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
3// found in the LICENSE file.
4
5#ifndef CHROME_UTILITY_IMPORTER_BOOKMARK_HTML_READER_H_
6#define CHROME_UTILITY_IMPORTER_BOOKMARK_HTML_READER_H_
7
8#include <string>
9#include <vector>
10
11#include "base/callback_forward.h"
12#include "base/strings/string16.h"
13
14class GURL;
15struct ImportedBookmarkEntry;
16struct ImportedFaviconUsage;
17
18namespace base {
19class FilePath;
20class Time;
21}
22
23namespace bookmark_html_reader {
24
25// Imports the bookmarks from the specified file.
26//
27// |cancellation_callback| is polled to query if the import should be cancelled;
28// if it returns |true| at any time the import will be cancelled. If
29// |cancellation_callback| is a null callback the import will run to completion.
30//
31// |valid_url_callback| is called to determine if a specified URL is valid for
32// import; it returns |true| if it is. If |valid_url_callback| is a null
33// callback, all URLs are considered to be valid.
34//
35// |file_path| is the path of the file on disk to import.
36//
37// |bookmarks| is a pointer to a vector, which is filled with the imported
38// bookmarks. It may not be NULL.
39//
40// |favicons| is a pointer to a vector, which is filled with the favicons of
41// imported bookmarks. It may be NULL, in which case favicons are not imported.
42void ImportBookmarksFile(
43    const base::Callback<bool(void)>& cancellation_callback,
44    const base::Callback<bool(const GURL&)>& valid_url_callback,
45    const base::FilePath& file_path,
46    std::vector<ImportedBookmarkEntry>* bookmarks,
47    std::vector<ImportedFaviconUsage>* favicons);
48
49namespace internal {
50
51// The file format that BookmarkHTMLReader parses starts with a heading
52// tag, which contains its title. All bookmarks and sub-folders follow,
53// bracketed by a <DL> tag:
54//   <DT><H3 PERSONAL_TOOLBAR_FOLDER="true" ...>title</H3>
55//   <DL><p>
56//      ... container ...
57//   </DL><p>
58// And a bookmark is presented by a <A> tag:
59//   <DT><A HREF="url" SHORTCUTURL="shortcut" ADD_DATE="11213014"...>name</A>
60// Reference: http://kb.mozillazine.org/Bookmarks.html
61
62bool ParseCharsetFromLine(const std::string& line,
63                          std::string* charset);
64bool ParseFolderNameFromLine(const std::string& line,
65                             const std::string& charset,
66                             base::string16* folder_name,
67                             bool* is_toolbar_folder,
68                             base::Time* add_date);
69// See above, this will also put the data: URL of the favicon into |*favicon|
70// if there is a favicon given. |post_data| is set for POST base keywords to
71// the contents of the actual POST (with %s for the search term).
72bool ParseBookmarkFromLine(const std::string& line,
73                           const std::string& charset,
74                           base::string16* title,
75                           GURL* url,
76                           GURL* favicon,
77                           base::string16* shortcut,
78                           base::Time* add_date,
79                           base::string16* post_data);
80// Save bookmarks imported from browsers with Firefox 2 compatible bookmark
81// systems such as Epiphany. This bookmark format is the same as that of the
82// basic Firefox 2 bookmark, but it misses additional properties and uses
83// lower-case tag:
84//   ...<h1>Bookmarks</h1><dl>
85//   <dt><a href="url">name</a></dt>
86//   <dt><a href="url">name</a></dt>
87//   </dl>
88bool ParseMinimumBookmarkFromLine(const std::string& line,
89                                  const std::string& charset,
90                                  base::string16* title,
91                                  GURL* url);
92
93}  // namespace internal
94
95}  // namespace bookmark_html_reader
96
97#endif  // CHROME_UTILITY_IMPORTER_BOOKMARK_HTML_READER_H_
98