I have to agree that issues with thr current 1.3.5 release are morr of
a showstopper for my projects as well. Having to apply manual patches
in order to get things working on JRuby has held back our internal
buildr experiments for quite some time now.
The patch is attached to the Jira issue Rhett created BTW.
Pepijn
Op 17-jun-2010 om 18:51 heeft Alex Boisvert <[email protected]>
het volgende geschreven:\
I agree. I think more people are affected by issues in 1.3.5 today
than
would potentially be affected with 1.4.0 as it is. The way forward
is to
release 1.4.0 and address issues promptly as they are reported. We
can't
keep pushing 1.4.0 out.
alex
On Thu, Jun 17, 2010 at 9:35 AM, Daniel Spiewak
<[email protected]> wrote:
+1 on cutting the release now, fixing later. Again, remember that
there
are
very few bugs which are *more* critical than our outstanding
rubygems issue
with 1.3.5.
Daniel
On Thu, Jun 17, 2010 at 11:26 AM, Antoine Toulme <[email protected]
wrote:
You mean, the patch you attached to the dev list.
I have learnt first hand that patches have harmful side effects.
I'm not sure I want to change anything there. I am not sure having a
cycle
because you try to call things in the wrong order should be
corrected.
The patch is not attached to a Jira bug, so I lost track of it.
And most important, I'm out of juice. I need this release out now
or I'll
give up. That's me drawing the line in the sand.
On Thu, Jun 17, 2010 at 09:15, Pepijn Van Eeckhoudt <
[email protected]> wrote:
On 17/6/2010 17:38, Antoine Toulme wrote:
I think we now have a good understanding of the problem. Mainly,
that
this
is not a good or valid approach.
Rhett pointed at a workaround ; I guess that using an enhance
block
would
also have fixed the problem.
But since there is a patch that solves the problem (at least in the
short
term) and doesn't break the current specs; why not include this
in the
1.4
release?
Pepijn
