-
Notifications
You must be signed in to change notification settings - Fork 90
New issue
Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.
By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.
Already on GitHub? Sign in to your account
Development install fails with package version conflict on shinyswatch
#1857
Comments
This part seems suspicious. Are you sure you're installing into a fresh virtual environment (it's easy to miss) because |
Got it: I had forked the repo, which doesn't carry over tags by default, but tags are used by
And the build gets farther, and hits a different error "Preparing metadata (pyproject.toml) did not run successfully."
For the git tags, I'll make a little PR for the README in case that would help anyone else, and then look at the new error. |
That'd be great, thanks for the offer. It should be included under the "Development" heading; it's definitely the kind of subtle detail with the automatic versioning that's easy to miss. |
See #1858 for proposed addition to README. Still looking at the rest of the install. Since the problem seems to be with an older pandas version, what happens if I require something more recent? diff --git a/pyproject.toml b/pyproject.toml
index 8d7fc886..993227a9 100644
--- a/pyproject.toml
+++ b/pyproject.toml
@@ -107,7 +107,7 @@ dev = [
"pre-commit>=2.15.0",
"wheel",
"matplotlib",
- "pandas",
+ "pandas>=2.2.0",
"pandas-stubs",
"polars",
"numpy", A different error:
Seems to be hitting this problem in modin: And this in the cpython changelog for 3.12.0:
I'll try installing on python 3.11. (I'll add notes on this issue as I debug, rather than filing lots of little issues along the way, but feel free to close this when the PR comes in. Your call.) |
Thanks @mccalluc. In general, our approach is to be lenient with versioning unless we're aware of a specific incompatibility with shiny that needs to be resolved by requiring a newer version of a dependency. In other words, we'd need to distinguish between a) it being difficult to install an older version of pandas (very likely) or b) a compatibility issue prevents using an older version of pandas with shiny. Please feel free to track your journey here and if you uncover an incompatibility we'll definitely address it. |
Downgraded to Python 3.11.7 and (At the end of this, I'm not sure what the pandas problem was, or whether it was related to modin.) Closing. |
I'm trying to follow the developer instructions in the README on a fresh checkout with a clean virtualenv:
I get the following error:
My environment:
I looked at the github workflows, but the install process there is different, so this process probably isn't checked by CI. If I sort this out, I'll file a PR.
The text was updated successfully, but these errors were encountered: