mirrors/ladybird
0
0
Fork 0
mirror of https://github.com/LadybirdBrowser/ladybird.git synced 2025-07-30 20:59:16 +00:00
Code Issues Projects Releases Packages Wiki Activity Actions 9

LibTextCodec: Implement euc-kr encoder

Browse source 
Implements the `euc-kr` encoder, as specified by
https://encoding.spec.whatwg.org/#euc-kr-encoder
This commit is contained in:
BenJilks 2024-08-05 20:00:47 +01:00 • committed by Tim Ledbetter
commit 826292536c
Notes: github-actions[bot] 2024-08-08 16:51:31 +00:00
4 changed files with 62 additions and 1 deletions
Download patch file Download diff file Expand all files Collapse all files

38
Userland/Libraries/LibTextCodec/Encoder.cpp
View file

@ -15,6 +15,7 @@ namespace TextCodec {
namespace {
UTF8Encoder s_utf8_encoder;
EUCJPEncoder s_euc_jp_encoder;
EUCKREncoder s_euc_kr_encoder;
}
Optional<Encoder&> encoder_for_exact_name(StringView encoding)
@ -23,6 +24,8 @@ Optional<Encoder&> encoder_for_exact_name(StringView encoding)
return s_utf8_encoder;
if (encoding.equals_ignoring_ascii_case("euc-jp"sv))
return s_euc_jp_encoder;
if (encoding.equals_ignoring_ascii_case("euc-kr"sv))
return s_euc_kr_encoder;
dbgln("TextCodec: No encoder implemented for encoding '{}'", encoding);
return {};
}
@ -100,4 +103,39 @@ ErrorOr<void> EUCJPEncoder::process(Utf8View input, Function<ErrorOr<void>(u8)>
return {};
}
// https://encoding.spec.whatwg.org/#euc-kr-encoder
ErrorOr<void> EUCKREncoder::process(Utf8View input, Function<ErrorOr<void>(u8)> on_byte)
{
for (u32 item : input) {
// 1. If code point is end-of-queue, return finished.
// 2. If code point is an ASCII code point, return a byte whose value is code point.
if (is_ascii(item)) {
TRY(on_byte(static_cast<u8>(item)));
continue;
}
// 3. Let pointer be the index pointer for code point in index EUC-KR.
auto pointer = code_point_euc_kr_index(item);
// 4. If pointer is null, return error with code point.
if (!pointer.has_value()) {
// TODO: Report error.
continue;
}
// 5. Let lead be pointer / 190 + 0x81.
auto lead = *pointer / 190 + 0x81;
// 6. Let trail be pointer % 190 + 0x41.
auto trail = *pointer % 190 + 0x41;
// 7. Return two bytes whose values are lead and trail.
TRY(on_byte(static_cast<u8>(lead)));
TRY(on_byte(static_cast<u8>(trail)));
}
return {};
}
}