rmaia / pavo

tools for the analysis of color data in R
http://pavo.colrverse.com
GNU General Public License v2.0
69 stars 17 forks source link

Slight fix test validation due to RNG update in upcoming future 1.21.0 #212

Closed HenrikBengtsson closed 3 years ago

HenrikBengtsson commented 3 years ago

In future 1.21.0 there's a fix to the 'sequential' backend such that the RNG kind is preserved also in that case (just like for other types of backends), cf. https://github.com/HenrikBengtsson/future/issues/446

I'm posting this issue just for the record. I'll following up with a PR shortly.

HenrikBengtsson commented 3 years ago

Hi, I'm hoping to submit future 1.21.0 to CRAN by the end of this week or so. What's your plan for releasing the next version of pavo?

If I submit before you, I need to figure out a way to work around the fact that the pavo checks will then fail and the patch here will probably not work and cause another error. In other words, if you submit after me, there needs to be more coordination. If you submit before me, things should work as-is. In https://github.com/rmaia/pavo/issues/211#issuecomment-736774002 you hinted at you could release an updated pavo before next future release; would that be possible for you?

PS. If before submitting to CRAN, please make sure you also check against a recent devel version future 1.20.1-9000.

thomased commented 3 years ago

Hi,

Yeah sorry for the delay — we're planning a release very shortly. We need to do so before the 16th anyway as one of our Imports is being orphaned. But we've hit a small snag when making the necessary fix, and are just chasing down the cause atm. Once that's sorted it'll be off asap.

HenrikBengtsson commented 3 years ago

Ok, thanks for the update. If you can, please give me a small heads up when you're getting closer so we're in sync and in case I decided to submit before you.

thomased commented 3 years ago

Hi again — I think we're all set on our end. I ran it against future 1.20.1-9000, no issues.

Let me know if there's anything you need, otherwise we can ship it whenever ready.

HenrikBengtsson commented 3 years ago

Brilliant. Let me run a quick check on my end too. It's the pavo 'master' branch I should check against, correct?

thomased commented 3 years ago

Yep - that's the one. Much appreciated.

HenrikBengtsson commented 3 years ago

All good. Validated on R 4.0.3 and latest R-devel on Linux. Thanks for pushing this new version out. Hoping it'll go smoothly for you.

thomased commented 3 years ago

Appreciate your help. It's submitted, so will hopefully be out soon.

HenrikBengtsson commented 3 years ago

It's submitted, so will hopefully be out soon.

Did you submit it because it's nowhere to find under ftp://cran.r-project.org/incoming/?

thomased commented 3 years ago

Was a bit slow in checking my emails to confirm — should be on its way now.

Screen Shot 2020-12-09 at 1 31 21 pm
HenrikBengtsson commented 3 years ago

Perfect!

HenrikBengtsson commented 3 years ago

I see pavo ended up being archived while your submitted pavo 2.6.0 still made it into the archives. Hopefully, this is just a weird timing issue but if not, I guess it passed the incoming tests and the "someone" decided to drop you when other OS tests became available. If so, I'm sorry about your struggles.

Just FYI, you don't have to worry about 'future' at this point, because if pavo makes it back on CRAN, I'm fine. And if it remains archived, I'm also fine for my next release.

thomased commented 3 years ago

Yes, it made it through, then was archived ~12 hours later because it apparently creates bin/phantomjs in the home dir, which is a bit mysterious to me, but we'll get it figured out.

Glad to hear we're not holding up 'future' though — that's great.