From 3c2565da9427aa245d23c7e7a9ebfeec59f02cf7 Mon Sep 17 00:00:00 2001
From: Max Wipfli <mail@maxwipfli.ch>
Date: Thu, 20 May 2021 12:56:38 +0200
Subject: [PATCH] AK: Add UnicodeUtils with Unicode-related helper functions

This introduces the UnicodeUtils file, which contains helper functions
related to Unicode. This is in contrast to StringUtils, whose functions
are not directly related to Unicode and are, in theory,
encoding-agnostic.
---
 AK/UnicodeUtils.cpp | 37 +++++++++++++++++++++++++++++++++++++
 AK/UnicodeUtils.h   | 20 ++++++++++++++++++++
 2 files changed, 57 insertions(+)
 create mode 100644 AK/UnicodeUtils.cpp
 create mode 100644 AK/UnicodeUtils.h

diff --git a/AK/UnicodeUtils.cpp b/AK/UnicodeUtils.cpp
new file mode 100644
index 0000000000..2db467065c
--- /dev/null
+++ b/AK/UnicodeUtils.cpp
@@ -0,0 +1,37 @@
+/*
+ * Copyright (c) 2021, Max Wipfli <mail@maxwipfli.ch>
+ *
+ * SPDX-License-Identifier: BSD-2-Clause
+ */
+
+#include <AK/Array.h>
+#include <AK/Optional.h>
+#include <AK/StringView.h>
+#include <AK/UnicodeUtils.h>
+
+namespace AK::UnicodeUtils {
+
+Optional<StringView> get_unicode_control_code_point_alias(u32 code_point)
+{
+    static constexpr Array<StringView, 32> ascii_controls_lookup_table = {
+        "NUL", "SOH", "STX", "ETX", "EOT", "ENQ", "ACK", "BEL",
+        "BS", "HT", "LF", "VT", "FF", "CR", "SO", "SI",
+        "DLE", "DC1", "DC2", "DC3", "DC4", "NAK", "SYN", "ETB",
+        "CAN", "EM", "SUB", "ESC", "FS", "GS", "RS", "US"
+    };
+
+    static constexpr Array<StringView, 32> c1_controls_lookup_table = {
+        "XXX", "XXX", "BPH", "NBH", "IND", "NEL", "SSA", "ESA",
+        "HTS", "HTJ", "VTS", "PLD", "PLU", "RI", "SS2", "SS3",
+        "DCS", "PU1", "PU2", "STS", "CCH", "MW", "SPA", "EPA",
+        "SOS", "XXX", "SCI", "CSI", "ST", "OSC", "PM", "APC"
+    };
+
+    if (code_point < 0x20)
+        return ascii_controls_lookup_table[code_point];
+    if (code_point >= 0x80 && code_point < 0xa0)
+        return c1_controls_lookup_table[code_point - 0x80];
+    return {};
+}
+
+}
diff --git a/AK/UnicodeUtils.h b/AK/UnicodeUtils.h
new file mode 100644
index 0000000000..e7211deaea
--- /dev/null
+++ b/AK/UnicodeUtils.h
@@ -0,0 +1,20 @@
+/*
+ * Copyright (c) 2021, Max Wipfli <mail@maxwipfli.ch>
+ *
+ * SPDX-License-Identifier: BSD-2-Clause
+ */
+
+#pragma once
+
+#include <AK/Forward.h>
+
+namespace AK::UnicodeUtils {
+
+constexpr bool is_unicode_control_code_point(u32 code_point)
+{
+    return code_point < 0x20 || (code_point >= 0x80 && code_point < 0xa0);
+}
+
+Optional<StringView> get_unicode_control_code_point_alias(u32);
+
+}