Gibbs Consulting

Added #490 for this work

@GarderesG the template tag - which incidentally can be overridden like any other Django template - doesn't do very much and doesn't introduce any css of its own so its...

It seems that restricting the channels package to a version below 3.0 provides a workaround, and also strongly suggests that changes in that package are responsible for this issue.

Telling curl to redirect (and as a POST) seems to work: ``` curl -L --post301 --post302 http://djangoplotlydash.com/dpd/views/poke/ -d'{"channel_name":"live_button_counter","label":"named_counts","value":{"click_colour":"cyan"}}' ``` This is possibly a bit safer than trying to work out...

If you are going to remain within a Jupyter notebook, then you can use the notebook environment to store state. This would avoid the need to use an external cache...

This is an [open issue](https://github.com/GibbsConsulting/django-plotly-dash/issues/141) with the underlying django-plotly-dash library

This has now been addressed in `django-plotly-dash` and so should now work if you update to the latest version. @lauraln can you try `pip install -U django-plotly-dash` and then retry?

The [LocalServing.ipynb](https://github.com/GibbsConsulting/jupyter-plotly-dash/blob/master/LocalServing.ipynb) notebook contains an example of multiple callback outputs

It should certainly possible to get it to work, if it doesn't already. Right now jupyter-plotly-dash is known to work with * jupyter-notebook * jupyterlab * binder but jupyterlab through...

I get 'Open in new window' URLs like `/app/endpoints/c1c51597b69d435198703c704ca51ab3/` but this is only a fragment of the full URI (jupyterlab and jupyterhub both insert an additional proxy). On my local...