Add a method to "reset" a figure

[swevrywhere]

I want the ability to reset a figure via code in the same way as the figure reset tool.

I asked a related question on Stack Overflow, and it was recommended that I open an issue to track the request.

http://stackoverflow.com/questions/39278110/whats-the-command-to-reset-a-bokeh-plot

[markjay4k]

I would like the same.

Looking for a way to easily clear a bokeh figure

[ross8711]

I also would need way of doing this.

[uweschmitt]

Would also need this.

[nanodan]

Also need this.

[pchavanne]

Would need this too.

[Milly]

Need this.

[malewick]

Would also need this.

[bryevdv]

After #7757 is merged, this will work:

from bokeh.io import show
from bokeh.layouts import column
from bokeh.models import Button, CustomJS
from bokeh.plotting import figure

p = figure(tools="reset,pan,wheel_zoom,lasso_select")
p.circle(list(range(10)), list(range(10)))

b = Button()
b.js_on_click(CustomJS(args=dict(p=p), code="""
    p.reset.emit()
"""))

show(column(p, b))

[nathan-vignal]

the code given by bryevdv doesn't work, i get this js error "
Uncaught TypeError: Cannot read property 'emit' of undefined at eval (eval at <anonymous> (bokeh-0.12.7.min.js:25), <anonymous>:4:13) at r.execute (bokeh-0.12.7.min.js:25) at e.<anonymous> (bokeh-0.12.7.min.js:22) at t.emit (bokeh-0.12.7.min.js:2) at e.t._setv (bokeh-0.12.7.min.js:1) at e.t.setv (bokeh-0.12.7.min.js:1) at e.set (bokeh-0.12.7.min.js:1) at e.change_input (bokeh-widgets-0.12.7.min.js:1) at e._button_click (bokeh-widgets-0.12.7.min.js:1) at HTMLButtonElement.<anonymous> (bokeh-widgets-0.12.7.min.js:1)
Any idea on how i can reset a figure ?

[bryevdv]

@nathan-vignal it's the opposite, your version of Bokeh is way too old. This feature landed in 0.12.16 and your output shows 0.12.7

[RainerHeintzmann]

If you are in a notebook, the simple solution can be to directly access the renderers in the figure p:

p.renderers[0].visible=False
hides a specific renderer in the figure, whereas

p.renderers = []
would clear all the renderers. Of course you may need to force update of the display. In a notebook by calling:

push_notebook()
I am not sure, how much of a hack this is, but it seems to work.

[CLocs]

If plotting using ColumnDataSource, one Python-only work-around is to reset the data. Here is a snippet example for a multi-line plot.

# Make initial plot
signal_plt_cds = ColumnDataSource(data={"xs": [[0,1,2], [0,1,2]], "ys": [[1,2,3], [4,5,6]]})
feature_plt = figure(background_fill_color="#fafafa")
_ = feature_plt.multi_line("xs", "ys", source=feature_plt_cds)

# Reset plot then update
signal_plt_cds.data = {"xs": [], "ys": []}  # reset work-around
signal_plt_cds.data = {"xs": [[0,1,2], [0,1,2]], "ys": [[10,20,30], [40,50,60]]}