-
-
Notifications
You must be signed in to change notification settings - Fork 18.4k
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
CI: clean environment.yml, comment that req-dev is generated #30392
Conversation
@@ -33,7 +33,8 @@ dependencies: | |||
- nbconvert>=5.4.1 | |||
- nbsphinx | |||
- pandoc | |||
# Dask and its dependencies | |||
|
|||
# Dask and its dependencies (that dont install with dask) |
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
IIUC conda install dask
does the right thing without needing to specify everything else, but pip install dask
does not. could we unify all this as dask[dataframe]
? also i guess put it in the "downstream tests" block?
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
@TomAugspurger might have ideas
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
dask[dataframe]
works for installing from PyPI, but there isn't a dask[dataframe]
or dask-dataframe
conda package.
|
||
pip_content = "\n".join(pip_deps) | ||
fname = os.path.split(conda_fname)[1] |
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
Is there a reason for the split here?
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
otherwise we get a full path including username
@@ -33,7 +33,8 @@ dependencies: | |||
- nbconvert>=5.4.1 | |||
- nbsphinx | |||
- pandoc | |||
# Dask and its dependencies | |||
|
|||
# Dask and its dependencies (that dont install with dask) |
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
@TomAugspurger might have ideas
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
Nice changes, just one small styling idea.
header = ( | ||
f"# This file is auto-generated from {fname}, do not modify.\n" | ||
"# See that file for comments about the need/usage of each depdendency.\n\n" | ||
) | ||
pip_content = header + "\n".join(pip_deps) |
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
Small personal preference, but I think this would be simpler if when defining pip_deps above, instead of pip_content = []
you write:
pip_content = [
f"# This file is auto-generated from {fname}, do not modify.",
"# See that file for comments about the need/usage of each depdendency.",
]
If you do that, all this can be deleted here (you'll have to define fname
before).
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
ATM the pip_deps
list has a specific meaning. id prefer to retain that
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
Not important, but headers
could still be a list, and use "\n".join(headers + pip_deps)
. I think it'll still be an improvement, avoiding the \n
in the strings. Just saying, happy to get this merge as is.
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
I would actually write this as a triple quoted string.
header = f"""
...
..."""
which avoids the embedded newlines
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
triple quoted strings that arent at the module-level mess with code-folding, so i prefer to avoid them
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
k that's what i would have done, but ok as is.
There is a little more id like to do in environment.yml to put docbuild things in the same place, but holding off to keep the diff tidier. Also it appears there are some unused things in environment.yml that we can look into removing, separate PR