From 50158abaf196e23e5996646bfc1b42f8ca0d3764 Mon Sep 17 00:00:00 2001
From: Timothy Flynn <trflynn89@pm.me>
Date: Sun, 5 Sep 2021 15:09:03 -0400
Subject: [PATCH] LibUnicode: Implement locale-aware BEFORE_DOT special casing

Note that the algorithm in the Unicode spec is for checking that a code
point precedes U+0307, but the special casing condition NotBeforeDot is
interested in the inverse of this rule.
---
 .../LibUnicode/TestUnicodeCharacterTypes.cpp  | 23 +++++++++++++++
 .../Libraries/LibUnicode/CharacterTypes.cpp   | 28 ++++++++++++++++++-
 2 files changed, 50 insertions(+), 1 deletion(-)

diff --git a/Tests/LibUnicode/TestUnicodeCharacterTypes.cpp b/Tests/LibUnicode/TestUnicodeCharacterTypes.cpp
index e53b951945..c20f0e4849 100644
--- a/Tests/LibUnicode/TestUnicodeCharacterTypes.cpp
+++ b/Tests/LibUnicode/TestUnicodeCharacterTypes.cpp
@@ -243,6 +243,29 @@ TEST_CASE(to_unicode_lowercase_special_casing_more_above)
     EXPECT_EQ(result, "\u012f\u0307\u0300"sv);
 }
 
+TEST_CASE(to_unicode_lowercase_special_casing_not_before_dot)
+{
+    // LATIN CAPITAL LETTER I
+    auto result = Unicode::to_unicode_lowercase_full("I"sv, "en"sv);
+    EXPECT_EQ(result, "i"sv);
+
+    result = Unicode::to_unicode_lowercase_full("I"sv, "az"sv);
+    EXPECT_EQ(result, "\u0131"sv);
+
+    result = Unicode::to_unicode_lowercase_full("I"sv, "tr"sv);
+    EXPECT_EQ(result, "\u0131"sv);
+
+    // LATIN CAPITAL LETTER I followed by COMBINING DOT ABOVE
+    result = Unicode::to_unicode_lowercase_full("I\u0307"sv, "en"sv);
+    EXPECT_EQ(result, "i\u0307"sv);
+
+    result = Unicode::to_unicode_lowercase_full("I\u0307"sv, "az"sv);
+    EXPECT_EQ(result, "i"sv);
+
+    result = Unicode::to_unicode_lowercase_full("I\u0307"sv, "tr"sv);
+    EXPECT_EQ(result, "i"sv);
+}
+
 TEST_CASE(to_unicode_uppercase_unconditional_special_casing)
 {
     // LATIN SMALL LETTER SHARP S
diff --git a/Userland/Libraries/LibUnicode/CharacterTypes.cpp b/Userland/Libraries/LibUnicode/CharacterTypes.cpp
index 347d2f0c04..0c2c7b2e90 100644
--- a/Userland/Libraries/LibUnicode/CharacterTypes.cpp
+++ b/Userland/Libraries/LibUnicode/CharacterTypes.cpp
@@ -135,6 +135,30 @@ static bool is_followed_by_combining_class_above(Utf8View const& string, size_t
     return false;
 }
 
+static bool is_followed_by_combining_dot_above(Utf8View const& string, size_t index, size_t byte_length)
+{
+    // C is followed by combining dot above (U+0307). Any sequence of characters with a combining class that is neither 0 nor 230 may
+    // intervene between the current character and the combining dot above.
+    auto following_view = ((index + byte_length) < string.byte_length())
+        ? string.substring_view(index + byte_length)
+        : Utf8View {};
+
+    for (auto code_point : following_view) {
+        if (code_point == 0x307)
+            return true;
+
+        auto unicode_data = Detail::unicode_data_for_code_point(code_point);
+        if (!unicode_data.has_value())
+            return false;
+        if (unicode_data->canonical_combining_class == 0)
+            return false;
+        if (unicode_data->canonical_combining_class == 230)
+            return false;
+    }
+
+    return false;
+}
+
 static SpecialCasing const* find_matching_special_case(Utf8View const& string, Optional<StringView> locale, size_t index, size_t byte_length, UnicodeData const& unicode_data)
 {
     auto requested_locale = Locale::None;
@@ -174,7 +198,9 @@ static SpecialCasing const* find_matching_special_case(Utf8View const& string, O
                 return special_casing;
             break;
 
-        default:
+        case Condition::NotBeforeDot:
+            if (!is_followed_by_combining_dot_above(string, index, byte_length))
+                return special_casing;
             break;
         }
     }