lilypond-devel
[Top][All Lists]
Advanced

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

Re: the latest convert-ly fiasco


From: David Kastrup
Subject: Re: the latest convert-ly fiasco
Date: Tue, 25 Oct 2011 13:44:46 +0200
User-agent: Gnus/5.13 (Gnus v5.13) Emacs/24.0.90 (gnu/linux)

Graham Percival <address@hidden> writes:

> I've lost track of where we are with convert-ly.
>
> I have just ran update-with-convert-ly.  I saw a flag change from
> 2.15.10.  I have pushed that.

There is a bunch of duplicate Flag overrides/reverts now.  I suppose
Mike and/or translators will have to clean those up manually.  Does not
make for a pretty history, but not a deal-breaker.

> I know that there are other syntax changes that people want to make.

There are several independent ones.  Since their respective pushability
depends on the order in which they will get applied, I don't see that
changing their status to Patch-push or Patch-countdown would be
appropriate.

You vetoed using dev/staging for preparation of a mergeable sequence of
independent patches with corresponding automatic convert-ly updates.

Since the automatic updates need to be combined with their respective
patches for disruptive syntax changes (changes that don't work without
running convert-ly), I don't see a sensible way around stepping the
version number in between the patches.

I don't see Rietveld up to the task of reviewing all that suitably.
I'll prepare such a sequence, but I have no idea how it can make it
through the policies.

> I am not going to try to make any general guidelines for convert-ly at
> the present time.  When I am feeling better, we will resume GOP, and
> we'll get a definite set of instructions for syntax changes.  Until
> then, just do whatever works, and if it's useful to have a couple of
> extra releases just to avoid combining multiple syntax changes in the
> same version number, so be it.

As I said: that will be the cleanest sequence.

-- 
David Kastrup




reply via email to

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