hey steve,
On Tue, May 17, 2005 at 07:21:31PM -0700, Steve Langasek wrote:
> Well, unpacking the two versions and applying the diffs still gives me 100
> lines of changes to upstream code that don't seem to be documented in the
> changelog, though perhaps this is what's referred to as "import of upstream
> packages was b0rken"? Either way, these are functional changes, and I don't
> see any explanation for why they're necessary.
to be more explicit, the patch was originally not cleanly applied (as
you yourself pointed out to me), so fixing it involved fixing the dpatch
patches, which involved unapplying them (02 and 04), removing reject
cruft files, and then reapplying the same upstream patches again (which
i did as one dpatch, 04). i guess i could have been more explicit in
the changelog, sorry about that.
anyway this resulted in a lot of noise in the interdiff, because the dpatch
file has superficially changed (mostly just the order of the lines, i
guess due to the file's being diffed in a different order by d-e-p).
so to quote the bard, "full of sound and fury, signifying nothing".
anyway, i've said my peace on the issue now, and explained things as
best i can. i'm sure you have much bigger fish to fry wrt to sarge, so
if you still don't think it should go in, life will go on i suppose :)
> > yeah, this is actually what we're doing. for some reason the fix isn't
> > working for woody mysql-server -> mysql-server-4.1 though, which is what's
> > holding things up.
>
> It would be nice to see an upload for mysql-server at least if you have that
> working, since that's the one that also still doesn't have a fix for 308762
> in testing.
we're still working on 4.1[1], but i'll try and get an upload of mysql-server
before i go to bed tonight. what? it's 1:30 already? damn...
sean
--
[1] the 4.1 will unfortunately also have to be a full source upload,
because some non-free documentation snuck its way into the orig.tar.gz :(
Attachment:
signature.asc
Description: Digital signature