Not sure if a known issue but using a spec-file with the below content in 1.26.0 results in a

Exception message: Both address specs and filesystem specs given. Please use only one type of spec.

wonders if there's a ticket for this:

// TODO: Run this by someone who truly understands software licensing rules.  Maybe this abundance
// of caution is unnecessary, and we can just re-use the name with no license considerations.

@witty-crayon-22786: ⚒️🐭 ...

“whack a mole” 😃

poor moles.

they're really misunderstood, and very cute 😕

@witty-crayon-22786: is this comment talking about the line above it or the line below it? https://github.com/pantsbuild/pants/blob/master/src/java/org/pantsbuild/tools/junit/impl/CompositeRequestRunner.java#L72

Hi all: If you're using Pants v1, and would like your needs and use-cases to inform the v2 roadmap, please fill out this brief survey: https://forms.gle/F1i43tRAyv2JEknz9

Thanks!

I have a project set up like

src/python/thing-i-am-working-on/module.py
src/python/some-other-thing/other_module.py

when I try to run

pants setup-py src/python/thing-i-am-working-on::

I get this error and I’m not sure what to do about it

FAILURE: No exported target owner found for PythonLibrary(BuildFileAddress(src/python/some-other-thing/BUILD, some-other-thing))

Any thoughts on what could be going on here?

i have a python project that uses another repo as a dependency. i have no problems importing this with pip via my requirements.txt file. my dependency is formatted like this:

<repo_name> @ <git+git://github.com/><path_to_repo>

. when i add that to my 3rdparty requirements.txt for pants i get a 'Could not satisfy all requirements' error. does anyone have suggestions?

https://twitter.com/pantsbuild/status/1263164046582312960 The Pants v2 engine is approaching beta for Python! If you're using Pants today for any language, the community would love your feedback to help build a roadmap for what "done" and 2.0 will mean: https://t.co/za6xeOxZ6S Twitter

When calling the register() api to register options:

register("--some-option", type=str, help="bla bla bla.")

Is there a way to specify that the option is required ? (I tried passing required=True) but I am getting an error:

InvalidKwarg: Invalid registration kwarg required.

looking here: https://github.com/pantsbuild/pants/blob/e59706d9aef5c0d6ca528478ccd8b2e6eb43c739/src/python/pants/option/parser.py#L549 nothing seems like an obvious choice.

@hundreds-father-404 are u sure this cover all the various pants releases https://github.com/pantsbuild/pants/pull/9838 ? For example, I see this: https://pypi.org/project/pantsbuild.pants/1.29.0.dev0/#files which seems to only include py 3.6

Thoughts on disabling the pip progress bars (

--progress-bar off

) in the pants bootstrap script? https://github.com/pantsbuild/setup/blob/gh-pages/pants#L173 Happy to submit a PR.. We use CircleCI which presents itself as a TTY, so it’s hard to navigate stdout if we don’t have a cached bootstrap of pants. Not really pants’ fault of course but I don’t think progress bars are strictly better even in interactive shells. The good news is that it’s easy for users to tweak the bootstrap script if they like.

The value of the dependencies field is a list of addresses of other targets. If code in target A directly depends on code in target B (e.g., by importing a symbol from B) then A should declare a dependency on B. There is no need to declare indirect dependencies (that is, transitive dependencies of your direct dependencies).

I think it would be helpful to mention that there are two types of dependencies: 3rdparty and other monorepo target.

UI observation: (it might just be me so engage the consultants to do human studies before making a decision) when I want a list of targets for my repo I type

pants targets ::

first. This gives me a list of target types. I then think or use the help to remember that the goal I want is

lists

. As needing a list of target types is less frequently needed then a list of targets, I would argue that

targets

is more appropriate for listing the targets in a build. Is there a perspective on this that I am missing?

Hi all, I am having some problems with wheels when building PEX for a wheel I have built. I get the following exception:

Exception message: Package SourcePackage('file:///Users/jay.lawrence/code/pantstheon/.pants.d/python-setup/resolved_requirements/CPython-3.7.7/PyYAML-5.3.1.tar.gz') is not translateable by ChainedTranslator(WheelTranslator, EggTranslator, SourceTranslator)

I have run into another issue:

09:31:56 00:13       [create-monolithic-jar]
09:31:56 00:13         [add-internal-classes]
09:31:56 00:13         [add-dependency-jars]
                       ==== stdout ====

                       ==== stderr ====
                       java.lang.IllegalAccessError: class org.pantsbuild.tools.jar.JarBuilder$7 tried to access method 'com.google.common.collect.TreeTraverser com.google.common.io.Files.fileTreeTraverser()' (org.pantsbuild.tools.jar.JarBuilder$7 and com.google.common.io.Files are in unnamed module of loader 'app')
                       	at org.pantsbuild.tools.jar.JarBuilder$7.execute(JarBuilder.java:743)
                       	at org.pantsbuild.tools.jar.JarBuilder.getAdditions(JarBuilder.java:1188)
                       	at org.pantsbuild.tools.jar.JarBuilder.getEntries(JarBuilder.java:1093)
                       	at org.pantsbuild.tools.jar.JarBuilder.write(JarBuilder.java:1009)
                       	at org.pantsbuild.tools.jar.Main.doRun(Main.java:401)
                       	at org.pantsbuild.tools.jar.Main.run(Main.java:367)
                       	at org.pantsbuild.tools.jar.Main.main(Main.java:481)
                       	at java.base/jdk.internal.reflect.NativeMethodAccessorImpl.invoke0(Native Method)
                       	at java.base/jdk.internal.reflect.NativeMethodAccessorImpl.invoke(NativeMethodAccessorImpl.java:62)
                       	at java.base/jdk.internal.reflect.DelegatingMethodAccessorImpl.invoke(DelegatingMethodAccessorImpl.java:43)
                       	at java.base/java.lang.reflect.Method.invoke(Method.java:566)
                       	at com.martiansoftware.nailgun.NGSession.run(NGSession.java:280)

FAILURE: jar-tool failed

This happens for all cases when I try to bundle/jar a JVM/Java target. I have recently upgraded from 1.17 to 1.26 (migrating py2 to py3). Hopefully there is something simple that I need to change.

The TOML web linter that I believe our documentation recommends to use chokes on valid

pants.toml

format:

[GLOBAL]

backend_packages2.add = [
	"loop",
    "de",
]

Returns

Expected "=", Whitespace, or [a-z], [A-Z], [0-9], "-", "_" but "." found.

for the third line. Removing

.add

produces valid TOML according to this linter. Is the linter broken? (https://www.toml-lint.com/)

is there a way to consistently force pants to use my locally-built pyenv python intepreter? i've set this line in pants.ini

interpreter_search_paths = ['<PYENV>', '<PEXRC>', '<PATH>']

is there a way to force pants to use a desired interpreter? or even reliably output which one it's using?

adding

-ldebug

to the command line should output the selected interpreter as a debug statement in the select interpreter task. i haven’t used the <PYENV> syntax so i’ll have to let someone else take that.

Hello! Does

interpreter_search_paths = ['<PYENV>']

work? That will only expose Pyenv paths to Pants. Also, which version of Pants are you using? We recently added

<PYENV_LOCAL>

so that running

pyenv local

will be picked up by Pants.

mm. i had also opened https://github.com/pantsbuild/pants/issues/9760 about this (if you’re trying to configure machine-wide vs per-repository)

i think it was caching the interpreter

i ran

./pants clean-all

and then the repl command again and got the one i wanted

Great, glad it’s working now.

Let us know if you run into any other issues or have any feedback 🙂

@flat-zoo-31952: gotcha. yea, the v1 python backend has a few rough edges in that department. if your repository is pure python, you might consider trying out v2

it's not pure python (or at least that's not the plan)