I suspected 399 might have something to do with the cycle stuff that's
been popping up. I still think the fix for 399 itself is correct
though; the only difference is cycles are now being correctly detected
where they previously might not have been.
This could come across as a regression to a user of buildr though. Any
idea how to fix it?
Pepijn
Op 12-jun-2010 om 21:11 heeft Antoine Toulme <anto...@lunar-ocean.com>
het volgende geschreven:\
Commenting on my own email:
On Sat, Jun 12, 2010 at 01:01, Antoine Toulme <anto...@lunar-
ocean.com>wrote:
I tried some functional testing of RC4 with Apache ODE.
It didn't play well.
First, here is still an issue with tag_name over git. I thought
this was
fixed, but our API obviously is still not ready. We should at the
very least
die with a deprecation message. I'll work on that this week-end.
Fixed now.
Second, with jruby, I got this annoying stacktrace:
http://jira.codehaus.org/browse/JRUBY-4867
Turns out our buildr script for jruby needed to be updated for
version
1.5.1. Fixed now.
Third, I ran into issues with cycle detection.
Commenting out the Rake monkey-patching in application.rb fixes the
issue -
I need to hack it some more.
Looks like BUILDR-399 causes the problem. Looks also like BUILDR-354
is
unrelated.
I'm wondering if the Rakefile of ODE is the problem - after all, it
was one
of the first Rakefile and may contain some stale stuff.
Fourth, compilation fails because jmock and junit are not found.
Not there yet.
I'm not happy with this mess. I intend to start having rounds of
functional
testing harnessed in Hudson.
Ongoing - I detailed my plan in BUILDR-456.
I'll probably sacrifice part of my week-end over this.
Thanks,
Antoine
