[Top][All Lists]

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

Re: bypassing the patch countdown

From: Urs Liska
Subject: Re: bypassing the patch countdown
Date: Wed, 07 Dec 2016 23:58:14 +0100
User-agent: K-9 Mail for Android

Am 7. Dezember 2016 23:34:39 MEZ, schrieb Graham Percival <address@hidden>:
>I was going to wait a month or two before suggesting this, just to
>make sure I was fully "up to date", but I'll jump in now.
>We instituted the policy of patch countdowns and Patchy after the
>lengthy wait for 2.14.0, which was due to a large number of
>regression bugs due to patches which either broke the compile, or
>broke previously-working output.
>However, even after that, I still pushed some commits directly to
>staging, bypassing the countdown.  Obviously I did this for
>updating the VERSION when making a release, but I also did it for
>a few typo fixes as well.
>Is this still an accepted practice?  If not, I suggest that it
>should be.  If I had to formalize it, I'd say something like "if
>two developers with push ability agree that a fix is trivial and
>obvious, it can go straight to staging".

Yes, this is still common practice.
Developers can and do take this liberty occasionally.
An alternative giving some extra safety is to upload a patch and wait for the 
first automated tests before pushing to staging.

In the current case the point is not that it would take a significant  effort 
to update the news but rather that it's not woth touching at all, given the 
temporary nature of the information. 


>(please note that I'm not suggesting that anybody should feel
>obligated to make such typo fixes -- instead, I'm checking that
>the "door is open".  So that if we manage to get 1 or 2 users who
>are able to fix typos, and those fixes are very obvious, they
>wouldn't need to wait 2-4 days.  In this case, the "two
>developers" would be "1 mentor, and the release or patch
>- Graham
>lilypond-devel mailing list

Diese Nachricht wurde von meinem Android-Mobiltelefon mit K-9 Mail gesendet.

reply via email to

[Prev in Thread] Current Thread [Next in Thread]