From 8cdfbfed4996b469bbaaaa62885d72948beb2dce Mon Sep 17 00:00:00 2001
From: Andreas Kling <andreas@ladybird.org>
Date: Sat, 10 Jan 2026 12:26:20 +0100
Subject: [PATCH] LibUnicode+LibWeb: Add fast path grapheme segmenter for ASCII
 text

For ASCII text, every character is its own grapheme - there are no
combining characters or emoji sequences. This means grapheme boundary
detection is trivial: next_boundary(i) is simply i+1.

This commit adds AsciiGraphemeSegmenter, a simple Segmenter subclass
that performs O(1) boundary lookups without any ICU overhead.

TextNode::grapheme_segmenter() now checks if the text is ASCII and uses
this fast path, avoiding expensive ICU BreakIterator cloning and
boundary detection for the common case of ASCII-only text.
---
 Libraries/LibUnicode/Segmenter.cpp   | 91 ++++++++++++++++++++++++++++
 Libraries/LibUnicode/Segmenter.h     |  1 +
 Libraries/LibWeb/Layout/TextNode.cpp | 11 +++-
 3 files changed, 101 insertions(+), 2 deletions(-)

diff --git a/Libraries/LibUnicode/Segmenter.cpp b/Libraries/LibUnicode/Segmenter.cpp
index 254103fb12..a834492c03 100644
--- a/Libraries/LibUnicode/Segmenter.cpp
+++ b/Libraries/LibUnicode/Segmenter.cpp
@@ -41,6 +41,92 @@ StringView segmenter_granularity_to_string(SegmenterGranularity segmenter_granul
     VERIFY_NOT_REACHED();
 }
 
+// Fast path segmenter for ASCII text where every character is its own grapheme.
+// This avoids all ICU overhead for the common case of ASCII-only text.
+class AsciiGraphemeSegmenter : public Segmenter {
+public:
+    explicit AsciiGraphemeSegmenter(size_t length)
+        : Segmenter(SegmenterGranularity::Grapheme)
+        , m_length(length)
+    {
+    }
+
+    virtual ~AsciiGraphemeSegmenter() override = default;
+
+    virtual NonnullOwnPtr<Segmenter> clone() const override
+    {
+        return make<AsciiGraphemeSegmenter>(m_length);
+    }
+
+    virtual void set_segmented_text(String text) override
+    {
+        m_length = text.byte_count();
+    }
+
+    virtual void set_segmented_text(Utf16View const& text) override
+    {
+        m_length = text.length_in_code_units();
+    }
+
+    virtual size_t current_boundary() override
+    {
+        return m_current;
+    }
+
+    virtual Optional<size_t> previous_boundary(size_t index, Inclusive inclusive) override
+    {
+        if (inclusive == Inclusive::Yes && index <= m_length)
+            return index;
+        if (index == 0)
+            return {};
+        return index - 1;
+    }
+
+    virtual Optional<size_t> next_boundary(size_t index, Inclusive inclusive) override
+    {
+        if (inclusive == Inclusive::Yes && index <= m_length)
+            return index;
+        if (index >= m_length)
+            return {};
+        return index + 1;
+    }
+
+    virtual void for_each_boundary(String text, SegmentationCallback callback) override
+    {
+        set_segmented_text(move(text));
+        for_each_boundary_impl(callback);
+    }
+
+    virtual void for_each_boundary(Utf16View const& text, SegmentationCallback callback) override
+    {
+        set_segmented_text(text);
+        for_each_boundary_impl(callback);
+    }
+
+    virtual void for_each_boundary(Utf32View const& text, SegmentationCallback callback) override
+    {
+        m_length = text.length();
+        for_each_boundary_impl(callback);
+    }
+
+    virtual bool is_current_boundary_word_like() const override
+    {
+        return false;
+    }
+
+private:
+    void for_each_boundary_impl(SegmentationCallback& callback)
+    {
+        for (size_t i = 0; i <= m_length; ++i) {
+            if (callback(i) == IterationDecision::Break)
+                return;
+        }
+    }
+
+    size_t m_length { 0 };
+    size_t m_current { 0 };
+};
+
 class SegmenterImpl : public Segmenter {
 public:
     SegmenterImpl(NonnullOwnPtr<icu::BreakIterator> segmenter, SegmenterGranularity segmenter_granularity)
@@ -247,6 +333,11 @@ NonnullOwnPtr<Segmenter> Segmenter::create(StringView locale, SegmenterGranulari
     return make<SegmenterImpl>(segmenter.release_nonnull(), segmenter_granularity);
 }
 
+NonnullOwnPtr<Segmenter> Segmenter::create_for_ascii_grapheme(size_t length)
+{
+    return make<AsciiGraphemeSegmenter>(length);
+}
+
 bool Segmenter::should_continue_beyond_word(Utf16View const& word)
 {
     for (auto code_point : word) {
diff --git a/Libraries/LibUnicode/Segmenter.h b/Libraries/LibUnicode/Segmenter.h
index 6acf13bd48..a80c0d2e72 100644
--- a/Libraries/LibUnicode/Segmenter.h
+++ b/Libraries/LibUnicode/Segmenter.h
@@ -25,6 +25,7 @@ class Segmenter {
 public:
     static NonnullOwnPtr<Segmenter> create(SegmenterGranularity segmenter_granularity);
     static NonnullOwnPtr<Segmenter> create(StringView locale, SegmenterGranularity segmenter_granularity);
+    static NonnullOwnPtr<Segmenter> create_for_ascii_grapheme(size_t length);
     virtual ~Segmenter() = default;
 
     static bool should_continue_beyond_word(Utf16View const&);
diff --git a/Libraries/LibWeb/Layout/TextNode.cpp b/Libraries/LibWeb/Layout/TextNode.cpp
index cdbd6ee20e..92559c27c4 100644
--- a/Libraries/LibWeb/Layout/TextNode.cpp
+++ b/Libraries/LibWeb/Layout/TextNode.cpp
@@ -396,8 +396,15 @@ void TextNode::compute_text_for_rendering()
 Unicode::Segmenter& TextNode::grapheme_segmenter() const
 {
     if (!m_grapheme_segmenter) {
-        m_grapheme_segmenter = document().grapheme_segmenter().clone();
-        m_grapheme_segmenter->set_segmented_text(text_for_rendering());
+        auto const& text = text_for_rendering();
+        // Fast path: For ASCII text, every character is its own grapheme.
+        // We can use a trivial segmenter that avoids all ICU overhead.
+        if (text.is_ascii()) {
+            m_grapheme_segmenter = Unicode::Segmenter::create_for_ascii_grapheme(text.length_in_code_units());
+        } else {
+            m_grapheme_segmenter = document().grapheme_segmenter().clone();
+            m_grapheme_segmenter->set_segmented_text(text);
+        }
     }
 
     return *m_grapheme_segmenter;