Cherry-pick a6c922e. rdar://problem/101518569

Punycode U+0E01 when in the context of non-Thai characters https://bugs.webkit.org/show_bug.cgi?id=247287 rdar://101434628 Reviewed by Tim Horton. * Source/WTF/wtf/URLHelpers.cpp: (WTF::URLHelpers::isLookalikeCharacterOfScriptType<USCRIPT_THAI>): (WTF::URLHelpers::isLookalikeCharacter): * Tools/TestWebKitAPI/Tests/WTF/cocoa/URLExtras.mm: (TestWebKitAPI::TEST): Canonical link: https://commits.webkit.org/256276@main Canonical link: https://commits.webkit.org/245886.841@safari-7613.4.1.0-branch
WebKit · Nov 9, 2022 · 04050e1 · 04050e1
1 parent 44e845a
commit 04050e1
Show file tree

Hide file tree

Showing 4 changed files with 19 additions and 3 deletions.
diff --git a/LayoutTests/fast/encoding/idn-security-expected.txt b/LayoutTests/fast/encoding/idn-security-expected.txt
@@ -30,8 +30,8 @@ PASS testIDNRoundTrip(0x30a1) is '%u30A1'
 PASS testIDNRoundTripNotFirstCharacter(0x30a1) is '%u30A1'
 PASS testIDNRoundTrip(0xb94) is '%u0B94'
 PASS testIDNRoundTripNotFirstCharacter(0xb94) is '%u0B94'
-PASS testIDNRoundTrip(0xe01) is '%u0E01'
-PASS testIDNRoundTripNotFirstCharacter(0xe01) is '%u0E01'
+PASS testIDNRoundTrip(0xe02) is '%u0E02'
+PASS testIDNRoundTripNotFirstCharacter(0xe02) is '%u0E02'
 PASS testIDNRoundTrip(0xa000) is '%uA000'
 PASS testIDNRoundTripNotFirstCharacter(0xa000) is '%uA000'
 PASS testIDNRoundTrip(0xff0f) is '/'

diff --git a/LayoutTests/fast/encoding/idn-security.html b/LayoutTests/fast/encoding/idn-security.html
@@ -130,7 +130,7 @@
 testIDNCharacter(0x3041, "allowed");
 testIDNCharacter(0x30A1, "allowed");
 testIDNCharacter(0x0B94, "allowed");
-testIDNCharacter(0x0E01, "allowed");
+testIDNCharacter(0x0E02, "allowed");
 testIDNCharacter(0xA000, "allowed");
 
 /* ICU converts these to other allowed characters, so the original character can't be used to get to a phishy domain name */

diff --git a/Source/WTF/wtf/URLHelpers.cpp b/Source/WTF/wtf/URLHelpers.cpp
@@ -102,6 +102,16 @@ template<> bool isLookalikeCharacterOfScriptType<USCRIPT_CANADIAN_ABORIGINAL>(UC
     }
 }
 
+template<> bool isLookalikeCharacterOfScriptType<USCRIPT_THAI>(UChar32 codePoint)
+{
+    switch (codePoint) {
+    case 0x0E01: // THAI CHARACTER KO KAI
+        return true;
+    default:
+        return false;
+    }
+}
+
 template <UScriptCode ScriptType>
 bool isOfScriptType(UChar32 codePoint)
 {
@@ -314,6 +324,7 @@ static bool isLookalikeCharacter(const std::optional<UChar32>& previousCodePoint
         return isLookalikeSequence<USCRIPT_ARMENIAN>(previousCodePoint, codePoint)
             || isLookalikeSequence<USCRIPT_TAMIL>(previousCodePoint, codePoint)
             || isLookalikeSequence<USCRIPT_CANADIAN_ABORIGINAL>(previousCodePoint, codePoint)
+            || isLookalikeSequence<USCRIPT_THAI>(previousCodePoint, codePoint)
             || isLookalikeSequence<USCRIPT_ARABIC>(previousCodePoint, codePoint);
     }
 }

diff --git a/Tools/TestWebKitAPI/Tests/WTF/cocoa/URLExtras.mm b/Tools/TestWebKitAPI/Tests/WTF/cocoa/URLExtras.mm
@@ -131,6 +131,8 @@
         "xn--o-10e", // 'o' U+0BE6
         "xn--a-53i", // U+15AF 'a'
         "xn--a-63i", // 'a' U+15AF
+        "xn--n-twf", // U+0E01 'n'
+        "xn--n-uwf", // 'n' U+0E01
         "xn--3hb112n", // U+065B
         "xn--a-ypc062v", // 'a' U+065B
     };
@@ -163,6 +165,9 @@
     // Canadian aboriginal
     EXPECT_STREQ("https://\u15AF\u1401abc/", userVisibleString(literalURL("https://\u15AF\u1401abc/")));
 
+    // Thai
+    EXPECT_STREQ("https://\u0E01\u0E02abc/", userVisibleString(literalURL("https://\u0E01\u0E02abc/")));
+
     // Arabic
     EXPECT_STREQ("https://\u0620\u065Babc/", userVisibleString(literalURL("https://\u0620\u065Babc/")));
 }