mirror of
https://github.com/SerenityOS/serenity
synced 2024-10-02 22:24:26 +00:00
5d3696174b
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
49 lines
921 B
C++
49 lines
921 B
C++
#pragma once
|
|
|
|
#include <AK/StringView.h>
|
|
#include <AK/Types.h>
|
|
|
|
namespace AK {
|
|
|
|
class Utf8View;
|
|
|
|
class Utf8CodepointIterator {
|
|
friend class Utf8View;
|
|
|
|
public:
|
|
~Utf8CodepointIterator() {}
|
|
|
|
bool operator==(const Utf8CodepointIterator&) const;
|
|
bool operator!=(const Utf8CodepointIterator&) const;
|
|
Utf8CodepointIterator& operator++();
|
|
u32 operator*() const;
|
|
|
|
private:
|
|
Utf8CodepointIterator(const unsigned char*, int);
|
|
const unsigned char* m_ptr { nullptr };
|
|
int m_length { -1 };
|
|
};
|
|
|
|
class Utf8View {
|
|
public:
|
|
explicit Utf8View(const StringView&);
|
|
~Utf8View() {}
|
|
|
|
const StringView& as_string() const { return m_string; }
|
|
|
|
Utf8CodepointIterator begin() const;
|
|
Utf8CodepointIterator end() const;
|
|
|
|
bool validate() const;
|
|
|
|
private:
|
|
const unsigned char* begin_ptr() const;
|
|
const unsigned char* end_ptr() const;
|
|
|
|
StringView m_string;
|
|
};
|
|
|
|
}
|
|
|
|
using AK::Utf8View;
|