Normalize scheme and reg-name elements to lower case

2 files changed, 43 insertions, 1 deletions

diff --git a/src/Uri.cpp b/src/Uri.cpp
index b30796c..08304d3 100644
--- a/src/Uri.cpp
+++ b/src/Uri.cpp
@@ -7,6 +7,7 @@
  */
 
 #include "IsCharacterInSet.hpp"
+#include "NormalizeCaseInsensitiveString.hpp"
 #include "PercentEncodedCharacterDecoder.hpp"
 
 #include <functional>
@@ -467,6 +468,7 @@ namespace Uri {
             int decodedCharacter = 0;
             host.clear();
             PercentEncodedCharacterDecoder pecDecoder;
+            bool hostIsRegName = false;
             for (const auto c: hostPortString) {
                 switch(decoderState) {
                     case 0: { // first character
@@ -476,6 +478,7 @@ namespace Uri {
                             break;
                         } else {
                             decoderState = 1;
+                            hostIsRegName = true;
                         }
                     }
 
@@ -555,6 +558,9 @@ namespace Uri {
                     } break;
                 }
             }
+            if (hostIsRegName) {
+                host = NormalizeCaseInsensitiveString(host);
+            }
             if (portString.empty()) {
                 hasPort = false;
             } else {
@@ -600,6 +606,7 @@ namespace Uri {
             ) {
                 return false;
             }
+            impl_->scheme = NormalizeCaseInsensitiveString(impl_->scheme);
             rest = uriString.substr(schemeEnd + 1);
         }
 
diff --git a/test/src/UriTests.cpp b/test/src/UriTests.cpp
index bb541d0..4242885 100644
--- a/test/src/UriTests.cpp
+++ b/test/src/UriTests.cpp
@@ -275,6 +275,23 @@ TEST(UriTests, ParseFromStringSchemeBarelyLegal) {
     }
 }
 
+TEST(UriTests, ParseFromStringSchemeMixedCase) {
+    const std::vector< std::string > testVectors{
+        {"http://www.example.com/"},
+        {"hTtp://www.example.com/"},
+        {"HTTP://www.example.com/"},
+        {"Http://www.example.com/"},
+        {"HttP://www.example.com/"},
+    };
+    size_t index = 0;
+    for (const auto& testVector : testVectors) {
+        Uri::Uri uri;
+        ASSERT_TRUE(uri.ParseFromString(testVector)) << index;
+        ASSERT_EQ("http", uri.GetScheme()) << ">>> Failed for test vector element " << index << " <<<";
+        ++index;
+    }
+}
+
 TEST(UriTests, ParseFromStringUserInfoIllegalCharacters) {
     const std::vector< std::string > testVectors{
         {"//%X@www.example.com/"},
@@ -331,7 +348,7 @@ TEST(UriTests, ParseFromStringHostBarelyLegal) {
         std::string host;
     };
     const std::vector< TestVector > testVectors{
-        {"//%41/", "A"},
+        {"//%41/", "a"},
         {"///", ""},
         {"//!/", "!"},
         {"//'/", "'"},
@@ -339,6 +356,7 @@ TEST(UriTests, ParseFromStringHostBarelyLegal) {
         {"//;/", ";"},
         {"//1.2.3.4/", "1.2.3.4"},
         {"//[v7.:]/", "[v7.:]"},
+        {"//[v7.aB]/", "[v7.aB]"},
     };
     size_t index = 0;
     for (const auto& testVector : testVectors) {
@@ -349,6 +367,23 @@ TEST(UriTests, ParseFromStringHostBarelyLegal) {
     }
 }
 
+TEST(UriTests, ParseFromStringHostMixedCase) {
+    const std::vector< std::string > testVectors{
+        {"http://www.example.com/"},
+        {"http://www.EXAMPLE.com/"},
+        {"http://www.exAMple.com/"},
+        {"http://www.example.cOM/"},
+        {"http://wWw.exampLe.Com/"},
+    };
+    size_t index = 0;
+    for (const auto& testVector : testVectors) {
+        Uri::Uri uri;
+        ASSERT_TRUE(uri.ParseFromString(testVector)) << index;
+        ASSERT_EQ("www.example.com", uri.GetHost()) << ">>> Failed for test vector element " << index << " <<<";
+        ++index;
+    }
+}
+
 TEST(UriTests, ParseFromStringDontMisinterpretColonInOtherPlacesAsSchemeDelimiter) {
     const std::vector< std::string > testVectors{
         {"//foo:bar@www.example.com/"},