Re: HTTPS RR side of things
Daniel Stenberg via curl-library <[email protected]>
| Newsgroups | gmane.comp.web.curl.library |
|---|---|
| Message-ID | <[email protected]> |
On Thu, 20 Feb 2025, Niall O'Reilly wrote: >> Goal: design data structure and related APIs for making HTTPS/SVCB data, as >> well as what getaddrinfo() provides, available, as simply as possible, to >> an application. Additionally: since getaddrinfo() is such a beast to replace reliably, curl will continue using that in addition to resolve additional records "on the side" using other means (c-ares in our case) [1]. This will of course ocasionally introduce some undesired side-effects down the line when they don't agree. That's effectively also the approach Firefox uses [2], so it's not only because we're a bunch of lazy incompetent developers here. :-) [1] = I mean in addition to a "pure" c-ares build that uses c-ares for everything. The problem is that the c-ares getaddrinfo implementation is not a 100% drop-in replacement for getaddrinfo(), only like 99.7% or something. [2] = they don't use c-ares though, and I've been told that they *are* going to attempt to do getaddrinfo() replacement but that's not what is used now. -- / daniel.haxx.se || https://rock-solid.curl.dev -- Unsubscribe: https://lists.haxx.se/mailman/listinfo/curl-library Etiquette: https://curl.se/mail/etiquette.html