I suppose this is as good a time as any to ask: why do single-core IOReactors support the Exception callback (for error logging), but not multi-core IOReactors? Also, how should this callback be configured when creating a client? (I wasn't able to immediately find the relevant config method to call.)
On Wed, Jan 1, 2020 at 11:46 AM Oleg Kalnichevski <[email protected]> wrote: > On Wed, 2020-01-01 at 11:25 -0800, Ryan Schmitt wrote: > > This is still a beta release, right? My understanding is that we're > > going > > to treat it as a release candidate and try to avoid breaking changes, > > but > > we can still make them if necessary (e.g. to fix some show-stopping > > defect). > > > > That goes without saying. But changes driven largely by a subjective > sense of aesthetics ought to come with a migration path. > > Oleg > > > > On Wed, Jan 1, 2020 at 11:08 AM Oleg Kalnichevski <[email protected]> > > wrote: > > > > > On Wed, 2020-01-01 at 14:00 -0500, Gary Gregory wrote: > > > > I think b11 can go ahead. > > > > > > > > Gary > > > > > > > > > Folks > > > > > > That effectively means no more renaming and moving things around > > > without deprecation and a migration path. > > > > > > I will proceed with the release later today or tomorrow. > > > > > > Oleg > > > > > > > > > > > On Wed, Jan 1, 2020, 12:33 Oleg Kalnichevski <[email protected]> > > > > wrote: > > > > > On Wed, 2020-01-01 at 09:54 -0500, Gary Gregory wrote: > > > > > > On Wed, Jan 1, 2020 at 9:43 AM Gary Gregory < > > > > > > > > > > [email protected]> > > > > > > wrote: > > > > > > > > > > > > > > On Wed, Jan 1, 2020 at 9:23 AM Gary Gregory < > > > > > > > > > > [email protected] > > > > > > > > wrote: > > > > > > > > On Wed, Jan 1, 2020 at 8:47 AM Michael Osipov < > > > > > > > > [email protected]> wrote: > > > > > > > > > Am 2020-01-01 um 14:45 schrieb Oleg Kalnichevski: > > > > > > > > > > Folks > > > > > > > > > > > > > > > > > > > > Shall I proceed with HttpCore 5.0-beta11 release or > > > > > > > > > > not? > > > > > > > > > > > > > > > > > > > > Do we still need to rename anything in core? > > > > > > > > > > > > > > > > > > Nothing left on my list. From my POV, you may go ahead. > > > > > > > > > > Gary, > > > > > > > > > Ryan? > > > > > > > > > > > > > > > > We did a good but partial job of cleaning up enums. > > > > > > > > > > Specifically, > > > > > > > > we did not circle back to our dual AuthSchemes and > > > > > > > > AuthScheme > > > > > > > > types. > > > > > > > > > > > > > > > > I will experiment now with something like renaming > > > > > > > > > > AuthSchemes to > > > > > > > > StandardAuthScheme and having it implement an interface > > > > > > > > lie AuthScheme. > > > > > > > > > > > > > > > > > > > > > > A very quick look suggests that such a design will not work > > > > > > > > > > since > > > > > > > AuthScheme implementations carry state. > > > > > > > > > > > > > > At the very least we should rename AuthSchemes to > > > > > > > StandardAuthScheme. > > > > > > > > > > > > > > WDYT? I'll make a PR. > > > > > > > > > > > > > > > > > > > This is for the client, not core, doh. > > > > > > > > > > > > > > > > Gary, > > > > > > > > > > Is there _anything_ that blocks the release of _core_ in your > > > > > opinion? > > > > > While core is being released there will still be time to work > > > > > on > > > > > client. > > > > > > > > > > Oleg > > > > > > > > > > > > > > > > > > > ----------------------------------------------------------------- > > > ---- > > > To unsubscribe, e-mail: [email protected] > > > For additional commands, e-mail: [email protected] > > > > > > > > > --------------------------------------------------------------------- > To unsubscribe, e-mail: [email protected] > For additional commands, e-mail: [email protected] > >
