Omnibox: Highlight Matches in URLs Properly

chrome/browser/autocomplete/history_quick_provider_unittest.cc

 // Copyright (c) 2012 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/autocomplete/history_quick_provider.h"
 
 #include <algorithm>
 #include <functional>
 #include <set>
 #include <string>
@@ skipping 72 matching lines @@
   {"http://abcxyzdefghijklmnopqrstuvw.com/a", "", 3, 1, 0},
   {"http://xyzabcdefghijklmnopqrstuvw.com/a", "", 3, 1, 0},
   {"http://cda.com/Dogs%20Cats%20Gorillas%20Sea%20Slugs%20and%20Mice",
    "Dogs & Cats & Mice & Other Animals", 1, 1, 0},
   {"https://monkeytrap.org/", "", 3, 1, 0},
   {"http://popularsitewithpathonly.com/moo",
    "popularsitewithpathonly.com/moo", 50, 50, 0},
   {"http://popularsitewithroot.com/", "popularsitewithroot.com", 50, 50, 0},
   {"http://testsearch.com/?q=thequery", "Test Search Engine", 10, 10, 0},
   {"http://testsearch.com/", "Test Search Engine", 9, 9, 0},
-  {"http://anotherengine.com/?q=thequery", "Another Search Engine", 8, 8, 0}
+  {"http://anotherengine.com/?q=thequery", "Another Search Engine", 8, 8, 0},
+  // The encoded stuff between /wiki/ and the # is 第二次世界大戦
+  {"http://ja.wikipedia.org/wiki/%E7%AC%AC%E4%BA%8C%E6%AC%A1%E4%B8%96%E7%95"
+   "%8C%E5%A4%A7%E6%88%A6#.E3.83.B4.E3.82.A7.E3.83.AB.E3.82.B5.E3.82.A4.E3."
+   "83.A6.E4.BD.93.E5.88.B6", "Title Unimportant", 2, 2, 0}
 };
 
 class HistoryQuickProviderTest : public testing::Test,
                                  public AutocompleteProviderListener {
  public:
   HistoryQuickProviderTest()
       : ui_thread_(BrowserThread::UI, &message_loop_),
         file_thread_(BrowserThread::FILE, &message_loop_) {}
 
   // AutocompleteProviderListener:
@@ skipping 247 matching lines @@
 }
 
 TEST_F(HistoryQuickProviderTest, EncodingMatch) {
   std::vector<std::string> expected_urls;
   expected_urls.push_back("http://spaces.com/path%20with%20spaces/foo.html");
   RunTest(ASCIIToUTF16("path with spaces"), false, expected_urls, false,
           ASCIIToUTF16("spaces.com/path with spaces/foo.html"),
           base::string16());
 }
 
+TEST_F(HistoryQuickProviderTest, ContentsClass) {
+  std::vector<std::string> expected_urls;
+  expected_urls.push_back(
+      "http://ja.wikipedia.org/wiki/%E7%AC%AC%E4%BA%8C%E6%AC%A1%E4%B8%96%E7"
+      "%95%8C%E5%A4%A7%E6%88%A6#.E3.83.B4.E3.82.A7.E3.83.AB.E3.82.B5.E3.82."
+      "A4.E3.83.A6.E4.BD.93.E5.88.B6");
+  RunTest(base::UTF8ToUTF16("第二 e3"), false, expected_urls, false,
+          base::UTF8ToUTF16("ja.wikipedia.org/wiki/第二次世界大戦#.E3.83.B4.E3."
+                            "82.A7.E3.83.AB.E3.82.B5.E3.82.A4.E3.83.A6.E4.BD."
+                            "93.E5.88.B6"),
+          base::string16());
+  ac_matches_[0].Validate();
+  // Verify that contents_class divides the string in the right places.
+  // [22, 24) is the "第二".  All the other pairs are the "e3".
+  ACMatchClassifications contents_class(ac_matches_[0].contents_class);
+  size_t expected_offsets[] = { 0, 22, 24, 31, 33, 40, 42, 49, 51, 58, 60, 67,
+                                69, 76, 78 };
+  // ScoredHistoryMatch may not highlight all the occurrences of these terms
+  // because it only highlights terms at word breaks, and it only stores word
+  // breaks up to some specified number of characters (50 at the time of this
+  // comment).  This test is written flexibly so it still will pass if we
+  // increase that number in the future.  Regardless, we require the first
+  // five offsets to be correct--in this example tthese cover at least one
+  // occurrence of each term.
+  EXPECT_GE(arraysize(expected_offsets), contents_class.size());
+  EXPECT_LE(5u, contents_class.size());

[Peter Kasting, 2014/04/23 23:18:01]

Nit: For EXPECT_XXX other than EQ and NE, write as (actual, expected) so the
expression reads like "expect a > 5" instead of "expect 5 < a".  In these cases,
the printed output won't assume that the first arg is the "expected" value, and
reading the code is usually easier in "natural" order.

[Mark P, 2014/04/24 14:05:02]

Good tip / reminder.  Done.

(This is the only place that needs to change in this changelist, right?)

+  for (size_t i = 0; i < contents_class.size(); ++i)
+    EXPECT_EQ(expected_offsets[i], contents_class[i].offset);
+}
+
 TEST_F(HistoryQuickProviderTest, VisitCountMatches) {
   std::vector<std::string> expected_urls;
   expected_urls.push_back("http://visitedest.com/y/a");
   expected_urls.push_back("http://visitedest.com/y/b");
   expected_urls.push_back("http://visitedest.com/x/c");
   RunTest(ASCIIToUTF16("visitedest"), false, expected_urls, true,
           ASCIIToUTF16("visitedest.com/y/a"),
                     ASCIIToUTF16(".com/y/a"));
 }
 
@@ skipping 298 matching lines @@
 
 TEST_F(HQPOrderingTest, TEAMatch) {
   std::vector<std::string> expected_urls;
   expected_urls.push_back("http://www.teamliquid.net/");
   expected_urls.push_back("http://www.teamliquid.net/tlpd");
   expected_urls.push_back("http://www.teamliquid.net/tlpd/korean/players");
   RunTest(ASCIIToUTF16("tea"), false, expected_urls, true,
           ASCIIToUTF16("www.teamliquid.net"),
                  ASCIIToUTF16("mliquid.net"));
 }