Three things to do:
1) If any changes were made to Og darcs in the last month, please port
those changes to Ogden's repo.
2) Remove use of facets/paramix.rb (its begin deprecated). This
effects Orderable and Hierarchical mixins (any others?)
3) Get the tests passing. Most of them are failing, and I need people
more familiar with the Og's internals to help fix.
I'm not sure how much the failing tests has to do with the switch from
test/unit to rspec. Who did this transition? Were the specs ever all
passing?
Ogden is usintg SVN and not darcs. Please see this page to check out a copy:
http://rubyforge.org/scm/?group_id=5023
And let me know if you need to be added as an team developer.
Thanks,
T.
_______________________________________________
Nitro-general mailing list
Nitro-...@rubyforge.org
http://rubyforge.org/mailman/listinfo/nitro-general
> I'm not sure how much the failing tests has to do with the switch from
> test/unit to rspec. Who did this transition? Were the specs ever all
> passing?
I translated all of Ogs specs and some of the Nitro ones.
The specs were passing about 80% as far as I can remember.
As a sidenote: I resented the change to rspec, it was introduced in
a hurry, was not executed by someone knowledgeable enough, did not have
full backing from George and as a result is now in it's usual morbid
half-state.
Before the switch to rspec my mind tells me that about 3 tests failed.
How about that. If you can personally assure me that the spec structure
is basically working, that only `rake test` or similar has to be done
and that a strict 'only patches passing all tests'-policy will be used
on Ogden - then I will have a look at the failing specs.
Jo
--
Using Opera's revolutionary e-mail client: http://www.opera.com/mail/
On Feb 27, 6:38 am, "Jonathan Buch" <j...@oxyliquit.de> wrote:
> Hi,
>
> > I'm not sure how much the failing tests has to do with the switch from
> > test/unit to rspec. Who did this transition? Were the specs ever all
> > passing?
>
> I translated all of Ogs specs and some of the Nitro ones.
> The specs were passing about 80% as far as I can remember.
>
> As a sidenote: I resented the change to rspec, it was introduced in
> a hurry, was not executed by someone knowledgeable enough, did not have
> full backing from George and as a result is now in it's usual morbid
> half-state.
I did too. Not that it was a bad idea in the long run, but it was very
much wrong time to be doing it.
> Before the switch to rspec my mind tells me that about 3 tests failed.
>
> How about that. If you can personally assure me that the spec structure
> is basically working,
I think so. I'm still trying to understand it myself. I assume it was
correct to just load all the .rb files within the test/ directory.
Since there is no spec_ prefix, or anything like that, I could only
assume this to be the case. It seems to be working for the most part --
with the exception of warnings about constants being reset. I will
keep working to improve this, and make sure it is working right Any
help you can offer in the way of this is appreciated.
> that only `rake test` or similar has to be done
Yes, just run "ruby task/test".
> and that a strict 'only patches passing all tests'-policy will be used
> on Ogden - then I will have a look at the failing specs.
Yes. After we get 0.50 out the door (which George has been asking me
do for weeks now), we will switch to a proper patch policy with
branches and passing tests. In fact, would you like to be in change of
overseeing that?
T.
> I think so. I'm still trying to understand it myself. I assume it was
> correct to just load all the .rb files within the test/ directory.
> Since there is no spec_ prefix, or anything like that, I could only
another annoyance; one gets what one asks for. :P
> assume this to be the case. It seems to be working for the most part --
> with the exception of warnings about constants being reset. I will
> keep working to improve this, and make sure it is working right Any
> help you can offer in the way of this is appreciated.
>
>> that only `rake test` or similar has to be done
>
> Yes, just run "ruby task/test".
jo:ogden jo:0$ ruby task/test
ruby: No such file or directory -- task/test (LoadError)
jo:ogden jo:1$ ls task/
clobber rdoc setup
>> and that a strict 'only patches passing all tests'-policy will be used
>> on Ogden - then I will have a look at the failing specs.
>
> Yes. After we get 0.50 out the door (which George has been asking me
> do for weeks now), we will switch to a proper patch policy with
> branches and passing tests.
But of course, until there is actually a Ogden which has no failing
tests, that point is moot. But I like the plan. :)
> In fact, would you like to be in change of overseeing that?
No, I don't see that coming yet. As I'm wrapping up my University
time this semester, I very much doubt I have the leasure time to
do maintainer work.
All I hope I can do is putting my Og knowledge to use to at least get
it in a half stable state.
Jo
--
Using Opera's revolutionary e-mail client: http://www.opera.com/mail/
:D If it's any consolation I'll buy a beer (or preferred beverage).
> > assume this to be the case. It seems to be working for the most part --
> > with the exception of warnings about constants being reset. I will
> > keep working to improve this, and make sure it is working right Any
> > help you can offer in the way of this is appreciated.
>
> >> that only `rake test` or similar has to be done
>
> > Yes, just run "ruby task/test".
>
> jo:ogden jo:0$ ruby task/test
> ruby: No such file or directory -- task/test (LoadError)
> jo:ogden jo:1$ ls task/
> clobber rdoc setup
Sorry. I worked on these last night and forgot to check them in. They
should be there now.
> >> and that a strict 'only patches passing all tests'-policy will be used
> >> on Ogden - then I will have a look at the failing specs.
>
> > Yes. After we get 0.50 out the door (which George has been asking me
> > do for weeks now), we will switch to a proper patch policy with
> > branches and passing tests.
>
> But of course, until there is actually a Ogden which has no failing
> tests, that point is moot. But I like the plan. :)
>
> > In fact, would you like to be in change of overseeing that?
>
> No, I don't see that coming yet. As I'm wrapping up my University
> time this semester, I very much doubt I have the leasure time to
> do maintainer work.
> All I hope I can do is putting my Og knowledge to use to at least get
> it in a half stable state.
Understood. Hopefully a couple of other interested and capable
programmers will turn up once we get 0.50 out the door. I'm pretty
busy too.
T.
On Mar 23, 4:34 pm, "Dan North" <tasta...@gmail.com> wrote:
> Hi folks.
>
> I just downloaded from the svn link and ran task/test and got the following:
>
> Finished in 7.722335 seconds
>
> 19 examples, 13 failures
>
> Is this the same thing you guys are getting or am I even more out of whack
> than you?
That's about the state of things right now. I just spent about 15
minutes updating my copy to the latest Facets and I end up with 11
failures.
There is some clear work that needs to be done before a new release --
mainly removing the use of paramix.rb and then getting the tests to
pass. But I haven't had any time lately to work on it. It would be
great if someone wanted to dive in.
On Mar 24, 4:00 pm, "George Moschovitis"
<george.moschovi...@gmail.com> wrote:
> Do you know what kind of changes are needed to make this compatible with the
> latest facets?
> -g.
I just committed my changes. The changes were mainly just adjusting
some require statements.
I just committed my changes. The changes were mainly just adjusting
some require statements.
T.
_______________________________________________
Nitro-general mailing list
Nitro-...@rubyforge.org
http://rubyforge.org/mailman/listinfo/nitro-general
On Mar 26, 4:24 am, "George Moschovitis"
<george.moschovi...@gmail.com> wrote:
> > I just committed my changes. The changes were mainly just adjusting
> > some require statements.
>
> can you remind me where they are commited?
http://rubyforge.org/scm/?group_id=5023