https://pantsbuild.org/ logo
Powered by
#development
  • a

    average-vr-56795

    04/24/2019, 12:30 PM
    @witty-crayon-22786 https://github.com/pantsbuild/pants/pull/6880 introduced a sneaky dep cycle that we didn’t notice because we don’t do separate pythonpaths for test executions… 
    src/python/pants/engine:goal
    should now depend on 
    src/python/pants/cache
    but can’t because that would cause a cycle 😞
    😢 1
  • w

    witty-crayon-22786

    04/24/2019, 4:53 PM
    whomp whomp. i think that 
    --no-fast
    would detect this?
  • w

    witty-crayon-22786

    04/24/2019, 4:53 PM
    @average-vr-56795: is it blocking you? is it possible to not add a dep somewhere for now?
  • a

    average-vr-56795

    04/24/2019, 6:03 PM
    Not blocked 🙂
  • a

    average-vr-56795

    04/24/2019, 6:03 PM
    And I’m sure we can cycle-break at some point
  • a

    average-vr-56795

    04/24/2019, 6:03 PM
    Just a thing I noticed when debugging a thing 🙂
  • a

    aloof-angle-91616

    04/26/2019, 8:55 AM
    https://robert.ocallahan.org/2019/04/update-to-rr-master-to-debug-firefox.html?m=1 (1) mentions LMDB (2) shows how a more frequent release cycle of development tools can work
  • a

    aloof-angle-91616

    04/28/2019, 7:37 PM
    watch @hundreds-father-404 body slam my bash knowledge into the ground https://github.com/pantsbuild/pants/pull/7624#discussion_r279208934
  • h

    hundreds-father-404

    04/28/2019, 7:39 PM
    *watch shellcheck body slam all of our bash knowledge into the ground One of the best tools I’ve ever used. Incredible how sophisticated it is with understanding edge cases and fantastic documentation (failures link to their relevant wiki page)
    😂 2
    💯 1
    🔥 1
  • a

    aloof-angle-91616

    04/28/2019, 7:40 PM
    ssssh don't tell anyone let them think you're one of the mystical bash gurus
    😂 2
  • a

    aloof-angle-91616

    04/28/2019, 7:40 PM
    i've never met a bash guru in my entire life
    😁 1
  • a

    average-vr-56795

    04/29/2019, 3:39 PM
    The native 
    setup_pantsd_logger
    requires that externs have been init’d before it can be called; this is done by initialising the memoized 
    native.context
    in Python… Is there a good way to encode this dependency in code?
    a
    w
    • 3
    • 23
  • a

    average-vr-56795

    04/29/2019, 10:18 PM
    @witty-crayon-22786 async/await in beta in July, maybe... https://internals.rust-lang.org/t/await-syntax-discussion-summary/9914
    🤞 2
  • a

    average-vr-56795

    05/01/2019, 5:06 PM
    Heads up, I’m making 
    Externs
    actually injected into each ffi function into 
    Native
    , rather than having them being global statics 🙂
    w
    • 2
    • 26
  • a

    aloof-angle-91616

    05/01/2019, 7:09 PM
    i'm gonna try to find time this week to open 
    --query
    back up (https://github.com/pantsbuild/pants/pull/7350) after i fix engine stack traces in https://github.com/pantsbuild/pants/pull/7532. may have to wait but that one was really close to being mergeable
    ❤️ 1
  • a

    aloof-angle-91616

    05/02/2019, 3:09 PM
    https://www.sublimetext.com/blog/articles/use-mmap-with-care featuring descriptions of signal handling and more normal operating system features which aren't quite fully supported across every environment
  • a

    average-vr-56795

    05/03/2019, 2:29 PM
    @witty-crayon-22786 Remind me, is there a good reason we still have a separate fs_pool and runtime on Core, or have we just not gotten around to unifying them (and maybe switching to tokio-fs) yet?
    w
    • 2
    • 12
  • w

    witty-crayon-22786

    05/03/2019, 3:14 PM
    noooot.... really? (edit: see above)
  • a

    average-vr-56795

    05/05/2019, 7:10 PM
    Looks like rust is getting something along the lines of outlining: https://github.com/rust-lang/cargo/pull/6883
    🔥 1
    a
    w
    • 3
    • 3
  • w

    witty-crayon-22786

    05/07/2019, 4:56 PM
    the native engine depends on libgit2... um, neat (wat)
    a
    • 2
    • 4
  • a

    average-vr-56795

    05/08/2019, 2:16 PM
    @witty-crayon-22786 @hundreds-father-404 is starting to bump into v2 target API problems, so we should probably sync up on those
    h
    w
    a
    • 4
    • 27
  • w

    witty-crayon-22786

    05/08/2019, 7:07 PM
    we should pretty clearly split this into two flags:
    Copy code
    --process-execution-parallelism=<int> (default: 4)
   Number of concurrent processes that may be executed either locally and
   remotely.
  • w

    witty-crayon-22786

    05/08/2019, 7:07 PM
    not sure why i didn't think about that when we were adding supporting for limiting
  • a

    aloof-angle-91616

    05/08/2019, 7:09 PM
    is there a reason we have this setting at all? why would pants be the one setting the number of remote workers?
  • w

    witty-crayon-22786

    05/08/2019, 7:09 PM
    yea, my feeling is that the remote setting should be very high by default. but i could still imagine capping it.
    a
    • 2
    • 22
  • w

    witty-crayon-22786

    05/08/2019, 7:09 PM
    but yea. two settings.
  • a

    average-vr-56795

    05/09/2019, 6:31 PM
    @witty-crayon-22786 Can you talk me through the enum python-to-rust thing?
    ⁉️ 1
    w
    • 2
    • 8
  • h

    hundreds-father-404

    05/09/2019, 11:56 PM
    Think this is the last major issue to fix before being able to use V2 Pytest with unit tests in CI: https://github.com/pantsbuild/pants/pull/7696 Ideas appreciated on how to approach a fix. Started with writing a failing integration test, and now trying to figure out how to fix it.
  • w

    witty-crayon-22786

    05/09/2019, 11:56 PM
    would take a look at what the failing workspace looks like
    h
    • 2
    • 13
  • w

    witty-crayon-22786

    05/10/2019, 4:28 PM
    @average-vr-56795: added feedback on the prefix-stripping PR. thanks for jumping on that!
    a
    h
    • 3
    • 3
1...624625626...906Latest