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
|
// Copyright (c) 2006-2008 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.
// This command-line program converts an effective-TLD data file in UTF-8 from
// the format provided by Mozilla to the format expected by Chrome. Any errors
// or warnings are recorded in tld_cleanup.log.
//
// In particular, it
// * Strips blank lines and comments, as well as notes for individual rules.
// * Changes all line endings to LF.
// * Strips a single leading and/or trailing dot from each rule, if present.
// * Logs a warning if a rule contains '!' or '*.' other than at the beginning
// of the rule. (This also catches multiple ! or *. at the start of a rule.)
// * Logs a warning if GURL reports a rule as invalid, but keeps the rule.
// * Canonicalizes each rule's domain by converting it to a GURL and back.
// * Adds explicit rules for true TLDs found in any rule.
#include <set>
#include <string>
#include "base/at_exit.h"
#include "base/file_util.h"
#include "base/icu_util.h"
#include "base/logging.h"
#include "base/file_path.h"
#include "base/file_util.h"
#include "base/path_service.h"
#include "base/process_util.h"
#include "base/string_util.h"
#include "googleurl/src/gurl.h"
#include "googleurl/src/url_parse.h"
typedef std::set<std::string> StringSet;
// Writes the list of domain rules contained in the 'rules' set to the
// 'outfile', with each rule terminated by a LF. The file must already have
// been created with write access.
bool WriteRules(const StringSet& rules, FilePath outfile) {
std::string data;
for (StringSet::const_iterator iter = rules.begin();
iter != rules.end();
++iter) {
data.append(*iter);
data.append("\n");
}
int written = file_util::WriteFile(outfile.ToWStringHack(), data.data(),
data.size());
return written == static_cast<int>(data.size());
}
// These result codes should be in increasing order of severity.
typedef enum {
kSuccess,
kWarning,
kError,
} NormalizeResult;
// Adjusts the rule to a standard form: removes single extraneous dots and
// canonicalizes it using GURL. Returns kSuccess if the rule is interpreted as
// valid; logs a warning and returns kWarning if it is probably invalid; and
// logs an error and returns kError if the rule is (almost) certainly invalid.
NormalizeResult NormalizeRule(std::string* rule) {
NormalizeResult result = kSuccess;
// Strip single leading and trailing dots.
if (rule->at(0) == '.')
rule->erase(0, 1);
if (rule->size() == 0) {
LOG(WARNING) << "Ignoring empty rule";
return kWarning;
}
if (rule->at(rule->size() - 1) == '.')
rule->erase(rule->size() - 1, 1);
if (rule->size() == 0) {
LOG(WARNING) << "Ignoring empty rule";
return kWarning;
}
// Allow single leading '*.' or '!', saved here so it's not canonicalized.
bool wildcard = false;
bool exception = false;
size_t start_offset = 0;
if (rule->at(0) == '!') {
rule->erase(0, 1);
exception = true;
} else if (rule->find("*.") == 0) {
rule->erase(0, 2);
wildcard = true;
}
if (rule->size() == 0) {
LOG(WARNING) << "Ignoring empty rule";
return kWarning;
}
// Warn about additional '*.' or '!'.
if (rule->find("*.", start_offset) != std::string::npos ||
rule->find('!', start_offset) != std::string::npos) {
LOG(WARNING) << "Keeping probably invalid rule: " << *rule;
result = kWarning;
}
// Make a GURL and normalize it, then get the host back out.
std::string url = "http://";
url.append(*rule);
GURL gurl(url);
const std::string& spec = gurl.possibly_invalid_spec();
url_parse::Component host = gurl.parsed_for_possibly_invalid_spec().host;
if (host.len < 0) {
LOG(ERROR) << "Ignoring rule that couldn't be normalized: " << *rule;
return kError;
}
if (!gurl.is_valid()) {
LOG(WARNING) << "Keeping rule that GURL says is invalid: " << *rule;
result = kWarning;
}
rule->assign(spec.substr(host.begin, host.len));
// Restore wildcard or exception marker.
if (exception)
rule->insert(0, 1, '!');
else if (wildcard)
rule->insert(0, "*.");
return result;
}
// Loads the file described by 'in_filename', converts it to the desired format
// (see the file comments above), and saves it into 'out_filename'. Returns
// the most severe of the result codes encountered when normalizing the rules.
NormalizeResult NormalizeFile(const std::wstring& in_filename,
const std::wstring& out_filename) {
std::string data;
if (!file_util::ReadFileToString(in_filename, &data)) {
fwprintf(stderr, L"Unable to read file %ls\n", in_filename.c_str());
// We return success since we've already reported the error.
return kSuccess;
}
// We do a lot of string assignment during parsing, but simplicity is more
// important than performance here.
std::string rule;
NormalizeResult result = kSuccess;
size_t line_start = 0;
size_t line_end = 0;
StringSet rules;
while (line_start < data.size()) {
// Skip comments.
if (line_start + 1 < data.size() &&
data[line_start] == '/' &&
data[line_start + 1] == '/') {
line_end = data.find_first_of("\r\n", line_start);
if (line_end == std::string::npos)
line_end = data.size();
} else {
// Truncate at first whitespace.
line_end = data.find_first_of("\r\n \t", line_start);
if (line_end == std::string::npos)
line_end = data.size();
rule.assign(data.data(), line_start, line_end - line_start);
NormalizeResult new_result = NormalizeRule(&rule);
if (new_result != kError) {
rules.insert(rule);
// Add true TLD for multi-level rules.
size_t tld_start = rule.find_last_of('.');
if (tld_start != std::string::npos && tld_start + 1 < rule.size())
rules.insert(rule.substr(tld_start + 1));
}
result = std::max(result, new_result);
}
// Find beginning of next non-empty line.
line_start = data.find_first_of("\r\n", line_end);
if (line_start == std::string::npos)
line_start = data.size();
line_start = data.find_first_not_of("\r\n", line_start);
if (line_start == std::string::npos)
line_start = data.size();
}
if (!WriteRules(rules, FilePath::FromWStringHack(out_filename))) {
LOG(ERROR) << "Error(s) writing " << out_filename;
result = kError;
}
return result;
}
int main(int argc, const char* argv[]) {
base::EnableTerminationOnHeapCorruption();
if (argc != 3) {
fprintf(stderr, "Normalizes and verifies UTF-8 TLD data files\n");
fprintf(stderr, "Usage: %s <input> <output>\n", argv[0]);
return 1;
}
// Manages the destruction of singletons.
base::AtExitManager exit_manager;
// Only use OutputDebugString in debug mode.
#ifdef NDEBUG
logging::LoggingDestination destination = logging::LOG_ONLY_TO_FILE;
#else
logging::LoggingDestination destination =
logging::LOG_TO_BOTH_FILE_AND_SYSTEM_DEBUG_LOG;
#endif
CommandLine::Init(argc, argv);
FilePath log_filename;
PathService::Get(base::DIR_EXE, &log_filename);
log_filename = log_filename.Append(FILE_PATH_LITERAL("tld_cleanup.log"));
logging::InitLogging(log_filename.value().c_str(),
destination,
logging::LOCK_LOG_FILE,
logging::DELETE_OLD_LOG_FILE);
icu_util::Initialize();
NormalizeResult result = NormalizeFile(UTF8ToWide(argv[1]),
UTF8ToWide(argv[2]));
if (result != kSuccess) {
fprintf(stderr,
"Errors or warnings processing file. See log in tld_cleanup.log.");
}
if (result == kError)
return 1;
return 0;
}
|
