pantsbuild #general

Channels

welcome

github-notifications

announce

billowy-alarm-80641

07/30/2015, 4:54 PM

is there a way to drive the help to make

--jvm-debug

appear?

billowy-alarm-80641

07/30/2015, 4:54 PM

happy-kitchen-89482

07/30/2015, 5:12 PM

I have it in my TODOs for the documentation fixit in a couple of weeks

billowy-alarm-80641

07/30/2015, 5:12 PM

Yeah, I'm not seeing an obvious hack to do in the short term

happy-kitchen-89482

07/30/2015, 5:13 PM

The long way is, e.g., --test-junit-jvm-debug

billowy-alarm-80641

07/30/2015, 5:13 PM

I'm okay with that

billowy-alarm-80641

07/30/2015, 5:13 PM

the issue is that the only way to know that is to use --help-advanced

happy-kitchen-89482

07/30/2015, 5:13 PM

The short way is, .e.g., ./pants test.junit --jvm-debug=...

billowy-alarm-80641

07/30/2015, 5:13 PM

which I don't want to tell the general populace to use

happy-kitchen-89482

07/30/2015, 5:13 PM

Can you file that as a github issue and assign to me? So I don't forget to specifically address that when I jazz up the help?

billowy-alarm-80641

07/30/2015, 5:14 PM

sure, will do

billowy-alarm-80641

07/30/2015, 5:14 PM

thanks

happy-kitchen-89482

07/30/2015, 5:15 PM

thx

enough-analyst-54434

07/30/2015, 5:20 PM

@gorgeous-fish-11040: I'm in favor of @ancient-stone-50795's approach, making the lowest level model true will make everything above it easier - or more clear - at the end of the day.

billowy-alarm-80641

07/30/2015, 5:23 PM

https://github.com/pantsbuild/pants/issues/1886

ancient-stone-50795

07/30/2015, 5:28 PM

@gorgeous-fish-11040: @enough-analyst-54434: I’m sorry I haven’t had time to look at the ivy refactoring projects yet, but I welcome anyone who wants to dig in and start off.

rough-minister-58256

07/30/2015, 5:29 PM

@ancient-stone-50795 is that going to be based off of 4sq’s ivy replacement or something else?

fast-megabyte-33579

07/30/2015, 5:29 PM

is there a roadmap of the planned changes to pants? There seem to be a bunch of google docs but I can't tell which are still 'current'

rough-minister-58256

07/30/2015, 5:30 PM

ah, or perhaps you’re referring to a different thing

ancient-stone-50795

07/30/2015, 5:30 PM

@rough-minister-58256: no, it is not an ivy replacement. It is a refactoring of how we pass information about artifacts around.

rough-minister-58256

07/30/2015, 5:30 PM

ah I see

ancient-stone-50795

07/30/2015, 5:31 PM

It will help with 4sq’s ivy replacement in that there wont be a bunch of ivy specific concept like ‘confs’ tied up in the datastructures.

gorgeous-fish-11040

07/30/2015, 5:33 PM

@ancient-stone-50795 @enough-analyst-54434 great! I prefer this option too. Will work on this simple smile

enough-analyst-54434

07/30/2015, 5:33 PM

@fast-megabyte-33579: The most recent pants summit notes are always probably the best window into what we'll all be working on

enough-analyst-54434

07/30/2015, 5:35 PM

Right now that's from the July 1 summit - notes here: https://docs.google.com/document/d/1tBVOaK2u1Q_-RmL2gG9qIoLpArHXiiEy-XcGRQHEElw/edit#heading=h.78ft8i783wgo

fast-megabyte-33579

07/30/2015, 5:37 PM

thanks

happy-kitchen-89482

07/30/2015, 5:40 PM

this work, if I understand it correctly, will allow our pom resolver to be a true drop-in replacement for ivy resolve, for codebases that can utilize its model (i.e., a single global static resolution of all jars)

happy-kitchen-89482

07/30/2015, 5:50 PM

Re go support:

happy-kitchen-89482

07/30/2015, 5:50 PM

One of go's big advantages is how blazingly fast its toolchain is

happy-kitchen-89482

07/30/2015, 5:50 PM

It might be a tough sell to add Pants's somewhat cumbersome startup time to that