#general
- e
enough-analyst-54434
10/20/2015, 6:05 PMIts ~no matter anyhow, the queue is shallow right now. - e
enough-analyst-54434
10/20/2015, 6:05 PMJust 1 prod CI and then you're 1 PR - c
cuddly-architect-27013
10/20/2015, 6:06 PMok, well can still kill…i see it’s a style issue i need to fix - c
cuddly-architect-27013
10/20/2015, 6:06 PMi should run the style checks before i push - e
enough-analyst-54434
10/20/2015, 6:06 PM./build-support/bin/setup.sh - e
enough-analyst-54434
10/20/2015, 6:07 PMAlthough that doesn't run checkstyle, but it does do isort and boilerplate checks right now - c
cuddly-architect-27013
10/20/2015, 6:08 PMok, cool…thx! - f
freezing-motorcycle-14222
10/20/2015, 6:48 PMSo my team has been interested in splitting up the large JAR generated by the
goal into smaller targets for better caching. I believe @witty-crayon-22786 told somebody on our team that./pants bundle
actually originally had the original smaller JARs before bundling them into a larger JAR and I wanted to learn more about this (to see if there was any way that I could help)pants - f
freezing-motorcycle-14222
10/20/2015, 6:48 PMs/smaller targets/smaller JARs/ - h
happy-kitchen-89482
10/20/2015, 6:52 PM@enough-analyst-54434: My bintray acct is benjyw. Thanks! - e
enough-analyst-54434
10/20/2015, 6:56 PM@happy-kitchen-89482: invited - e
enough-analyst-54434
10/20/2015, 6:57 PMSo @freezing-motorcycle-14222 - the winds of fancy.... this is how it used to be at least once. It confused the hell out of people (at Twitter!) - somethings doesn't work right ... where are my classfiles? Their is nothing in the main jar (except a classpath manifest - which they don't notice / know about as a thing) - f
freezing-motorcycle-14222
10/20/2015, 6:58 PMsimple smile - e
enough-analyst-54434
10/20/2015, 6:58 PMSo - I give up. - f
freezing-motorcycle-14222
10/20/2015, 6:58 PMIn our case the motivation is that this JAR has become very large with the advent of the monorepo, and it's difficult for our tools to cache - f
freezing-motorcycle-14222
10/20/2015, 6:59 PMThis gives very slow JAR upload times and inefficient JAR localization on Hadoop - e
enough-analyst-54434
10/20/2015, 6:59 PMYour jars are big because of scala's poor mapping to the jvm - but that is another thing - e
enough-analyst-54434
10/20/2015, 6:59 PMI'm not opposed - f
freezing-motorcycle-14222
10/20/2015, 7:00 PMDo you have any strong opinions on the API for this or would it be okay to just add a flag to
to toggle this behavior?bundle - e
enough-analyst-54434
10/20/2015, 7:00 PMI'd prefer no flag, but given history - introducing that complication may make most sense. - f
freezing-motorcycle-14222
10/20/2015, 7:01 PMOkay, I'll take a look at this more closely, then. I can't promise anything (this is just investigatory) but if I make progress then I'll put up a review board - e
enough-analyst-54434
10/20/2015, 7:01 PMSG - e
enough-analyst-54434
10/20/2015, 7:03 PMI'm glad to hear btw that using the hadoop distributed cache is now a thing! I think that's what I hear - e
enough-analyst-54434
10/20/2015, 7:04 PMFought for this a long time - e
enough-analyst-54434
10/20/2015, 7:07 PMHeads up that I'm going to TBR https://rbcommons.com/s/twitter/r/3000/ to get in a few more scheduling examples and then send out notes/homework suggestions for the engine summit so we can be efficient and actually get things planned. - w
witty-crayon-22786
10/20/2015, 7:51 PM@enough-analyst-54434: thanks a lot john: read through it this morning, and didn't see anything objectionable. - w
witty-crayon-22786
10/20/2015, 7:51 PMthe 3000 is pretty amazing really - e
enough-analyst-54434
10/20/2015, 7:52 PMYeah - that was whimsical - e
enough-analyst-54434
10/20/2015, 7:52 PMI just figured out the 1 wart with pickling so thats about to be patched up - w
witty-crayon-22786
10/20/2015, 7:52 PMYour jars are big because of scala's poor mapping to the jvm - but that is another thing