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

    witty-crayon-22786

    03/30/2016, 9:12 PM
    @melodic-kite-14017: to pass that to zinc, you probably want it in:
    Copy code
    [compile.zinc]
args: [..]
  • w

    witty-crayon-22786

    03/30/2016, 9:12 PM
    see 
    ./pants help-advanced compile.zinc
  • m

    melodic-kite-14017

    03/30/2016, 9:14 PM
    alright, i'll try that out
  • w

    witty-crayon-22786

    03/30/2016, 9:14 PM
    as an example, our 
    args
    , 
    warning_args
    etc sections look like:
  • w

    witty-crayon-22786

    03/30/2016, 9:15 PM
    Copy code
    args: [
    '-C-encoding', '-CUTF-8',
    '-S-encoding', '-SUTF-8',
    '-S-g:vars',
    # see <https://reviewboard.twitter.biz/r/446503/>
    '-S-language:higherKinds',

    # This is intentionally in the default args section: never warn for generated files.
    '-file-filter', '/\.pants\.d/gen/',
    # Ditto: Suppress the warning for annotations with no processor.
    '-C-Xlint:-processing',
  ]
warning_args: [
    '-S-deprecation',
    '-S-unchecked',
    # request warnings for <http://www.scala-lang.org/api/2.10.4/index.html#scala.language$>
    '-S-feature',
    '-S-Xlint',
    '-C-Xlint:all', '-C-Xlint:-serial', '-C-Xlint:-path',
    '-C-deprecation',
    '-S-Ywarn-unused-import',
  ]
no_warning_args: [
    '-S-nowarn',
    '-C-Xlint:none',
    '-C-nowarn',
  ]
  • w

    witty-crayon-22786

    03/30/2016, 9:15 PM
    where 
    -C
    prefixed are flags for java, and 
    -S
    prefixed are flags for scala
  • w

    wide-energy-11069

    03/30/2016, 9:26 PM
    change the 
    —verify-configs
    flag after https://rbcommons.com/s/twitter/r/3623?
  • w

    witty-crayon-22786

    03/30/2016, 9:27 PM
    hm?
  • w

    wide-energy-11069

    03/30/2016, 9:33 PM
    will put a RB to set 
    —verify-config
    true by default once https://rbcommons.com/s/twitter/r/3623 has landed
  • m

    melodic-kite-14017

    03/30/2016, 9:38 PM
    I got a team member who had never run pants before to run it with my 
    -Xmax-classfile-name: 254
    change and it didn't generate. as soon as he removed that line and reran gen, generate worked, so I think pants might be picking up that change
  • b

    bored-art-40741

    03/30/2016, 9:40 PM
    [jvm_options]
    isn't a thing
  • w

    witty-crayon-22786

    03/30/2016, 9:40 PM
    right.
  • b

    bored-art-40741

    03/30/2016, 9:40 PM
    I assume you want
    Copy code
    [gen.scrooge]
jvm_options: ...
  • w

    witty-crayon-22786

    03/30/2016, 9:40 PM
    @bored-art-40741: it's a scalac option
  • b

    bored-art-40741

    03/30/2016, 9:40 PM
    oh
  • m

    melodic-kite-14017

    03/30/2016, 9:42 PM
    that option wasn't actually for the scrooge generation, it was for a different package which relies on the scrooge generated package
  • b

    bored-art-40741

    03/30/2016, 9:42 PM
    So unrelated question for the peanut gallery here: does anyone actually use the Haskell integration?
  • w

    witty-crayon-22786

    03/30/2016, 9:42 PM
    @bored-art-40741: no.
  • b

    bored-art-40741

    03/30/2016, 9:43 PM
    I think it's a nice proof of concept, but right now this crazy stack bug is blocking the much more important Jenkins integration
  • w

    witty-crayon-22786

    03/30/2016, 9:43 PM
    there isn't much there currently.
  • b

    bored-art-40741

    03/30/2016, 9:43 PM
    And it's unclear to me if this is fixable on our end, or if it's some stack race condition
  • w

    witty-crayon-22786

    03/30/2016, 9:44 PM
    @wide-energy-11069: yea, sounds good
  • b

    bored-art-40741

    03/30/2016, 9:44 PM
    Soooo... how do we feel about deleting it? Or I guess disabling it aggressively
  • w

    wide-energy-11069

    03/30/2016, 9:44 PM
    @witty-crayon-22786: not sure if i understand the comment on https://rbcommons.com/s/twitter/r/3626/, does
    Copy code
    synthetic_jar = safe_classpath(classpath, basedir)
rm_rf(synthetic_jar_path)
os.rename(synthetic_jar[0], synthetic_jar_path)
    have race condition? my understanding is each is line a sync operation
  • w

    witty-crayon-22786

    03/30/2016, 9:45 PM
    @bored-art-40741: could give @freezing-motorcycle-14222 an opportunity to work on fixing it before we disable
  • w

    witty-crayon-22786

    03/30/2016, 9:45 PM
    @wide-energy-11069: the point is just: what you're doing there manually you can do with a helper function instead.
  • b

    bored-art-40741

    03/30/2016, 9:46 PM
    Okay, I'd be happy with that too; didn't know who to ping. Ideally this should be fixed sooner than later though, or at least worked around
  • w

    wide-energy-11069

    03/30/2016, 9:46 PM
    @witty-crayon-22786: gotcha. thx
  • w

    witty-crayon-22786

    03/30/2016, 9:46 PM
    @wide-energy-11069: ... er. wait. it doesn't give you a choice of where the jar goes?
  • b

    bored-art-40741

    03/30/2016, 9:46 PM
    I want to finish this Jenkins proof of concept and get it reliable now that it's coming to decision time
1...111011111112...2420Latest