mapix

@yigitsoy Great job, I will assist you in the follow-up development and test this feature together.

@krafftta You can explicitly use Celery for asynchronous processing in the callback after `dcc.Uploader` has completed the upload, or, if you want to simplify, you can use something like `dcc.Store`...

This issue is very important. In my scenario, I use some `dcc.Store` objects to store user session and some important user Tokens. Adding State can help verify the validity of...

Thanks to @Sisyphus235 's contribution, this issue will be fixed in the next version release. Ref https://github.com/fohrloop/dash-uploader/pull/128

@fohrloop Yes ， Could you please help add the release permissions on PyPI? I am currently unable to publish. `mapix` or `[email protected]` on pypi

@fohrloop Understood. I have gathered some necessary information. We can track it at https://github.com/fohrloop/dash-uploader/pull/135 .

@fohrloop The reason I didn't add more tests here is because the current tests are a bit difficult to run on my local machine. If feasible, I will concentrate on...

The current modification has been running well in my scene for quite a long time, for example: ```python @du.callback( output=[ PREVIEW_IMAGE_ID.get_output("children"), UPLOAD_IMAGE_LOADING_ID.get_output("children"), LOGIN_MODAL_ID.get_output("is_open"), ], state=[ SESSION_STORE_ID.get_state("data"), TOKEN_STORE_ID.get_state("data"), ], id=UPLOAD_IMAGE_ID.get_identifier(), )...

@fohrloop The reason for the difficulty with tests is that I always have a critical Case failure on my Mac, but I cannot determine for a while whether it is...

``` pip list | grep torch | awk '{print $1}' | xargs pip uninstall -y ``` wdb not compatible with pytorch, uninstall all pytorch related packages works for me