Secure core import #483
Secure core import #483
Conversation
setup.py
Outdated
| @@ -88,7 +88,7 @@ def run(self): | |||
| ], | |||
| install_requires = [ | |||
| 'traitlets', | |||
| 'jupyter_core', | |||
| 'jupyter_core>=4.5.0', | |||
There was a problem hiding this comment.
Since we're already building a 4.6.0 jupyter_core release - which is waiting on the final resolution for secure_write, I think we should just move to 4.6.0 with this change.
|
Now that core is at 4.6.0, it seems like we should pump out a jupyter_client release that depends on 4.6.0. Then, I believe many of the Windows issues should be resolved - although there's something fishing about the |
|
@kevin-bates the pywin32==225 build sounds like exclusively an anaconda build issue where the recipe is doing bad things. I've opened an issue in conda-forge/pywin32-feedstock#18 and linked to the thread in winpy32 |
|
Gonna merge and release to get this fixed |
|
5.3.4 released! |
|
Does this mean that the |
|
@MSeal - thank you! That's actually great new about Since there's a NB patch release brewin', does it make sense to bump the min versions of core and client to 4.6.0 and 5.3.4, respectively? I'm happy to submit that PR if you agree. What is the timeframe to get |
|
@vidartf Yes, assuming the dependency being arch specific doesn't force the downstream package to do so (don't think it does). If you want to make the PR I can help get that merged. @kevin-bates +1. |
Its use was removed in jupyter#483, so we can drop it now.
Its use was removed in jupyter#483, so we can drop it now.
Its use was removed in jupyter#483, so we can drop it now.
Its use was removed in jupyter#483, so we can drop it now.
Bug fixes for windows issues are going into jupyter_core for secure_write -- we can move to depending on that location to pick up fixes.
@kevin-bates We can bump the setup requirement to 4.5.1 once jupyter/jupyter_core#166 merges and releases.