From a72bb34970e5f1f0ed3d28790431e1df37b7dac7 Mon Sep 17 00:00:00 2001
From: Max Wipfli <mail@maxwipfli.ch>
Date: Tue, 18 May 2021 16:09:20 +0200
Subject: [PATCH] AK: Add Utf8View::iterator_at_byte_offset method

This implements a method to get a Utf8CodepointIterator at a specified
byte offset.
---
 AK/Utf8View.cpp | 11 +++++++++++
 AK/Utf8View.h   |  1 +
 2 files changed, 12 insertions(+)

diff --git a/AK/Utf8View.cpp b/AK/Utf8View.cpp
index 02944a42a8..0826eb0503 100644
--- a/AK/Utf8View.cpp
+++ b/AK/Utf8View.cpp
@@ -46,6 +46,17 @@ Utf8CodepointIterator Utf8View::end() const
     return { end_ptr(), 0 };
 }
 
+Utf8CodepointIterator Utf8View::iterator_at_byte_offset(size_t byte_offset) const
+{
+    size_t current_offset = 0;
+    for (auto iterator = begin(); !iterator.done(); ++iterator) {
+        if (current_offset >= byte_offset)
+            return iterator;
+        current_offset += iterator.code_point_length_in_bytes();
+    }
+    return end();
+}
+
 size_t Utf8View::byte_offset_of(const Utf8CodepointIterator& it) const
 {
     VERIFY(it.m_ptr >= begin_ptr());
diff --git a/AK/Utf8View.h b/AK/Utf8View.h
index d9239e6b62..7324bd2984 100644
--- a/AK/Utf8View.h
+++ b/AK/Utf8View.h
@@ -54,6 +54,7 @@ public:
 
     Utf8CodepointIterator begin() const;
     Utf8CodepointIterator end() const;
+    Utf8CodepointIterator iterator_at_byte_offset(size_t) const;
 
     const unsigned char* bytes() const { return begin_ptr(); }
     size_t byte_length() const { return m_string.length(); }