RGBCube/serenity
1
Fork 0
mirror of https://github.com/RGBCube/serenity synced 2025-07-25 06:37:43 +00:00
Code Issues Activity Actions

LibJS: Implement String.prototype.replace with UTF-16 code units

Browse source
This commit is contained in:
Timothy Flynn 2021-07-19 16:53:35 -04:00 • committed by Andreas Kling
parent bdbe716547
commit 06208aaa15
3 changed files with 64 additions and 23 deletions
Download patch file Download diff file Expand all files Collapse all files

50
Userland/Libraries/LibJS/Runtime/AbstractOperations.cpp
View file

@ -10,6 +10,7 @@
#include <AK/Optional.h>
#include <AK/Result.h>
#include <AK/TemporaryChange.h>
#include <AK/Utf16View.h>
#include <LibJS/Interpreter.h>
#include <LibJS/Parser.h>
#include <LibJS/Runtime/AbstractOperations.h>
@ -583,36 +584,51 @@ String get_substitution(GlobalObject& global_object, String const& matched, Stri
if (vm.exception())
return {};
// FIXME: Once RegExp.prototype supports UTF-16, this AO can take UTF-16 strings as parameters instead of having to transcode here.
auto utf16_matched = AK::utf8_to_utf16(matched);
auto match_length = utf16_matched.size();
auto utf16_string = AK::utf8_to_utf16(str);
Utf16View utf16_string_view { utf16_string };
auto string_length = utf16_string_view.length_in_code_units();
auto utf16_replace = AK::utf8_to_utf16(replace_string);
Utf16View utf16_replace_view { utf16_replace };
auto replace_length = utf16_replace_view.length_in_code_units();
StringBuilder result;
for (size_t i = 0; i < replace_string.length(); ++i) {
char curr = replace_string[i];
for (size_t i = 0; i < replace_length; ++i) {
u16 curr = utf16_replace_view.code_unit_at(i);
if ((curr != '$') || (i + 1 >= replace_string.length())) {
if ((curr != '$') || (i + 1 >= replace_length)) {
result.append(curr);
continue;
}
char next = replace_string[i + 1];
u16 next = utf16_replace_view.code_unit_at(i + 1);
if (next == '$') {
result.append(next);
result.append('$');
++i;
} else if (next == '&') {
result.append(matched);
++i;
} else if (next == '`') {
result.append(str.substring_view(0, position));
auto substring = utf16_string_view.substring_view(0, position);
result.append(substring.to_utf8(Utf16View::AllowInvalidCodeUnits::Yes));
++i;
} else if (next == '\'') {
auto tail_pos = position + matched.length();
if (tail_pos < str.length())
result.append(str.substring_view(tail_pos));
auto tail_pos = position + match_length;
if (tail_pos < string_length) {
auto substring = utf16_string_view.substring_view(tail_pos);
result.append(substring.to_utf8(Utf16View::AllowInvalidCodeUnits::Yes));
}
++i;
} else if (is_ascii_digit(next)) {
bool is_two_digits = (i + 2 < replace_string.length()) && is_ascii_digit(replace_string[i + 2]);
bool is_two_digits = (i + 2 < replace_length) && is_ascii_digit(utf16_replace_view.code_unit_at(i + 2));
auto capture_postition_string = replace_string.substring_view(i + 1, is_two_digits ? 2 : 1);
auto capture_postition_string = utf16_replace_view.substring_view(i + 1, is_two_digits ? 2 : 1).to_utf8();
auto capture_position = capture_postition_string.to_uint();
if (capture_position.has_value() && (*capture_position > 0) && (*capture_position <= captures.size())) {
@ -632,12 +648,20 @@ String get_substitution(GlobalObject& global_object, String const& matched, Stri
}
} else if (next == '<') {
auto start_position = i + 2;
auto end_position = replace_string.find('>', start_position);
Optional<size_t> end_position;
for (size_t j = start_position; j < replace_length; ++j) {
if (utf16_replace_view.code_unit_at(j) == '>') {
end_position = j;
break;
}
}
if (named_captures.is_undefined() || !end_position.has_value()) {
result.append(curr);
} else {
auto group_name = replace_string.substring(start_position, *end_position - start_position);
auto group_name_view = utf16_replace_view.substring_view(start_position, *end_position - start_position);
auto group_name = group_name_view.to_utf8(Utf16View::AllowInvalidCodeUnits::Yes);
auto capture = named_captures.as_object().get(group_name);
if (vm.exception())