Legend layout problems when embedding in iframe

I totally realize that the details of layout elements can be beyond the scope of Bokeh itself, and often depends on the html/css framework in which Bokeh plots are rendered. However, I thought I would throw this out there for any suggestions.

I have a radial plot figure I recently designed using Bokeh. You can see this here:

The organization that we built this tool for is trying to embed this page directly into their website using an iframe (navigate to the “Radial Plot” tab):

This seems to work fine, except the legend layouts get weird. Particularly the “Depth” legend in the upper-right. The legend seems to increase all the row heights for each legend item, so that it overlaps the figure. Testing with Chrome and Firefox, it only occurs with Chrome. The legend in the upper-left also has some minor layout issues that occur:

The relevant code to create the upper right “Depth” legend is:

  # This is a dummy glyph just to have consistent colors for a custom legend
  event_scatter_dummy = p.scatter(
    fill_color=['green','yellow','red'], fill_alpha=0.5,
    line_color='black', line_alpha=0.5,
  depth_legend = Legend(items=[
    LegendItem(label='Depth < 12"', renderers=[event_scatter_dummy], index=0),
    LegendItem(label='Depth 12-44"', renderers=[event_scatter_dummy], index=1),
    LegendItem(label='Depth > 44"', renderers=[event_scatter_dummy], index=2),
    border_line_alpha = 0,

The only way I could figure out to add the upper-left “R-size” legend was to create separate legend instances and add them individually:

  # ============================================================================
  event_radius_dummy_1 = p.circle(
    fill_alpha=0.0, line_color='black', 
  # location is absolute location in screen coordinates
  # (pixels from the bottom-left corner)
  offset = 13
  event_legend1 = Legend(items=[
    LegendItem(label='R1', renderers=[event_radius_dummy_1])],
    location = (20, 553-offset),
    label_standoff = 10
  event_legend2 = Legend(items=[
    LegendItem(label='R2', renderers=[event_radius_dummy_1])],
    location = (14, 532-offset),
    label_standoff = 5
  event_legend3 = Legend(items=[
    LegendItem(label='R3', renderers=[event_radius_dummy_1])],
    location = (8, 507-offset),
    label_standoff = 0
  event_legend4 = Legend(items=[
    LegendItem(label='R4', renderers=[event_radius_dummy_1])],
    location = (2, 479-offset),
    label_standoff = -5
  event_legend5 = Legend(items=[
    LegendItem(label='R5', renderers=[event_radius_dummy_1])],
    location = (-4, 447-offset),
    label_standoff = -10
  event_legend_list = [event_legend1,event_legend2,event_legend3,event_legend4,event_legend5]
  for legend in event_legend_list:
  size_list = [15,26,37,48,59]
  index_list = [1,2,3,4,5]
  for index, size in zip(index_list, size_list):
    p.legend[index].glyph_height = size
    p.legend[index].glyph_width = size
    p.legend[index].padding = 0
    p.legend[index].margin = 0
    p.legend[index].border_line_alpha = 0
    p.legend[index].background_fill_alpha = 0
    p.legend[index].label_text_baseline = "middle"
    p.legend[index].label_text_font_size = "10pt"

The code for the scatter plot figure is:

  cds = create_cds(event_data)
  p = figure(
      width=575, height=575,
  p.toolbar.logo = None
  p.x_range = Range1d(-15, 15)
  p.y_range = Range1d(-15, 15)
  p.xgrid.visible = False
  p.ygrid.visible = False
  p.xaxis.visible = False
  p.yaxis.visible = False
  # Project data into polar coordinates
  # customjs from bryevdv (https://github.com/bokeh/bokeh/issues/657)
  polarx = CustomJSTransform(args=dict(source=cds), v_func='''
  const new_xs = new Array(source.data.elev_project_vectors.length)
  for(var i = 0; i < new_xs.length; i++) {
      new_xs[i] = source.data.elev_project_vectors[i] * Math.sin(source.data.elev_angles[i] )
  return new_xs

  polary = CustomJSTransform(args=dict(source=cds), v_func='''
  const new_ys = new Array(source.data.elev_project_vectors.length)
  for(var i = 0; i < new_ys.length; i++) {
      new_ys[i] = source.data.elev_project_vectors[i] * Math.cos(source.data.elev_angles[i] )
  return new_ys
  event_scatter = p.scatter(
    x=transform('elev_project_vectors', polarx),
    y=transform('elev_project_vectors', polary),
    fill_color='DepthClass', fill_alpha=0.5,
    line_color='black', line_alpha=0.5,

I am currently using Bokeh 1.3.4, and we are providing the content to embed via iframe using a Heroku instance that is running the Bokeh server. The div for the iframe is:

<div class="cell ">
    <iframe src="https://invlabs-bokehserver.herokuapp.com/bkapp_event_radial_plot" scrolling="yes" style="width:100%;min-height:600px;margin:0 auto; z-index:1;"></iframe>

Any ideas or feedback much appreciated. Thanks again to the core Bokeh developers! We have found Bokeh to be a very powerful and high quality tool. When we get this web app polished and in a final form, we hope to include it in the Showcase page.

Patrick Wright

@Patrick_Wright Just checking preliminaries, have you attempted to set the properties listed here that control various legend dimensions?


Another option to consider might be to make the legend orientation horizontal and placing them above the main plot drawing.

@Bryan yes I have tried adjusting/defining many of the Legend properties, although I may keep poking there. I assume if these are not defined, they will acquire default settings. I have not tried defining spacing yet, I will try that.

The horizontal layout above the plot is a good idea, I may try that also. Thanks!

Yes all the defaults are viewable in the class definition:


I thought defaults were rendered in the refguide entries too but it seems not.

Another idea you might try, if it is only ever just that corner that overlaps, is to set the legend background to be transparent.

@Bryan thanks for all the suggestions. For now, I decided to put the “Depth” legend below the plot area, with a horizontal layout. I also decided it was ridiculous how I was creating 5 individual legends with different glyph sizes to get different radius circles in the legend (see code above). I tried also putting these below the plot area, but it kept messing with the layout and was difficult to make work. When Issue #2603 is closed, I will be able to make my second legend normally. (I also have an SO question about having different sized glyphs in the legend, but I didn’t see this open issue at the time…) For now, I dropped this second legend entirely.

You can see my updated layout in the links included above.

There is still some slight weirdness on the legend label position in the iframe view (despite having label_text_baseline defined), but this is relatively minor:


1 Like