0ct0pu5/ladybird
1
0
Fork 0
mirror of https://github.com/LadybirdBrowser/ladybird.git synced 2024-12-04 05:20:30 +00:00
Code Issues Projects Releases Packages Wiki Activity Actions 7

LibTextCodec: Return Optional<Decoder&> from bom_sniff_to_decoder()

Browse source
This commit is contained in:
Sam Atkins 2023-02-17 19:53:51 +00:00 committed by Andreas Kling
parent f2a9426885
commit 3c5090e172
Notes: sideshowbarker 2024-07-17 00:01:03 +09:00 
Author: https://github.com/AtkinsSJ
Commit: https://github.com/SerenityOS/serenity/commit/3c5090e172
Pull-request: https://github.com/SerenityOS/serenity/pull/17511
Reviewed-by: https://github.com/ADKaster
Reviewed-by: https://github.com/nico
2 changed files with 18 additions and 12 deletions
Show stats Download patch file Download diff file Expand all files Collapse all files

26
Userland/Libraries/LibTextCodec/Decoder.cpp
View file

@ -157,7 +157,7 @@ Optional<StringView> get_standardized_encoding(StringView encoding)
}
// https://encoding.spec.whatwg.org/#bom-sniff
Decoder* bom_sniff_to_decoder(StringView input)
Optional<Decoder&> bom_sniff_to_decoder(StringView input)
{
// 1. Let BOM be the result of peeking 3 bytes from ioQueue, converted to a byte sequence.
// 2. For each of the rows in the table below, starting with the first one and going down,
@ -172,22 +172,28 @@ Decoder* bom_sniff_to_decoder(StringView input)
auto bytes = input.bytes();
if (bytes.size() < 2)
return nullptr;
return {};
auto first_byte = bytes[0];
switch (first_byte) {
case 0xEF: // UTF-8
if (bytes.size() < 3)
return nullptr;
return bytes[1] == 0xBB && bytes[2] == 0xBF ? &s_utf8_decoder : nullptr;
return {};
if (bytes[1] == 0xBB && bytes[2] == 0xBF)
return s_utf8_decoder;
return {};
case 0xFE: // UTF-16BE
return bytes[1] == 0xFF ? &s_utf16be_decoder : nullptr;
if (bytes[1] == 0xFF)
return s_utf16be_decoder;
return {};
case 0xFF: // UTF-16LE
return bytes[1] == 0xFE ? &s_utf16le_decoder : nullptr;
if (bytes[1] == 0xFE)
return s_utf16le_decoder;
return {};
}
return nullptr;
return {};
}
// https://encoding.spec.whatwg.org/#decode
@ -197,13 +203,13 @@ DeprecatedString convert_input_to_utf8_using_given_decoder_unless_there_is_a_byt
// 1. Let BOMEncoding be the result of BOM sniffing ioQueue.
// 2. If BOMEncoding is non-null:
if (auto* unicode_decoder = bom_sniff_to_decoder(input); unicode_decoder) {
if (auto unicode_decoder = bom_sniff_to_decoder(input); unicode_decoder.has_value()) {
// 1. Set encoding to BOMEncoding.
actual_decoder = unicode_decoder;
actual_decoder = &unicode_decoder.value();
// 2. Read three bytes from ioQueue, if BOMEncoding is UTF-8; otherwise read two bytes. (Do nothing with those bytes.)
// FIXME: I imagine this will be pretty slow for large inputs, as it's regenerating the input without the first 2/3 bytes.
input = input.substring_view(unicode_decoder == &s_utf8_decoder ? 3 : 2);
input = input.substring_view(&unicode_decoder.value() == &s_utf8_decoder ? 3 : 2);
}
VERIFY(actual_decoder);

4
Userland/Libraries/LibTextCodec/Decoder.h
View file

@ -89,8 +89,8 @@ public:
Optional<Decoder&> decoder_for(StringView encoding);
Optional<StringView> get_standardized_encoding(StringView encoding);
// This returns the appropriate Unicode decoder for the sniffed BOM or nullptr if there is no appropriate decoder.
Decoder* bom_sniff_to_decoder(StringView);
// This returns the appropriate Unicode decoder for the sniffed BOM or nothing if there is no appropriate decoder.
Optional<Decoder&> bom_sniff_to_decoder(StringView);
// NOTE: This has an obnoxious name to discourage usage. Only use this if you absolutely must! For example, XHR in LibWeb uses this.
// This will use the given decoder unless there is a byte order mark in the input, in which we will instead use the appropriate Unicode decoder.