Let url filter test the scheme in urlContains/Equals/Prefix/Suffix criteria

chrome/common/extensions/matcher/url_matcher.cc

Index: chrome/common/extensions/matcher/url_matcher.cc
diff --git a/chrome/common/extensions/matcher/url_matcher.cc b/chrome/common/extensions/matcher/url_matcher.cc
index 722efd9c3c633e6898c7f4462cdc181fbd387dac..678983aa5e64ec533005096298bea1ae4eb9f81b 100644
--- a/chrome/common/extensions/matcher/url_matcher.cc
+++ b/chrome/common/extensions/matcher/url_matcher.cc
@@ -8,6 +8,7 @@
 #include <iterator>
 
 #include "base/logging.h"
+#include "content/public/common/url_constants.h"
 #include "googleurl/src/gurl.h"
 
 namespace extensions {
@@ -88,19 +89,18 @@ namespace extensions {
 // Case 2: url_{prefix,suffix,equals,contains} searches.
 // =====================================================
 //
-// Step 1: as above
+// Step 1: as above, except that the scheme is not removed.
 //
 // Step 2:
 // Translate URL to String and add the following position markers:
 // - BU = Beginning of URL
 // - EU = End of URL
-// Furthermore, the hostname is canonicalized to start with a ".".
 //
-// -> www.example.com/index.html?search=foo becomes
-// BU .www.example.com/index.html?search=foo EU
+// -> http://www.example.com:8080/index.html?search=foo#first_match becomes
+// BU http://www.example.com/index.html?search=foo EU
 //
-// url_prefix(prefix) = BU add_missing_dot_prefix(prefix)
-// -> url_prefix("www.example") = BU .www.example
+// url_prefix(prefix) = BU prefix
+// -> url_prefix("http://www.example") = BU http://www.example
 //
 // url_contains(substring) = substring
 // -> url_contains("index") = index
@@ -113,7 +113,7 @@ namespace extensions {
 // by a combination of a url_contains() query followed by an explicit test:
 //
 // host_contains(str) = url_contains(str) followed by test whether str occurs
-//   in host comonent of original URL.
+//   in host component of original URL.
 // -> host_contains("example.co") = example.co
 //    followed by gurl.host().find("example.co");
 //
@@ -160,7 +160,7 @@ bool URLMatcherCondition::operator<(const URLMatcherCondition& rhs) const {
 
 bool URLMatcherCondition::IsFullURLCondition() const {
   // For these criteria the SubstringMatcher needs to be executed on the
-  // GURL that is canonlizaliced with
+  // GURL that is canonicalized with
   // URLMatcherConditionFactory::CanonicalizeURLForFullSearches.
   switch (criterion_) {
     case HOST_CONTAINS:
@@ -314,14 +314,15 @@ URLMatcherConditionFactory::CreateHostEqualsPathPrefixCondition(
 
 std::string URLMatcherConditionFactory::CanonicalizeURLForFullSearches(
     const GURL& url) {
-  return kBeginningOfURL + CanonicalizeHostname(url.host()) + url.path() +
-      (url.has_query() ? "?" + url.query() : "") + kEndOfURL;
+  return kBeginningOfURL + url.scheme() + content::kStandardSchemeSeparator +

[Yoyo Zhou, 2012/08/15 17:53:12]

By the way, I just learned about GURL::Replacements yesterday; it could be
useful here.

[battre, 2012/08/16 13:16:25]

I was aware of this. I cannot use the GURL::Replacements for
URLMatcherConditionFactory::CanonicalizeURLForComponentSearches (I cannot get
rid of the scheme nor am I sure my host name normalization would be acceptable).
I'd rather have the two URLMatcherConditionFactory::CanonicalizeURLFor...()
functions look similar than use GURL::Replacements here.

+      url.host() + url.path() + (url.has_query() ? "?" + url.query() : "") +
+      kEndOfURL;
 }
 
 URLMatcherCondition URLMatcherConditionFactory::CreateURLPrefixCondition(
     const std::string& prefix) {
   return CreateCondition(URLMatcherCondition::URL_PREFIX,
-      kBeginningOfURL + CanonicalizeHostname(prefix));
+      kBeginningOfURL + prefix);
 }
 
 URLMatcherCondition URLMatcherConditionFactory::CreateURLSuffixCondition(
@@ -337,7 +338,7 @@ URLMatcherCondition URLMatcherConditionFactory::CreateURLContainsCondition(
 URLMatcherCondition URLMatcherConditionFactory::CreateURLEqualsCondition(
     const std::string& str) {
   return CreateCondition(URLMatcherCondition::URL_EQUALS,
-      kBeginningOfURL + CanonicalizeHostname(str) + kEndOfURL);
+      kBeginningOfURL + str + kEndOfURL);
 }
 
 void URLMatcherConditionFactory::ForgetUnusedPatterns(