Record language usage as UMA histograms.

chrome/browser/language_usage_metrics_unittest.cc

+// Copyright (c) 2011 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 "chrome/browser/language_usage_metrics.h"
+
+#include <algorithm>
+#include <vector>
+#include "testing/gtest/include/gtest/gtest.h"
+
+namespace {
+void ToSortedList(const std::set<Language>& language_set,

[Yusuke Sato, 2011/07/13 06:52:06]

The word "List" reminds me std::list. It's up to you but how about using "Array"
or "Vector"?

[Yuzo, 2011/07/13 11:30:20]

Renamed to language_vector.

+                  std::vector<Language>* language_list) {
+  language_list->clear();
+  language_list->assign(language_set.begin(), language_set.end());
+  sort(language_list->begin(), language_list->end());

[Yusuke Sato, 2011/07/13 06:52:06]

std:: seems to be missing.

I'm also wondering why this line is necessary. I guess the language list is
sorted from the beginning (since std::set is an ordered set).

[Yuzo, 2011/07/13 11:30:20]

Removed the function as a whole. I admit I didn't know that it is always sorted.

+}
+}  // namespace
+
+class LanguageUsageMetricsTest : public testing::Test {
+};
+
+TEST_F(LanguageUsageMetricsTest, ParseAcceptLanguages) {

[Yusuke Sato, 2011/07/13 06:52:06]

I slightly prefer using TEST() and removing line 20-21 (since it's shorter.)

[Yuzo, 2011/07/13 11:30:20]

Done.

+  std::set<Language> language_set;
+  std::vector<Language> language_list;
+
+  // Basic single language case.
+  LanguageUsageMetrics::ParseAcceptLanguages("ja", &language_set);
+  ToSortedList(language_set, &language_list);
+  EXPECT_EQ(1U, language_list.size());
+  EXPECT_EQ(JAPANESE, language_list[0]);
+
+  // Empty language.
+  LanguageUsageMetrics::ParseAcceptLanguages("", &language_set);
+  ToSortedList(language_set, &language_list);
+  EXPECT_EQ(0U, language_list.size());
+
+  // Country code is ignored.
+  LanguageUsageMetrics::ParseAcceptLanguages("ja-JP", &language_set);
+  ToSortedList(language_set, &language_list);
+  EXPECT_EQ(1U, language_list.size());
+  EXPECT_EQ(JAPANESE, language_list[0]);
+
+  // Case is ignored.
+  LanguageUsageMetrics::ParseAcceptLanguages("Ja_jP", &language_set);
+  ToSortedList(language_set, &language_list);
+  EXPECT_EQ(1U, language_list.size());
+  EXPECT_EQ(JAPANESE, language_list[0]);
+
+  // The result contains a same language code only once.
+  LanguageUsageMetrics::ParseAcceptLanguages("ja-JP,ja", &language_set);
+  ToSortedList(language_set, &language_list);
+  EXPECT_EQ(1U, language_list.size());
+  EXPECT_EQ(JAPANESE, language_list[0]);
+
+  // Basic two languages case.
+  LanguageUsageMetrics::ParseAcceptLanguages("en,ja", &language_set);
+  ToSortedList(language_set, &language_list);
+  EXPECT_EQ(2U, language_list.size());
+  EXPECT_EQ(ENGLISH, language_list[0]);
+  EXPECT_EQ(JAPANESE, language_list[1]);
+
+  // Multiple languages case.
+  LanguageUsageMetrics::ParseAcceptLanguages("ja-JP,en,es,ja,en-US",
+                                             &language_set);
+  ToSortedList(language_set, &language_list);
+  EXPECT_EQ(3U, language_list.size());
+  EXPECT_EQ(ENGLISH, language_list[0]);
+  EXPECT_EQ(JAPANESE, language_list[1]);
+  EXPECT_EQ(SPANISH, language_list[2]);
+}

[Yusuke Sato, 2011/07/13 06:52:06]

What about adding tests for bad inputs like ",", "en,", or ",en"?

[Yuzo, 2011/07/13 11:30:20]

Done.

+
+TEST_F(LanguageUsageMetricsTest, ToLanguage) {
+  // Basic case.
+  EXPECT_EQ(JAPANESE, LanguageUsageMetrics::ToLanguage("ja"));
+
+  // Case is ignored.
+  EXPECT_EQ(SPANISH, LanguageUsageMetrics::ToLanguage("Es"));
+
+  // Coutry code is ignored.
+  EXPECT_EQ(JAPANESE, LanguageUsageMetrics::ToLanguage("ja-JP"));
+
+  // Invalid locales are considered as unknown language.
+  EXPECT_EQ(UNKNOWN_LANGUAGE, LanguageUsageMetrics::ToLanguage(""));
+  EXPECT_EQ(UNKNOWN_LANGUAGE, LanguageUsageMetrics::ToLanguage("xx"));
+}