Re: What's cooking in git.git (Apr 2019, #03; Tue, 16)

[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

 



On Wed, Apr 17, 2019 at 03:14:28PM +0900, Junio C Hamano wrote:
> Taylor Blau <me@xxxxxxxxxxxx> writes:
>
> >> That's this one
> >>
> >>     Date: Tue, 9 Apr 2019 19:13:06 -0700 (1 week, 3 hours, 23 minutes ago)
> >>     Subject: [PATCH v2 0/7] harden unexpected object types checks
> >>
> >> which I think is what has been queued and what is listed in the
> >> message you are responding to.
> >
> > Ah, perhaps you could clarify some confusion I have about the "What's
> > Cooking" emails (or at least point me somewhere I could un-confuse
> > myself).
> >
> > This topic is in the "Cooking" section with a "-" (which I recall means
> > that it's in 'pu'), but there is no "Will merge to ..." line below it
> > from you.
> >
> > That makes sense to me, but I'm not sure whether or not that means it's
> > queued. Do you say that a topic is queued once it's on your pu, or once
> > you have written "Will merge to..."?
> >
> > Thanks in advance for your clarification.
>
> Hmph, I guess I shouldn't use the verb "queue" if it implies a lot
> more than there is to it.  I create a topic branch out of a patch
> series on the list at some point in the iteration of a series and
> merge it to 'pu'; I can say "I queued the topic to 'pu'" after doing
> so.  The verb to me does not mean anything more than that.

I thought that there was much more to it than it seems there actually
is. Thank you for clarifying what is meant here.

> The fact that a topic is queued on 'pu' does not mean much.  It can
> be taken as a sign "Gitster thought that it may become 'next' worthy
> material, either as-is or with further polishing and replacing."

This matches my understanding that 'pu' is a precursor for 'next' and so
on.

> Once a topic gets discussed on list and it seems apparent that there
> is a concensus that supports it, I may mark the topic as "Will merge
> to 'next'" in the What's cooking report, but I may not realize that
> the list already reached such a concensus and may leave it unmarked
> in the report.
>
> So "this has been battle tested in such and such environment" and/or
> "this round was reviwed by the thread here and they were supportive"
> etc. is a very appreciated response to the "what's cooking" report
> to help me merge the topic down to 'next'.

That's quite helpful to know in the future.

For this topic in particular, we've been running a nearly-identical
version of it at GitHub for a couple of weeks now. It went out smoothly,
and hasn't caused any trouble since. (In fact, quite the opposite: it
fixed the bug that caused me to look into this in the first place. That
is, the repository that SIGSEV'd after a `git rev-list --all --objects`
no longer does).

So, for clarity I think that this can be considered "battle-tested" and
ready to merge onto next.

> "I have sent a reroll at ..." I typically hear when I miss a
> rerolled version and what is listed is still from an old iteration,
> hence my "Eh, you are pointing at what I have queued" reaction.

Understood, and thank you again :-).

> Thanks.

Thanks,
Taylor



[Index of Archives]     [Linux Kernel Development]     [Gcc Help]     [IETF Annouce]     [DCCP]     [Netdev]     [Networking]     [Security]     [V4L]     [Bugtraq]     [Yosemite]     [MIPS Linux]     [ARM Linux]     [Linux Security]     [Linux RAID]     [Linux SCSI]     [Fedora Users]

  Powered by Linux