rake spec:extensions EXT=multi_site exits with 0 although there are failing specs

[Johannes Fahrenkrug]

Could this be a radiant bug or a multi_site bug?

See: http://travis-ci.org/#!/radiant/radiant-multi-site-extension/jobs/1276398

[Jim Gay]

There's something wrong. The specs return 0 when they fail.
I've seen this in other extensions as well.

This commit in RSpec is related, but only mentions the before :all
scenario https://github.com/rspec/rspec-core/commit/79273a74a83ea3210c7abc219d840da4f1fdb0d0

> --
> Radiant CMS Dev Mailing List
> Post:        radiant...@googlegroups.com
> Unsubscribe: radiantcms-de...@googlegroups.com
> Group Site:  http://groups.google.com/group/radiantcms-dev/



--
Write intention revealing code #=> http://www.clean-ruby.com

Jim Gay
Saturn Flyer LLC
571-403-0338

[Johannes Fahrenkrug]

I see, so it might be an RSpec bug?

springenwerk.com | github.com/jfahrenkrug | twitter.com/jfahrenkrug

[Jim Gay]

Yes.

And since we're still on an old version of both rails and rspec, we
might be stuck with it... I haven't taken the time to track it down to
see if we could fix it.

On Tue, May 8, 2012 at 11:30 AM, Johannes Fahrenkrug

[Johannes Fahrenkrug]

Ugh. That would be ugly. Maybe we can find a way (monkey patch if
worse comes to worse).

Springenwerk Consulting
Johannes Fahrenkrug
Muehlenredder 21c, 24582 Bordesholm, Germany
+49 (0)151 22 634 651

[Johannes Fahrenkrug]

This is also related: https://github.com/rspec/rspec-core/pull/410

Do you know since when this is happening in radiant projects?

[Johannes Fahrenkrug]

Hi,

this seems to be exactly what we need:
http://spectator.in/2011/02/06/rspecs-1-dot-3-1-exit-codes/

Where exactly should we put the monkey patch?

- Johannes

On Wed, May 9, 2012 at 12:05 PM, Johannes Fahrenkrug

--
springenwerk.com | github.com/jfahrenkrug | twitter.com/jfahrenkrug

[Jim Gay]

Does this actually fix the error on Travis?

I'm confused by the comment at the end of that article "This issue has
been fixed in Ruby 1.9.2-p180."
Is this a Ruby problem? Why would the exit codes be incorrect on 1.9.3 if it is?

On Wed, May 9, 2012 at 6:57 AM, Johannes Fahrenkrug

[Johannes Fahrenkrug]

That message confused me too. But this (more elaborate) ticket doesn't
mention Ruby 1.9.2: https://github.com/rspec/rspec-core/pull/410

I haven't tried it with Travis yet: Monkey-patching RSpec proves to be
difficult.