From f1b79e0cd30b9f55275eeb73ac766433f98ef227 Mon Sep 17 00:00:00 2001
From: Gurkirat Singh <tbhaxor@gmail.com>
Date: Wed, 27 Sep 2023 22:41:57 +0530
Subject: [PATCH] AK: Implement `slugify` function for URL slug generation

The slugify function is used to convert input into URL-friendly slugs.
It processes each character in the input, keeping ascii alpha characters
after lowercase and replacing non-alphanum characters with the glue
character or a space if multiple spaces are encountered consecutively.
The resulting string is trimmed of leading and trailing whitespace, and
any internal whitespace is replaced with the glue character.

It is currently used in LibMarkdown headings generation code.
---
 AK/CMakeLists.txt        |  1 +
 AK/Slugify.cpp           | 33 ++++++++++++++++++++++++++++++
 AK/Slugify.h             | 17 ++++++++++++++++
 Tests/AK/CMakeLists.txt  |  1 +
 Tests/AK/TestSlugify.cpp | 43 ++++++++++++++++++++++++++++++++++++++++
 5 files changed, 95 insertions(+)
 create mode 100644 AK/Slugify.cpp
 create mode 100644 AK/Slugify.h
 create mode 100644 Tests/AK/TestSlugify.cpp

diff --git a/AK/CMakeLists.txt b/AK/CMakeLists.txt
index c5eb4c37e5..366b1d271f 100644
--- a/AK/CMakeLists.txt
+++ b/AK/CMakeLists.txt
@@ -24,6 +24,7 @@ set(AK_SOURCES
     OptionParser.cpp
     Random.cpp
     SipHash.cpp
+    Slugify.cpp
     StackInfo.cpp
     Stream.cpp
     String.cpp
diff --git a/AK/Slugify.cpp b/AK/Slugify.cpp
new file mode 100644
index 0000000000..51feb5ef78
--- /dev/null
+++ b/AK/Slugify.cpp
@@ -0,0 +1,33 @@
+/*
+ * Copyright (c) 2023, Gurkirat Singh <tbhaxor@gmail.com>
+ *
+ * SPDX-License-Identifier: BSD-2-Clause
+ */
+
+#include <AK/CharacterTypes.h>
+#include <AK/Slugify.h>
+#include <AK/StringView.h>
+
+namespace AK {
+ErrorOr<String> slugify(String const& input, char const glue)
+{
+    StringBuilder sb;
+    bool just_processed_space = false;
+
+    for (auto const& code_point : input.code_points()) {
+        if (is_ascii_alphanumeric(code_point)) {
+            sb.append_code_point(to_ascii_lowercase(code_point));
+            just_processed_space = false;
+        } else if ((code_point == static_cast<u32>(glue) || is_ascii_space(code_point)) && !just_processed_space) {
+            sb.append_code_point(glue);
+            just_processed_space = true;
+        }
+    }
+
+    auto output = TRY(sb.to_string());
+    if (output.ends_with(static_cast<u32>(glue))) {
+        return output.trim(StringView { &glue, 1 }, TrimMode::Right);
+    }
+    return output;
+}
+}
diff --git a/AK/Slugify.h b/AK/Slugify.h
new file mode 100644
index 0000000000..477f09a9a1
--- /dev/null
+++ b/AK/Slugify.h
@@ -0,0 +1,17 @@
+/*
+ * Copyright (c) 2023, Gurkirat Singh <tbhaxor@gmail.com>
+ *
+ * SPDX-License-Identifier: BSD-2-Clause
+ */
+
+#pragma once
+
+#include <AK/String.h>
+
+namespace AK {
+ErrorOr<String> slugify(String const& input, char glue = '-');
+}
+
+#if USING_AK_GLOBALLY
+using AK::slugify;
+#endif
diff --git a/Tests/AK/CMakeLists.txt b/Tests/AK/CMakeLists.txt
index 177fd6cf7b..02dcc4ebb9 100644
--- a/Tests/AK/CMakeLists.txt
+++ b/Tests/AK/CMakeLists.txt
@@ -65,6 +65,7 @@ set(AK_TEST_SOURCES
     TestRefPtr.cpp
     TestSIMD.cpp
     TestSinglyLinkedList.cpp
+    TestSlugify.cpp
     TestSourceGenerator.cpp
     TestSourceLocation.cpp
     TestSpan.cpp
diff --git a/Tests/AK/TestSlugify.cpp b/Tests/AK/TestSlugify.cpp
new file mode 100644
index 0000000000..afa8db73ee
--- /dev/null
+++ b/Tests/AK/TestSlugify.cpp
@@ -0,0 +1,43 @@
+/*
+ * Copyright (c) 2023, Gurkirat Singh <tbhaxor@gmail.com>
+ *
+ * SPDX-License-Identifier: BSD-2-Clause
+ */
+
+#include <AK/Slugify.h>
+#include <LibTest/TestCase.h>
+
+TEST_CASE(ignore_unicode_characters)
+{
+    EXPECT_EQ(MUST(slugify("Hello World!🎉"_string)), "hello-world"_string);
+}
+
+TEST_CASE(all_whitespace_empty_string)
+{
+    EXPECT_EQ(MUST(slugify("  "_string)), ""_string);
+}
+
+TEST_CASE(squeeze_multiple_whitespace)
+{
+    EXPECT_EQ(MUST(slugify("Hello   World"_string)), "hello-world"_string);
+}
+
+TEST_CASE(trim_trailing_whitelist)
+{
+    EXPECT_EQ(MUST(slugify("Hello   World    "_string)), "hello-world"_string);
+}
+
+TEST_CASE(lowercase_all_result)
+{
+    EXPECT_EQ(MUST(slugify("HelloWorld"_string)), "helloworld"_string);
+}
+
+TEST_CASE(slug_glue_change)
+{
+    EXPECT_EQ(MUST(slugify("Hello World"_string, '|')), "hello|world"_string);
+}
+
+TEST_CASE(multiple_glue_squeeze)
+{
+    EXPECT_EQ(MUST(slugify("Hello_ World"_string, '_')), "hello_world"_string);
+}