Add TitledUrlIndex for indexing arbitrary title/URL pairs

components/bookmarks/browser/titled_url_index.h

Index: components/bookmarks/browser/titled_url_index.h
diff --git a/components/bookmarks/browser/titled_url_index.h b/components/bookmarks/browser/titled_url_index.h
new file mode 100644
index 0000000000000000000000000000000000000000..97397174f8c7f742df86cc45077f2acee33dffe0
--- /dev/null
+++ b/components/bookmarks/browser/titled_url_index.h
@@ -0,0 +1,89 @@
+// Copyright 2016 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.
+
+#ifndef COMPONENTS_BOOKMARKS_BROWSER_TITLED_URL_INDEX_H_
+#define COMPONENTS_BOOKMARKS_BROWSER_TITLED_URL_INDEX_H_
+
+#include <stddef.h>
+
+#include <map>
+#include <set>
+#include <string>
+#include <vector>
+
+#include "base/macros.h"
+#include "base/strings/string16.h"
+#include "components/bookmarks/browser/titled_url_node.h"
+#include "components/query_parser/query_parser.h"
+
+namespace bookmarks {
+
+struct TitledUrlMatch;
+
+// TitledUrlIndex maintains an index of paired titles and URLs for quick lookup.
+//
+// TitledUrlIndex stores the index (index_) as a map of sets. The map (type
+// Index) maps from a lower case string to the set (type NodeSet) of
+// TitledUrlNodes that contain that string in their title or URL.
+class TitledUrlIndex {

[sky, 2016/12/01 20:24:51]

Can you convert BookmarkIndex in place and then rename as your last step? To do
otherwise means you lose history and the review is much harder.

[mattreynolds, 2016/12/01 23:33:18]

Done.

+ public:
+  TitledUrlIndex();
+  ~TitledUrlIndex();
+
+  // Invoked when a title/URL pair has been added to the model.
+  void Add(const TitledUrlNode* node);
+
+  // Invoked when a title/URL pair has been removed from the model.
+  void Remove(const TitledUrlNode* node);
+
+  // Returns up to |max_count| of matches containing each term from the text
+  // |query| in either the title or the URL.
+  void GetResultsMatching(const base::string16& query,
+                          size_t max_count,
+                          query_parser::MatchingAlgorithm matching_algorithm,
+                          std::vector<TitledUrlMatch>* results);
+
+ protected:
+  typedef std::vector<const TitledUrlNode*> Nodes;
+  typedef std::set<const TitledUrlNode*> NodeSet;
+  typedef std::map<base::string16, NodeSet> Index;
+
+ private:
+  // Constructs |sorted_nodes| by copying the matches in |matches|. If the order
+  // of matches is relevant, override this method to specify the sorting and
+  // deduplication behavior. The default implementation performs no sorting.
+  virtual void SortMatches(const NodeSet& matches, Nodes* sorted_nodes) const;
+
+  // Add |node| to |results| if the node matches the query.
+  void AddMatchToResults(const TitledUrlNode* node,
+                         query_parser::QueryParser* parser,
+                         const query_parser::QueryNodeVector& query_nodes,
+                         std::vector<TitledUrlMatch>* results);
+
+  // Populates |matches| for the specified term. If |first_term| is true, this
+  // is the first term in the query. Returns true if there is at least one node
+  // matching the term.
+  bool GetResultsMatchingTerm(
+      const base::string16& term,
+      bool first_term,
+      query_parser::MatchingAlgorithm matching_algorithm,
+      NodeSet* matches);
+
+  // Returns the set of query words from |query|.
+  std::vector<base::string16> ExtractQueryWords(const base::string16& query);
+
+  // Adds |node| to |index_|.
+  void RegisterNode(const base::string16& term, const TitledUrlNode* node);
+
+  // Removes |node| from |index_|.
+  void UnregisterNode(const base::string16& term, const TitledUrlNode* node);
+
+  Index index_;
+
+  DISALLOW_COPY_AND_ASSIGN(TitledUrlIndex);
+};
+
+}  // namespace bookmarks
+
+#endif  // COMPONENTS_BOOKMARKS_BROWSER_TITLED_URL_INDEX_H_