F821 false positive in notebook when using magic command as variables #9648
Assignees
Labels
bug
Something isn't working
Comments
|
Yeah we can probably improve our heuristics here to specifically detect assignments. The automagics are actually really complicated because it means that (in general) cells may or may not be unsafe depending on the order of their execution (which we can't know). For example: #8398 (comment). |
|
I'll try to fix this specific case today :) |
|
But I need to audit the list of automagics which I dread 😂 |
|
Fixed in #9653. |
charliermarsh
added a commit
that referenced
this issue
Jan 29, 2024
## Summary Given a statement like `colors = 6`, we currently treat the cell as an automagic (since `colors` is an automagic) -- i.e., we assume it's equivalent to `%colors = 6`. This PR adds some additional detection whereby if the statement is an _assignment_, we avoid treating it as such. I audited the list of automagics, and I believe this is safe for all of them. Closes #8526. Closes #9648. ## Test Plan `cargo test`
|
Thanks! |
|
Of course! If you see more of these, let us know, since we have to rely on some heuristics for this given the flexibility of notebooks. |
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
ruff gives false positives for F821 (Undefined name) in notebooks when using magic command as variables. Ok, this is very specific and potentially related to #9644.
In a jupyter notebook when defining a variable which has the same name as a ipython magic command and using this variable in a following cell and not as the first command triggers F821 (wrongly).
Only classify magic commands as magic if they are preceded with a
%?command
version
MVCE
(I am not sure how to create a MVCE for a jupyter notebook, but I guess it's easy enough...)
The text was updated successfully, but these errors were encountered: