Retrieving the range of the x-axis of a zoomed in plot

📊 Plotly Python
1

I’m trying to retrieve the range of the x-axis when you zoom in a plotly scatter plot, but when I tried to access it using:

figure['layout']['xaxis']['range']

it just returns ‘None’.
This is the code that I am using to create the graph:

# self.plot contains the dataframe passed to the function to be plotted
self.plot = df
# Creates the plotly plot figure
self.fig = self.plot.iplot(asFigure=True,kind='scatter', xTitle='Date', yTitle='Temperature')
# Displays the plot
iplot(self.fig, show_link=False)

I’m using cufflinks to create the plot from a pandas DataFrame so I don’t explicitly set the layout.xaxis.range to anything. Also I’m using a Jupyter Notebook to display the graph, if that helps at all. So is there any way of getting the range of the x-axis of the current view window of the plot? Thanks in advance!

2

Graph parameters don’t dynamically change with chart actions. However, as you zoom, the plot does emit data regarding the new x-axis range, you’re unable to to access this information directly in Python. Instead you’d have to use javascript: https://plot.ly/javascript/zoom-events/

1 Like
3

Is this still up-to-date information, is it possible to access x-axis range within python (maybe though some interpreter of JS)?

4

Hi @davzup89, there is a way to do this in python, it’s not recommended though:

5

Thanks. Just being curious, for which reasons it is not recommended? Overload due to user activity?

6

@davzup89 I actually don’t know, that’s just the warning message you get while using it. :rofl:

Might be interesting to know, though. @adamschroeder

7

hey @davzup89 hi @AIMPED

The issue with full_figure_for_development() is that it spins up Kaleido from Python; basically it’s running a whole browser on the server to render the figure, which is very resource-hungry in terms of RAM, CPU, boot-time, etc.

So it’s very much not recommended for production, hence the name for_development. in the comment, the figure is already being rendered in a browser in a Dash app, so you can use the client’s browser and a client-side callback to grab the relevant info if you like. It much cheaper, resource-wise.

2 Likes