How can I pass

PYTHONOPTIMIZE

to a pex file? Some of our code depends on

__debug__

- when I run

PYTHONOPTIMIZE=1 python

, then

__debug__

becomes false. But it seems that when I run

PYTHONOPTIMIZE=1 file.pex

,

__debug__

remains true. The same result with

python -O file.pex

🙂 Not sure how this is supposed to work... But I would like to turn on optimizations for production images based on pex files. Thanks for hints.

If you either build your PEX file with

--venv

or else you install your PEX as a venv using pex-tools in your image, this works. You want to be doing the latter anyhow! For more on installing a PEX file in an image instead of just running it - which is not optimal in a number of dimensions - see: + https://pex.readthedocs.io/en/v2.1.121/recipes.html#pex-app-in-a-container + https://blog.pantsbuild.org/optimizing-python-docker-deploys-using-pants/

Thanks for your reply. I used to just copy a pex file into the image. Recently, I followed the link you posted and switched to the "Simple Multi-Stage Build" approach. But still, I can't get optimizations "on". Example of dockerfile here: https://github.com/ZdenekM/arcor2/blob/master/src/docker/arcor2_execution/Dockerfile and the macro I use for pex targets is here: https://github.com/ZdenekM/arcor2/blob/master/pants-plugins/macros.py#L73

Ah, ok. The problem here is the recipes are too formulaic. You need to step back and understand what's going on. Look at your Dockerfile. The optimization is too late! You already compiled your src and deps in two separate RUN statements above! That's when the pyc get created and

__debug__

sealed in.

Ah, ok... I see 🙂 Thanks!

@narrow-activity-17405 it's been a while, can you please re-state a brief summary of the steps I follow to reproduce your issue? For example, is it using the tip of master for https://github.com/ZdenekM/arcor2 and what exact commands should I run to witness the issue.