Hey peeps! Just curious, will pants 2.16's local Py make resolution of Python binary close to 'native' installs? I run Black on save for example, and have to maintain a dev env in parallel because waiting 2+ seconds just to start formatting is pretty awkward for on-save runs. Also had another issue but I'll post that later when I get to work.

can pants work with conda environments?

Is there a way to pass additional CLI arguments to the

docker build

command generated by a

docker image

target? For some reasons I’m installing dependencies into my image with

pip

. I’d like to be able to cache those layers in the GitHub Actions cache, but to do that I need to set the

--cache-to

and

--cache-from

options on the CLI. See: https://docs.docker.com/build/cache/backends/local/ Is that possible in the

BUILD

or

pants.toml

file? I didn’t see an option that looked like a generic

args

thing, although I do see the appropriate mixin classes in

backend/docker/target_types.py

Good morning/evening, I'm starting to make some attempts at the

incremental adoption

method to get pants into one of our internal repositories. @hundreds-father-404 was kind enough to get me started with an example plugin. I'm moving my questions here though to not bog them down too much. The example plugin exists in its own repo, and I have a seperate target repo I'd like to start integrating pants into, I'm thinking I start a new branch in the target repo and add the pants.toml, but I've got a couple questions 1. When I set up my pants.toml and spell out the new plugin, is pants "system aware" of my plugin? Do I need to add the plugin to a "pants environment" or something to that effect? 2. When I run pants the first time in a repo it initializes some things in the project right? Is there anything that gets generated I need to commit. I'm thinking I should just try this but I'm so new to pants I don't want to break anything about the "pants install". I assume that's not a problem really but better safe to ask first 🙂

By default, does running

./pants fix

run formatters that would’ve been invoked via

./pants fmt

? Another way to ask the question: if I first run

./pants fix

, will running

./pants fmt

afterwards be a no-op?

Hello there! I am trying to do remote caching only to accelerate my pipeline, I am trying to integrate with Buildbuddy, and no matter what I try, Pants doesn't even try to do a request to the remote grpcs host. My

pants.toml

looks like this

[GLOBAL]
pants_version = "2.15.0"

backend_packages = [
    "pants.backend.docker",
    "pants.backend.python",
    "pants.backend.python.lint.flake8",
]

remote_cache_read = true
remote_cache_write = true
remote_store_address = "<grpcs://remote.buildbuddy.io>"
remote_store_headers = "{'x-buildbuddy-api-key': '<token>'}"
remote_instance_name = "main"
remote_store_chunk_upload_timeout_seconds = 3600

When I add the keys for the

remote_execution

I see errors and stuff that actually trying to do something with the remote host, but for caching only, nothing is happening (even if I put a random

remote_store_address

)

Is it expected that every time I change the file of the entry_point of a pex_binary that it will rebuild every requirement each time? I was hoping for something like Docker where the build process is cached at each stage and altering code (which does not require different packages) is very quick if kept as last layer. But when I use

pants run :cli

it will rebuild all the package each time I just change the source code.

does Pants automatically include

conftest.py

in the same way pytest does, even if a test target doesn't refer to any fixtures in

conftest.py

? I am transitioning a repo to use pants, and when I run a target that doesn't use any fixtures it still seems to want to include the conftest file and all its dependencies. Is the only way around this to put the test in a folder that doesn't have conftest.py in one of its parent folders? for example the tests are structured like this: tests/ tests/confest.py tests/foo/bar/test_file.py #doesn't refer to any fixtures in tests/conftest.py

Is there an example on how to use

run_shell_command

with a shell script? I can't seem to get it to load a .sh file next to the BUILD file

Where would I find an updated pants bash script that I can call to run pants. I have a bash script that retrieves the tool and executes whatever pants goal you specify. Last one I have is from a year ago, and I was hoping there was a newer one which has direct support for python 3.11 and drops requirements for 3.7

If I have a pip package that installs a cli tool.. how can I use that in pants? I'm attempting to run meson to do some builds via shell_command

How do I debug system_binary?

system_binary(
    name="git",
    binary_name="git",
)

Leads to

pants run .build:build-libcamera
17:37:36.50 [INFO] Completed: Testing candidate for `git` at `/bin/git`
17:37:36.50 [INFO] Completed: Testing candidate for `git` at `/usr/bin/git`
17:37:36.51 [ERROR] 1 Exception encountered:

Engine traceback:
  in `run` goal

ValueError: Could not find a binary with name `git`. The following paths were searched: /usr/bin, /bin, /usr/local/bin, /opt/homebrew/bin.

But..

which git
/usr/bin/git

Okay okay... I need to run a script with root via

run_shell_command

.. it is to build a c library with ninja and link it with ldconfig. If I have

system_binary(
    name="git",
    binary_name="git",
    fingerprint_args=["version"],
)

adhoc_tool(
    name = "meson",
    runnable=".build/requirements:reqs-libcamera#meson",
)

run_shell_command(
    name="libcamera",
    command="./libcamera.sh",
    runnable_dependencies=[".build/requirements:reqs-libcamera#meson", ":git"]
)

shell_sources(name="shell-scripts")

Where libcamera.sh is

#!/usr/bin/env bash

rm -rf .lc_build_temp

mkdir -p .lc_build_temp
cd .lc_build_temp \
    && git clone <https://github.com/bloop/libcamera.git> \
    && cd libcamera \
    && git checkout v0.0.4 \
    && meson setup build \
    && ninja -C build
cd .lc_build_temp/libcamera && sudo ninja -C build install
sudo ldconfig || echo "Finished ldconfig"

rm -rf .lc_build_temp

It totally works up until it needs a sudo password 🤔 Anyway to have it ask me?

Hello! I'm having some trouble with remote caching. After reading the Pants documentation and browsing the internet, I'm under the impression that to enable remote caching of build results, one has to set up a gRPC solution. Such solutions allow for not only remote caching, but also remote execution, which I don't need (I'm running the builds in my GCP environment). It feels strange to have to set up and maintain a complex solution only to use a small part of its capability (the caching). Is it in some way possible to simply use a file system for storing the cache? In Bazel, for instance, one can simply pass a GCP bucket as a location for storing the cache. Would a similar solution be possible with Pants? Thanks!

hey guys, we're in the middle of switching to using Pants for building Helm artifacts and just wanted to make sure we're not missing something obvious 🙂 • Can anyone confirm that there's currently no way to authenticate with a registry? • Has anyone managed to include a README file into the resulting Hlem artifact?

I'm trying to build a pex_binary for aws lambda on a MacOS machine. Accordingly, I'm using a docker_environment to deal with dependencies.

docker_environment(
    name="aws_lambda_python39",
    platform="linux_x86_64",
    image="public.ecr.aws/lambda/python:3.9",
)

pex_binary(
    name="flagger",
    layout="zipapp",
    inherit_path="fallback",
    dependencies=["src/dz/anomaly_flagger/handler.py"],
    environment="aws_lambda_python39",
)

pants package is failing with

Failed to create Docker execution in container: RequestTimeoutError

I am able to to

docker run --platform=linux/x86_64 public.ecr.aws/lambda/python:3.9

on the machine, and I've got no clue what request is timing out, nor why. Any idea how to proceed?

Let's say I have a

run_shell_command

"foo". Is there anyway I can say that foo must run before bar? Or do I need to use shell_command with outputs?

Is it possible to do something like

./pants help <my macro>

I have another cache related question 🧵

Is there a way to pass environment variables to an

adhoc_tool

target? I’m using

pnpm

as a

system_binary

and I need to pass an argument to the build script. Since I can only do that by using environment variables, I was hoping to be able to add it somehow to

system_binary

or

adhoc_tool

.

Hello, reposting this issue here https://github.com/pantsbuild/pex/issues/2154

Is there a way I can run an adhoc tool or shell_command as a test? In particular, I want to run

npm test

I have npm run build| and npm start working 😃

Hi, I started getting this error when trying to bootstrap Pants:

pants.base.exceptions.BackendConfigurationError: Failed to load the pants.core.register backend: ImportError("urllib3 v2.0 only supports OpenSSL 1.1.1+, currently the 'ssl' module is compiled with LibreSSL 2.8.3. See: <https://github.com/urllib3/urllib3/issues/2168>")

Seems to be related to breaking changes in

urllib3==2.x

, but downgrading it to

1.x

doesn't help. The error appears when I try to run any

pants

command using a binary created before this urrlib update. How do I fix it? Thanks!

When trying to run

pants package

, I'm getting the following error:

#3 [internal] load metadata for <http://docker.io/tensorflow/tensorflow:2.9.1-gpu|docker.io/tensorflow/tensorflow:2.9.1-gpu>
#3 sha256:8d3bd1ccb65ab89bd2786298e6e6e88a2a5e6f1a26f13dec1757448577beca32
Failed to fire hook: while creating logrus local file hook: user: Current requires cgo or $USER, $HOME set in environment
[2023-05-17T11:16:32.965266000Z][docker-credential-desktop][F] get system info: exec: "sw_vers": executable file not found in $PATH
[goroutine 1 [running, locked to thread]:
[common/pkg/system.init.0()
[ common/pkg/system/os_info.go:32 +0x1bc
#3 ERROR: rpc error: code = Unknown desc = error getting credentials - err: exit status 1, out: ``

[internal] load metadata for <http://docker.io/tensorflow/tensorflow:2.9.1-gpu|docker.io/tensorflow/tensorflow:2.9.1-gpu>:

failed to solve with frontend dockerfile.v0: failed to create LLB definition: rpc error: code = Unknown desc = error getting credentials - err: exit status 1, out: ``

This happens for every docker image I'm trying to build, the tensorflow above is just an example. The corresponding

docker build

command to build the same works fine. How do I fix it? Thanks!

I’m trying to run a pants job in a

matrix

on GitHub Actions. However, the keys for the lmdb store are based on the git sha which means they conflict and my caches aren’t merged properly. Is there any way to merge these appropriately or do I just need to convince my company to buy a remote caching service? 😄

Is there a guide anywhere on migrating away from the Python tool lockfiles? Just spent an hour debugging an issue caused by a dependency mismatch between our main lockfiles and pytest's lockfile so I'm feeling motivated to rip them out asap.

Hiii guys, I am trying to deploy a

FastAPI

application packed into a pex executable. It uses

SQLAlchemy

+

Alembic

. In my native dev environment, I'd use alembic to make migration files and apply the migration before starting the app. But since the pex only has one entrypoint which is the main function of the fastapi app, I'm not so sure how to ensure migrations are applied before the app starts up in production.. any ideas?

Ah no worries already found the solution 😬

Hello everyone. I am trying to install pants inside a docker image so when the image is used, pants is ready. During installation (running

pants

after

get-pants.sh

) the

pants

command returns the code 1 and docker will crash due to the non 0 return. However, it seems that the command run successfully. I do get however the

No goals specified.

which could be the code 1? Is this expected? If yes, why does it return that code? Did others managed to install pants and initialize it successfully inside docker?

Hey there! I am trying to accelerate my python projects (we already using pants to package everything with Docker) I would like to see what is happening in the Docker build phase, even if I run

pants -ldebug package ...

I get the

[DEBUG] spawned local process as Some(2605) for Process { argv: ["/usr/bin/docker", "build" ...

message and then

⠓ 79.65s Building docker image

during the build Process, is there a way to expose the logs to see the steps and the layers of the build to see what is taking time and how I can optimize it ? Thanks in advance.