Thanks to visit codestin.com
Credit goes to github.com

Skip to content

wrong intersphinx prefix for many numpy functions #23032

Closed
@drammock

Description

@drammock

in the current objects.inv file (https://numpy.org/devdocs/objects.inv) it appears that many of the functions have changed from py:function prefix to the py:data prefix. For example:

numpy.arange py:function 1 reference/generated/numpy.arange.html#$ -
numpy.arccos py:data 1 reference/generated/numpy.arccos.html#$ -
numpy.arccosh py:data 1 reference/generated/numpy.arccosh.html#$ -
numpy.arcsin py:data 1 reference/generated/numpy.arcsin.html#$ -
numpy.arcsinh py:data 1 reference/generated/numpy.arcsinh.html#$ -
numpy.arctan py:data 1 reference/generated/numpy.arctan.html#$ -
numpy.arctan2 py:data 1 reference/generated/numpy.arctan2.html#$ -
numpy.arctanh py:data 1 reference/generated/numpy.arctanh.html#$ -
numpy.argmax py:data 1 reference/generated/numpy.argmax.html#$ -
numpy.argmin py:data 1 reference/generated/numpy.argmin.html#$ -
numpy.argpartition py:data 1 reference/generated/numpy.argpartition.html#$ -
numpy.argsort py:data 1 reference/generated/numpy.argsort.html#$ -
numpy.argwhere py:data 1 reference/generated/numpy.argwhere.html#$ -
numpy.around py:data 1 reference/generated/numpy.around.html#$ -
numpy.array py:function 1 reference/generated/numpy.array.html#$ -
numpy.array2string py:data 1 reference/generated/numpy.array2string.html#$ -
numpy.array_equal py:data 1 reference/generated/numpy.array_equal.html#$ -
numpy.array_equiv py:data 1 reference/generated/numpy.array_equiv.html#$ -
numpy.array_repr py:data 1 reference/generated/numpy.array_repr.html#$ -
numpy.array_split py:data 1 reference/generated/numpy.array_split.html#$ -
numpy.array_str py:data 1 reference/generated/numpy.array_str.html#$ -
numpy.asanyarray py:function 1 reference/generated/numpy.asanyarray.html#$ -

This random chunk of 22 entries has 3 py:function entries and 19 py:data entries. But aren't all of them functions? This is causing the MNE-Python CIs to fail because we have lots of things like

see :func:`numpy.where` for details

which as of today no longer resolve.

Metadata

Metadata

Assignees

No one assigned

    Type

    No type

    Projects

    No projects

    Milestone

    No milestone

    Relationships

    None yet

    Development

    No branches or pull requests

    Issue actions