[svn.haxx.se] · SVN Dev · SVN Users · SVN Org · TSVN Dev · TSVN Users · Subclipse Dev · Subclipse Users · this month's index

Re: Backport of r22311 to 1.4.3.

From: Daniel Rall <dlr_at_collab.net>
Date: 2007-01-19 20:01:07 CET

On Sat, 20 Jan 2007, Chia-Liang Kao wrote:

> On 20/01/07, Malcolm Rowe <malcolm-svn-dev@farside.org.uk> wrote:
> >I just noticed that your backport of r22311 to 1.4.3 in r22694 didn't
> >make use of the backport branch that clkao updated in r22570
> >(/branches/1.4.x-r22311/) - not too surprising, since it wasn't
> >mentioned in STATUS.
> >
> >Anyway, I notice that there is a small difference between the two merges
> >- clkao's one changes (genericises?) a Perl-specific thingy above the
> >changed general typemap.

And that's not the only difference between r22311 and clkao's branch
-- r22311 only adds handling for switch and status, while the branch
also provides support for diff (which is not part of r22311).

> >Could someone confirm what the effect of the current state on 1.4.3 is?
> >I suspect that the answer is that the Perl bindings are still in the
> >same state as they were in 1.4.2, and so we still need a merge of that
> >final hunk in STATUS for 1.4.4. (Ironically, I think one result of
> >this change might been that we managed to wrap the _Ruby_ bindings,
> >something we didn't intend).
>
> Actually I found a problem about this yesterday when i was trying a
> fresh 1.4.x build, and was about to mail the list about the kind-of
> regression.

It would've been really nice to have a correct entry in STATUS, or
failing that, earlier notice of this problem (say, before 6 developers
take their time to test the release tarball).

> The change to generalise the EDITOR/BATON signature is required, as
> the typemaps (which the patch was to fix) are applied to such
> signature.

The change is definitely required. It's interesting that the tests
for both the Perl and Ruby bindings pass with the change which was
backported. There must be no tests for the Perl bindings...

So what now? We have to re-roll the 1.4.3 tarball. I wish I could
say we only have to re-test the SWIG bindings, but since the tests
apparently don't even cover this area of the bindings, that would be
pointless...

  • application/pgp-signature attachment: stored
Received on Fri Jan 19 20:01:23 2007

This is an archived mail posted to the Subversion Dev mailing list.