For questions, use https://discourse.holoviz.org/ instead! This chat is for informal discussions and debugging sessions with developers of the HoloViz.org tools (HoloViews, Datashader, Panel, hvPlot, GeoViews, Param, etc.), but user questions should go to the HoloViz Discourse site so that others won't have to ask them again.
Hi! When I plot an hv.Curve of cftime.DatetimeNoLeap timeseries data, I get the warning 'Converting cftime.datetime from a non-standard calendar (noleap) to a standard calendar for plotting. This may lead to subtle errors in formatting dates, for accurate tick formatting switch to the matplotlib backend.'. However, I am okay with the output and want to suppress the warning. The warning is set internally using param.warning rather than the normal warning package, so I can't suppress it using warnings.filterwarnings. Does anyone know if it is possible to suppress this warning, or is that a bad practice?
Param's warnings use the logging module...
Hope there is a video link somewhere. I really need to learn this.
Regarding Bokeh and Extensions. I'm trying to create a custom button by inheriting from the Bokeh and Panel Button. So far without luck. If you have time please send me a few helpful hints via https://discourse.holoviz.org/t/how-do-i-implement-a-custom-button-by-extending-the-existing-button/1169. Thanks. FYI. @MattEttus
A little inspiration for improving the code base of HoloViews and moving it a little bit into the future :-) can be found in https://fastpages.fast.ai/fastcore/. Just all the kwargs and params that are all over the place and gives not insight into what can actually be provided in a modern IDE could be solved via some of the tools described in that blog.
For whom might be interested. I've added a FastButton to the awesome-panel-extensions package. It's implemented by extending the existing Panel and Bokeh Button classes. Worked really well. See https://discourse.holoviz.org/t/awesome-panel-extensions-package-change-log/1009/12?u=marc
All the Fast widgets are implemented as true Bokeh Extensions extending or reusing the code of the existing Bokeh widget.
If you care about which components Fast currently plan to support they have a "spec". You can check it out here https://github.com/microsoft/fast/tree/master/specs. I've checked it versus the Panel Reference Gallery and it seems they will be supporting most of the components you would expect.
I actually learned a lot from reusing all the Bokeh and Panel code. I've been used to rewriting my own from scratch. That has taken a lot longer time.
:-)
I have been thinking about Material vs Fast widgets. There are a lot of users on Material in general. But not on the MWC widgets in particular. It seems they are not that high a priority for Google. And if you look at the implementation of the web components (investigate the shadow dom on a page where they are used it looks very ugly and looks like just a wrapper around the existing js and css material design components. Not a simplified from the bottom up rewrite. That is very different from Fast. A lot of marketing has been put together and there is already a lot of people (thousands) signed up of their twitter account on their discord channel. The Fast components just seem simpler. And they also come with a very simple framework for writing web components. Something that could maybe also be wrapped in Python on day. And the light vs dark mode and change of style settings is just very, very simple and powerful. And they openly claim that their web components should implement the same attributes and properties as the "old" html elements and then just extend on top. Making it very, very easy to drop them in. This is the right way to do things.
Eventually I would hope to implement both one day as I believe they will both be major as time goes by.
And the fast specs describe a range slider :-)
@MarcSkovMadsen sorry...im just jumping into the discussion like this... But wouldn't it be more neat to do a proper integration between panel/bokeh and react instead? That way you get a lot more power than if just integrating with a design system.
Right now you need to wrap any js library whether vanilla js, web components, react, vua or angular into a Bokeh Model Extension to be able to use it in Panel.
So for each react component there would be some work. I have also look at Vuetify and Material UI. But so far I've not been able to find any nice and easy ways to incorporate it. It's one component at the time. And then you need to understand React as well.
FYI. Regarding Fast I've created some Feature Requests for "missing" components. See https://github.com/microsoft/fast/issues/created_by/MarcSkovMadsen
@bsdis . Regarding the integration with React. You can integrate with React or Vue via Bokeh extensions. https://awesome-panel.readthedocs.io/en/latest/guides/awesome-panel-extensions-guide/bokeh-extensions.html. You can use anything you can use in Javascript/ Typescript. I've just not yet seen a lot of people looking to integrate React. I don't know of Panel developers or users that have these skills? There is one user who has a PR on adding the React Grid Layout though. See https://discourse.holoviz.org/t/react-with-panel/1001 and holoviz/panel#1535.
But @bsdis I would really, really like to provide some documentation and examples for integrating React. If you know how to do it you are welcome to show me a small working POC. That is often what I need to get moving :-)
Another thing I would like to dive into one day is actually wrapping Lit-Element or Fast-Element into Python. I think that should be very possible and make it very easy to create fast and performant things that run in the front end but are defined in the back end using Python. See an example of a Fast-Element here. https://fast.design/docs/fast-element/leveraging-css
For inspiration when developing Panel there is a Layout Guide for Shiny https://shiny.rstudio.com/articles/layout-guide.html. I can see some of the layout stuff I'm still missing from Panel is there. This includes a
NavigationList. I've created an issue here holoviz/panel#1571
Just wondering if there's a guide that shows how to use DynamicMaps with custom Panel layouts? In particular, we have three components we need to scrub through, one is a video, the other two are timeseries plots for which we have a dynamic map VLine that shows us at which time step we're at. We want to put the video on the right, the timeseries plots on the left, and the scrubber at the bottom, while taking advantage of the syntactic ease and composability of using DynamicMaps. I put up a reprex here: https://gist.github.com/ericmjl/3d05c15339bd0cd26e07165b43035e7a, do you all have advice on how we can go about this?
Question on naming conventions for alternative widgets in Panel. Should I add the name of the framework as a prefix to the widget class or just call them the same as in Panel?
For example: Button (Panel) vs MaterialButton, FastButton, WiredButton, SmartButton, UI5Button, ...?
or
For example: panel.widgets.Button, material.Button, fast.Button, wired.Button, smart.Button, ui5.Button, ...?
The reason for asking is that I would like to make it easy some day in the future to switch between the differents ones. For example if you are experimenting with the look and Feel. This includes trying out different frameworks as widgets parameter to pn.Param.
And I am asking now because renaming is just tedious later on :-)
And then always refer to them as fast.Button, wired.Button, etc in examples and scripts if it matters what the button comes from, and refer to them (i.e. import them as) just Button if it doesn't matter.
Hello Every1,
A Gentle Request.
Guys why we are facing so difficulties in making our dynamic plot embedded with flask with our own server.
I am not able to findout good resources to learn ,
How we can make our dynamic plot dynamic in nature when embedded with flask,
whereas it was very easy to see the dynamic property of plot when running the bokeh server.
I really appreciate for bokeh server but we want plot embedded with flask in our own server there we are facing difficulty .
we are not able to find the good resource to learn and do.
Thank you.
A Gentle Request.
Guys why we are facing so difficulties in making our dynamic plot embedded with flask with our own server.
I am not able to findout good resources to learn ,
How we can make our dynamic plot dynamic in nature when embedded with flask,
whereas it was very easy to see the dynamic property of plot when running the bokeh server.
I really appreciate for bokeh server but we want plot embedded with flask in our own server there we are facing difficulty .
we are not able to find the good resource to learn and do.
Thank you.
@philippjfr any estimate when Holoviz 1.13.4 will arrive? https://github.com/holoviz/hvplot/issues/462#issuecomment-681076375