pantsbuild #general

Channels

welcome

github-notifications

announce

witty-crayon-22786

01/11/2021, 11:14 PM

doing it in a v1 Target constructor is safe as well, actually. it’s basically just that you can’t do it in a macro.

loud-stone-83419

01/11/2021, 11:18 PM

That gives me food for thoughts, thx!

witty-crayon-22786

01/11/2021, 11:19 PM

yep, sorry for totally misleading you at the start 😃

loud-stone-83419

01/11/2021, 11:19 PM

no worries 🙂

enough-postman-51783

01/12/2021, 3:51 PM

Hi, I have some question. Is it possible to have two

requirements.txt

files in separated folders for separated packages and with the same dependency (3rd part)? Because when I want to add the same dependency in second

requirements.txt

I lose the first one. I saw in docs

Dependency inference will no-op if >2 targets refer to the same file

section.

loud-stone-83419

01/12/2021, 4:51 PM

Hi guys, Other day, new issue 🙂 We realized that we weren't processing the coverage tests properly: we are trying to have the CI always run test with coverage. We realized that the scoverage plugin is outputing several reports for the same class, we upload all of them to codecov but they only pick up one per class. If that one is not the one of the last batch of test, then it will show 0 coverage for the files that got tested in a different test batch. Looking through the code it looks like this is something known, ie the

--output-as-cobertura"

flag describes as

Export cobertura formats which would allow users to merge with cobertura coverage for java targets.

The question thus is the following, historically how have you been solving this? Do you have a tool that can do merges of reports that would prevent the batches from being a problem? thank you

echoing-nightfall-72095

01/13/2021, 11:39 AM

Hi, I'm trying to set up pants for a mono repo. The apps are written python and using the flask framework. I have a directory called

apps

and for now a single app in a subdirectory

sample

in this a

app.py

When executing

./pants run apps/sample/app.py

pants. build as expected but, I'm not able to run flask in debug mode.

Copy code

if __name__ == "__main__":
   app.run(host='localhost', port=7002, debug=True)

Then, I get the following error message:

Copy code

No module named apps.__main__; 'apps' is a package and cannot be directly executed

It runs without errors when:

Copy code

if __name__ == "__main__":
   app.run(host='localhost', port=7002)

I've tried googling for this issue, but with no luck. Hence now reaching out. It would be nice not having to

ctr-c

and re-run whenever changes are made to the app when developing. Hope to hear from you! BR Alex

polite-vase-75369

01/13/2021, 5:26 PM

Has anyone ever had problems running a pex when your PEX_PYTHON_PATH points to a pyenv virtualenv?

polite-vase-75369

01/13/2021, 5:28 PM

I see

Copy code

File "job-scheduler.pex/.bootstrap/pex/bootstrap.py", line 68, in imported_from_bootstrap
  File "//home/ubuntu/.pyenv/versions/2.7.14/lib/python2.7/email/__init__.py", line 79, in __getattr__
    __import__(self.__name__)
  File "/home/ubuntu/.pyenv/versions/2.7.14/envs/pants_1_16/lib/python2.7/site-packages/_virtualenv.py", line 112, in find_module
    if fullname in _DISTUTILS_PATCH:
TypeError: argument of type 'NoneType' is not iterable

polite-vase-75369

01/13/2021, 5:36 PM

I added a println to that virtualenv file, something is unsetting

_DISTUTILS_PATCH

enough-analyst-54434

01/13/2021, 5:48 PM

What version of Pex are you using? 2.1.22 added support for virtualenv detection that should avoid issues like this: https://github.com/pantsbuild/pex/releases/tag/v2.1.22

polite-vase-75369

01/13/2021, 5:53 PM

2.1.9 😞

polite-vase-75369

01/13/2021, 5:53 PM

can i upgrade pex without upgrading pants?

enough-analyst-54434

01/13/2021, 5:55 PM

I'm AFK but it's unlikely. What version of Pants? I can dig a bit this afternoon.

polite-vase-75369

01/13/2021, 6:00 PM

1.27

polite-vase-75369

01/13/2021, 6:00 PM

Copy code

ERROR: Cannot install pantsbuild-pants==1.27.0 and pex==2.1.22 because these package versions have conflicting dependencies.

The conflict is caused by:
    The user requested pex==2.1.22
    pantsbuild-pants 1.27.0 depends on pex==2.1.9

enough-analyst-54434

01/13/2021, 6:37 PM

I backported Pex 2.1.24 to Pants 1.30.x if you are in a position to upgrade Pants.

polite-vase-75369

01/13/2021, 6:37 PM

haha just upgraded to 27 and that broke this so not in the mood to upgrade again 😛

polite-vase-75369

01/13/2021, 6:40 PM

if i wanted to backport to 27, what would i need to do?

loud-stone-83419

01/13/2021, 6:44 PM

Hi! Is there a way to tell pants to ignore cache for running tests? I see that I can pass flags for specific test runners, not sure if there is a global one

enough-analyst-54434

01/13/2021, 7:30 PM

You'd be doing your own private release of Pants 1st off. We don't support older than 1.30.x with backports. Assuming you're willing to do that, you'd probably best start by looking at the 1.30.x backport PR here: https://github.com/pantsbuild/pants/commit/4ea3a2ed14c49e0b80180f9d883674f89392a8f3

enough-analyst-54434

01/13/2021, 7:31 PM

I think your change wouldn't be substantially different from that one. I'd be happy to code review / answer questions that come up.

polite-vase-75369

01/13/2021, 7:34 PM

let me see what i can do. kinda burnt out from upgrading to 1.27 tbh

enough-analyst-54434

01/13/2021, 8:10 PM

I'm guessing the upgrade from 1.27 -> 1.30 is easier than the backport and manual release, but I can understand the burnout.

polite-vase-75369

01/13/2021, 8:12 PM

i think it might be easier for me to tell people to not use pyenv virtualenvs… which seems like an env within an env

polite-vase-75369

01/13/2021, 8:13 PM

and upgrade to 1.30 in the future

enough-analyst-54434

01/13/2021, 8:14 PM

SGTM

square-oxygen-75288

01/14/2021, 10:05 AM

Hello! I have a question regarding Scalatest 3.1.0+ on PANTS 1.30 Everything was working fine with Scalatest 3.0.8, but now, having it bumped to 3.1.0 (or 3.2.0, it does not matter), we are experiencing the following issues:

Copy code

11:01:39 00:12       [run]
                     Auto-detected 32 processors, using -parallel-threads=32
                     .E.E.E.E
                     Time: 0.021
                     There were 4 failures:
                     1) initializationError
                     java.lang.NullPointerException
                     	at __shaded_by_pants__.org.pantsbuild.tools.junit.impl.ScalaTestUtil.getJUnitRunner(ScalaTestUtil.java:62)
                     	at __shaded_by_pants__.org.pantsbuild.tools.junit.impl.CustomAnnotationBuilder$ScalaTestAnnotatedBuilder.runnerForClass(CustomAnnotationBuilder.java:52)
                     	at org.junit.runners.model.RunnerBuilder.safeRunnerForClass(RunnerBuilder.java:59)
                     	at org.junit.internal.builders.AllDefaultPossibilitiesBuilder.runnerForClass(AllDefaultPossibilitiesBuilder.java:26)
                     	at org.junit.runners.model.RunnerBuilder.safeRunnerForClass(RunnerBuilder.java:59)
                     	at __shaded_by_pants__.org.pantsbuild.tools.junit.impl.AnnotatedClassRequest.getRunner(AnnotatedClassRequest.java:41)
                     	at __shaded_by_pants__.org.pantsbuild.tools.junit.impl.CompositeRequestRunner.runChild(CompositeRequestRunner.java:53)
                     	at __shaded_by_pants__.org.pantsbuild.tools.junit.impl.ConcurrentCompositeRequestRunner$1$1.run(ConcurrentCompositeRequestRunner.java:37)
                     	at __shaded_by_pants__.org.pantsbuild.tools.junit.impl.ConcurrentRunnerScheduler.finished(ConcurrentRunnerScheduler.java:89)
                     	at __shaded_by_pants__.org.pantsbuild.tools.junit.impl.ConcurrentCompositeRequestRunner$1.evaluate(ConcurrentCompositeRequestRunner.java:46)
                     	at org.junit.runners.ParentRunner.run(ParentRunner.java:363)
                     	at org.junit.runner.JUnitCore.run(JUnitCore.java:137)
                     	at __shaded_by_pants__.org.pantsbuild.tools.junit.impl.ConsoleRunnerImpl.runLegacy(ConsoleRunnerImpl.java:613)
                     	at __shaded_by_pants__.org.pantsbuild.tools.junit.impl.ConsoleRunnerImpl.run(ConsoleRunnerImpl.java:503)
                     	at __shaded_by_pants__.org.pantsbuild.tools.junit.impl.ConsoleRunnerImpl.main(ConsoleRunnerImpl.java:790)

is it something that just needs some additional configuration which we are missing currently?

echoing-nightfall-72095

01/14/2021, 1:39 PM

Hi again, I’ve a question regarding the repo structure considering importing and linting: my repo has mainly two folders “apps” that contain subfolders of all apps let call them “a”, “b and “c”. Additionally, a libs folder containing “z”, “y”, “x”. in my

pants.toml

i’ve defined roots

Copy code

[source]
root_patterns = [
    "/apps/*",
    "/libs"
]

I can run my apps with:

./pants run apps/a

However, all imports cannot be found using linting (IDE is not happy either). I have include in dependencies

Copy code

python_library(
    dependencies=[
        "3rdparty/python:flask",,
        "libs/z",
        "libs/y",
        "libs/x"
        ],
)

I import as:

Copy code

from z import something

and

Copy code

from sublib import somthingelse

where the sublib is defined in a

BUILD

file using

python_library()

Everything runs great but I would like full support. When I look at the documentation I can see an example of multiple top level project https://www.pantsbuild.org/docs/source-roots#multiple-top-level-projects. Is there a way to make project subtop level and still have linters finding imports and stuff? BR Alex

hundreds-father-404

01/14/2021, 3:42 PM

Hello! One of the most common gotchas with new users is confusion with

python_library

, e.g. thinking it's more meaningful than it really is like that it corresponds to a Python distribution (setup.py). Proposal to fix this by renaming some targets https://docs.google.com/document/d/1afm8NhCuE19YyNZHJCrig9EnaV0eJrSsIVWXL7eDflA/edit#. tl;dr: -

python_library

python_sources

protobuf_library

protobuf_sources

python_requirement_library

python_requirement

Ack that this is disruptive; we'd provide a snippet to fix this automatically, like when we renamed

python_binary

pex_binary

. Feedback appreciated!

🙌 2

🤔 3