Version 0.8 release planning

[Paul Romano]

Hi all,

It's been quite some time since our last release and we've had many significant changes come into the code since 0.7.1, including:

- Multi-group mode

- Many, many additions/improvements to the Python API

- Improved documentation and new notebooks

- Windowed multipole capability

- Periodic boundary conditions

- Refactor of CE reaction data

- Expanded source sampling options

- Distributed materials

- Improved method for reproducible URR table sampling

- Subcritical multiplication support

- Numerous bug fixes

I'd like to get a new release out prior to my new big pull request, which will propose a shift from using ACE data to a new native HDF5 format. Please speak up now if you plan to make any PRs which you intend to get into this release. Otherwise, we'll go ahead and start doing some extra testing on a release branch.

I should really reiterate -- the Python API has undergone so many changes since the last release that it may be unrecognizable to someone who was using the API in the 0.7.1 release. Fortunately, I think the number of people who are using version 0.7.1 and the Python API is close to zero. In any event, if you're not familiar with the changes in the Python API and this is something you care about, I'd encourage you to familiarize yourself with it now.

Best regards,

Paul

[Adam Nelson]

Wow those are some huge changes!

I had wanted to get in the HDF5 MGXS Library in to v0.8 so we didn't switch it out on the users too quickly.  But I think it makes the most sense to release without it and I will include a script to convert from the old to new format when I do get that PR in.

[Paul Romano]

That's a good point Adam. When do you think you'll be able to have that implemented by? I agree it would be nice to not have to change things too quickly for users.

[Adam Nelson]

It's implemented and seems to work fine right now.  However, that is based off your pyapi-data branch.  So to get it in I'd have to rewrite everything that sits above the actual initialize routine itself.

[Adam Nelson]

That being said, I do have the XML -> HDF5 script written and it seems to work pretty well.  So I don't know if it matters too much if we get this in for the v0.8 release or not.

[Paul Romano]

Ah, ok- in that case we can just defer. Thanks!

[Paul Romano]

I think develop is in a pretty good state so I just created a release-0.8 branch that increments the version number, adds release notes, and fixes a few compiler flags. I'd appreciate if some of you could do some testing and if all is well, we'll target the release for next week.

Thanks,

Paul

[Paul Romano]

I've been doing some testing and found a few minor bugs that I've since fixed on the release-0.8 branch. Unless I hear objections from anyone, I'll plan on going forward with the release in the next few days.

Thanks,

Paul