@enough-analyst-54434: can

pex-tools … extract

be convinced to extract for multiple interpreters at once?

I think we should probably disable our

@skip_unless_python27_present

et al decorators in CI - we should eagerly error in that case, rather than skipping a test. Thoughts on simply deleting them outright? It will make it less likely

./pants test ::

works for desktop developers, but maybe that's fine? Otherwise, we can add a check

if not "CI" in os.environ

Hi guys, we are currently evaluating using pants as company’s build system, our code comprised of multiple services that have different lock files that use shared libraries with their own requirements. I was reading the section on python 3rd party dependencies (https://www.pantsbuild.org/docs/python-third-party-dependencies) and saw that pants doesn’t support locking multiple libraries and it says to to talk our use case

Is this something that could be supported?

We are currently using poetry as the basis for our locking with relative paths for specifying dependencies

is there a way to get more information on a dependency cycle? the error message I’m getting is:

Engine traceback:
  in select
  in pants.backend.go.build.run_go_build
  in pants.backend.go.build.build_target
Traceback (no traceback):
  <pants native internals>
Exception: Dependency graph contained a cycle:
  @rule(pants.backend.go.build.build_target) <-
  @rule(pants.backend.go.build.build_target) <-

which is not very helpful

TIL: python requirements are very cyclic. unconfirmed!

@happy-kitchen-89482: re: the config parsing changes: it would be really great at some point soon to migrate to consuming the rust code for this

argh. last issue for https://github.com/pantsbuild/pants/pull/12675 seems to be a failing test with identical venv and sources as

main

Hmm... Any thoughts on the right release category for "Just some stuff"? e.g. https://github.com/pantsbuild/pants/pull/12673 - it's not a feature, it's not a bug fix, it's just... An internal change, but not internal enough to mark as internal?

If anyone can answer why https://github.com/pantsbuild/pants/pull/12609 is generated in the release notes as

* Embrace that help strings are markdown. ([#12609](<https://github.com/pantsbuild/pants/pull/12609))>

which happens to be the PR title of the next PR in the list (https://github.com/pantsbuild/pants/pull/12606), I'd be curious...

What should I clean up, after upgrading my python version?

@average-vr-56795: how would you feel about replacing

DirectoryMaterializeMetadata

and

LoadMetadata

with metrics instead? we could store histograms of size or load time by source, for example.

Hi Guys, I want do the following, tell me if it’s possible, I want to create a custom rule named

poetry_requirements_from_lock

that parses the

poetry.lock

file and generates

python_requirement_library

, I couldn’t find any documentation on how to do that. Is this API exported?

What's been going on with CI recently? Seeing some flakiness.

fyi new design doc for an “immutable caches” feature for

Process

invocations. comments welcome! tracking issue is https://github.com/pantsbuild/pants/issues/12716 and design doc is https://docs.google.com/document/d/1e4djdm2D9L6IeeJ8Q1M0_yJU-kF4tTHopBomwzdDRpE/edit#heading=h.n4lrkjmz3x2s

@hundreds-father-404: it seems like the two halves of

build-support/bin/generate_all_lockfiles.sh

should be split… regenerating our own lockfile is independent of tool lockfiles

Hi Guys, I’m developing on my mac and want to deploy a python library (with native dependencies) to a linux server. Is there a “run pants in docker” options or it’s something we need to orchestrate?

hm. looks like the console UI is lightly broken on master. maybe the futures update? will check after getting current PR posted.

Hi there 👋 I want to create a VSCode language plugin for Pants BUILD file (if there's one already, plz let me know), but I'm not familiar with Pants code source yet. I want to extract the functions/metadata available for the BUILD file, any suggestions to save some time on my side?

Hm, I realized we can let users have a choice between Pex/pip lockfile generation vs Poetry generation. Poetry generation has caveats like not working with --find-links. But it may help current poetry users to adopt Pants. Worth it? Unclear

Hi, I’m trying to use pants for a dummy hello world target: BUILD:

python_library()

pex_binary(entry_point="main.py", name="bin")

main.py:

print("Hello world")

However running it takes ~2 seconds:

./pants run bins:bin  0.59s user 0.15s system 37% cpu 1.984 total

Am I doing something wrong or should I expect a big penalty from pex binaries?

This is a general question, what's your local workflow with pants? What's your ide? Debug workflow? Do you have any interesting wrappers or integrations?

Just out of curiosity, how much of the rust code in pants relates to bazel? Or rather, I’m curious about the origins of the code.. have we been able to borrow code from bazel, or is it just that it mimics some bazel protocols, and hence the name comes from that.. ? (spotted a few references to bazel protos, digest and remote stuff…)

bunch ’o releases prepped, whenever someone gets the chance to stamp them: • https://github.com/pantsbuild/pants/pull/12755 • https://github.com/pantsbuild/pants/pull/12754 • ~https://github.com/pantsbuild/pants/pull/12753~

i seem to have gotten myself locked out of readme.com: if someone else has time to create the

2.8

version of the site as per https://www.pantsbuild.org/docs/release-process#dev0---set-up-the-new-version, that would be appreciated.

I think the

PythonToolBase

bits of: https://www.pantsbuild.org/docs/rules-api-installing-tools is out of date

entry_point

seems to have changed to

default_main

I’m looking at juggling with paths in digests, using RemovePrefix mainly. Any reason not to make

RemovePrefix

to err optional in case the prefix does not exist, with a flag or so? Otherwise, I’ll do a

DigestSubset

with path globs to pick out the files that have, and another subset that doesn’t have, the prefix, then remove prefix from the former, and then merge them back together again. However, it feels a bit clumsy and inefficient..

@curved-television-6568: do you have things you need picked to

2.7.x

?

@handsome-dawn-63164 has left the channel