Allow HEXDIG to include lower-case 'a'..'f'

3 files changed, 40 insertions, 4 deletions

diff --git a/src/PercentEncodedCharacterDecoder.cpp b/src/PercentEncodedCharacterDecoder.cpp
index 890d392..dc7cd58 100644
--- a/src/PercentEncodedCharacterDecoder.cpp
+++ b/src/PercentEncodedCharacterDecoder.cpp
@@ -21,7 +21,13 @@ namespace {
      * This is the character set containing just the upper-case
      * letters 'A' through 'F', used in upper-case hexadecimal.
      */
-    const Uri::CharacterSet HEX('A', 'F');
+    const Uri::CharacterSet HEX_UPPER('A', 'F');
+
+    /**
+     * This is the character set containing just the lower-case
+     * letters 'a' through 'f', used in lower-case hexadecimal.
+     */
+    const Uri::CharacterSet HEX_LOWER('a', 'f');
 
 }
 
@@ -58,8 +64,10 @@ namespace Uri {
                 impl_->decodedCharacter <<= 4;
                 if (IsCharacterInSet(c, DIGIT)) {
                     impl_->decodedCharacter += (int)(c - '0');
-                } else if (IsCharacterInSet(c, HEX)) {
+                } else if (IsCharacterInSet(c, HEX_UPPER)) {
                     impl_->decodedCharacter += (int)(c - 'A') + 10;
+                } else if (IsCharacterInSet(c, HEX_LOWER)) {
+                    impl_->decodedCharacter += (int)(c - 'a') + 10;
                 } else {
                     return false;
                 }
@@ -70,8 +78,10 @@ namespace Uri {
                 impl_->decodedCharacter <<= 4;
                 if (IsCharacterInSet(c, DIGIT)) {
                     impl_->decodedCharacter += (int)(c - '0');
-                } else if (IsCharacterInSet(c, HEX)) {
+                } else if (IsCharacterInSet(c, HEX_UPPER)) {
                     impl_->decodedCharacter += (int)(c - 'A') + 10;
+                } else if (IsCharacterInSet(c, HEX_LOWER)) {
+                    impl_->decodedCharacter += (int)(c - 'a') + 10;
                 } else {
                     return false;
                 }
diff --git a/src/Uri.cpp b/src/Uri.cpp
index 3bb3a59..b30796c 100644
--- a/src/Uri.cpp
+++ b/src/Uri.cpp
@@ -38,7 +38,8 @@ namespace {
      */
     const Uri::CharacterSet HEXDIG{
         Uri::CharacterSet('0', '9'),
-        Uri::CharacterSet('A', 'F')
+        Uri::CharacterSet('A', 'F'),
+        Uri::CharacterSet('a', 'f')
     };
 
     /**
diff --git a/test/src/UriTests.cpp b/test/src/UriTests.cpp
index 6d59697..bb541d0 100644
--- a/test/src/UriTests.cpp
+++ b/test/src/UriTests.cpp
@@ -524,3 +524,28 @@ TEST(UriTests, ParseFromStringFragmentBarelyLegal) {
         ++index;
     }
 }
+
+TEST(UriTests, ParseFromStringPathsWithPercentEncodedCharacters) {
+    struct TestVector {
+        std::string uriString;
+        std::string pathFirstSegment;
+    };
+    const std::vector< TestVector > testVectors{
+        {"%41", "A"},
+        {"%4A", "J"},
+        {"%4a", "J"},
+        {"%bc", "\xbc"},
+        {"%Bc", "\xbc"},
+        {"%bC", "\xbc"},
+        {"%BC", "\xbc"},
+        {"%41%42%43", "ABC"},
+        {"%41%4A%43%4b", "AJCK"},
+    };
+    size_t index = 0;
+    for (const auto& testVector : testVectors) {
+        Uri::Uri uri;
+        ASSERT_TRUE(uri.ParseFromString(testVector.uriString)) << index;
+        ASSERT_EQ(testVector.pathFirstSegment, uri.GetPath()[0]);
+        ++index;
+    }
+}