proper way to do bokeh server development?

So, I’m running into documentation issues with bokeh_server, as I don’t know the proper route to develop. I am trying to make a web visualization applet, using both bokeh plots and controls.

A lot of the server examples show using just the “plot style” with using output_server() as the target.

But then, I look at something like the stock applet it has a completely different design using things like Instance(VBox), etc., which I think is the correct route to use the widgets.

What are the drawbacks of using one versus the other, and which do you recommend? Also, if it is the latter is there any documentation on how to make these applets?

In particular, prior to the 0.7 release I could add properties that I wanted to store in the applet class during init, but now I can’t because setattr is overridden to fail if I didn’t initially provide those properties in the class, i.e., Instance(foo).

Hi Jon,

So, I’m running into documentation issues with bokeh_server, as I don’t know the proper route to develop. I am trying to make a web visualization applet, using both bokeh plots and controls.

A lot of the server examples show using just the “plot style” with using output_server() as the target.

But then, I look at something like the stock applet it has a completely different design using things like Instance(VBox), etc., which I think is the correct route to use the widgets.

What are the drawbacks of using one versus the other, and which do you recommend?

I defer this to Hugo who actually wrote that app and can give you a better explanation… Hugo, can you chime in?

Also, if it is the latter is there any documentation on how to make these applets?

Nop, not documentation yet. We will try to improve this soon.

In particular, prior to the 0.7 release I could add properties that I wanted to store in the applet class during init, but now I can’t because setattr is overridden to fail if I didn’t initially provide those properties in the class, i.e., Instance(foo).

Damian

···

On Thursday, December 11, 2014 5:26:10 PM UTC-3, Jon Woodring wrote:

Loading the script into the Bokeh Server allows the bokeh server to
create a new document with the app in it per page load. Bokeh’s
object model is a global cross javascript/python object model - if
you just define the events in a script then really only one person
can consume the application at a time.

we'll be trying to unify these 2 approaches in the future - see

this:

(very brief still)
···

https://github.com/bokeh/bokeh/wiki/BEP-6-Bokeh-Applications

  On 12/12/2014 01:45 PM, Damian Avila

wrote:

Hi Jon,

    On Thursday, December 11, 2014 5:26:10 PM UTC-3, Jon Woodring

wrote:

        So, I'm running into documentation issues with

bokeh_server, as I don’t know the proper route to develop. I
am trying to make a web visualization applet, using both
bokeh plots and controls.

          A lot of the server examples show using just the "plot

style" with using output_server() as the target.

But then, I look at something like the stock applet it has a completely
different design using things like Instance(VBox) , etc., which I
think is the correct route to use the widgets.

          What are the drawbacks of using one versus the other,

and which do you recommend?

      I defer this to Hugo who actually wrote that app and can

give you a better explanation… Hugo, can you chime in?

          Also, if it is the latter is there any documentation on

how to make these applets?

      Nop, not documentation yet. We will try to improve this

soon.

          In particular, prior to the 0.7 release I could add

properties that I wanted to store in the applet class
during init, but now I can’t because setattr is
overridden to fail if I didn’t initially provide those
properties in the class, i.e., Instance(foo).

Damian

  You received this message because you are subscribed to the Google

Groups “Bokeh Discussion - Public” group.

  To unsubscribe from this group and stop receiving emails from it,

send an email to [email protected].

  To post to this group, send email to [email protected].

  To view this discussion on the web visit [https://groups.google.com/a/continuum.io/d/msgid/bokeh/96cfe43a-37b8-4d6b-9a72-83bdacb0acc0%40continuum.io](https://groups.google.com/a/continuum.io/d/msgid/bokeh/96cfe43a-37b8-4d6b-9a72-83bdacb0acc0%40continuum.io?utm_medium=email&utm_source=footer).

  For more options, visit [https://groups.google.com/a/continuum.io/d/optout](https://groups.google.com/a/continuum.io/d/optout).