https://pantsbuild.org/ logo
Powered by
#general
  • h

    happy-kitchen-89482

    08/13/2015, 4:40 PM
    my break, presumably, so I'll fix
  • h

    happy-kitchen-89482

    08/13/2015, 4:46 PM
    @ancient-stone-50795: I have a fix if you've not gotten to this yet
  • a

    ancient-stone-50795

    08/13/2015, 4:47 PM
    @happy-kitchen-89482: The fix was trivial, but I added some integration tests to the different help flavors in https://rbcommons.com/s/twitter/r/2616/
  • h

    happy-kitchen-89482

    08/13/2015, 4:47 PM
    thanks, sorry for the break
  • a

    ancient-stone-50795

    08/13/2015, 4:47 PM
    Might have been John, I didn’t investigate.
  • h

    happy-kitchen-89482

    08/13/2015, 4:49 PM
    pretty sure that was mine
  • h

    happy-kitchen-89482

    08/13/2015, 4:51 PM
    anyone know why python-eval is registered under 'compile'
  • h

    happy-kitchen-89482

    08/13/2015, 4:51 PM
    ?
  • a

    ancient-stone-50795

    08/13/2015, 4:51 PM
    I thought it was akin to a compile step in that it validates your python code syntax.
  • h

    happy-kitchen-89482

    08/13/2015, 4:52 PM
    makes sense
  • a

    ancient-stone-50795

    08/13/2015, 4:52 PM
    Got this today: "Dear Eric, Congratulations! We are pleased to inform you that the session(s) below on which you are a speaker has been accepted…"
  • a

    ancient-stone-50795

    08/13/2015, 4:52 PM
    From JavaOne
  • h

    happy-kitchen-89482

    08/13/2015, 4:52 PM
    got the same
  • h

    happy-kitchen-89482

    08/13/2015, 4:52 PM
    yay!
  • h

    happy-kitchen-89482

    08/13/2015, 4:52 PM
    thanks for pushing on that Eric
  • f

    fast-megabyte-33579

    08/13/2015, 4:53 PM
    awesome simple smile
  • w

    witty-crayon-22786

    08/13/2015, 5:02 PM
    @jolly-chef-92794: JvmPlatformValidate takes ~20 seconds to run on a medium sized project
  • w

    witty-crayon-22786

    08/13/2015, 5:03 PM
    and it doesn't use 
    self.invalidated
    to avoid redoing work on followup builds =/
  • a

    ancient-stone-50795

    08/13/2015, 5:05 PM
    @witty-crayon-22786: want to file it as an issue? Or should this be pretty easy to fix with the new support Cody added recently?
  • w

    witty-crayon-22786

    08/13/2015, 5:18 PM
  • g

    gorgeous-fish-11040

    08/13/2015, 5:55 PM
    I was postponding this change for long time and seems after @cold-lizard-88792’s change it’s finally a right time to address it because plugin is broken after compiler output paths were changed: https://github.com/pantsbuild/pants/issues/1973. Unfortunately I’m not sure how to do it. Any suggestions will be appreciated. Commented on the issue with my thoughts.
  • w

    witty-crayon-22786

    08/13/2015, 5:59 PM
    @gorgeous-fish-11040: using the compile_classpath product probably
  • w

    witty-crayon-22786

    08/13/2015, 5:59 PM
    ...oh. except that's generated by compile.
  • g

    gorgeous-fish-11040

    08/13/2015, 5:59 PM
    @witty-crayon-22786: then I need to compile. I want to know these paths before compiling
  • g

    gorgeous-fish-11040

    08/13/2015, 5:59 PM
    yeah 😞
  • w

    witty-crayon-22786

    08/13/2015, 6:00 PM
    should be easier to solve it more generally once the global strategy is gone/
  • w

    witty-crayon-22786

    08/13/2015, 6:00 PM
    ?
  • w

    witty-crayon-22786

    08/13/2015, 6:01 PM
    i can't think of any other reason for task shuffling until we remove it
  • g

    gorgeous-fish-11040

    08/13/2015, 6:09 PM
    even if it will be removed then the plugin won’t guarded from changes like @cold-lizard-88792 did with "omnivorous” ZincCompile where java and scala class files ends up in the same zinc dir instead of java and scala dirs regardless. I wanted to move this logic to export goal but had no time and know it seems to be a good clean up.
  • w

    witty-crayon-22786

    08/13/2015, 6:15 PM
    @gorgeous-fish-11040: right, but my point is that i think that is definitely the last task change until after grouptask is gone
1...165166167...2420Latest