Replace ICU encoding detection with CED

base/i18n/encoding_detection.cc

Index: base/i18n/encoding_detection.cc
diff --git a/base/i18n/encoding_detection.cc b/base/i18n/encoding_detection.cc
new file mode 100644
index 0000000000000000000000000000000000000000..a3e3be828e2fac8ebc49cd4d8677d1291420f99b
--- /dev/null
+++ b/base/i18n/encoding_detection.cc
@@ -0,0 +1,41 @@
+// Copyright (c) 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.
+
+#include "base/i18n/encoding_detection.h"
+
+#include "third_party/ced/src/compact_enc_det/compact_enc_det.h"
+
+namespace base {
+
+bool DetectEncoding(const std::string& text, std::string* encoding) {
+  int consumedBytes;

[Lei Zhang, 2016/08/08 22:09:55]

nit: foo_bar, not fooBar.

[Jinsuk Kim, 2016/08/09 01:16:19]

Done.

+  bool isReliable;
+  Encoding enc = CompactEncDet::DetectEncoding(
+      text.c_str(), text.length(), nullptr, nullptr, nullptr,
+      UNKNOWN_ENCODING,
+      UNKNOWN_LANGUAGE,
+      CompactEncDet::QUERY_CORPUS, // plain text

[Lei Zhang, 2016/08/08 22:09:55]

two spaces in front of comments.

[Jinsuk Kim, 2016/08/09 01:16:19]

Done.

+      false, // Include 7-bit encodings
+      &consumedBytes,
+      &isReliable);
+
+  if (enc == UNKNOWN_ENCODING)
+    return false;
+
+  // 7-bit encodings (except ISO-2022-JP) are not supported in web standard.
+  // Mark them as ascii to keep the raw bytes intact.
+  switch (enc) {
+    case HZ_GB_2312:
+    case ISO_2022_KR:
+    case ISO_2022_CN:
+    case UTF7:
+      enc = ASCII_7BIT;
+      break;
+    default:
+      break;
+  }
+  *encoding = MimeEncodingName(enc);
+  return true;
+}
+}  // namespace base