[prev in list] [next in list] [prev in thread] [next in thread] 

List:       perl5-porters
Subject:    Re: [perl #132594] BBC smartmatch da4e040f42421764ef069371d77c008e6b801f45
From:       Karen Etheridge <perl () froods ! org>
Date:       2017-12-29 21:59:31
Message-ID: CAPJsHfBPJDm_4f9FVJ0vEVgq6hAVTGu+M=u5LsROyHcJKBx11g () mail ! gmail ! com
[Download RAW message or body]

On Thu, Dec 28, 2017 at 8:23 PM, Craig A. Berry <craig.a.berry@gmail.com>
wrote:

> I would just point out that as far as I know, a
> smoke-me branch doesn't automatically trigger any downstream testing.
> It takes a lot of cycles (people and computer) to do both the smoke
> testing and the BBC reports and I'm skeptical whether the downstream
> part will ever happen consistently and thoroughly for any branch
> except blead.  Even the core smoke testing has pretty limited coverage
> of platforms, branches, and configurations.  Which is just to say that
> I'm not sure how much of what broke with dumb_match was anticipated or
> could have been anticipated without merging it.  In any case, now we
> know.
>
​
Very good point; I was sloppy and conflated smoke-me branches and smoking
the core tests across different​ architectures with testing against
high-river cpan distributions. Do we have any process for requesting cpan
testing of branches? Is it even possible to do this without a lot of manual
steps? It would be fabulous if we could set up some sort of automated
process for cpan testing of contentious changes, perhaps with a naming
convention for branches akin to $user/smoke-me/$foo.

[Attachment #3 (text/html)]

<div dir="ltr"><div class="gmail_default" style="font-size:small"><br></div><div \
class="gmail_extra"><br><div class="gmail_quote">On Thu, Dec 28, 2017 at 8:23 PM, \
Craig A. Berry <span dir="ltr">&lt;<a href="mailto:craig.a.berry@gmail.com" \
target="_blank">craig.a.berry@gmail.com</a>&gt;</span> wrote:<br><blockquote \
class="gmail_quote" style="margin:0 0 0 .8ex;border-left:1px #ccc \
solid;padding-left:1ex">I would just point out that as far as I know, a<br> smoke-me \
branch doesn&#39;t automatically trigger any downstream testing.<br> It takes a lot \
of cycles (people and computer) to do both the smoke<br> testing and the BBC reports \
and I&#39;m skeptical whether the downstream<br> part will ever happen consistently \
and thoroughly for any branch<br> except blead.   Even the core smoke testing has \
pretty limited coverage<br> of platforms, branches, and configurations.   Which is \
just to say that<br> I&#39;m not sure how much of what broke with dumb_match was \
anticipated or<br> could have been anticipated without merging it.   In any case, now \
we<br> know.<br>
</blockquote></div></div><div class="gmail_extra"><div style="font-size:small" \
class="gmail_default">​</div><div style="font-size:small" \
class="gmail_default">Very good point; I was sloppy and conflated smoke-me branches \
and smoking the core tests across different​ architectures with testing against \
high-river cpan distributions. Do we have any process for requesting cpan testing of \
branches? Is it even possible to do this without a lot of manual steps? It would be \
fabulous if we could set up some sort of automated process for cpan testing of \
contentious changes, perhaps with a naming convention for branches akin to \
$user/smoke-me/$foo.</div><div style="font-size:small" \
class="gmail_default"><br></div><br></div></div>



[prev in list] [next in list] [prev in thread] [next in thread]