`runpy.run_module` for a non-existent module raises an `ImportError` without `name` set to `None`

[godlygeek]

Bug report

Bug description:

When import fails to import a module, it raises an ImportError with the name attribute set to the module it failed to import.

When runpy.run_module fails to import a module, it raises an ImportError with the name attribute set to None.

This seems to be an accidental difference, or at least an undocumented one:

>>> try:
...     import foobar
... except ImportError as e:
...     print(f"{e=} {e.name=}")
...
e=ModuleNotFoundError("No module named 'foobar'") e.name='foobar'

>>> try:
...     import runpy
...     runpy.run_module("foobar")
... except ImportError as e:
...     print(f"{e=} {e.name=}")
...
e=ImportError('No module named foobar') e.name=None

CPython versions tested on:

3.14

Operating systems tested on:

Linux

Linked PRs

• gh-149117: Set ImportError.name on errors from runpy.run_module/run_path #149159

[picnixz]

I guess it is not really important but I see why it could be annoying for introspection. Do you want to submit a PR? we also have a different error raised so I guess it is actually raised manually (I haven't checked).