On Fri, 2006-04-07 at 09:53 -0700, Dom Lachowicz wrote:
> > Mind that the breakage of 2.4.3 was not a result of
> > any of my backport
> > actions, or lack of testing.
>
> Also mind that the breakage of 2.4.3 was not a result
> of me breaking anything in AbiWord.
Indeed.
> link-grammar's API
> changed (as it should have), and the plugin was not
> updated to work with the new API. As the library's
> maintainer, I'm not going to correct all of the
> downstream apps when API changes.
Nor should you.
> I'm also not the plugin's author, so fixing the plugin wasn't my
> responsibility.
Indeed not.
> You also didn't want to do the simple step of making
> the STABLE plugin require version link-grammar-4.1.4,
> which would have been a trivial and correct fix.
Because from what I understood from the irc discussions, was that we
really should be using the (soon to be 4.2.1 version). Iirc, it fixes
some dictionary issues; i have no problems postponing a release for that
a bit to be able to include it.
> link-4.1.4 + AbiWord are tested on Windows, while
> link-4.2.1 + AbiWord is not.
4.1.4 + Abi on windows should be fine indeed, but I wanted to included
your new version, hence the lil' wait for abi 2.4.4.
> All of this could've been avoided if the release
> process included something like "make distcheck" on
> the plugins before tagging, to make sure that they at
> least built.
I do the usual build checking, but I don't have all versions of every
lib installed :) It worked fine on my FC3-5 machines, and since I didn't
remember that the that the link-grammar API could have been broken,
link-grammar 4.2.0 was not tested in combination with Abi until Ryan
tried to use it.
Anyway, this is offtopic, as I don't 'blame' you for fixing link-grammar
*at all*.
I _only_ want to be careful what goes in so soon before a release (the
mailmerge change), as in the past we had some bad experiences with it.
The rest of my previous mail was to explain that my backporting is not
the simple auto-pilot action as you claimed it was. I put a good amount
of time and energy in it (hence the massive delay in 2.4.3).
Cheers,
Marc
Received on Fri Apr 7 19:14:27 2006
This archive was generated by hypermail 2.1.8 : Fri Apr 07 2006 - 19:14:27 CEST