mirror of
https://github.com/RGBCube/serenity
synced 2025-07-26 23:37:36 +00:00
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
This commit is contained in:
parent
970e0147f7
commit
5d3696174b
4 changed files with 241 additions and 1 deletions
48
AK/Utf8View.h
Normal file
48
AK/Utf8View.h
Normal file
|
@ -0,0 +1,48 @@
|
|||
#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;
|
Loading…
Add table
Add a link
Reference in a new issue