| OLD | NEW |
| 1 // Copyright (c) 2012 The Chromium Authors. All rights reserved. | 1 // Copyright (c) 2012 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/browser/history/url_index_private_data.h" | 5 #include "chrome/browser/history/url_index_private_data.h" |
| 6 | 6 |
| 7 #include <functional> | 7 #include <functional> |
| 8 #include <iterator> | 8 #include <iterator> |
| 9 #include <limits> | 9 #include <limits> |
| 10 #include <numeric> | 10 #include <numeric> |
| 11 #include <string> | 11 #include <string> |
| 12 #include <vector> | 12 #include <vector> |
| 13 | 13 |
| 14 #include "base/basictypes.h" | 14 #include "base/basictypes.h" |
| 15 #include "base/file_util.h" | 15 #include "base/file_util.h" |
| 16 #include "base/i18n/break_iterator.h" | 16 #include "base/i18n/break_iterator.h" |
| 17 #include "base/i18n/case_conversion.h" | 17 #include "base/i18n/case_conversion.h" |
| 18 #include "base/metrics/histogram.h" | 18 #include "base/metrics/histogram.h" |
| 19 #include "base/strings/string_util.h" | 19 #include "base/strings/string_util.h" |
| 20 #include "base/strings/utf_string_conversions.h" | 20 #include "base/strings/utf_string_conversions.h" |
| 21 #include "base/time/time.h" | 21 #include "base/time/time.h" |
| 22 #include "chrome/browser/autocomplete/autocomplete_provider.h" | 22 #include "chrome/browser/autocomplete/autocomplete_provider.h" |
| 23 #include "chrome/browser/autocomplete/url_prefix.h" | 23 #include "chrome/browser/autocomplete/url_prefix.h" |
| 24 #include "chrome/browser/history/history_database.h" | 24 #include "chrome/browser/history/history_database.h" |
| 25 #include "chrome/browser/history/history_db_task.h" | 25 #include "chrome/browser/history/history_db_task.h" |
| 26 #include "chrome/browser/history/history_service.h" | 26 #include "chrome/browser/history/history_service.h" |
| 27 #include "chrome/browser/history/in_memory_url_index.h" | 27 #include "chrome/browser/history/in_memory_url_index.h" |
| 28 #include "components/bookmarks/core/browser/bookmark_service.h" | |
| 29 #include "components/bookmarks/core/browser/bookmark_utils.h" | 28 #include "components/bookmarks/core/browser/bookmark_utils.h" |
| 29 #include "components/history/core/browser/history_client.h" |
| 30 #include "net/base/net_util.h" | 30 #include "net/base/net_util.h" |
| 31 | 31 |
| 32 #if defined(USE_SYSTEM_PROTOBUF) | 32 #if defined(USE_SYSTEM_PROTOBUF) |
| 33 #include <google/protobuf/repeated_field.h> | 33 #include <google/protobuf/repeated_field.h> |
| 34 #else | 34 #else |
| 35 #include "third_party/protobuf/src/google/protobuf/repeated_field.h" | 35 #include "third_party/protobuf/src/google/protobuf/repeated_field.h" |
| 36 #endif | 36 #endif |
| 37 | 37 |
| 38 using google::protobuf::RepeatedField; | 38 using google::protobuf::RepeatedField; |
| 39 using google::protobuf::RepeatedPtrField; | 39 using google::protobuf::RepeatedPtrField; |
| (...skipping 72 matching lines...) Expand 10 before | Expand all | Expand 10 after Loading... |
| 112 URLID url_id) | 112 URLID url_id) |
| 113 : private_data_(private_data), | 113 : private_data_(private_data), |
| 114 url_id_(url_id), | 114 url_id_(url_id), |
| 115 succeeded_(false) { | 115 succeeded_(false) { |
| 116 } | 116 } |
| 117 | 117 |
| 118 bool UpdateRecentVisitsFromHistoryDBTask::RunOnDBThread( | 118 bool UpdateRecentVisitsFromHistoryDBTask::RunOnDBThread( |
| 119 HistoryBackend* backend, | 119 HistoryBackend* backend, |
| 120 HistoryDatabase* db) { | 120 HistoryDatabase* db) { |
| 121 // Make sure the private data is going to get as many recent visits as | 121 // Make sure the private data is going to get as many recent visits as |
| 122 // ScoredHistoryMatch::GetFrecency() hopes to use. | 122 // ScoredHistoryMatch::GetFrequency() hopes to use. |
| 123 DCHECK_GE(kMaxVisitsToStoreInCache, ScoredHistoryMatch::kMaxVisitsToScore); | 123 DCHECK_GE(kMaxVisitsToStoreInCache, ScoredHistoryMatch::kMaxVisitsToScore); |
| 124 succeeded_ = db->GetMostRecentVisitsForURL(url_id_, | 124 succeeded_ = db->GetMostRecentVisitsForURL(url_id_, |
| 125 kMaxVisitsToStoreInCache, | 125 kMaxVisitsToStoreInCache, |
| 126 &recent_visits_); | 126 &recent_visits_); |
| 127 if (!succeeded_) | 127 if (!succeeded_) |
| 128 recent_visits_.clear(); | 128 recent_visits_.clear(); |
| 129 return true; // Always claim to be done; do not retry failures. | 129 return true; // Always claim to be done; do not retry failures. |
| 130 } | 130 } |
| 131 | 131 |
| 132 void UpdateRecentVisitsFromHistoryDBTask::DoneRunOnMainThread() { | 132 void UpdateRecentVisitsFromHistoryDBTask::DoneRunOnMainThread() { |
| (...skipping 12 matching lines...) Expand all Loading... |
| 145 saved_cache_version_(kCurrentCacheFileVersion), | 145 saved_cache_version_(kCurrentCacheFileVersion), |
| 146 pre_filter_item_count_(0), | 146 pre_filter_item_count_(0), |
| 147 post_filter_item_count_(0), | 147 post_filter_item_count_(0), |
| 148 post_scoring_item_count_(0) { | 148 post_scoring_item_count_(0) { |
| 149 } | 149 } |
| 150 | 150 |
| 151 ScoredHistoryMatches URLIndexPrivateData::HistoryItemsForTerms( | 151 ScoredHistoryMatches URLIndexPrivateData::HistoryItemsForTerms( |
| 152 base::string16 search_string, | 152 base::string16 search_string, |
| 153 size_t cursor_position, | 153 size_t cursor_position, |
| 154 const std::string& languages, | 154 const std::string& languages, |
| 155 BookmarkService* bookmark_service) { | 155 HistoryClient* history_client) { |
| 156 // If cursor position is set and useful (not at either end of the | 156 // If cursor position is set and useful (not at either end of the |
| 157 // string), allow the search string to be broken at cursor position. | 157 // string), allow the search string to be broken at cursor position. |
| 158 // We do this by pretending there's a space where the cursor is. | 158 // We do this by pretending there's a space where the cursor is. |
| 159 if ((cursor_position != base::string16::npos) && | 159 if ((cursor_position != base::string16::npos) && |
| 160 (cursor_position < search_string.length()) && | 160 (cursor_position < search_string.length()) && |
| 161 (cursor_position > 0)) { | 161 (cursor_position > 0)) { |
| 162 search_string.insert(cursor_position, base::ASCIIToUTF16(" ")); | 162 search_string.insert(cursor_position, base::ASCIIToUTF16(" ")); |
| 163 } | 163 } |
| 164 pre_filter_item_count_ = 0; | 164 pre_filter_item_count_ = 0; |
| 165 post_filter_item_count_ = 0; | 165 post_filter_item_count_ = 0; |
| (...skipping 71 matching lines...) Expand 10 before | Expand all | Expand 10 after Loading... |
| 237 // Don't score matches when there are no terms to score against. (It's | 237 // Don't score matches when there are no terms to score against. (It's |
| 238 // possible that the word break iterater that extracts words to search | 238 // possible that the word break iterater that extracts words to search |
| 239 // for in the database allows some whitespace "words" whereas Tokenize | 239 // for in the database allows some whitespace "words" whereas Tokenize |
| 240 // excludes a long list of whitespace.) One could write a scoring | 240 // excludes a long list of whitespace.) One could write a scoring |
| 241 // function that gives a reasonable order to matches when there | 241 // function that gives a reasonable order to matches when there |
| 242 // are no terms (i.e., all the words are some form of whitespace), | 242 // are no terms (i.e., all the words are some form of whitespace), |
| 243 // but this is such a rare edge case that it's not worth the time. | 243 // but this is such a rare edge case that it's not worth the time. |
| 244 return scored_items; | 244 return scored_items; |
| 245 } | 245 } |
| 246 scored_items = std::for_each(history_id_set.begin(), history_id_set.end(), | 246 scored_items = std::for_each(history_id_set.begin(), history_id_set.end(), |
| 247 AddHistoryMatch(*this, languages, bookmark_service, lower_raw_string, | 247 AddHistoryMatch(*this, languages, history_client, lower_raw_string, |
| 248 lower_raw_terms, base::Time::Now())).ScoredMatches(); | 248 lower_raw_terms, base::Time::Now())).ScoredMatches(); |
| 249 | 249 |
| 250 // Select and sort only the top kMaxMatches results. | 250 // Select and sort only the top kMaxMatches results. |
| 251 if (scored_items.size() > AutocompleteProvider::kMaxMatches) { | 251 if (scored_items.size() > AutocompleteProvider::kMaxMatches) { |
| 252 std::partial_sort(scored_items.begin(), | 252 std::partial_sort(scored_items.begin(), |
| 253 scored_items.begin() + | 253 scored_items.begin() + |
| 254 AutocompleteProvider::kMaxMatches, | 254 AutocompleteProvider::kMaxMatches, |
| 255 scored_items.end(), | 255 scored_items.end(), |
| 256 ScoredHistoryMatch::MatchScoreGreater); | 256 ScoredHistoryMatch::MatchScoreGreater); |
| 257 scored_items.resize(AutocompleteProvider::kMaxMatches); | 257 scored_items.resize(AutocompleteProvider::kMaxMatches); |
| (...skipping 225 matching lines...) Expand 10 before | Expand all | Expand 10 after Loading... |
| 483 data_copy->word_id_history_map_ = word_id_history_map_; | 483 data_copy->word_id_history_map_ = word_id_history_map_; |
| 484 data_copy->history_id_word_map_ = history_id_word_map_; | 484 data_copy->history_id_word_map_ = history_id_word_map_; |
| 485 data_copy->history_info_map_ = history_info_map_; | 485 data_copy->history_info_map_ = history_info_map_; |
| 486 data_copy->word_starts_map_ = word_starts_map_; | 486 data_copy->word_starts_map_ = word_starts_map_; |
| 487 return data_copy; | 487 return data_copy; |
| 488 // Not copied: | 488 // Not copied: |
| 489 // search_term_cache_ | 489 // search_term_cache_ |
| 490 // pre_filter_item_count_ | 490 // pre_filter_item_count_ |
| 491 // post_filter_item_count_ | 491 // post_filter_item_count_ |
| 492 // post_scoring_item_count_ | 492 // post_scoring_item_count_ |
| 493 }; | 493 } |
| 494 | 494 |
| 495 bool URLIndexPrivateData::Empty() const { | 495 bool URLIndexPrivateData::Empty() const { |
| 496 return history_info_map_.empty(); | 496 return history_info_map_.empty(); |
| 497 } | 497 } |
| 498 | 498 |
| 499 void URLIndexPrivateData::Clear() { | 499 void URLIndexPrivateData::Clear() { |
| 500 last_time_rebuilt_from_history_ = base::Time(); | 500 last_time_rebuilt_from_history_ = base::Time(); |
| 501 word_list_.clear(); | 501 word_list_.clear(); |
| 502 available_words_.clear(); | 502 available_words_.clear(); |
| 503 word_map_.clear(); | 503 word_map_.clear(); |
| (...skipping 213 matching lines...) Expand 10 before | Expand all | Expand 10 after Loading... |
| 717 word_starts_map_[history_id] = word_starts; | 717 word_starts_map_[history_id] = word_starts; |
| 718 | 718 |
| 719 // Update the recent visits information or schedule the update | 719 // Update the recent visits information or schedule the update |
| 720 // as appropriate. | 720 // as appropriate. |
| 721 if (history_db) { | 721 if (history_db) { |
| 722 // We'd like to check that we're on the history DB thread. | 722 // We'd like to check that we're on the history DB thread. |
| 723 // However, unittest code actually calls this on the UI thread. | 723 // However, unittest code actually calls this on the UI thread. |
| 724 // So we don't do any thread checks. | 724 // So we don't do any thread checks. |
| 725 VisitVector recent_visits; | 725 VisitVector recent_visits; |
| 726 // Make sure the private data is going to get as many recent visits as | 726 // Make sure the private data is going to get as many recent visits as |
| 727 // ScoredHistoryMatch::GetFrecency() hopes to use. | 727 // ScoredHistoryMatch::GetFrequency() hopes to use. |
| 728 DCHECK_GE(kMaxVisitsToStoreInCache, ScoredHistoryMatch::kMaxVisitsToScore); | 728 DCHECK_GE(kMaxVisitsToStoreInCache, ScoredHistoryMatch::kMaxVisitsToScore); |
| 729 if (history_db->GetMostRecentVisitsForURL(row_id, | 729 if (history_db->GetMostRecentVisitsForURL(row_id, |
| 730 kMaxVisitsToStoreInCache, | 730 kMaxVisitsToStoreInCache, |
| 731 &recent_visits)) | 731 &recent_visits)) |
| 732 UpdateRecentVisits(row_id, recent_visits); | 732 UpdateRecentVisits(row_id, recent_visits); |
| 733 } else { | 733 } else { |
| 734 DCHECK(history_service); | 734 DCHECK(history_service); |
| 735 ScheduleUpdateRecentVisits(history_service, row_id); | 735 ScheduleUpdateRecentVisits(history_service, row_id); |
| 736 } | 736 } |
| 737 | 737 |
| (...skipping 519 matching lines...) Expand 10 before | Expand all | Expand 10 after Loading... |
| 1257 : used_(true) {} | 1257 : used_(true) {} |
| 1258 | 1258 |
| 1259 URLIndexPrivateData::SearchTermCacheItem::~SearchTermCacheItem() {} | 1259 URLIndexPrivateData::SearchTermCacheItem::~SearchTermCacheItem() {} |
| 1260 | 1260 |
| 1261 | 1261 |
| 1262 // URLIndexPrivateData::AddHistoryMatch ---------------------------------------- | 1262 // URLIndexPrivateData::AddHistoryMatch ---------------------------------------- |
| 1263 | 1263 |
| 1264 URLIndexPrivateData::AddHistoryMatch::AddHistoryMatch( | 1264 URLIndexPrivateData::AddHistoryMatch::AddHistoryMatch( |
| 1265 const URLIndexPrivateData& private_data, | 1265 const URLIndexPrivateData& private_data, |
| 1266 const std::string& languages, | 1266 const std::string& languages, |
| 1267 BookmarkService* bookmark_service, | 1267 HistoryClient* history_client, |
| 1268 const base::string16& lower_string, | 1268 const base::string16& lower_string, |
| 1269 const String16Vector& lower_terms, | 1269 const String16Vector& lower_terms, |
| 1270 const base::Time now) | 1270 const base::Time now) |
| 1271 : private_data_(private_data), | 1271 : private_data_(private_data), |
| 1272 languages_(languages), | 1272 languages_(languages), |
| 1273 bookmark_service_(bookmark_service), | 1273 history_client_(history_client), |
| 1274 lower_string_(lower_string), | 1274 lower_string_(lower_string), |
| 1275 lower_terms_(lower_terms), | 1275 lower_terms_(lower_terms), |
| 1276 now_(now) { | 1276 now_(now) { |
| 1277 // Calculate offsets for each term. For instance, the offset for | 1277 // Calculate offsets for each term. For instance, the offset for |
| 1278 // ".net" should be 1, indicating that the actual word-part of the term | 1278 // ".net" should be 1, indicating that the actual word-part of the term |
| 1279 // starts at offset 1. | 1279 // starts at offset 1. |
| 1280 lower_terms_to_word_starts_offsets_.resize(lower_terms_.size(), 0u); | 1280 lower_terms_to_word_starts_offsets_.resize(lower_terms_.size(), 0u); |
| 1281 for (size_t i = 0; i < lower_terms_.size(); ++i) { | 1281 for (size_t i = 0; i < lower_terms_.size(); ++i) { |
| 1282 base::i18n::BreakIterator iter(lower_terms_[i], | 1282 base::i18n::BreakIterator iter(lower_terms_[i], |
| 1283 base::i18n::BreakIterator::BREAK_WORD); | 1283 base::i18n::BreakIterator::BREAK_WORD); |
| (...skipping 15 matching lines...) Expand all Loading... |
| 1299 HistoryInfoMap::const_iterator hist_pos = | 1299 HistoryInfoMap::const_iterator hist_pos = |
| 1300 private_data_.history_info_map_.find(history_id); | 1300 private_data_.history_info_map_.find(history_id); |
| 1301 if (hist_pos != private_data_.history_info_map_.end()) { | 1301 if (hist_pos != private_data_.history_info_map_.end()) { |
| 1302 const URLRow& hist_item = hist_pos->second.url_row; | 1302 const URLRow& hist_item = hist_pos->second.url_row; |
| 1303 const VisitInfoVector& visits = hist_pos->second.visits; | 1303 const VisitInfoVector& visits = hist_pos->second.visits; |
| 1304 WordStartsMap::const_iterator starts_pos = | 1304 WordStartsMap::const_iterator starts_pos = |
| 1305 private_data_.word_starts_map_.find(history_id); | 1305 private_data_.word_starts_map_.find(history_id); |
| 1306 DCHECK(starts_pos != private_data_.word_starts_map_.end()); | 1306 DCHECK(starts_pos != private_data_.word_starts_map_.end()); |
| 1307 ScoredHistoryMatch match(hist_item, visits, languages_, lower_string_, | 1307 ScoredHistoryMatch match(hist_item, visits, languages_, lower_string_, |
| 1308 lower_terms_, lower_terms_to_word_starts_offsets_, | 1308 lower_terms_, lower_terms_to_word_starts_offsets_, |
| 1309 starts_pos->second, now_, bookmark_service_); | 1309 starts_pos->second, now_, history_client_); |
| 1310 if (match.raw_score() > 0) | 1310 if (match.raw_score() > 0) |
| 1311 scored_matches_.push_back(match); | 1311 scored_matches_.push_back(match); |
| 1312 } | 1312 } |
| 1313 } | 1313 } |
| 1314 | 1314 |
| 1315 | 1315 |
| 1316 // URLIndexPrivateData::HistoryItemFactorGreater ------------------------------- | 1316 // URLIndexPrivateData::HistoryItemFactorGreater ------------------------------- |
| 1317 | 1317 |
| 1318 URLIndexPrivateData::HistoryItemFactorGreater::HistoryItemFactorGreater( | 1318 URLIndexPrivateData::HistoryItemFactorGreater::HistoryItemFactorGreater( |
| 1319 const HistoryInfoMap& history_info_map) | 1319 const HistoryInfoMap& history_info_map) |
| (...skipping 18 matching lines...) Expand all Loading... |
| 1338 // recently visited (within the last 12/24 hours) as highly important. Get | 1338 // recently visited (within the last 12/24 hours) as highly important. Get |
| 1339 // input from mpearson. | 1339 // input from mpearson. |
| 1340 if (r1.typed_count() != r2.typed_count()) | 1340 if (r1.typed_count() != r2.typed_count()) |
| 1341 return (r1.typed_count() > r2.typed_count()); | 1341 return (r1.typed_count() > r2.typed_count()); |
| 1342 if (r1.visit_count() != r2.visit_count()) | 1342 if (r1.visit_count() != r2.visit_count()) |
| 1343 return (r1.visit_count() > r2.visit_count()); | 1343 return (r1.visit_count() > r2.visit_count()); |
| 1344 return (r1.last_visit() > r2.last_visit()); | 1344 return (r1.last_visit() > r2.last_visit()); |
| 1345 } | 1345 } |
| 1346 | 1346 |
| 1347 } // namespace history | 1347 } // namespace history |
| OLD | NEW |