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

    bored-art-40741

    04/19/2016, 7:08 PM
    Oh yeah actually I think this works really cleanly
  • b

    bored-art-40741

    04/19/2016, 7:18 PM
    @witty-crayon-22786 @ancient-stone-50795 Updated that CR quite a bit: https://rbcommons.com/s/twitter/r/3722/
  • b

    bored-art-40741

    04/19/2016, 7:19 PM
    I think this works out to be cleaner, and it's definitely much more generic
  • b

    bored-art-40741

    04/19/2016, 7:20 PM
    Now any task can get eager cache writes just by calling 
    vt.update()
    themselves
  • w

    witty-crayon-22786

    04/19/2016, 7:26 PM
    very nice.
  • a

    ancient-stone-50795

    04/19/2016, 7:27 PM
    I really like it - I had a comment though
  • b

    bored-art-40741

    04/19/2016, 7:37 PM
    @ancient-stone-50795: I assume you meant to ship given your followup comment?
  • b

    bored-art-40741

    04/19/2016, 7:48 PM
    hmmm, looks like this is going to mess up output a bit
  • a

    ancient-stone-50795

    04/19/2016, 7:49 PM
    Yes, ship it
  • b

    bored-art-40741

    04/19/2016, 7:52 PM
    I'm going to propose that maybe we silence this line?
    Copy code
    src/python/pants/task/task.py:538
  • b

    bored-art-40741

    04/19/2016, 7:52 PM
    Copy code
    self._report_targets('Caching artifacts for ', list(targets), '.')
  • b

    bored-art-40741

    04/19/2016, 7:53 PM
    It's now always called with a single target
  • b

    bored-art-40741

    04/19/2016, 7:54 PM
    Any thoughts on that?
  • w

    witty-crayon-22786

    04/19/2016, 8:02 PM
    works for me.
  • w

    witty-crayon-22786

    04/19/2016, 8:03 PM
    well... honestly, i think we need a more holistic approach there
  • w

    witty-crayon-22786

    04/19/2016, 8:03 PM
    there is a ton of output
  • w

    witty-crayon-22786

    04/19/2016, 8:03 PM
    coming from all over the place.
  • b

    bored-art-40741

    04/19/2016, 8:29 PM
    Yeah I'm not sure what to do with it
  • b

    bored-art-40741

    04/19/2016, 8:29 PM
    But honestly I'm trying to keep this change relatively tight
  • b

    bored-art-40741

    04/19/2016, 8:29 PM
    I really want the perf win from this, but I also don't have a ton of time to sink into a big refactor
  • w

    witty-crayon-22786

    04/19/2016, 8:39 PM
    yep
  • b

    bored-art-40741

    04/19/2016, 8:40 PM
    So you're okay with punting on the output issue for now, @witty-crayon-22786 ?
  • w

    witty-crayon-22786

    04/19/2016, 8:40 PM
    yes
  • b

    bored-art-40741

    04/19/2016, 8:40 PM
    Cool
  • b

    bored-art-40741

    04/19/2016, 8:40 PM
    Will be landing this shortly then, CI is about to pass
  • w

    witty-crayon-22786

    04/19/2016, 8:40 PM
    (did it get worse? or just reordered?)
  • b

    bored-art-40741

    04/19/2016, 8:40 PM
    It's reordered and noisier
  • b

    bored-art-40741

    04/19/2016, 8:41 PM
    Now there's going to be one line per artifact written, instead of one line per block of them
  • w

    witty-crayon-22786

    04/19/2016, 8:41 PM
    ah. yea.
  • w

    witty-crayon-22786

    04/19/2016, 8:41 PM
    i'm fine dropping it to 
    debug
1...117811791180...2420Latest