OLD | NEW |
---|---|
1 // Copyright 2013 The Chromium Authors. All rights reserved. | 1 // Copyright 2013 The Chromium Authors. All rights reserved. |
2 // Use of this source code is governed by a BSD-style license that can be | 2 // Use of this source code is governed by a BSD-style license that can be |
3 // found in the LICENSE file. | 3 // found in the LICENSE file. |
4 | 4 |
5 #include "chrome/utility/importer/bookmark_html_reader.h" | 5 #include "chrome/utility/importer/bookmark_html_reader.h" |
6 | 6 |
7 #include "base/callback.h" | 7 #include "base/callback.h" |
8 #include "base/files/file_util.h" | 8 #include "base/files/file_util.h" |
9 #include "base/i18n/icu_string_conversions.h" | 9 #include "base/i18n/icu_string_conversions.h" |
10 #include "base/strings/string_number_conversions.h" | 10 #include "base/strings/string_number_conversions.h" |
(...skipping 101 matching lines...) Expand 10 before | Expand all | Expand 10 after Loading... | |
112 (cancellation_callback.is_null() || !cancellation_callback.Run()); | 112 (cancellation_callback.is_null() || !cancellation_callback.Run()); |
113 ++i) { | 113 ++i) { |
114 std::string line; | 114 std::string line; |
115 base::TrimString(lines[i], " ", &line); | 115 base::TrimString(lines[i], " ", &line); |
116 | 116 |
117 // Remove "<HR>" if |line| starts with it. "<HR>" is the bookmark entries | 117 // Remove "<HR>" if |line| starts with it. "<HR>" is the bookmark entries |
118 // separator in Firefox that Chrome does not support. Note that there can be | 118 // separator in Firefox that Chrome does not support. Note that there can be |
119 // multiple "<HR>" tags at the beginning of a single line. | 119 // multiple "<HR>" tags at the beginning of a single line. |
120 // See http://crbug.com/257474. | 120 // See http://crbug.com/257474. |
121 static const char kHrTag[] = "<HR>"; | 121 static const char kHrTag[] = "<HR>"; |
122 while (base::StartsWithASCII(line, kHrTag, false)) { | 122 while (base::StartsWith(line, kHrTag, |
123 base::CompareCase::INSENSITIVE_ASCII)) { | |
123 line.erase(0, arraysize(kHrTag) - 1); | 124 line.erase(0, arraysize(kHrTag) - 1); |
124 base::TrimString(line, " ", &line); | 125 base::TrimString(line, " ", &line); |
125 } | 126 } |
126 | 127 |
127 // Get the encoding of the bookmark file. | 128 // Get the encoding of the bookmark file. |
128 if (internal::ParseCharsetFromLine(line, &charset)) | 129 if (internal::ParseCharsetFromLine(line, &charset)) |
129 continue; | 130 continue; |
130 | 131 |
131 // Get the folder name. | 132 // Get the folder name. |
132 if (internal::ParseFolderNameFromLine(line, | 133 if (internal::ParseFolderNameFromLine(line, |
(...skipping 65 matching lines...) Expand 10 before | Expand all | Expand 10 after Loading... | |
198 | 199 |
199 // Save the favicon. DataURLToFaviconUsage will handle the case where | 200 // Save the favicon. DataURLToFaviconUsage will handle the case where |
200 // there is no favicon. | 201 // there is no favicon. |
201 if (favicons) | 202 if (favicons) |
202 DataURLToFaviconUsage(url, favicon, favicons); | 203 DataURLToFaviconUsage(url, favicon, favicons); |
203 | 204 |
204 continue; | 205 continue; |
205 } | 206 } |
206 | 207 |
207 // Bookmarks in sub-folder are encapsulated with <DL> tag. | 208 // Bookmarks in sub-folder are encapsulated with <DL> tag. |
208 if (base::StartsWithASCII(line, "<DL>", false)) { | 209 if (base::StartsWith(line, "<DL>", base::CompareCase::INSENSITIVE_ASCII)) { |
209 has_subfolder = true; | 210 has_subfolder = true; |
210 if (!last_folder.empty()) { | 211 if (!last_folder.empty()) { |
211 path.push_back(last_folder); | 212 path.push_back(last_folder); |
212 last_folder.clear(); | 213 last_folder.clear(); |
213 } | 214 } |
214 if (last_folder_on_toolbar && !toolbar_folder_index) | 215 if (last_folder_on_toolbar && !toolbar_folder_index) |
215 toolbar_folder_index = path.size(); | 216 toolbar_folder_index = path.size(); |
216 | 217 |
217 // Mark next folder empty as initial state. | 218 // Mark next folder empty as initial state. |
218 last_folder_is_empty = true; | 219 last_folder_is_empty = true; |
219 } else if (base::StartsWithASCII(line, "</DL>", false)) { | 220 } else if (base::StartsWith(line, "</DL>", |
221 base::CompareCase::INSENSITIVE_ASCII)) { | |
220 if (path.empty()) | 222 if (path.empty()) |
221 break; // Mismatch <DL>. | 223 break; // Mismatch <DL>. |
222 | 224 |
223 base::string16 folder_title = path.back(); | 225 base::string16 folder_title = path.back(); |
224 path.pop_back(); | 226 path.pop_back(); |
225 | 227 |
226 if (last_folder_is_empty) { | 228 if (last_folder_is_empty) { |
227 // Empty folder should be added explicitly. | 229 // Empty folder should be added explicitly. |
228 ImportedBookmarkEntry entry; | 230 ImportedBookmarkEntry entry; |
229 entry.is_folder = true; | 231 entry.is_folder = true; |
(...skipping 41 matching lines...) Expand 10 before | Expand all | Expand 10 after Loading... | |
271 TemplateURLData data; | 273 TemplateURLData data; |
272 data.SetURL(url_spec); | 274 data.SetURL(url_spec); |
273 *search_engine_url = url_spec; | 275 *search_engine_url = url_spec; |
274 return TemplateURL(data).SupportsReplacement(SearchTermsData()); | 276 return TemplateURL(data).SupportsReplacement(SearchTermsData()); |
275 } | 277 } |
276 | 278 |
277 namespace internal { | 279 namespace internal { |
278 | 280 |
279 bool ParseCharsetFromLine(const std::string& line, std::string* charset) { | 281 bool ParseCharsetFromLine(const std::string& line, std::string* charset) { |
280 const char kCharset[] = "charset="; | 282 const char kCharset[] = "charset="; |
281 if (base::StartsWithASCII(line, "<META", false) && | 283 if (base::StartsWith(line, "<META", base::CompareCase::INSENSITIVE_ASCII) && |
282 (line.find("CONTENT=\"") != std::string::npos || | 284 (line.find("CONTENT=\"") != std::string::npos || |
283 line.find("content=\"") != std::string::npos)) { | 285 line.find("content=\"") != std::string::npos)) { |
284 size_t begin = line.find(kCharset); | 286 size_t begin = line.find(kCharset); |
285 if (begin == std::string::npos) | 287 if (begin == std::string::npos) |
286 return false; | 288 return false; |
287 begin += std::string(kCharset).size(); | 289 begin += std::string(kCharset).size(); |
288 size_t end = line.find_first_of('\"', begin); | 290 size_t end = line.find_first_of('\"', begin); |
289 *charset = line.substr(begin, end - begin); | 291 *charset = line.substr(begin, end - begin); |
290 return true; | 292 return true; |
291 } | 293 } |
292 return false; | 294 return false; |
293 } | 295 } |
294 | 296 |
295 bool ParseFolderNameFromLine(const std::string& line, | 297 bool ParseFolderNameFromLine(const std::string& line, |
296 const std::string& charset, | 298 const std::string& charset, |
297 base::string16* folder_name, | 299 base::string16* folder_name, |
298 bool* is_toolbar_folder, | 300 bool* is_toolbar_folder, |
299 base::Time* add_date) { | 301 base::Time* add_date) { |
300 const char kFolderOpen[] = "<DT><H3"; | 302 const char kFolderOpen[] = "<DT><H3"; |
301 const char kFolderClose[] = "</H3>"; | 303 const char kFolderClose[] = "</H3>"; |
302 const char kToolbarFolderAttribute[] = "PERSONAL_TOOLBAR_FOLDER"; | 304 const char kToolbarFolderAttribute[] = "PERSONAL_TOOLBAR_FOLDER"; |
303 const char kAddDateAttribute[] = "ADD_DATE"; | 305 const char kAddDateAttribute[] = "ADD_DATE"; |
304 | 306 |
305 if (!base::StartsWithASCII(line, kFolderOpen, true)) | 307 if (!base::StartsWith(line, kFolderOpen, base::CompareCase::SENSITIVE)) |
Nico
2015/07/06 17:55:00
sure, let's treat tags as case-sensitive starting
| |
306 return false; | 308 return false; |
307 | 309 |
308 size_t end = line.find(kFolderClose); | 310 size_t end = line.find(kFolderClose); |
309 size_t tag_end = line.rfind('>', end) + 1; | 311 size_t tag_end = line.rfind('>', end) + 1; |
310 // If no end tag or start tag is broken, we skip to find the folder name. | 312 // If no end tag or start tag is broken, we skip to find the folder name. |
311 if (end == std::string::npos || tag_end < arraysize(kFolderOpen)) | 313 if (end == std::string::npos || tag_end < arraysize(kFolderOpen)) |
312 return false; | 314 return false; |
313 | 315 |
314 base::CodepageToUTF16(line.substr(tag_end, end - tag_end), charset.c_str(), | 316 base::CodepageToUTF16(line.substr(tag_end, end - tag_end), charset.c_str(), |
315 base::OnStringConversionError::SKIP, folder_name); | 317 base::OnStringConversionError::SKIP, folder_name); |
(...skipping 38 matching lines...) Expand 10 before | Expand all | Expand 10 after Loading... | |
354 const char kAddDateAttribute[] = "ADD_DATE"; | 356 const char kAddDateAttribute[] = "ADD_DATE"; |
355 const char kPostDataAttribute[] = "POST_DATA"; | 357 const char kPostDataAttribute[] = "POST_DATA"; |
356 | 358 |
357 title->clear(); | 359 title->clear(); |
358 *url = GURL(); | 360 *url = GURL(); |
359 *favicon = GURL(); | 361 *favicon = GURL(); |
360 shortcut->clear(); | 362 shortcut->clear(); |
361 post_data->clear(); | 363 post_data->clear(); |
362 *add_date = base::Time(); | 364 *add_date = base::Time(); |
363 | 365 |
364 if (!base::StartsWithASCII(line, kItemOpen, true)) | 366 if (!base::StartsWith(line, kItemOpen, base::CompareCase::SENSITIVE)) |
365 return false; | 367 return false; |
366 | 368 |
367 size_t end = line.find(kItemClose); | 369 size_t end = line.find(kItemClose); |
368 size_t tag_end = line.rfind('>', end) + 1; | 370 size_t tag_end = line.rfind('>', end) + 1; |
369 if (end == std::string::npos || tag_end < arraysize(kItemOpen)) | 371 if (end == std::string::npos || tag_end < arraysize(kItemOpen)) |
370 return false; // No end tag or start tag is broken. | 372 return false; // No end tag or start tag is broken. |
371 | 373 |
372 std::string attribute_list = line.substr(arraysize(kItemOpen), | 374 std::string attribute_list = line.substr(arraysize(kItemOpen), |
373 tag_end - arraysize(kItemOpen) - 1); | 375 tag_end - arraysize(kItemOpen) - 1); |
374 | 376 |
(...skipping 55 matching lines...) Expand 10 before | Expand all | Expand 10 after Loading... | |
430 GURL* url) { | 432 GURL* url) { |
431 const char kItemOpen[] = "<DT><A"; | 433 const char kItemOpen[] = "<DT><A"; |
432 const char kItemClose[] = "</"; | 434 const char kItemClose[] = "</"; |
433 const char kHrefAttributeUpper[] = "HREF"; | 435 const char kHrefAttributeUpper[] = "HREF"; |
434 const char kHrefAttributeLower[] = "href"; | 436 const char kHrefAttributeLower[] = "href"; |
435 | 437 |
436 title->clear(); | 438 title->clear(); |
437 *url = GURL(); | 439 *url = GURL(); |
438 | 440 |
439 // Case-insensitive check of open tag. | 441 // Case-insensitive check of open tag. |
440 if (!base::StartsWithASCII(line, kItemOpen, false)) | 442 if (!base::StartsWith(line, kItemOpen, base::CompareCase::INSENSITIVE_ASCII)) |
Nico
2015/07/06 17:55:00
…or don't! I guess <DT><A is different from <DT><A
| |
441 return false; | 443 return false; |
442 | 444 |
443 // Find any close tag. | 445 // Find any close tag. |
444 size_t end = line.find(kItemClose); | 446 size_t end = line.find(kItemClose); |
445 size_t tag_end = line.rfind('>', end) + 1; | 447 size_t tag_end = line.rfind('>', end) + 1; |
446 if (end == std::string::npos || tag_end < arraysize(kItemOpen)) | 448 if (end == std::string::npos || tag_end < arraysize(kItemOpen)) |
447 return false; // No end tag or start tag is broken. | 449 return false; // No end tag or start tag is broken. |
448 | 450 |
449 std::string attribute_list = line.substr(arraysize(kItemOpen), | 451 std::string attribute_list = line.substr(arraysize(kItemOpen), |
450 tag_end - arraysize(kItemOpen) - 1); | 452 tag_end - arraysize(kItemOpen) - 1); |
(...skipping 18 matching lines...) Expand all Loading... | |
469 *url = GURL(value); | 471 *url = GURL(value); |
470 } | 472 } |
471 } | 473 } |
472 | 474 |
473 return true; | 475 return true; |
474 } | 476 } |
475 | 477 |
476 } // namespace internal | 478 } // namespace internal |
477 | 479 |
478 } // namespace bookmark_html_reader | 480 } // namespace bookmark_html_reader |
OLD | NEW |