RGBCube/serenity
1
Fork 0
mirror of https://github.com/RGBCube/serenity synced 2025-05-31 13:38:11 +00:00
Code Issues Activity Actions

LibWeb: Correct logic when consuming a CSS number in scientific notation

Browse source 
Before, we were classifying the number as a "number" type if it had an
"E", even if that was not followed by an exponent.
This commit is contained in:
Sam Atkins 2023-03-20 17:31:36 +00:00 • committed by Andreas Kling
parent a8cd6c396b
commit a3d6d9db37
1 changed files with 3 additions and 2 deletions
Download patch file Download diff file Expand all files Collapse all files

5
Userland/Libraries/LibWeb/CSS/Parser/Tokenizer.cpp
View file

@ -534,10 +534,11 @@ Number Tokenizer::consume_a_number()
// U+0065 LATIN SMALL LETTER E (e), optionally followed by U+002D HYPHEN-MINUS (-)
// or U+002B PLUS SIGN (+), followed by a digit, then:
auto maybe_exp = peek_triplet();
if (is_E(maybe_exp.first) || is_e(maybe_exp.first)) {
if ((is_E(maybe_exp.first) || is_e(maybe_exp.first))
&& (((is_plus_sign(maybe_exp.second) || is_hyphen_minus(maybe_exp.second)) && is_ascii_digit(maybe_exp.third))
|| (is_ascii_digit(maybe_exp.second)))) {
// 1. Consume them.
// 2. Append them to repr.
// FIXME: These conditions should be part of step 5 above.
if (is_plus_sign(maybe_exp.second) || is_hyphen_minus(maybe_exp.second)) {
if (is_ascii_digit(maybe_exp.third)) {
repr.append_code_point(next_code_point());