Ergonomic, hermetic, user-friendly software build system for Python, Go, Java, Kotlin, Protobuf, Scala, Shell. Pants lets you fearlessly scale up your codebase!

Pants

<@U0607UFLH>: which process are we talking about

i still have the one from the last report, but have been busy with other things so haven't dug deeper on that one yet

Yes, we dumped the jstack but it had minimal information, just that it was waiting on a socket. I will go track down the tails and post them.

jstack_dump.txt

yea, that would be nailgun/pants related.

<@U04S45AHA>: it's funny... we just got a feature request internally to stop building monolithic jars for internal dependencies (there are a lot more post-monobuild). do you remember why we started doing that? related to the existence of `classes_by_target`?

Do <@U071M9KNE> or <@U071MJHQD> want to comment on <https://rbcommons.com/s/twitter/r/2946/> ?

<@U06A03HV1>: <@U04S45AHA> im gonna get back on the shrinkwrap thing… did either of you do anything on that since friday?

<@U053H6LQS>: can you explain `--infer-test-from-siblings` (in IdeaGen) to me? Are you currently using it? Is anyone?

What is a "sibling source_root() entry" ?

<@U071M9KNE>: ^ we don't need that anymore, right?

<@U051221NF>: <@U06A03HV1>: This was a change to the behavior of idea_gen.  It is for determining whether a directory should be marked as a regular source or a target source.  We have BUILD files with both java_library() and junit_tests() defined in them.  If this flag is enabled, then it will always choose to mark the directory as a TEST source root.    I thought we used this, but I don’t see it turned on in our pants.ini!

I see that option is a lot older than I thought (October 2014)  The logic has been superseded by the logic in dedup_sources() in ide_gen, so yes, I think it can go.
```
34ab8524 src/python/pants/backend/project_info/tasks/ide_gen.py (Eric Ayers        2015-08-17 17:21:27 -0400 495)   def dedup_sources(cls, source_set_list):
```

I just put up an RB to clean that up. <https://rbcommons.com/s/twitter/r/2966/>

hm... so it looks like coveralls hasn't gotten coverage data since <https://coveralls.io/github/pantsbuild/pants?page=38>

hm. no additional information from coveralls. wonder if the return code is hidden

This is ready to go, if someone could merge it for me! <https://rbcommons.com/s/twitter/r/2948/>

<@U098KD4HJ>: I see CI passed. I’ll do it.

the issue can be closed also <https://github.com/pantsbuild/pants/issues/1766>