnwantedUrlList;
break;
case INCLUSIONWHITELIST:
*list = kInclusionWhitelist;
break;
case MODULEWHITELIST:
*list = kModuleWhitelist;
case RESOURCEBLACKLIST:
*list = kResourceBlacklist;
break;
default:
return false;
}
DCHECK(IsKnownList(*list));
return true;
}
SBFullHash SBFullHashForString(const base::StringPiece& str) {
SBFullHash h;
crypto::SHA256HashString(str, &h.full_hash, sizeof(h.full_hash));
return h;
}
SBFullHash StringToSBFullHash(const std::string& hash_in) {
DCHECK_EQ(crypto::kSHA256Length, hash_in.size());
SBFullHash hash_out;
memcpy(hash_out.full_hash, hash_in.data(), crypto::kSHA256Length);
return hash_out;
}
std::string SBFullHashToString(const SBFullHash& hash) {
DCHECK_EQ(crypto::kSHA256Length, sizeof(hash.full_hash));
return std::string(hash.full_hash, sizeof(hash.full_hash));
}
std::string Unescape(const std::string& url) {
std::string unescaped_str(url);
const int kMaxLoopIterations = 1024;
size_t old_size = 0;
int loop_var = 0;
do {
old_size = unescaped_str.size();
unescaped_str = net::UnescapeURLComponent(
unescaped_str, net::UnescapeRule::SPOOFING_AND_CONTROL_CHARS |
net::UnescapeRule::SPACES |
net::UnescapeRule::URL_SPECIAL_CHARS);
} while (old_size != unescaped_str.size() &&
++loop_var <= kMaxLoopIterations);
return unescaped_str;
}
std::string Escape(const std::string& url) {
std::string escaped_str;
// The escaped string is larger so allocate double the length to reduce the
// chance of the string being grown.
escaped_str.reserve(url.length() * 2);
const char* kHexString = "0123456789ABCDEF";
for (size_t i = 0; i < url.length(); i++) {
unsigned char c = static_cast<unsigned char>(url[i]);
if (c <= ' ' || c > '~' || c == '#' || c == '%') {
escaped_str += '%';
escaped_str += kHexString[c >> 4];
escaped_str += kHexString[c & 0xf];
} else {
escaped_str += c;
}
}
return escaped_str;
}
std::string RemoveConsecutiveChars(base::StringPiece str, const char c) {
std::string output;
// Output is at most the length of the original string.
output.reserve(str.size());
size_t i = 0;
while (i < str.size()) {
output.append(1, str[i++]);
if (str[i - 1] == c) {
while (i < str.size() && str[i] == c) {
i++;
}
}
}
return output;
}
// Canonicalizes url as per Google Safe Browsing Specification.
// See section 6.1 in
// http://code.google.com/p/google-safe-browsing/wiki/Protocolv2Spec.
void CanonicalizeUrl(const GURL& url,
std::string* canonicalized_hostname,
std::string* canonicalized_path,
std::string* canonicalized_query) {
DCHECK(url.is_valid());
// We only canonicalize "normal" URLs.
if (!url.IsStandard())
return;
// Following canonicalization steps are excluded since url parsing takes care
// of those :-
// 1. Remove any tab (0x09), CR (0x0d), and LF (0x0a) chars from url.
// (Exclude escaped version of these chars).
// 2. Normalize hostname to 4 dot-seperated decimal values.
// 3. Lowercase hostname.
// 4. Resolve path sequences "/../" and "/./".
// That leaves us with the following :-
// 1. Remove fragment in URL.
GURL url_without_fragment;
GURL::Replacements f_replacements;
f_replacements.ClearRef();
f_replacements.ClearUsername();
f_replacements.ClearPassword();
url_without_fragment = url.ReplaceComponents(f_replacements);
// 2. Do URL unescaping until no more hex encoded characters exist.
std::string url_unescaped_str(Unescape(url_without_fragment.spec()));
url::Parsed parsed;
url::ParseStandardURL(url_unescaped_str.data(), url_unescaped_str.length(),
&parsed);
// 3. In hostname, remove all leading and trailing dots.
base::StringPiece host;
if (parsed.host.len > 0)
host.set(url_unescaped_str.data() + parsed.host.begin, parsed.host.len);
base::StringPiece host_without_end_dots =
base::TrimString(host, ".", base::TrimPositions::TRIM_ALL);
// 4. In hostname, replace consecutive dots with a single dot.
std::string host_without_consecutive_dots(RemoveConsecutiveChars(
host_without_end_dots, '.'));
// 5. In path, replace runs of consecutive slashes with a single slash.
base::StringPiece path;
if (parsed.path.len > 0)
path.set(url_unescaped_str.data() + parsed.path.begin, parsed.path.len);
std::string path_without_consecutive_slash(RemoveConsecutiveChars(path, '/'));
url::Replacements<char> hp_replacements;
hp_replacements.SetHost(
host_without_consecutive_dots.data(),
url::Component(0, host_without_consecutive_dots.length()));
hp_replacements.SetPath(
path_without_consecutive_slash.data(),
url::Component(0, path_without_consecutive_slash.length()));
std::string url_unescaped_with_can_hostpath;
url::StdStringCanonOutput output(&url_unescaped_with_can_hostpath);
url::Parsed temp_parsed;
url::ReplaceComponents(url_unescaped_str.data(),
url_unescaped_str.length(),
parsed,
hp_replacements,
NULL,
&output,
&temp_parsed);
output.Complete();
// 6. Step needed to revert escaping done in url::ReplaceComponents.
url_unescaped_with_can_hostpath = Unescape(url_unescaped_with_can_hostpath);
// 7. After performing all above steps, percent-escape all chars in url which
// are <= ASCII 32, >= 127, #, %. Escapes must be uppercase hex characters.
std::string escaped_canon_url_str(Escape(url_unescaped_with_can_hostpath));
url::Parsed final_parsed;
url::ParseStandardURL(escaped_canon_url_str.data(),
escaped_canon_url_str.length(),
&final_parsed);
if (canonicalized_hostname && final_parsed.host.len > 0) {
*canonicalized_hostname =
escaped_canon_url_str.substr(final_parsed.host.begin,
final_parsed.host.len);
}
if (canonicalized_path && final_parsed.path.len > 0) {
*canonicalized_path = escaped_canon_url_str.substr(final_parsed.path.begin,
final_parsed.path.len);
}
if (canonicalized_query && final_parsed.query.len > 0) {
*canonicalized_query = escaped_canon_url_str.substr(
final_parsed.query.begin, final_parsed.query.len);
}
}
void UrlToFullHashes(const GURL& url,
bool include_whitelist_hashes,
std::vector<SBFullHash>* full_hashes) {
// Include this function in traces because it's not cheap so it should be
// called sparingly.
TRACE_EVENT2("loader", "safe_browsing::UrlToFullHashes", "url", url.spec(),
"include_whitelist_hashes", include_whitelist_hashes);
std::vector<std::string> hosts;
if (url.HostIsIPAddress()) {
hosts.push_back(url.host());
} else {
GenerateHostsToCheck(url, &hosts);
}
std::vector<std::string> paths;
GeneratePathsToCheck(url, &paths);
for (const std::string& host : hosts) {
for (const std::string& path : paths) {
full_hashes->push_back(
SBFullHashForString(host + path));
// We may have /foo as path-prefix in the whitelist which should
// also match with /foo/bar and /foo?bar. Hence, for every path
// that ends in '/' we also add the path without the slash.
if (include_whitelist_hashes && path.size() > 1 &&
path[path.size() - 1] == '/') {
full_hashes->push_back(SBFullHashForString(
host + path.substr(0, path.size() - 1)));
}
}
}
}
void GenerateHostsToCheck(const GURL& url, std::vector<std::string>* hosts) {
hosts->clear();
std::string canon_host;
CanonicalizeUrl(url, &canon_host, NULL, NULL);
const std::string host = canon_host; // const sidesteps GCC bugs below!
if (host.empty())
return;
// Per the Safe Browsing Protocol v2 spec, we try the host, and also up to 4
// hostnames formed by starting with the last 5 components and successively
// removing the leading component. The last component isn't examined alone,
// since it's the TLD or a subcomponent thereof.
//
// Note that we don't need to be clever about stopping at the "real" eTLD --
// the data on the server side has been filtered to ensure it will not
// blacklist a whole TLD, and it's not significantly slower on our side to
// just check too much.
//
// Also note that because we have a simple blacklist, not some sort of complex
// whitelist-in-blacklist or vice versa, it doesn't matter what order we check
// these in.
const size_t kMaxHostsToCheck = 4;
bool skipped_last_component = false;
for (std::string::const_reverse_iterator i(host.rbegin());
i != host.rend() && hosts->size() < kMaxHostsToCheck; ++i) {
if (*i == '.') {
if (skipped_last_component)
hosts->push_back(std::string(i.base(), host.end()));
else
skipped_last_component = true;
}
}
hosts->push_back(host);
}
void GeneratePathsToCheck(const GURL& url, std::vector<std::string>* paths) {
paths->clear();
std::string canon_path;
std::string canon_query;
CanonicalizeUrl(url, NULL, &canon_path, &canon_query);
const std::string path = canon_path; // const sidesteps GCC bugs below!
const std::string query = canon_query;
if (path.empty())
return;
// Per the Safe Browsing Protocol v2 spec, we try the exact path with/without
// the query parameters, and also up to 4 paths formed by starting at the root
// and adding more path components.
//
// As with the hosts above, it doesn't matter what order we check these in.
const size_t kMaxPathsToCheck = 4;
for (std::string::const_iterator i(path.begin());
i != path.end() && paths->size() < kMaxPathsToCheck; ++i) {
if (*i == '/')
paths->push_back(std::string(path.begin(), i + 1));
}
if (!paths->empty() && paths->back() != path)
paths->push_back(path);
if (!query.empty())
paths->push_back(path + "?" + query);
}
void GeneratePatternsToCheck(const GURL& url, std::vector<std::string>* urls) {
std::vector<std::string> hosts, paths;
GenerateHostsToCheck(url, &hosts);
GeneratePathsToCheck(url, &paths);
for (size_t h = 0; h < hosts.size(); ++h) {
for (size_t p = 0; p < paths.size(); ++p) {
urls->push_back(hosts[h] + paths[p]);
}
}
}
} // namespace safe_browsing
|