Closed kyleam closed 1 year ago
Can you cc me when you release this new dev version?
Even once this is merged, things are held up by alpha2 being at the tip of the unmerged gh-268. It'd be confusing to have an alpha3 that doesn't have alpha2 as an ancestor, so I'd like to avoid making any alpha3 tag until that is brought in. (And in general would prefer to batch these alpha releases a bit more, though when we decided to bring alpha1 into the main line makes things trickier given how skip_if_old_bbi
handles dev versions).
Anyway, a lot of this depends on the specifics, but for now I can at least help you get set up locally to build this bbi and point bbr to it, if you'd like.
Can you cc me when you release this new dev version?
Even once this is merged, things are held up by alpha2 being at the tip of the unmerged gh-268. It'd be confusing to have an alpha3 that doesn't have alpha2 as an ancestor, so I'd like to avoid making any alpha3 tag until that is brought in. (And in general would prefer to batch these alpha releases a bit more, though when we decided to bring alpha1 into the main line makes things trickier given how
skip_if_old_bbi
handles dev versions).Anyway, a lot of this depends on the specifics, but for now I can at least help you get set up locally to build this bbi and point bbr to it, if you'd like.
No worries, had a hunch that might be the case anyways. I can wait till the official next release of bbi. I mainly just wanted to see if check_run_times
/model_summary
still worked as expected, but im fairly confident that will be the case given what was changed.
I just approved #268 (and looking at #272) now. Feel free to merge and tag however y'all think is best.
With merge of gh-268, there was direct conflict in validation/requirements.yaml
and semantic conflicts in validation/stories.yaml
and acop-nan.golden.json
. Rebased.
There are CI failures, but I don't think they're specific to this PR. The build triggered with the merge of gh-268 looks to have the same failures. Hmm, and I don't think it's a matter of a bad interaction with the new base from the merge, because the base (f090539) hasn't changed since 268 had a successful run.
There are CI failures, but I don't think they're specific to this PR.
Issues now resolved on Drone executor side. Jobs for this PR restarted and are green.
As @barrettk noted in gh-273,
bbi nonmem summary
doesn't reportpostprocess_time
if it has a value of zero. (There are a few cases of this in both bbi's and bbr's test suite.) The same applies tocpu_time
.This PR fixes that by dropping the
omitempty
tag from theRunDetails
struct.