Re: [cdi-unit] Committers

23 views
Skip to first unread message

Sean Flanigan

unread,
Jul 11, 2016, 2:51:55 AM7/11/16
to cdi-...@googlegroups.com
Hi Bryn,

On 6 July 2016 at 02:49, Bryn Cooke <bryn...@gmail.com> wrote:

As you have probably noticed, I haven't made any changes to CDI-Unit in a while, this is because I no longer work for a company that uses CDI. As such I'll need pull requests to take things forward.

If you don't mind my asking, is that because this company has achieved post-injection enlightenment? What does that look like? Or are you just using a non-CDI variety of injection framework?
 
I can do the bare minimum to keep this ticking over, but if folks are interested in becoming a committers then I am more than happy for this to happen. This project needs people who are using CDI in their day jobs to decide on new features.

I'm interested. In the Zanata project, we used to maintain an internal feature called SeamAutowire to simulate Seam injection in unit tests, but we switched to CDI-Unit after moving to CDI, so we'll certainly be making some use of it.

And maybe I could release a new version which includes https://github.com/BrynCooke/cdi-unit/pull/92 , instead of hassling you to do it! ;-) (But that would be up to you, of course.)

Regards,

Sean.


Bryn Cooke

unread,
Jul 11, 2016, 3:23:28 PM7/11/16
to CDI-Unit
Hi Sean,
 
If you don't mind my asking, is that because this company has achieved post-injection enlightenment? What does that look like? Or are you just using a non-CDI variety of injection framework?

It's simply that I joined a company that already had an injection strategy based around an alternative framework. I still think that CDI one of the better DI alternatives.


I'm interested. In the Zanata project, we used to maintain an internal feature called SeamAutowire to simulate Seam injection in unit tests, but we switched to CDI-Unit after moving to CDI, so we'll certainly be making some use of it.

That's great, it looks you've been involved in CDI for a long time, I've invited you to collaborate.
 
And maybe I could release a new version which includes https://github.com/BrynCooke/cdi-unit/pull/92 , instead of hassling you to do it! ;-) (But that would be up to you, of course.)


I'm not sure how to add another user to deploy to maven central off the top of my head. So in the sort term I'll just knock out another release.

Longer term:
  • I'll shift the documentation to github so that collaborators can edit it 
  • We can work out how others can do releases.
Many thanks for your interest in this project.

Bryn

 

Sean Flanigan

unread,
Jul 11, 2016, 11:09:30 PM7/11/16
to cdi-...@googlegroups.com
On 12 July 2016 at 05:23, Bryn Cooke <bryn...@gmail.com> wrote:
Hi Sean,
 
If you don't mind my asking, is that because this company has achieved post-injection enlightenment? What does that look like? Or are you just using a non-CDI variety of injection framework?

It's simply that I joined a company that already had an injection strategy based around an alternative framework. I still think that CDI one of the better DI alternatives.
 
Okay, good to know! 

I'm interested. In the Zanata project, we used to maintain an internal feature called SeamAutowire to simulate Seam injection in unit tests, but we switched to CDI-Unit after moving to CDI, so we'll certainly be making some use of it.

That's great, it looks you've been involved in CDI for a long time, I've invited you to collaborate.

Well, it's a long time if you count the slow and painful Seam-to-CDI planning and migration. Zanata 3.9 is our first CDI release, but it was in development for quite a while.

Got the invite. Thank you, Bryn. 
 
And maybe I could release a new version which includes https://github.com/BrynCooke/cdi-unit/pull/92 , instead of hassling you to do it! ;-) (But that would be up to you, of course.)


I'm not sure how to add another user to deploy to maven central off the top of my head. So in the sort term I'll just knock out another release.

Great, thanks.

Longer term:
  • I'll shift the documentation to github so that collaborators can edit it 
  • We can work out how others can do releases.

Going by the parent pom, it looks like CDI-Unit is using Sonatype OSS, the same as Zanata. We usually just add a comment to our original ticket in https://issues.sonatype.org/browse/OSSRH requesting repository access for the relevant username (I'm seanf). This looks like your ticket: https://issues.sonatype.org/browse/OSSRH-2949

But there's no hurry at all. You probably want to control access to org.jglue.cdi-unit separately from org.jglue.fluent-json. I'm sure we would need to nut out the full release process anyway, among other things. 

Then there's the pull request policy. Let me know what you're thinking.

Perhaps a cdi-unit-dev Google Group would be worth having at some point. I suppose that depends on the amount of traffic though.

 
Many thanks for your interest in this project.

Thanks for a very handy testing tool!
 
Bryn

 

--
You received this message because you are subscribed to the Google Groups "CDI-Unit" group.
To unsubscribe from this group and stop receiving emails from it, send an email to cdi-unit+u...@googlegroups.com.
For more options, visit https://groups.google.com/d/optout.

Reply all
Reply to author
Forward
0 new messages