Move Firefox importer's INI parser to c/browser/common.

Media Galleries API Picasa Import will require reading some INI files. Rather than re-invent the wheel and duplicate code, we are moving the INI parser used by Firefox import into chrome/browser/common.

BUG=151701

Review URL: https://chromiumcodereview.appspot.com/16982004

git-svn-id: svn://svn.chromium.org/chrome/trunk/src@207111 0039d316-1c4b-4281-b951-d872f2087c98

6 files changed, 275 insertions, 47 deletions

diff --git a/base/base.gyp b/base/base.gyp
index aaead7d..e35e6b1 100644
--- a/base/base.gyp
+++ b/base/base.gyp
@@ -494,6 +494,7 @@
         'i18n/rtl_unittest.cc',
         'i18n/string_search_unittest.cc',
         'i18n/time_formatting_unittest.cc',
+        'ini_parser_unittest.cc',
         'ios/device_util_unittest.mm',
         'json/json_parser_unittest.cc',
         'json/json_reader_unittest.cc',
diff --git a/base/base.gypi b/base/base.gypi
index 7a5879b..2ccb2bb 100644
--- a/base/base.gypi
+++ b/base/base.gypi
@@ -201,6 +201,8 @@
           'hi_res_timer_manager_win.cc',
           'hi_res_timer_manager.h',
           'id_map.h',
+          'ini_parser.cc',
+          'ini_parser.h',
           'ios/device_util.h',
           'ios/device_util.mm',
           'ios/ios_util.h',
diff --git a/base/ini_parser.cc b/base/ini_parser.cc
new file mode 100644
index 0000000..7a2bd1c
--- /dev/null
+++ b/base/ini_parser.cc
@@ -0,0 +1,66 @@
+// Copyright 2013 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.
+
+#include "base/ini_parser.h"
+
+#include "base/logging.h"
+#include "base/strings/string_tokenizer.h"
+
+namespace base {
+
+INIParser::INIParser() : used_(false) {}
+
+INIParser::~INIParser() {}
+
+void INIParser::Parse(const std::string& content) {
+  DCHECK(!used_);
+  used_ = true;
+  base::StringTokenizer tokenizer(content, "\r\n");
+
+  std::string current_section;
+  while (tokenizer.GetNext()) {
+    std::string line = tokenizer.token();
+    if (line.empty()) {
+      // Skips the empty line.
+      continue;
+    }
+    if (line[0] == '#' || line[0] == ';') {
+      // This line is a comment.
+      continue;
+    }
+    if (line[0] == '[') {
+      // It is a section header.
+      current_section = line.substr(1);
+      size_t end = current_section.rfind(']');
+      if (end != std::string::npos)
+        current_section.erase(end);
+    } else {
+      std::string key, value;
+      size_t equal = line.find('=');
+      if (equal != std::string::npos) {
+        key = line.substr(0, equal);
+        value = line.substr(equal + 1);
+        HandleTriplet(current_section, key, value);
+      }
+    }
+  }
+}
+
+DictionaryValueINIParser::DictionaryValueINIParser() {}
+
+DictionaryValueINIParser::~DictionaryValueINIParser() {}
+
+void DictionaryValueINIParser::HandleTriplet(const std::string& section,
+                                             const std::string& key,
+                                             const std::string& value) {
+
+  // Checks whether the section and key contain a '.' character.
+  // Those sections and keys break DictionaryValue's path format when not
+  // using the *WithoutPathExpansion methods.
+  if (section.find('.') == std::string::npos &&
+      key.find('.') == std::string::npos)
+    root_.SetString(section + "." + key, value);
+}
+
+}  // namespace base
diff --git a/base/ini_parser.h b/base/ini_parser.h
new file mode 100644
index 0000000..0aa7754
--- /dev/null
+++ b/base/ini_parser.h
@@ -0,0 +1,69 @@
+// Copyright 2013 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 BASE_INI_PARSER_H_
+#define BASE_INI_PARSER_H_
+
+#include <string>
+
+#include "base/base_export.h"
+#include "base/basictypes.h"
+#include "base/values.h"
+
+namespace base {
+
+// Parses INI files in a string. Users should in inherit from this class.
+// This is a very basic INI parser with these characteristics:
+//  - Ignores blank lines.
+//  - Ignores comment lines beginning with '#' or ';'.
+//  - Duplicate key names in the same section will simply cause repeated calls
+//    to HandleTriplet with the same |section| and |key| parameters.
+//  - No escape characters supported.
+//  - Global properties result in calls to HandleTriplet with an empty string in
+//    the |section| argument.
+//  - Section headers begin with a '[' character. It is recommended, but
+//    not required to close the header bracket with a ']' character. All
+//    characters after a closing ']' character is ignored.
+//  - Key value pairs are indicated with an '=' character. Whitespace is not
+//    ignored. Quoting is not supported. Everything before the first '='
+//    is considered the |key|, and everything after is the |value|.
+class BASE_EXPORT INIParser {
+ public:
+  INIParser();
+  virtual ~INIParser();
+
+  // May only be called once per instance.
+  void Parse(const std::string& content);
+
+ private:
+  virtual void HandleTriplet(const std::string& section,
+                             const std::string& key,
+                             const std::string& value) = 0;
+
+  bool used_;
+};
+
+// Parsed values are stored as strings at the "section.key" path. Triplets with
+// |section| or |key| parameters containing '.' are ignored.
+class BASE_EXPORT DictionaryValueINIParser : public INIParser {
+ public:
+  DictionaryValueINIParser();
+  virtual ~DictionaryValueINIParser();
+
+  const DictionaryValue& root() const { return root_; }
+
+ private:
+  // INIParser implementation.
+  virtual void HandleTriplet(const std::string& section,
+                             const std::string& key,
+                             const std::string& value) OVERRIDE;
+
+  DictionaryValue root_;
+
+  DISALLOW_COPY_AND_ASSIGN(DictionaryValueINIParser);
+};
+
+}  // namespace base
+
+#endif  // BASE_INI_PARSER_H_
diff --git a/base/ini_parser_unittest.cc b/base/ini_parser_unittest.cc
new file mode 100644
index 0000000..ff90f6c
--- /dev/null
+++ b/base/ini_parser_unittest.cc
@@ -0,0 +1,131 @@
+// Copyright 2013 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.
+
+#include <string>
+#include <vector>
+
+#include "base/ini_parser.h"
+#include "testing/gtest/include/gtest/gtest.h"
+
+namespace base {
+
+namespace {
+
+struct TestTriplet {
+  TestTriplet(const std::string& section,
+          const std::string& key,
+          const std::string& value)
+      : section(section),
+        key(key),
+        value(value) {
+  }
+
+  std::string section;
+  std::string key;
+  std::string value;
+};
+
+class TestINIParser : public INIParser {
+ public:
+  explicit TestINIParser(
+      const std::vector<TestTriplet>& expected_triplets)
+      : expected_triplets_(expected_triplets),
+        pair_i_(0) {
+  }
+  virtual ~TestINIParser() {}
+
+  size_t pair_i() {
+    return pair_i_;
+  }
+
+ private:
+  virtual void HandleTriplet(const std::string& section, const std::string& key,
+                             const std::string& value) OVERRIDE {
+    EXPECT_EQ(expected_triplets_[pair_i_].section, section);
+    EXPECT_EQ(expected_triplets_[pair_i_].key, key);
+    EXPECT_EQ(expected_triplets_[pair_i_].value, value);
+    ++pair_i_;
+  }
+
+  std::vector<TestTriplet> expected_triplets_;
+  size_t pair_i_;
+};
+
+TEST(INIParserTest, BasicValid) {
+  std::vector<TestTriplet> expected_triplets;
+  expected_triplets.push_back(TestTriplet("section1", "key1", "value1"));
+  expected_triplets.push_back(TestTriplet("section1", "key2", "value2"));
+  expected_triplets.push_back(TestTriplet("section1", "key3", "value3"));
+  expected_triplets.push_back(TestTriplet("section2", "key4", "value4"));
+  expected_triplets.push_back(TestTriplet("section2", "key5",
+                                          "value=with=equals"));
+  expected_triplets.push_back(TestTriplet("section2", "key6", "value6"));
+  TestINIParser test_parser(expected_triplets);
+
+  test_parser.Parse(
+      "[section1]\n"
+      "key1=value1\n"
+      "key2=value2\r\n"  // Testing DOS "\r\n" line endings.
+      "key3=value3\n"
+      "[section2\n"      // Testing omitted closing bracket.
+      "key4=value4\r"    // Testing "\r" line endings.
+      "key5=value=with=equals\n"
+      "key6=value6");    // Testing omitted final line ending.
+}
+
+TEST(INIParserTest, IgnoreBlankLinesAndComments) {
+  std::vector<TestTriplet> expected_triplets;
+  expected_triplets.push_back(TestTriplet("section1", "key1", "value1"));
+  expected_triplets.push_back(TestTriplet("section1", "key2", "value2"));
+  expected_triplets.push_back(TestTriplet("section1", "key3", "value3"));
+  expected_triplets.push_back(TestTriplet("section2", "key4", "value4"));
+  expected_triplets.push_back(TestTriplet("section2", "key5", "value5"));
+  expected_triplets.push_back(TestTriplet("section2", "key6", "value6"));
+  TestINIParser test_parser(expected_triplets);
+
+  test_parser.Parse(
+      "\n"
+      "[section1]\n"
+      "key1=value1\n"
+      "\n"
+      "\n"
+      "key2=value2\n"
+      "key3=value3\n"
+      "\n"
+      ";Comment1"
+      "\n"
+      "[section2]\n"
+      "key4=value4\n"
+      "#Comment2\n"
+      "key5=value5\n"
+      "\n"
+      "key6=value6\n");
+}
+
+TEST(INIParserTest, DictionaryValueINIParser) {
+  DictionaryValueINIParser test_parser;
+
+  test_parser.Parse(
+      "[section1]\n"
+      "key1=value1\n"
+      "key.2=value2\n"
+      "key3=va.lue3\n"
+      "[se.ction2]\n"
+      "key.4=value4\n"
+      "key5=value5\n");
+
+  const DictionaryValue& root = test_parser.root();
+  std::string value;
+  EXPECT_TRUE(root.GetString("section1.key1", &value));
+  EXPECT_EQ("value1", value);
+  EXPECT_FALSE(root.GetString("section1.key.2", &value));
+  EXPECT_TRUE(root.GetString("section1.key3", &value));
+  EXPECT_EQ("va.lue3", value);
+  EXPECT_FALSE(root.GetString("se.ction2.key.4", &value));
+  EXPECT_FALSE(root.GetString("se.ction2.key5", &value));
+}
+
+}  // namespace
+
+}  // namespace base
diff --git a/chrome/browser/importer/firefox_importer_utils.cc b/chrome/browser/importer/firefox_importer_utils.cc
index 70c11f1..4553c11 100644
--- a/chrome/browser/importer/firefox_importer_utils.cc
+++ b/chrome/browser/importer/firefox_importer_utils.cc
@@ -9,10 +9,10 @@
 #include <string>
 
 #include "base/file_util.h"
+#include "base/ini_parser.h"
 #include "base/logging.h"
 #include "base/strings/string_number_conversions.h"
 #include "base/strings/string_split.h"
-#include "base/strings/string_util.h"
 #include "base/strings/stringprintf.h"
 #include "base/strings/utf_string_conversions.h"
 #include "base/values.h"
@@ -50,9 +50,12 @@ class FirefoxURLParameterFilter : public TemplateURLParser::ParameterFilter {
 }  // namespace
 
 base::FilePath GetFirefoxProfilePath() {
-  DictionaryValue root;
   base::FilePath ini_file = GetProfilesINI();
-  ParseProfileINI(ini_file, &root);
+  std::string content;
+  file_util::ReadFileToString(ini_file, &content);
+  base::DictionaryValueINIParser ini_parser;
+  ini_parser.Parse(content);
+  const DictionaryValue& root = ini_parser.root();
 
   base::FilePath source_path;
   for (int i = 0; ; ++i) {
@@ -130,50 +133,6 @@ bool GetFirefoxVersionAndPathFromProfile(const base::FilePath& profile_path,
   return ret;
 }
 
-void ParseProfileINI(const base::FilePath& file, DictionaryValue* root) {
-  // Reads the whole INI file.
-  std::string content;
-  file_util::ReadFileToString(file, &content);
-  ReplaceSubstringsAfterOffset(&content, 0, "\r\n", "\n");
-  std::vector<std::string> lines;
-  base::SplitString(content, '\n', &lines);
-
-  // Parses the file.
-  root->Clear();
-  std::string current_section;
-  for (size_t i = 0; i < lines.size(); ++i) {
-    std::string line = lines[i];
-    if (line.empty()) {
-      // Skips the empty line.
-      continue;
-    }
-    if (line[0] == '#' || line[0] == ';') {
-      // This line is a comment.
-      continue;
-    }
-    if (line[0] == '[') {
-      // It is a section header.
-      current_section = line.substr(1);
-      size_t end = current_section.rfind(']');
-      if (end != std::string::npos)
-        current_section.erase(end);
-    } else {
-      std::string key, value;
-      size_t equal = line.find('=');
-      if (equal != std::string::npos) {
-        key = line.substr(0, equal);
-        value = line.substr(equal + 1);
-        // Checks whether the section and key contain a '.' character.
-        // Those sections and keys break DictionaryValue's path format,
-        // so we discard them.
-        if (current_section.find('.') == std::string::npos &&
-            key.find('.') == std::string::npos)
-          root->SetString(current_section + "." + key, value);
-      }
-    }
-  }
-}
-
 bool CanImportURL(const GURL& url) {
   const char* kInvalidSchemes[] = {"wyciwyg", "place", "about", "chrome"};