davidot

@AtkinsSJ You might know whether I have used proper whitespace trimming and implementation for the LibWeb new double parser usage Also if you happen to know more places where we...

> I'm still a bit iffy on replacing individual spec-based implementations with a single one. I get that, are you more worried about the supported format or the actual conversion...

 These are the new `strtod` and `strtof` performance numbers, so no regression for common cases is visible anymore and the long cases are even faster!

Sorry for the many names :sweat_smile: I made `to_double` and `to_float` on strings non templated, hopefully making it more clear. And then StringUtils does have a templated version in case...

CI was flaky, just rebased to rerun it

> I dont know where to put it, so I threw it right into the home directory! There is a Tests directory in the home folder, perhaps a slightly nicer...

> Looks like can be closed now. If you're talking about #15377 being merged indeed this only fixed parsing doubles not to_string the opposite direction as Idan shows

Also in case you're interested there are even newer algorithms like dragonbox: https://github.com/jk-jeon/dragonbox which is used by libfmt

> Well, I hadn't seen it when I was doing the initial research. Anyway, I believe that what is currently implemented would already be a significant improvement over a for...

I don't see any way in the ECMA spec to call evaluate on the same module twice. But this is a real issue for a browser. From the HTML spec:...