Bug: 0.9.0 ModuleNotFoundError: No module named 'goblet.infrastructures'

[rasnes]

Hi. I just tried to update from version 0.8.4 to 0.9.0, and I get this error when I try to run goblet deploy:

Run poetry run goblet deploy --location europe-north1
Traceback (most recent call last):
  File "/home/runner/.cache/pypoetry/virtualenvs/bucket-to-bigquery-85QkY9Aq-py3.10/bin/goblet", line 5, in <module>
    from goblet.cli import main
  File "/home/runner/.cache/pypoetry/virtualenvs/bucket-to-bigquery-85QkY9Aq-py3.10/lib/python3.10/site-packages/goblet/__init__.py", line 1, in <module>
    from goblet.app import Goblet, jsonify, goblet_entrypoint  # noqa: F401
  File "/home/runner/.cache/pypoetry/virtualenvs/bucket-to-bigquery-85QkY9Aq-py3.10/lib/python3.10/site-packages/goblet/app.py", line 7, in <module>
    from goblet.decorators import Register_Handlers
  File "/home/runner/.cache/pypoetry/virtualenvs/bucket-to-bigquery-85QkY9Aq-py3.10/lib/python3.10/site-packages/goblet/decorators.py", line 8, in <module>
    from goblet.infrastructures.redis import Redis
ModuleNotFoundError: No module named 'goblet.infrastructures'
Error: Process completed with exit code 1.

Anything I can do about this, or is this something that should be fixed on your end?

[anovis]

let me take a look into it. what were you trying to deploy?

[anovis]

this looks like it could be a poetry caching issue, or a pythonpath issue potentially. I use conda and i was able to deploy a cloudfuntion, cloudrun, and cloudrun job ok. will continue to look into it

[rasnes]

Thanks for looking at this so quickly. I checked the caching in my Github Actions workflow, and it seemed to work as intended, i.e. that poetry did a fresh install (not from cache) of the updated dependencies, including goblet 0.9.0. Then it runs poetry export -f requirements.txt --output requirements.txt to produce

goblet-gcp==0.9.0 ; python_version >= "3.10" and python_version < "3.11" \
    --hash=sha256:75b1a1e7cdc096cf4b9a76d14f957ac71b821b6b61532ab5c7920b8e39336583 \
    --hash=sha256:7acbff7413805f046d23e140d4fe236aec3674cf6632f65abc0e48f6e65b1e08

in my requirements.txt file that goblet deploy will fetch to deploy my Cloud Run Jobs. UPDATE: the requirements.txt should not matter in this case I think, this issue is on deploying the container, not running it. My main.py code looks like:

from goblet import Goblet, goblet_entrypoint
import logging

app = Goblet(
    function_name="bucket-to-bigquery",
    backend="cloudrun",
)
goblet_entrypoint(app)
app.log.setLevel(logging.DEBUG)

@app.job("ingest-routes")
def ingest_routes(id):
    app.log.info(f"executes {id}")
    foo(bar)
    return "200"

But I will try to look more closely at my end of course. I might have been a bit fast to report this as a bug, but these things should, ideally, just work 😆 (Sorry for the close and reopen, I had my kids running on my keyboard.)

[anovis]

thanks for the example. i will try with poetry locally as well and see if i can replicate.

[AdeelK93]

I fully uninstalled goblet and reinstalled, and I'm still having this issue when installing with poetry.
Here's a guess - is it because __init__.py is missing from infrastructures?

[AdeelK93]

@jissac seems we're on the same page! with #263

[anovis]

yup! appreciate the help. will merge and create a new release.

Sorry about the bug

[anovis]

released 0.9.1 so hopefully this is fixed.

[rasnes]

This is fixed now with 0.9.1. Thanks!