From d50724956e3e9fec55fb9528cf129f00e9ce5843 Mon Sep 17 00:00:00 2001
From: Timothy Flynn <trflynn89@pm.me>
Date: Sun, 22 Jan 2023 09:24:12 -0500
Subject: [PATCH] AK: Add a method to find the byte offset of a code point

---
 AK/String.cpp           | 14 ++++++++++++++
 AK/String.h             |  2 ++
 Tests/AK/TestString.cpp | 42 +++++++++++++++++++++++++++++++++++++++++
 3 files changed, 58 insertions(+)

diff --git a/AK/String.cpp b/AK/String.cpp
index 828e165eed0..8aa3d7dbcd0 100644
--- a/AK/String.cpp
+++ b/AK/String.cpp
@@ -275,6 +275,20 @@ ErrorOr<Vector<String>> String::split_limit(u32 separator, size_t limit, SplitBe
     return result;
 }
 
+Optional<size_t> String::find_byte_offset(u32 code_point, size_t from_byte_offset) const
+{
+    auto code_points = this->code_points();
+    if (from_byte_offset >= code_points.byte_length())
+        return {};
+
+    for (auto it = code_points.iterator_at_byte_offset(from_byte_offset); it != code_points.end(); ++it) {
+        if (*it == code_point)
+            return code_points.byte_offset_of(it);
+    }
+
+    return {};
+}
+
 bool String::operator==(String const& other) const
 {
     if (is_short_string())
diff --git a/AK/String.h b/AK/String.h
index cf34bcac503..1a7c01b1a97 100644
--- a/AK/String.h
+++ b/AK/String.h
@@ -127,6 +127,8 @@ public:
     ErrorOr<Vector<String>> split_limit(u32 separator, size_t limit, SplitBehavior = SplitBehavior::Nothing) const;
     ErrorOr<Vector<String>> split(u32 separator, SplitBehavior = SplitBehavior::Nothing) const;
 
+    Optional<size_t> find_byte_offset(u32 code_point, size_t from_byte_offset = 0) const;
+
     [[nodiscard]] bool operator==(String const&) const;
     [[nodiscard]] bool operator!=(String const& other) const { return !(*this == other); }
 
diff --git a/Tests/AK/TestString.cpp b/Tests/AK/TestString.cpp
index f90b7bf393d..872a9446678 100644
--- a/Tests/AK/TestString.cpp
+++ b/Tests/AK/TestString.cpp
@@ -315,3 +315,45 @@ TEST_CASE(split)
         EXPECT_EQ(parts[2], "ω"sv);
     }
 }
+
+TEST_CASE(find_byte_offset)
+{
+    {
+        String string {};
+        auto index = string.find_byte_offset(0);
+        EXPECT(!index.has_value());
+    }
+    {
+        auto string = MUST(String::from_utf8("foo"sv));
+
+        auto index1 = string.find_byte_offset('f');
+        EXPECT_EQ(index1, 0u);
+
+        auto index2 = string.find_byte_offset('o');
+        EXPECT_EQ(index2, 1u);
+
+        auto index3 = string.find_byte_offset('o', *index2 + 1);
+        EXPECT_EQ(index3, 2u);
+
+        auto index4 = string.find_byte_offset('b');
+        EXPECT(!index4.has_value());
+    }
+    {
+        auto string = MUST(String::from_utf8("ωΣωΣω"sv));
+
+        auto index1 = string.find_byte_offset(0x03C9U);
+        EXPECT_EQ(index1, 0u);
+
+        auto index2 = string.find_byte_offset(0x03A3u);
+        EXPECT_EQ(index2, 2u);
+
+        auto index3 = string.find_byte_offset(0x03C9U, 2);
+        EXPECT_EQ(index3, 4u);
+
+        auto index4 = string.find_byte_offset(0x03A3u, 4);
+        EXPECT_EQ(index4, 6u);
+
+        auto index5 = string.find_byte_offset(0x03C9U, 6);
+        EXPECT_EQ(index5, 8u);
+    }
+}