I'll take a look in the evening.
LieGrue,
Strub
> Am 09.08.2017 um 14:00 schrieb John D. Ament :
>
> Ok, I got tests passing on all three. It actually looks like its simpler
> (and more correct) to not use a custom provider. The provider approach
> wasn't supporting
Ok, I got tests passing on all three. It actually looks like its simpler
(and more correct) to not use a custom provider. The provider approach
wasn't supporting default values. I've committed changes, please take a
look and hopefully we can plan to reroll 1.0.
John
On Tue, Aug 8, 2017 at
Ok, I've readded some of the Weld3 support, 3 tests are failing.
Some interesting notes:
- If I remove Provider support, it generally works. However, I suddenly
get duplicate beans of type Provider, which is confusing (but probably
because of the alternative flag).
- I can get everything to
I can get to some of it.
We have a couple of issues now though:
https://issues.jboss.org/browse/WELD-2411
https://issues.jboss.org/browse/CDI-712
Martin doesn't feel the spec allows you to override Provider impl's.
Please feel free to weigh in :-)
John
On Tue, Aug 8, 2017 at 7:33 AM Mark
Oh I see what you mean.
Do you mind to re-apply your fix or should I?
Btw, I also tested with the -PWeld3 but I was not able to resolve
arquillian-weld-embedded:jar:2.0.0-SNAPSHOT
We might need to add a for it in the Weld3 profile.
I also would love to add a Weld profile. I thought we have
One note - hardcoding "isAlternative" to true breaks in Weld. Unless there
is a base bean of same type, the alternative is ignored. Hence why I only
did isAlternative on the provider type, all others are just regular beans.
John
On Tue, Aug 8, 2017 at 6:55 AM wrote:
>