pantsbuild #development

Channels

welcome

github-notifications

development

average-vr-56795

11/16/2017, 4:02 PM

@witty-crayon-22786 I think I managed to completely replace tar files with Directories on the plane before my laptop battery died. I'm just splitting it out into some reviewable chunks now 🙂

witty-crayon-22786

11/16/2017, 5:16 PM

metal

average-vr-56795

11/16/2017, 5:18 PM

Yeah, the stack of PRs rolling up to https://github.com/pantsbuild/pants/pull/5106 🙂 the top PR has a lot of topics for discussion. I think the stuff below that PR is pretty much good to go

witty-crayon-22786

11/16/2017, 6:09 PM

@average-vr-56795: looks great! https://github.com/pantsbuild/pants/pull/5104 will need a rebase... happy to do that if you're busy.

average-vr-56795

11/17/2017, 3:20 PM

@witty-crayon-22786 Finally got around to reviewing https://github.com/pantsbuild/pants/pull/5097 (sorry for the delay). Looks good, but noticed that https://github.com/pantsbuild/pants/pull/5112 could be done too 🙂

average-vr-56795

11/17/2017, 3:20 PM

(Also filed a bug against the rust type checker in the process: https://github.com/rust-lang/rust/issues/46061)

average-vr-56795

11/17/2017, 3:34 PM

Ooh, the other bug I was about to file has been fixed in 1.21.0! Going to upgrade us 🙂

average-vr-56795

11/20/2017, 11:10 AM

@witty-crayon-22786 I filed https://github.com/alexcrichton/futures-rs/issues/649 because it makes me sad that we need to track forks explicitly 😞

rough-minister-58256

11/20/2017, 7:27 PM

hmm

wide-energy-11069

11/20/2017, 7:27 PM

can anyone help running the s3-sync for pantsbuild/binaries?

rough-minister-58256

11/20/2017, 7:28 PM

seem to be seeing a daemon deadlock that afaict only happens in source

rough-minister-58256

11/20/2017, 7:28 PM

Copy code

Current thread 0x000070000a983000 (most recent call first):
      File "/Users/kwilson/Python/CPython-2.7.13/lib/python2.7/logging/__init__.py", line 882 in emit
      File "/Users/kwilson/Python/CPython-2.7.13/lib/python2.7/logging/__init__.py", line 759 in handle
      File "/Users/kwilson/Python/CPython-2.7.13/lib/python2.7/logging/__init__.py", line 1336 in callHandlers
      File "/Users/kwilson/Python/CPython-2.7.13/lib/python2.7/logging/__init__.py", line 1296 in handle
      File "/Users/kwilson/Python/CPython-2.7.13/lib/python2.7/logging/__init__.py", line 1286 in _log
      File "/Users/kwilson/Python/CPython-2.7.13/lib/python2.7/logging/__init__.py", line 1155 in debug
      File "/Users/kwilson/dev/release/twitter-pants/build-support/pants_dev_deps.venv/lib/python2.7/site-packages/urllib3/connectionpool.py", line 208 in _new_conn
      File "/Users/kwilson/dev/release/twitter-pants/build-support/pants_dev_deps.venv/lib/python2.7/site-packages/urllib3/connectionpool.py", line 251 in _get_conn
      File "/Users/kwilson/dev/release/twitter-pants/build-support/pants_dev_deps.venv/lib/python2.7/site-packages/urllib3/connectionpool.py", line 589 in urlopen
      File "/Users/kwilson/dev/release/twitter-pants/build-support/pants_dev_deps.venv/lib/python2.7/site-packages/requests/adapters.py", line 440 in send
      File "/Users/kwilson/dev/release/twitter-pants/build-support/pants_dev_deps.venv/lib/python2.7/site-packages/requests/sessions.py", line 618 in send
      File "/Users/kwilson/dev/release/twitter-pants/build-support/pants_dev_deps.venv/lib/python2.7/site-packages/requests/sessions.py", line 508 in request
      File "/Users/kwilson/dev/release/twitter-pants/build-support/pants_dev_deps.venv/lib/python2.7/site-packages/requests/api.py", line 58 in request
      File "/Users/kwilson/dev/release/twitter-pants/build-support/pants_dev_deps.venv/lib/python2.7/site-packages/requests/api.py", line 112 in post
      File "/Users/kwilson/dev/release/twitter-pants/src/python/pants/goal/run_tracker.py", line 280 in post_stats
      File "/Users/kwilson/dev/release/twitter-pants/src/python/pants/goal/run_tracker.py", line 329 in store_stats
      File "/Users/kwilson/dev/release/twitter-pants/src/python/pants/goal/run_tracker.py", line 377 in end
      File "/Users/kwilson/dev/release/twitter-pants/src/python/pants/bin/local_pants_runner.py", line 85 in _run
      File "/Users/kwilson/dev/release/twitter-pants/src/python/pants/bin/local_pants_runner.py", line 37 in run
      File "/Users/kwilson/dev/release/twitter-pants/src/python/pants/bin/daemon_pants_runner.py", line 167 in post_fork_child
      File "/Users/kwilson/dev/release/twitter-pants/src/python/pants/pantsd/process_manager.py", line 451 in daemonize
      File "/Users/kwilson/dev/release/twitter-pants/src/python/pants/bin/daemon_pants_runner.py", line 128 in run
      File "/Users/kwilson/dev/release/twitter-pants/src/python/pants/pantsd/pailgun_server.py", line 58 in _run_pants
      File "/Users/kwilson/dev/release/twitter-pants/src/python/pants/pantsd/pailgun_server.py", line 81 in handle
      File "/Users/kwilson/dev/release/twitter-pants/src/python/pants/pantsd/pailgun_server.py", line 41 in handle_request
      File "/Users/kwilson/dev/release/twitter-pants/src/python/pants/pantsd/pailgun_server.py", line 135 in process_request
      File "/Users/kwilson/Python/CPython-2.7.13/lib/python2.7/SocketServer.py", line 290 in _handle_request_noblock
      File "/Users/kwilson/Python/CPython-2.7.13/lib/python2.7/SocketServer.py", line 275 in handle_request
      File "/Users/kwilson/dev/release/twitter-pants/src/python/pants/pantsd/service/pailgun_service.py", line 94 in run
      ...

rough-minister-58256

11/20/2017, 7:30 PM

wonder if this relates to the stdin/repl patch

witty-crayon-22786

11/20/2017, 7:55 PM

TIL:

does the right thing in the context of a closure that returns a

Result

witty-crayon-22786

11/20/2017, 7:56 PM

and

Future::and_then

supports closures that have a return type of

IntoFuture

witty-crayon-22786

11/20/2017, 7:56 PM

meaning that your

and_then

closure can return either a

Result

or a

Future

witty-crayon-22786

11/20/2017, 8:04 PM

so, doesn't allow you to use

on a

Future

(wishlist item!) but does make it less awkward to do synchronous error handling with

Result

in the context of a

Future

average-vr-56795

11/22/2017, 3:45 PM

Is there a nice Futuresy pattern for: (foo, bar) -> (foo, Future<baz>) -> merged(foo, baz) ?

average-vr-56795

11/22/2017, 3:45 PM

Right now I'm creating a Vec<foo> and a Vec<Future<baz>> and join_all-ing on the second vec, and then zipping the iters

average-vr-56795

11/22/2017, 3:46 PM

Which feels fiddly

average-vr-56795

11/22/2017, 3:46 PM

Sorry, that was a really unclear example

average-vr-56795

11/22/2017, 3:46 PM

I have a list of Foos

average-vr-56795

11/22/2017, 3:47 PM

For each Foo, I want to extract part of the Foo, do some Futurey computation on it, then create a derived value which is composed of some of the original Foo, and the resolved future value

witty-crayon-22786

11/27/2017, 5:48 PM

responded on the review, but: fully converting to one unified type (

(foo, bar)

i guess), and only then `Future::join`ing and

Iter::concat

ing

average-vr-56795

11/27/2017, 7:54 PM

@witty-crayon-22786 I'm guessing the rust style generally leans more heavily towards "each of my futures returns a Vec, and I merge them" than "Each of my futures acquires a Mutex<Arc<Vec>> and appends to it", right?

witty-crayon-22786

11/27/2017, 8:51 PM

@average-vr-56795: the functional style would in general, yea.

witty-crayon-22786

11/27/2017, 8:51 PM

hard to say whether that would be more or less performant though.

aloof-angle-91616

11/29/2017, 7:52 AM

aloof-angle-91616

11/29/2017, 7:52 AM

excuse that, new wireless keyboard

average-vr-56795

11/30/2017, 11:42 AM

looks forward to impl Trait