Current firefox2 importer import bookmarks with strict rules.

This manner failed to import lazy html-based bookmarks like Epiphany's one.

Add hierarchical structure checs to import tests.

Patch contributed by Takashi Toyoshima <toyoshim@google.com>.

BUG=28517
TEST=Add epiphany bookmark import test into unit_tests and pass it

Review URL: http://codereview.chromium.org/6246024

git-svn-id: svn://svn.chromium.org/chrome/trunk/src@73255 0039d316-1c4b-4281-b951-d872f2087c98

4 files changed, 170 insertions, 17 deletions

diff --git a/chrome/browser/importer/firefox2_importer.cc b/chrome/browser/importer/firefox2_importer.cc
index 2ecccb0..bf68af0 100644
--- a/chrome/browser/importer/firefox2_importer.cc
+++ b/chrome/browser/importer/firefox2_importer.cc
@@ -195,7 +195,9 @@ void Firefox2Importer::ImportBookmarksFile(
     //                keywords yet.
     is_bookmark = ParseBookmarkFromLine(line, charset, &title,
                                         &url, &favicon, &shortcut, &add_date,
-                                        &post_data);
+                                        &post_data) ||
+        ParseMinimumBookmarkFromLine(line, charset, &title, &url);
+
     if (is_bookmark)
       last_folder_is_empty = false;
 
@@ -243,7 +245,7 @@ void Firefox2Importer::ImportBookmarksFile(
     }
 
     // Bookmarks in sub-folder are encapsulated with <DL> tag.
-    if (StartsWithASCII(line, "<DL>", true)) {
+    if (StartsWithASCII(line, "<DL>", false)) {
       path.push_back(last_folder);
       last_folder.clear();
       if (last_folder_on_toolbar && !toolbar_folder)
@@ -251,7 +253,7 @@ void Firefox2Importer::ImportBookmarksFile(
 
       // Mark next folder empty as initial state.
       last_folder_is_empty = true;
-    } else if (StartsWithASCII(line, "</DL>", true)) {
+    } else if (StartsWithASCII(line, "</DL>", false)) {
       if (path.empty())
         break;  // Mismatch <DL>.
 
@@ -400,8 +402,9 @@ void Firefox2Importer::GetSearchEnginesXMLFiles(
 bool Firefox2Importer::ParseCharsetFromLine(const std::string& line,
                                             std::string* charset) {
   const char kCharset[] = "charset=";
-  if (StartsWithASCII(line, "<META", true) &&
-      line.find("CONTENT=\"") != std::string::npos) {
+  if (StartsWithASCII(line, "<META", false) &&
+      (line.find("CONTENT=\"") != std::string::npos ||
+          line.find("content=\"") != std::string::npos)) {
     size_t begin = line.find(kCharset);
     if (begin == std::string::npos)
       return false;
@@ -550,6 +553,58 @@ bool Firefox2Importer::ParseBookmarkFromLine(const std::string& line,
 }
 
 // static
+bool Firefox2Importer::ParseMinimumBookmarkFromLine(const std::string& line,
+                                                    const std::string& charset,
+                                                    std::wstring* title,
+                                                    GURL* url) {
+  const char kItemOpen[] = "<DT><A";
+  const char kItemClose[] = "</";
+  const char kHrefAttributeUpper[] = "HREF";
+  const char kHrefAttributeLower[] = "href";
+
+  title->clear();
+  *url = GURL();
+
+  // Case-insensitive check of open tag.
+  if (!StartsWithASCII(line, kItemOpen, false))
+    return false;
+
+  // Find any close tag.
+  size_t end = line.find(kItemClose);
+  size_t tag_end = line.rfind('>', end) + 1;
+  if (end == std::string::npos || tag_end < arraysize(kItemOpen))
+    return false;  // No end tag or start tag is broken.
+
+  std::string attribute_list = line.substr(arraysize(kItemOpen),
+      tag_end - arraysize(kItemOpen) - 1);
+
+  // Title
+  base::CodepageToWide(line.substr(tag_end, end - tag_end), charset.c_str(),
+                       base::OnStringConversionError::SKIP, title);
+  HTMLUnescape(title);
+
+  // URL
+  std::string value;
+  if (GetAttribute(attribute_list, kHrefAttributeUpper, &value) ||
+      GetAttribute(attribute_list, kHrefAttributeLower, &value)) {
+    if (charset.length() != 0) {
+      std::wstring w_url;
+      base::CodepageToWide(value, charset.c_str(),
+                           base::OnStringConversionError::SKIP, &w_url);
+      HTMLUnescape(&w_url);
+
+      string16 url16 = WideToUTF16Hack(w_url);
+
+      *url = GURL(url16);
+    } else {
+      *url = GURL(value);
+    }
+  }
+
+  return true;
+}
+
+// static
 bool Firefox2Importer::GetAttribute(const std::string& attribute_list,
                                     const std::string& attribute,
                                     std::string* value) {
diff --git a/chrome/browser/importer/firefox2_importer.h b/chrome/browser/importer/firefox2_importer.h
index 8901e6f..7fef5e6 100644
--- a/chrome/browser/importer/firefox2_importer.h
+++ b/chrome/browser/importer/firefox2_importer.h
@@ -98,6 +98,18 @@ class Firefox2Importer : public Importer {
                                     std::wstring* shortcut,
                                     base::Time* add_date,
                                     std::wstring* post_data);
+  // Save bookmarks imported from browsers with Firefox2 compatible bookmark
+  // systems such as Epiphany. This bookmark format is the same as that of the
+  // basic Firefox bookmark, but it misses additional properties and uses
+  // lower-case tag:
+  //   ...<h1>Bookmarks</h1><dl>
+  //   <dt><a href="url">name</a></dt>
+  //   <dt><a href="url">name</a></dt>
+  //   </dl>
+  static bool ParseMinimumBookmarkFromLine(const std::string& line,
+                                           const std::string& charset,
+                                           std::wstring* title,
+                                           GURL* url);
 
   // Fetches the given attribute value from the |tag|. Returns true if
   // successful, and |value| will contain the value.
diff --git a/chrome/browser/importer/firefox_importer_unittest.cc b/chrome/browser/importer/firefox_importer_unittest.cc
index 12e75f4..15bdb14 100644
--- a/chrome/browser/importer/firefox_importer_unittest.cc
+++ b/chrome/browser/importer/firefox_importer_unittest.cc
@@ -176,6 +176,14 @@ TEST(FirefoxImporterTest, Firefox2BookmarkParse) {
   EXPECT_EQ(L"", shortcut);
   EXPECT_EQ(L"", post_data);
   EXPECT_TRUE(Time() == add_date);
+
+  // Epiphany format.
+  result = Firefox2Importer::ParseMinimumBookmarkFromLine(
+      "<dt><a href=\"http://www.google.com/\">Google</a></dt>",
+      charset, &title, &url);
+  EXPECT_TRUE(result);
+  EXPECT_EQ(L"Google", title);
+  EXPECT_EQ("http://www.google.com/", url.spec());
 }
 
 TEST(FirefoxImporterTest, Firefox2BookmarkFileImport) {
@@ -194,13 +202,46 @@ TEST(FirefoxImporterTest, Firefox2BookmarkFileImport) {
                                 first_folder_name, importer, &bookmarks,
                                 NULL, NULL);
   EXPECT_EQ(3, static_cast<int>(bookmarks.size()));
-  std::vector<ProfileWriter::BookmarkEntry>::iterator it = bookmarks.begin();
-  ProfileWriter::BookmarkEntry entry = *it++;
-  EXPECT_EQ(L"Empty", entry.title);
-  entry = *it++;
-  EXPECT_EQ(L"[Tamura Yukari.com]", entry.title);
-  entry = *it++;
-  EXPECT_EQ(L"Google", entry.title);
+  std::vector<ProfileWriter::BookmarkEntry>::iterator it;
+  ProfileWriter::BookmarkEntry entry;
+  std::vector<std::wstring>::iterator path_it;
+  if (bookmarks.size() == 3) {
+    it = bookmarks.begin();
+    entry = *it++;
+    EXPECT_EQ(L"Empty", entry.title);
+    EXPECT_TRUE(entry.is_folder);
+    EXPECT_EQ(Time::FromTimeT(1295938143), entry.creation_time);
+    EXPECT_EQ(2, static_cast<int>(entry.path.size()));
+    if (entry.path.size() == 2) {
+      path_it = entry.path.begin();
+      EXPECT_EQ(L"", *path_it++);
+      EXPECT_EQ(L"Empty's Parent", *path_it);
+    }
+
+    entry = *it++;
+    EXPECT_EQ(L"[Tamura Yukari.com]", entry.title);
+    EXPECT_FALSE(entry.is_folder);
+    EXPECT_EQ(Time::FromTimeT(1234567890), entry.creation_time);
+    EXPECT_EQ(2, static_cast<int>(entry.path.size()));
+    if (entry.path.size() == 2) {
+      path_it = entry.path.begin();
+      EXPECT_EQ(L"", *path_it++);
+      EXPECT_EQ(L"Not Empty", *path_it);
+    }
+    EXPECT_EQ("http://www.tamurayukari.com/", entry.url.spec());
+
+    entry = *it++;
+    EXPECT_EQ(L"Google", entry.title);
+    EXPECT_FALSE(entry.is_folder);
+    EXPECT_EQ(Time::FromTimeT(0000000000), entry.creation_time);
+    EXPECT_EQ(2, static_cast<int>(entry.path.size()));
+    if (entry.path.size() == 2) {
+      path_it = entry.path.begin();
+      EXPECT_EQ(L"", *path_it++);
+      EXPECT_EQ(L"Not Empty But Default", *path_it);
+    }
+    EXPECT_EQ("http://www.google.com/", entry.url.spec());
+  }
 
   // Import non-default bookmarks from a file.
   bookmarks.clear();
@@ -209,11 +250,51 @@ TEST(FirefoxImporterTest, Firefox2BookmarkFileImport) {
                                 first_folder_name, importer, &bookmarks,
                                 NULL, NULL);
   EXPECT_EQ(2, static_cast<int>(bookmarks.size()));
-  it = bookmarks.begin();
-  entry = *it++;
-  EXPECT_EQ(L"Empty", entry.title);
-  entry = *it++;
-  EXPECT_EQ(L"[Tamura Yukari.com]", entry.title);
+  if (bookmarks.size() == 2) {
+    it = bookmarks.begin();
+    entry = *it++;
+    EXPECT_EQ(L"Empty", entry.title);
+    EXPECT_TRUE(entry.is_folder);
+    EXPECT_EQ(Time::FromTimeT(1295938143), entry.creation_time);
+    EXPECT_EQ(2, static_cast<int>(entry.path.size()));
+    if (entry.path.size() == 2) {
+      path_it = entry.path.begin();
+      EXPECT_EQ(L"", *path_it++);
+      EXPECT_EQ(L"Empty's Parent", *path_it);
+    }
+
+    entry = *it++;
+    EXPECT_EQ(L"[Tamura Yukari.com]", entry.title);
+    EXPECT_FALSE(entry.is_folder);
+    EXPECT_EQ(Time::FromTimeT(1234567890), entry.creation_time);
+    EXPECT_EQ(2, static_cast<int>(entry.path.size()));
+    if (entry.path.size() == 2) {
+      path_it = entry.path.begin();
+      EXPECT_EQ(L"", *path_it++);
+      EXPECT_EQ(L"Not Empty", *path_it);
+    }
+    EXPECT_EQ("http://www.tamurayukari.com/", entry.url.spec());
+  }
+
+  // Import Epiphany bookmarks from a file
+  FilePath epiphany_path = path.AppendASCII("epiphany.html");
+  bookmarks.clear();
+  default_urls.clear();
+  importer->ImportBookmarksFile(epiphany_path, default_urls, false,
+                                first_folder_name, importer, &bookmarks,
+                                NULL, NULL);
+  EXPECT_EQ(2, static_cast<int>(bookmarks.size()));
+  if (bookmarks.size() == 2) {
+    it = bookmarks.begin();
+    entry = *it++;
+    EXPECT_EQ(L"[Tamura Yukari.com]", entry.title);
+    EXPECT_EQ("http://www.tamurayukari.com/", entry.url.spec());
+    EXPECT_EQ(0, static_cast<int>(entry.path.size()));
+    entry = *it++;
+    EXPECT_EQ(L"Google", entry.title);
+    EXPECT_EQ("http://www.google.com/", entry.url.spec());
+    EXPECT_EQ(0, static_cast<int>(entry.path.size()));
+  }
 
   importer->Release();
 }
diff --git a/chrome/test/data/firefox2_importer/epiphany.html b/chrome/test/data/firefox2_importer/epiphany.html
new file mode 100644
index 0000000..8fa3bbc
--- /dev/null
+++ b/chrome/test/data/firefox2_importer/epiphany.html
@@ -0,0 +1,5 @@
+<!DOCTYPE NETSCAPE-Bookmark-file-1>
+<meta http-equiv="Content-Type" content="text/html; charset=UTF-8"><title>Bookmarks</title><h1>Bookmarks</h1><dl>
+<dt><a href="http://www.tamurayukari.com/">[Tamura Yukari.com]</a></dt>
+<dt><a href="http://www.google.com">Google</a></dt>
+</dl>