Unpunycode search keywords and short names.

components/search_engines/template_url.cc

 // Copyright 2014 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 "components/search_engines/template_url.h"
 
 #include <string>
 #include <vector>
 
 #include "base/basictypes.h"
@@ skipping 1192 matching lines @@
   if (data_.search_terms_replacement_key ==
       "{google:instantExtendedEnabledKey}") {
     data_.search_terms_replacement_key = google_util::kInstantExtendedAPIParam;
   }
 }
 
 TemplateURL::~TemplateURL() {
 }
 
 // static
-base::string16 TemplateURL::GenerateKeyword(const GURL& url) {
+base::string16 TemplateURL::GenerateKeyword(
+    const GURL& url,
+    const std::string& accept_languages) {
   DCHECK(url.is_valid());
+  // Search engine's keyword is a user facing value and therefore
+  // should be IDN-decoded using user's accepted languages.

[Peter Kasting, 2015/07/20 19:30:52]

This rationale seems a bit inaccurate.  How about adding this to the comment
below (see the next two nits about how to reorder the code/comments here for
this):

"Since the underlying hostname may be punycoded, convert to Unicode if possible,
so the keyword will look like the Unicode hostname the user is used to seeing."

[alshabalin, 2015/07/21 08:44:19]

Done.

But I worded it like this:
  // Since keyword is generated from hostname, it might be IDN-encoded.
  // If so, decode it and convert to Unicode using user's accepted languages,
  // making it look like Unicode hostname the user is used to seeing.

+  base::string16 keyword = net::IDNToUnicode(url.host(), accept_languages);

[Peter Kasting, 2015/07/20 19:30:52]

Nit: Combine this with the line below.

   // Strip "www." off the front of the keyword; otherwise the keyword won't work
   // properly.  See http://code.google.com/p/chromium/issues/detail?id=6984 .
   // Special case: if the host was exactly "www." (not sure this can happen but

[Peter Kasting, 2015/07/20 19:30:52]

Nit: Move this "Special case" comment below the StripWWW() call so you can
append the commentary about IDN-decoding here instead.

   // perhaps with some weird intranet and custom DNS server?), ensure we at
   // least don't return the empty string.
-  base::string16 keyword(net::StripWWWFromHost(url));
+  keyword = net::StripWWW(keyword);
   return keyword.empty() ? base::ASCIIToUTF16("www") : keyword;
 }
 
 // static
 GURL TemplateURL::GenerateFaviconURL(const GURL& url) {
   DCHECK(url.is_valid());
   GURL::Replacements rep;
 
   const char favicon_path[] = "/favicon.ico";
   int favicon_path_len = arraysize(favicon_path) - 1;
@@ skipping 247 matching lines @@
   instant_url_ref_.prepopulated_ = prepopulated;
 }
 
 void TemplateURL::ResetKeywordIfNecessary(
     const SearchTermsData& search_terms_data,
     bool force) {
   if (IsGoogleSearchURLWithReplaceableKeyword(search_terms_data) || force) {
     DCHECK(GetType() != OMNIBOX_API_EXTENSION);
     GURL url(GenerateSearchURL(search_terms_data));
     if (url.is_valid())
-      data_.SetKeyword(GenerateKeyword(url));
+      data_.SetKeyword(
+          GenerateKeyword(url, search_terms_data.GetAcceptedLanguages()));
   }
 }
 
 bool TemplateURL::FindSearchTermsInURL(
     const GURL& url,
     const SearchTermsData& search_terms_data,
     base::string16* search_terms,
     url::Parsed::ComponentType* search_term_component,
     url::Component* search_terms_position) const {
   DCHECK(search_terms);
@@ skipping 10 matching lines @@
       // patterns. This means that given patterns
       //    [ "http://foo/#q={searchTerms}", "http://foo/?q={searchTerms}" ],
       // calling ExtractSearchTermsFromURL() on "http://foo/?q=bar#q=' would
       // return false. This is important for at least Google, where such URLs
       // are invalid.
       return !search_terms->empty();
     }
   }
   return false;
 }