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

    witty-crayon-22786

    04/07/2017, 12:04 AM
    i didn't think that they should have that effect no. that sounds like they're getting pushed all the way down to ivy
  • p

    plain-sundown-25537

    04/07/2017, 12:15 AM
    hmm~
  • p

    plain-sundown-25537

    04/07/2017, 12:15 AM
    how can I debug this
  • p

    plain-sundown-25537

    04/07/2017, 12:18 AM
    on 1.2.1
  • p

    plain-sundown-25537

    04/07/2017, 12:19 AM
    Copy code
    idea.exclude_folders = ['.pants.d/compile', '.pants.d/ivy', '.pants.d/python', '.pants.d/resources'] (from HARDCODED)
imports.ivy-imports.jvm_options = ['-Xmx256m'] (from HARDCODED)
imports.ivy-imports.nailgun_connect_attempts = 5 (from HARDCODED)
imports.ivy-imports.nailgun_server = //:nailgun-server (from HARDCODED)
imports.ivy-imports.nailgun_timeout_seconds = 10.0 (from HARDCODED)
imports.ivy-imports.soft_excludes = False (from HARDCODED)
imports.ivy-imports.use_nailgun = True (from HARDCODED)
ivy.bootstrap_fetch_timeout_secs = 10 (from HARDCODED)
ivy.bootstrap_ivy_settings = None (from NONE)
ivy.bootstrap_jar_url = <https://repo1.maven.org/maven2/org/apache/ivy/ivy/2.4.0/ivy-2.4.0.jar> (from HARDCODED)
ivy.cache_dir = /Users/leo/.ivy2/pants (from HARDCODED)
ivy.http_proxy = None (from NONE)
ivy.https_proxy = None (from NONE)
ivy.ivy_profile = 2.4.0 (from HARDCODED)
ivy.ivy_settings = build_support/ivy/ivysettings.xml (from CONFIG in pants.ini)
publish.jar.ivy_settings = None (from NONE)
resolve.ivy.args = [] (from NONE)
resolve.ivy.confs = ['default'] (from HARDCODED)
resolve.ivy.jvm_options = ['-Xmx256m'] (from HARDCODED)
resolve.ivy.mutable_pattern = None (from NONE)
resolve.ivy.nailgun_connect_attempts = 5 (from HARDCODED)
resolve.ivy.nailgun_server = //:nailgun-server (from HARDCODED)
resolve.ivy.nailgun_timeout_seconds = 10.0 (from HARDCODED)
resolve.ivy.open = False (from HARDCODED)
resolve.ivy.outdir = None (from NONE)
resolve.ivy.override = [] (from NONE)
resolve.ivy.report = False (from HARDCODED)
resolve.ivy.soft_excludes = False (from HARDCODED)
resolve.ivy.use_nailgun = True (from HARDCODED)
resolve.ivy.xalan = //:xalan (from HARDCODED)
  • f

    fancy-queen-20734

    04/07/2017, 12:28 AM
    soft_excludes False looks promising
  • f

    fancy-queen-20734

    04/07/2017, 12:28 AM
    with it, ivy will be told about the mid graph excludes
  • w

    witty-crayon-22786

    04/07/2017, 12:28 AM
    oh yea!... is that still false by default?
  • f

    fancy-queen-20734

    04/07/2017, 12:29 AM
    apparently
  • w

    witty-crayon-22786

    04/07/2017, 12:29 AM
    good eye.
  • w

    witty-crayon-22786

    04/07/2017, 12:32 AM
  • p

    plain-sundown-25537

    04/07/2017, 12:42 AM
    Should this value be true? I saw it ticket
  • h

    high-advantage-47550

    04/07/2017, 12:42 AM
    (this might be outside the scope of pants) i don’t suppose it’s possible to “reload” a scala repl?
  • h

    high-advantage-47550

    04/07/2017, 12:42 AM
    using ammonite or otherwise
  • l

    lively-nest-25467

    04/07/2017, 1:37 PM
    Hi everyone. Still haven't found a good answer to my question from yesterday. Is it possible to instruct pants to only run java/scala tests with certain annotations, similar to junit categories or Scalatest tags? If not possible out-of-the box, what alternatives would you recommend?
  • e

    enough-analyst-54434

    04/07/2017, 1:41 PM
    It is not. Beyond target tags, which are much coarser as you pointed out yesterday, you'd need to dig in to the pants junit runner Java code to add the capability and then wire up to the python side (pants shells out to a junit runner main).
  • e

    enough-analyst-54434

    04/07/2017, 1:41 PM
  • l

    lively-nest-25467

    04/07/2017, 1:57 PM
    @enough-analyst-54434 Thanks! As I mentioned yesterday, tags are probably enough for our use case, so we'll likely go that route. Is annotation-based testing somewhere on your planned feature list?
  • h

    high-advantage-47550

    04/07/2017, 2:17 PM
    is there a new way to use scalac plugins? i’m getting
    Copy code
    ScalaLibrary received unknown arguments:
  scalac_plugins
  • h

    high-advantage-47550

    04/07/2017, 2:17 PM
    when using the 
    scalac_plugins
    key on a 
    scala_library
    target
  • h

    high-advantage-47550

    04/07/2017, 2:23 PM
    oh, does it not work with 1.2?
  • h

    high-advantage-47550

    04/07/2017, 3:12 PM
    got it working with 
    scalac_plugins: ['macro-paradise-plugin']
    in pants.ini, but can’t seem to get per-target plugins to work
  • s

    straight-waiter-86090

    04/07/2017, 3:39 PM
    hi - does anyone have any tips for improving speed of java compilation? with an empty cache, compiling our java codebase with pants can easily take an hour or more. Our previous ant build took ~10-13 minutes. With pants we have mostly one BUILD per package, where the ant build would compile all .java files in a directory in one step. With the cache enabled, things are better, but change a file with lots of dependencies and building is slow.
  • s

    straight-waiter-86090

    04/07/2017, 3:40 PM
    just wondering if there are common settings that could help, or a method to identify bottlenecks. The timings show time in zinc:compile, but not much other detail.
  • s

    straight-waiter-86090

    04/07/2017, 3:40 PM
    Copy code
    [Cumulative timings]
20480.695	main:compile:zinc:compile
20462.613	main:compile:zinc:compile:zinc
2660.344	main:compile
2660.116	main:compile:zinc
2621.346	background
44.905	background:cache
40.148	background:cache:insert
18.045	main:compile:zinc:unused-check
8.202	main:gen
7.776	main:gen:messagegen
4.712	main:gen:messagegen:execute
3.135	main:setup
  • f

    full-kilobyte-93429

    04/07/2017, 3:45 PM
    one thing we have good experience with is stripping off unused deps and turning on strict deps.
  • f

    full-kilobyte-93429

    04/07/2017, 3:46 PM
    but that won’t give you 5x speed up, hmm
  • s

    straight-waiter-86090

    04/07/2017, 3:46 PM
    we definitely have some unused/incorrect deps logged.
  • f

    full-kilobyte-93429

    04/07/2017, 3:47 PM
    so ant build is less fine grained but smaller number of compiles?
  • b

    billions-airport-86951

    04/07/2017, 3:48 PM
    is anyone using grpc in java with pants?
1...169716981699...2420Latest