FTP: fix navigating to files listed under non-ASCII characters

We need to convert the file name back to server encoding.

BUG=38016
TEST=see bug

Review URL: http://codereview.chromium.org/1857002

git-svn-id: svn://svn.chromium.org/chrome/trunk/src@46900 0039d316-1c4b-4281-b951-d872f2087c98

4 files changed, 16 insertions, 33 deletions

diff --git a/net/base/net_error_list.h b/net/base/net_error_list.h
index c42549d..3338229 100644
--- a/net/base/net_error_list.h
+++ b/net/base/net_error_list.h
@@ -335,6 +335,9 @@ NET_ERROR(INVALID_AUTH_CREDENTIALS, -338)
 // machine.
 NET_ERROR(UNSUPPORTED_AUTH_SCHEME, -339)
 
+// Detecting the encoding of the response failed.
+NET_ERROR(ENCODING_DETECTION_FAILED, -340)
+
 // The cache does not have the requested entry.
 NET_ERROR(CACHE_MISS, -400)
 
diff --git a/net/base/net_util.h b/net/base/net_util.h
index 77bd69a..5c3e37e 100644
--- a/net/base/net_util.h
+++ b/net/base/net_util.h
@@ -210,10 +210,12 @@ std::string GetDirectoryListingHeader(const string16& title);
 // Currently, it's a script tag containing a call to a Javascript function
 // |addRow|.
 //
-// Its 1st parameter is derived from |name| and is the Javascript-string
-// escaped form of |name| (i.e \uXXXX). The 2nd parameter is the url-escaped
-// |raw_bytes| if it's not empty. If empty, the 2nd parameter is the
-// url-escaped |name| in UTF-8.
+// |name| is the file name to be displayed. |raw_bytes| will be used
+// as the actual target of the link (so for example, ftp links should use
+// server's encoding). If |raw_bytes| is an empty string, UTF-8 encoded |name|
+// will be used.
+//
+// Both |name| and |raw_bytes| are escaped internally.
 std::string GetDirectoryListingEntry(const string16& name,
                                      const std::string& raw_bytes,
                                      bool is_dir, int64 size,
diff --git a/net/ftp/ftp_directory_listing_buffer.cc b/net/ftp/ftp_directory_listing_buffer.cc
index bc2db9c..41daeb4 100644
--- a/net/ftp/ftp_directory_listing_buffer.cc
+++ b/net/ftp/ftp_directory_listing_buffer.cc
@@ -4,6 +4,7 @@
 
 #include "net/ftp/ftp_directory_listing_buffer.h"
 
+#include "base/i18n/icu_encoding_detection.h"
 #include "base/i18n/icu_string_conversions.h"
 #include "base/stl_util-inl.h"
 #include "base/string_util.h"
@@ -13,33 +14,6 @@
 #include "net/ftp/ftp_directory_listing_parser_netware.h"
 #include "net/ftp/ftp_directory_listing_parser_vms.h"
 #include "net/ftp/ftp_directory_listing_parser_windows.h"
-#include "unicode/ucsdet.h"
-
-namespace {
-
-// A very simple-minded character encoding detection.
-// TODO(jungshik): We can apply more heuristics here (e.g. using various hints
-// like TLD, the UI language/default encoding of a client, etc). In that case,
-// this should be pulled out of here and moved somewhere in base because there
-// can be other use cases.
-std::string DetectEncoding(const std::string& text) {
-  if (IsStringASCII(text))
-    return std::string();
-  UErrorCode status = U_ZERO_ERROR;
-  UCharsetDetector* detector = ucsdet_open(&status);
-  ucsdet_setText(detector, text.data(), static_cast<int32_t>(text.length()),
-                 &status);
-  const UCharsetMatch* match = ucsdet_detect(detector, &status);
-  const char* encoding = ucsdet_getName(match, &status);
-  ucsdet_close(detector);
-  // Should we check the quality of the match? A rather arbitrary number is
-  // assigned by ICU and it's hard to come up with a lower limit.
-  if (U_FAILURE(status))
-    return std::string();
-  return encoding;
-}
-
-}  // namespace
 
 namespace net {
 
@@ -109,8 +83,10 @@ bool FtpDirectoryListingBuffer::ConvertToDetectedEncoding(
 }
 
 int FtpDirectoryListingBuffer::ExtractFullLinesFromBuffer() {
-  if (encoding_.empty())
-    encoding_ = DetectEncoding(buffer_);
+  if (encoding_.empty()) {
+    if (!base::DetectEncoding(buffer_, &encoding_))
+      return ERR_ENCODING_DETECTION_FAILED;
+  }
 
   int cut_pos = 0;
   // TODO(phajdan.jr): This code accepts all endlines matching \r*\n. Should it
diff --git a/net/ftp/ftp_directory_listing_buffer.h b/net/ftp/ftp_directory_listing_buffer.h
index 7aead49..4123cf0 100644
--- a/net/ftp/ftp_directory_listing_buffer.h
+++ b/net/ftp/ftp_directory_listing_buffer.h
@@ -45,6 +45,8 @@ class FtpDirectoryListingBuffer {
   // time, although it will return SERVER_UNKNOWN if it doesn't know the answer.
   FtpServerType GetServerType() const;
 
+  const std::string& encoding() const { return encoding_; }
+
  private:
   typedef std::set<FtpDirectoryListingParser*> ParserSet;