Backport fixes from PR #210.

[lohedges]

This PR backports the fixes from #210 into main.

• I confirm that I have merged the latest version of main into this branch before issuing this pull request (e.g. by running git pull origin main): [y]
• I confirm that I have permission to release this code under the GPL3 license: [y]

[lohedges]

CI is failing since it looks like some of the Sire fixes were PRed as features, thus never backported. In particular, this and this.

[chryswoods]

Sorry - I don't know what I was thinking when I made them features... Do you still want them backported, or is it ok to have these fixes in just 2023.5.0?

[chryswoods]

I see what happened - I had them as issue branches, but combined several together into the merge into devel with a feature, as I wanted to save the CI. Because they went in merged with the feature, I forgot to backport the individual fixes to main.

[lohedges]

No problem, it's easily done. If I don't do them immediately then I forget. I'm happy to wait until the release unless it's needed by Exs before, e.g. perhaps they will be using 2023.4 for a bit and need a working version. I wasn't planning on doing another BioSimSpace patch release. At present, I think the existing main branch of BioSimSpace is broken, since there is no compatible version of Sire, i.e. I merged in my fixes assuming that the Sire updates to which they were related were already backported. Guess it might be worth checking with @xiki-tempula.

[chryswoods]

I looked at it and it wasn't that much work. Each issue fix was quite self-contained.

I've pulled in all of the issue fixes into this PR.

I think there is enough to make it worth releasing a 2023.4.2 at the same time as 2023.5.0. I guess @xiki-tempula that this would make it easier for you to have both a "latest fix update" patch release for the last major version that goes along with the first release of the next major version.

[lohedges]

Thanks. I actually only spotted this because I accidentally ran CI for the PR. I normally don't for trivial backports since it's already been validated on devel. It was only this that made me realise that Sire was out of sync with the BioSimSpace fixes.

[xiki-tempula]

Thanks for mentioning me. So we are currently basing our main to the Sire main (as we need to some fix from the main), which is not the Sire 2023.4.1. I wonder when you make them sync, would it cause trouble if we are using the main (potentially 2023.4.2)? Thanks.

[lohedges]

I don't think there will be any issue. Currently BIoSimSpace main is broken since there are backports that rely on functionality that isn't in Sire main. (These are minor updates to do with rotatiing boxes and the Sire trajectory analysis.) Once these are merged into Sire main, then all is okay.

[chryswoods]

It shouldn't - the PR for the fixes into main passes all CI. If @lohedges is happy, I will merge that in today and will prepare the 2023.4.2 release. We can then merge this PR into main on BioSimSpace and maybe do a patch release too?

[xiki-tempula]

Great. Thanks.

[lohedges]

Great. I'm off for part of the day but will try to review when I'm back. I've stacked a few other backports into this PR too. @xiki-tempula, do you want this one added too. (The one related to the AMBER mask.)

[xiki-tempula]

Yes, please. Thanks.

[lohedges]

All passing now that I'm building against Sire 2023.4.2 👍 I'd just like to figure out what to do about the backport for #226, then I'll be able to create a BioSimSpace 2023.4.1 release. @chryswoods, do you have any opinions on that?