Skip to content

Better show for numpy array #361

New issue
New issue
Open
#595
Open
Better show for numpy array#361
#595
Labels
enhancementNew feature or request
@xgdgsc

Description

@xgdgsc
xgdgsc
opened on Sep 19, 2023

Is your feature request related to a problem? Please describe.

PythonCall.jl/src/Py.jl

Line 186 in ad67c04

prefix = hasprefix ? compact ? "Py:$(multiline ? '\n' : ' ')" : "Python:$(multiline ? '\n' : ' ')" : ""
When it shows a python array in vscode inline results. It would only show Python when unexpanded because of the line end. Would it be better to show more informative like Julia Array shows 4×4 Matrix{Float64}: on the first line?

Describe the solution you'd like
shows numpy array shape like 4×4 Matrix{Float64}:

Describe alternatives you've considered

Additional context

Activity

xgdgsc
added
enhancementNew feature or request
on Sep 19, 2023
cjdoris

cjdoris commented on Sep 21, 2023

@cjdoris
cjdoris
on Sep 21, 2023
Collaborator

I'm hesitant to make this code depend on the actual Python type - currently this basically just shows repr(x). I'm not totally against the idea though - I agree it's useful info.

Maybe you could suggest this change to numpy itself?

xgdgsc

xgdgsc commented on Sep 22, 2023

@xgdgsc
xgdgsc
on Sep 22, 2023
Author

Would there still be a first line Python: ? How feasible is it to add something like numpy/numpy#9029 (comment) ?

cjdoris

cjdoris commented on Sep 22, 2023

@cjdoris
cjdoris
on Sep 22, 2023
Collaborator

Oh I see your point now.

To explain for myself: when a Py is shown, if it has multiple lines then the Python: prefix occurs on its own line, such as with numpy arrays:

Python:
array(
  [ ... ], dtype=int64)

If you execute code line-by-line from a Julia script in VS Code then it only displays the first line ("Python:") which is not very informative. So this question is about adding extra info into the first line.

xgdgsc
linked a pull request that will close this issue on Feb 28, 2025
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment

Metadata

Metadata

Assignees

No one assigned

    Labels

    enhancementNew feature or request

    Type

    No type

    Projects

    No projects

    Milestone

    No milestone

    Relationships

    None yet

      Development

      Participants

      @xgdgsc@cjdoris

      Issue actions
        Better show for numpy array · Issue #361 · JuliaPy/PythonCall.jl