AK: Add a Utf8View type for iterating over UTF-8 codepoints

Utf8View wraps a StringView and implements begin() and end() that
return a Utf8CodepointIterator, which parses UTF-8-encoded Unicode
codepoints and returns them as 32-bit integers.

This is the first step towards supporting emojis in Serenity ^)
https://github.com/SerenityOS/serenity/issues/490

AK/Utf8View.cpp

@@ -0,0 +1,130 @@
+#include <AK/Utf8View.h>
+
+namespace AK {
+
+Utf8View::Utf8View(const StringView& string)
+    : m_string(string)
+{
+}
+
+const unsigned char* Utf8View::begin_ptr() const
+{
+    return (const unsigned char*)m_string.characters_without_null_termination();
+}
+
+const unsigned char* Utf8View::end_ptr() const
+{
+    return (const unsigned char*)m_string.characters_without_null_termination() + m_string.length();
+}
+
+Utf8CodepointIterator Utf8View::begin() const
+{
+    return { begin_ptr(), m_string.length() };
+}
+
+Utf8CodepointIterator Utf8View::end() const
+{
+    return { end_ptr(), 0 };
+}
+
+static inline bool decode_first_byte(
+    unsigned char byte,
+    int& out_codepoint_length_in_bytes,
+    u32& out_value)
+{
+    if ((byte & 128) == 0) {
+        out_value = byte;
+        out_codepoint_length_in_bytes = 1;
+        return true;
+    }
+    if ((byte & 64) == 0) {
+        return false;
+    }
+    if ((byte & 32) == 0) {
+        out_value = byte & 31;
+        out_codepoint_length_in_bytes = 2;
+        return true;
+    }
+    if ((byte & 16) == 0) {
+        out_value = byte & 15;
+        out_codepoint_length_in_bytes = 3;
+        return true;
+    }
+    if ((byte & 8) == 0) {
+        out_value = byte & 7;
+        out_codepoint_length_in_bytes = 4;
+        return true;
+    }
+
+    return false;
+}
+
+bool Utf8View::validate() const
+{
+    for (auto ptr = begin_ptr(); ptr < end_ptr(); ptr++) {
+        int codepoint_length_in_bytes;
+        u32 value;
+        bool first_byte_makes_sense = decode_first_byte(*ptr, codepoint_length_in_bytes, value);
+        if (!first_byte_makes_sense)
+            return false;
+
+        for (int i = 1; i < codepoint_length_in_bytes; i++) {
+            ptr++;
+            if (ptr >= end_ptr())
+                return false;
+            if (*ptr >> 6 != 2)
+                return false;
+        }
+    }
+
+    return true;
+}
+
+Utf8CodepointIterator::Utf8CodepointIterator(const unsigned char* ptr, int length)
+    : m_ptr(ptr)
+    , m_length(length)
+{
+}
+
+bool Utf8CodepointIterator::operator==(const Utf8CodepointIterator& other) const
+{
+    return m_ptr == other.m_ptr && m_length == other.m_length;
+}
+
+bool Utf8CodepointIterator::operator!=(const Utf8CodepointIterator& other) const
+{
+    return !(*this == other);
+}
+
+Utf8CodepointIterator& Utf8CodepointIterator::operator++()
+{
+    do {
+        ASSERT(m_length > 0);
+        m_length--;
+        m_ptr++;
+    } while (m_ptr[0] >> 6 == 2);
+
+    return *this;
+}
+
+u32 Utf8CodepointIterator::operator*() const
+{
+    ASSERT(m_length > 0);
+
+    u32 codepoint_value_so_far;
+    int codepoint_length_in_bytes;
+
+    bool first_byte_makes_sense = decode_first_byte(m_ptr[0], codepoint_length_in_bytes, codepoint_value_so_far);
+    ASSERT(first_byte_makes_sense);
+    ASSERT(codepoint_length_in_bytes <= m_length);
+
+    for (int offset = 1; offset < codepoint_length_in_bytes; offset++) {
+        ASSERT(m_ptr[offset] >> 6 == 2);
+        codepoint_value_so_far <<= 6;
+        codepoint_value_so_far |= m_ptr[offset] & 63;
+    }
+
+    return codepoint_value_so_far;
+}
+
+}