Actually on the mypy cache, we can skip mtime checks, which makes a named cache possible... 🧵

Outside of the documentation, are there any blog posts, or good design docs/github issues regarding field sets, targets, and the various types of targets (Wrapped, Coarsened, etc) and when to use them? I've noticed that I spend more time trying to dig around the pants repo for examples of using these concepts, than I am actually writing the plugin/whatever code itself. And inevitably, when I come back to something after a week, I need to re-do a lot of that

hey folks: i’d like to land a change that re-enables “lazy fetch” from the remote cache: https://github.com/pantsbuild/pants/pull/16277 … in cases where the relevant issue reproduces, a test run might hang in a loop rather than completing. but that appears to be the easiest way to get a repro, unfortunately.

am preparing

2.14.0.dev3

! let me know if there is anything that i should wait for.

This looks interesting for our webdev friends, especially the Django and Flask folks. https://til.simonwillison.net/pytest/playwright-pytest Thoughts on potentially adding Playwright support...?

Getting a weird error on pants latest master branch when running

./pants filter src/python/pants/backend/python/lint/::

(for example, count-loc also causes this error).

Exception: Could not identify a process to backtrack to for: Missing digest: Could not walk unknown directory at "pants_plugins.pex/.deps": Was not present in the local store: Digest { hash: Fingerprint<e65d4afa767c4c7fdf67e21c2fe4e316c27575497a10f8d2d33a2798f7277750>, size_bytes: 1113 }

hey folks! i’d like to help @witty-family-13337 and @narrow-vegetable-37489 make progress on their Helm-deploy and OpenAPI reviews. but despite the great design docs and RFCs, i’m personally missing some of the context necessary to give the best reviews there (and a bit too busy/lazy to learn). so maybe we can split up responsibility for tool-semantics vs implementation? if someone(s) is able to be the tool-semantics reviewer for each of these, i’d be happy to review implementation

I noticed Readme's anchors are including code HTML tag 😞 https://www.pantsbuild.org/docs/reference-python_tests#codeoverridescode

(I think we ought to fix it, but also that breaks existing links. Thanks Readme)

FYI: folks are starting to ask technical questions more in #welcome And it's totally fine both for them to ask it there, and for anyone to answer there. However note that since there are a lot more people in #general vs #welcome, gently steering folks into picking up the question over in #general has the benefit of exposing the conversation/answer to more readers...

How come we already have 2.14.x docs? We haven't cut an rc for 2.14 yet

The macos-10.15 runners hosted by GitHub are going away:https://github.com/actions/virtual-environments/issues/5583 I'm prepping a PR to stop main CI bleeding. I already fixed this for Pex last night.

@bitter-ability-32190 Did your pex_binary running changes alter the CWD of the sandboxed run? It should be the repo root, but I'm not sure it is now?

Or rather it should be the user's CWD (almost always the repo root in practice)

Could use some additional brain power regarding dependencies and dependency inference in the OpenAPI backend -> 🧵

hey folks, this MR has been sitting around for a few weeks now. I've rebased and did a bit of cleanup. What's the next step for it? Is there something I still need to do?

Any thoughts on populating the

dependencies

of a

python_requirement

from a PEX lockfile? I'm trying to debug where a transitive 3rdparty Python dep come in, but Pants can't help me because it stops at surface-level dependencies

seeing a mypy cache related non breaking warning in CI https://github.com/pantsbuild/pants/runs/7563783872?check_suite_focus=true#step:9:268

/usr/bin/cp: cannot stat '.cache/mypy_cache/b965f334ce3fc02912aa7a74077b16748a44e843b35d9f1d89c15db9a9754891/3.7/cache.db': No such file or directory
Success: no issues found in 1157 source files
/usr/bin/mv: cannot move '.tmp_cache/mypy_cache/3.7/cache.db' to '.cache/mypy_cache/b965f334ce3fc02912aa7a74077b16748a44e843b35d9f1d89c15db9a9754891/3.7/cache.db': No such file or directory

also seen it locally (ubuntu/macos) a few times. just any FYI since nothing is broken.

Hi folks, as some of you know, GitHub Actions is discontinuing support for macOS 10.15 runners, so we have had to switch CI to build binaries for macOS 11 and up. This means that unless we take the trouble to set up an alternative, we will not be able to release 2.13 (or any point releases in earlier versions) for 10.15. Thoughts? For context, 10.15 (Catalina) was released in October 2019, and has been superseded by 11 (Big Sur) and 12 (Monterey), and the upcoming 13 (Ventura), to be released late 2022. Thoughts on this? Specifically, if you need macOS 10.15 support, let us know now, and let us know if you're able to put time and/or $ into getting 10.15 CI working for longer.

@witty-crayon-22786: re https://github.com/pantsbuild/pants/pull/16300#issuecomment-1195643396, any thoughts on whether I should land your workaround for being able to upgrade async-trait crate?

🧵 on JVM documentation. I'd like to (1) rename that category to "JVM" instead of "Java & Scala" (since we have Kotlin); (2) split the

jvm-overview.md

into Java and Scala specific pages; and (3) move BSP to its own page. Thoughts?

A few of our documentation pages are not versioned, so we need some convention for publishing those, presumably from main, and not cherrypicking them to stable branches

How do y'all feel about renaming

lang-foo

labels to

backend-foo

on GitHub? Docker isn't techincally a language, but I want to make a label 🙂

I am trying to re-run failed test

help_formatter_test.py

in order to debug and fix it but pytest doesn't see it for some reason? what am I doing wrong ?

I think I found a valid use for uncacheable_rule: https://github.com/pantsbuild/pants/issues/16356

Here’s a semi-exciting way to end the week:

chrisjrn@chrisjrns-MacBook-Pro example-python % _PANTS_VERSION_OVERRIDE="2.14.0.dev3" ../pants/dist/src.python.pants.bin/pants_oxidized_experimental/aarch64-apple-darwin/debug/install/pants_oxidized_experimental --no-pantsd help

Pants 2.14.0.dev3

Usage:

  ./pants [option ...] [goal ...] [file/target ...]   Attempt the specified goals on the specified files/targets.
  ./pants help                                        Display this usage message.
… etc etc etc etc …
… etc etc etc etc …

Download at <https://pypi.org/pypi/pantsbuild.pants/2.14.0.dev3>

hey folks, I was looking at an older project of mine that used Tox and was appreciating how easy it was to "add" a "plugin":

[testenv:radon]
deps = radon
commands = 
    radon cc -s --total-average --no-assert -nb src/

I thought it would be great if it were that easy to create a linter in Pants. So I hacked together a prototype here, and it looks like it "works" and all you need for a linter is:

def rules():
    radoncc = make_linter(
        "radoncc",
        "radon",
        "radon==5.1.0",
        ["cc", "-s", "--total-average", "--no-assert", "-n"],
    )

    return radoncc.rules()

2 questions: 1. is this a horrible hack that will explode? (I know it's a horrible hack that needs a lot of cleanup) 2. is this (or these sorts of helpers) something we'd like to include in Pants?

Working on the C/C++

check

backend and ran into a mild quandry. I have the following folder structure in most/all of my projects:

mylibrary/include/mylibrary/foo.h
mylibrary/src/foo.c

app/bar.h
app/bar.c
app/main.cpp

Inside of app/main.cpp, I have something like:

#include "mylibrary/foo.h"
#include "bar.h"

As of today's dependency inference, the

mylibrary

header may or may not be picked up depending on where the BUILD files are located and how they were generated (e.g. 111, recursive globs, etc). I can design a set of build files to make this pass or fail, or I can modify my header include to make

check

pass or fail. If I run

tailor

on my sample repos, they'll fail - because the dependency inference can't presume the additional

include

folder with no additional information, so they expect the headers to be filename only. If I put a build file with a recursive glob in the "include" folder, it works perfectly. In CMake (or other tools), it's pretty trivial to provide a list of include directories for compilation, but that doesn't feel very pants-y and almost feels like it's circumventing dep inference and BUILD files. I'm creating a

cc_library

target which will export headers + a shared/static library, but they will still run into the issue of how to lay out BUILD files in order to ensure the namespace is what I want. I, personally, have no problem creating BUILD files to that effect, but again, not very Pants-y, or automatic. Thoughts?

https://github.com/sethmlarson/pypi-scorecards fyi

👋 - couple of quick PRs for the

pants-jupyter-plugin

repo if anyone has cycles for a review: • Bump black version to fixup master breakage. #45 • Break out an EnvManager class for adhoc path generation cases. #44 (depends on #45 for green CI)