Has any one used cache on nfs server. I just configured Jenkins with kubernetes pods as agents and /root/.cache/pants on nfs server but I am disapointed by the performance. Maybe NFS is not such a good use for caching due to many files in tha cache?

If I am using

grpc

with both python and golang code, and I run

./pants export-codegen ::

it only exports the Go files. Before I added Go, it was exporting python files. Is it possible to get it to import both?

The

~/.cache/pants/lmdb_store

portion of the cache definitely won't work over NFS. See more here: https://github.com/pantsbuild/pants/issues/14599#issuecomment-1051182009

Thanks John, Is the recomended way of cache reuse to implement remote caching or there are different options.

It is. The remote caching protocol is open (https://github.com/bazelbuild/remote-apis) and there are several open source implementations. You can also use several service providers, Toolchain being the one many of the core maintainers work at: https://toolchain.com/product

Q. Is there a way to programmatically retrieve the Python interpreter that Pants is using? (version, path, etc.)

Thanks

Managed to setup buildbarn, testing how it performs

I don’t use the IDE debugger functionality, but it’s been requested by one of my devs that we continue to support it. Has anyone managed to get this functionality to work with pants? /currently/ they can run tests with their IDE, but I think that’s just the result of moving from non-pants to pants and not changing the source. But given I’ve never used it, I’m really not sure how it works, or how it’d hook in with pants-managed systems.

What programming language?

And which IDE?

I see that https://www.pantsbuild.org/docs/welcome-to-pants#which-languages-and-frameworks-does-pants-support mentions Avro, but that’s the only mention in the docs, and the list of targets doesn’t have anything that sounds like avro. Is this just a leftover in the docs, or is there really something hidden in the source?

Hi everyone! I am trying to package a Python application used for manufacturing HW devices into a PEX file. One module of the app calls a makefile that in turn calls some small Python scripts. The problem is that these scripts require Python packages, that are available in the PEX environment, but it appears that they are not on python include path when

/usr/bin/env python

is called from the makefile. Is there a way to give the same PEX environment ("installed" packages) to python scripts called this way? 🤔

After switching to using pex generated lockfile, my tests take too long to run. It's like 5 times (went from ~6 mins to ~29 mins). I am using pants 2.11.0 It's the exact set of dependencies as before but now I am using multiple resolves. Any debugging pointers to figure out why? This is the earlier thread https://pantsbuild.slack.com/archives/C046T6T9U/p1653655364539299

https://github.com/lablup/backend.ai/commit/f43b7966e9c14f83fba203f532bc9d93e504d900 Now, it's done. (But I think there will be more follow-up works of course as always. 😉 ) Big thanks to @curved-television-6568, @happy-kitchen-89482, @hundreds-father-404, @witty-crayon-22786, and all others.

I've had success using PyCharm with Python remote debugging against a Pants-run test or process

See https://www.pantsbuild.org/docs/python-run-goal#debugging

Pants-powered open-source projects are a bit of a pain to hack on in a corporate environment as brought to light here: https://github.com/pantsbuild/pants/issues/15675

I'm going to be building various docker images based on an entry point of a given module in a Python project, but I also need to build a pair of images that only rely on third party dependencies. I could just write a Dockerfile that does that, but I would prefer to rely on the poetry lock file in the repo, so I was going to try to have Pants handle that build as well. How do I tell it to install those third-party requirements? What I've got so far (doesn't do what I need yet) is this:

docker_image(
    name="ol-dagster-daemon",
    dependencies=[
        "//:requirements#dagster",
        "//:requirements#dagster-postgres",
    ],
    instructions=[
        "FROM python:3.9-slim",
        "WORKDIR /opt",
        "COPY . .",
    ]
)

We'd like to get Pants maintainers guesting on one or two fintech software engineering podcasts. If there are podcasts that you especially appreciate as an engineer working in fintech, we'd be grateful to hear your suggestions. Thanks so much!

Hey Pants team, I'm looking to improve the tailoring process for a project by ensuring dependencies are set for targets that are not imported. For example, a case like

api.py

in a module, we want the tests for this module to have this as a dependency

python_tests(dependencies=[app/module_1/api.py])

. I was thinking either a custom plugin (maybe somehow hooks into to tailor) or possibly a macro. Curious if anyone has done something similar or have a suggestion on best approach for this.

I submitted a new ticket where pants seemed to have issues determining the tag version of a pytho wheel I need to provide as a dependency. I was advised to open a ticket for this issue as it's currently blocking me from upgrading the pants version to 2.11. Ticket: https://github.com/pantsbuild/pants/issues/15724 Thread: https://pantsbuild.slack.com/archives/C046T6T9U/p1653679339273349

That wheel tag is still bad. It combines non platform specific

py36

with a specific platform

linux...

. It might be good to include the full auditwheel command line and output you used to repair the wheel on the ticket.

From the output of auditwheel in the prior thread it looks like the wheel tag triple should be

cp36-cp36m-manylinux_2_24_x86_64

.

Is it possible to get pants to use my existing

poetry.lock

as the lockfile when pulling dependencies? For example I'm trying to generate a

pex_binary

from a dependency specified in

pyproject.toml

and locked in

poetry.lock

, but when I run the package goal it seems to be trying to resolve the versions directly. When I looked at trying to use the

resolves

and set the lockfile type to

poetry

it doesn't seem to be using the same Python interpreter as what exists in the Poetry venv...

Anyone have some thoughts on how to make the test output blob more legible?

✓ path/to/my/project/and/my/module/test_that_my_module_works.py:tests succeeded in 54.06s (cached locally).

At the very least I'm thinking we could apply colors to a few select atoms of data here (like the test name and the process metadata source) Maybe

succeeded

and

failed

could also be color at least to break the monocolor line. (The check or X on the waaaay left doesn't help split the line visually)

Trying to avoid another channel in general, but after the suggestion to destroy lmdb_store (which solved my initial issue), the python code dependent on java.io now fails saying it can't access

libjava.so

which was never an issue before. I can run the same utests in pants 1.30 and have them work, so I'm curious if pants 2.11 needs some extra care when having python interact with java

pycharm running pytest tests.

Is it possible to configure

./pants tailor

,

./pants test ::

, etc to only operate on files tracked in Git?

is there a way to make (some) tests run serially? we have some tests that run against a live system and can't be run in parallel. At the moment we're doing two calls to

./pants test

with the non-parallel ones being called with

--test-debug

or whatever the flag is. But we're adding more live-systems that can be run in parallel with each other, and as things grow everything's getting slower by multiples. e.g. tests

a_1, a_2, a_3

must be run serially. and

b_1, b_2, b_3

must be run serially. But

a_1

and

b_1

could be run in parallel. If not supported OOTB, is there a path to implementing this with a plugin?