implement URLPattern

[jimmywarting]

Don't know if you have heard of this new URLPattern
Deno shipped it, and browser supports it, would be a cool util to have globally avalible to pear up with web standards and for usage in express routing for instance

[DerekNonGeneric]

Thanks for the heads up — obviously our feature request template is not very well-suited for these kinds of issues, but given that the specification for URLPattern is still in draft, I imagine that we would be setting the Webcompat Priority value of this to either ? or revisit if we were to use a triage system similar to what Mozilla uses for their WebCompat Tracking and Triage workflow.

Mainly being a volunteer-run organization, we have no real priorities (whatever gets done, gets done); that being said, this issue cannot be represented by one of the N values in that system…

I would say to perhaps set it as revisit and come back later once spec becomes a recommendation.

[targos]

Spec is at https://github.com/WICG/urlpattern
WPT: https://github.com/web-platform-tests/wpt/tree/master/urlpattern

[targos]

@nodejs/url what do you think?

[Jamesernator]

Deno shipped it, and browser supports it, would be a cool util to have globally avalible to pear up with web standards and for usage in express routing for instance

So I opened the original issue for support in Deno, I opened it on Deno as Deno tends to support considerably more cutting edge things.

Having said this, having it Node would be particularly valuable for libraries that are currently using path-to-regexp for patterns. The syntax is essentially the same (aside from a few bug fixes to work with URLs more generally), so libraries that are currently using path-to-regexp could use the builtin URLPattern instead, allowing them to remove a external dependency.

And having a common pattern syntax for things like routing would just generally be nice to have, so that libraries don't have a bunch of different microsyntaxes for essentially the same thing.

[jasnell]

@targos ... I'm +1 on an having an implementation in core. I have some concerns about the performance as spec'd but having an implementation makes sense.

[github-actions]

There has been no activity on this feature request for 5 months and it is unlikely to be implemented. It will be closed 6 months after the last non-automated comment.

For more information on how the project manages feature requests, please consult the feature request management document.