I'm trying to embed an interactive plotly (or bokeh) plot into excel.
To do this I've tried the following three things:
embed a Microsoft Web Browser UserForm into excel, following:
How do I embed a browser in an Excel VBA form?
This works and enables both online and offline html to be loaded
creating a plotly html
'''
import plotly
import plotly.graph_objects as go
x = [0.1, 0.5, 1.0, 1.5, 2.0, 2.5, 3.0]
y = [i**2 for i in x]
fig = go.Figure()
fig.add_trace(go.Scatter(x=x, y=x, mode='markers', name="y=x", marker=dict(color='royalblue', size=8)))
fig.add_trace(go.Scatter(x=x, y=y, name="y=x^2", line=dict(width=3)))
plotly.offline.plot(fig, filename='C:/Users/.../pythonProject/test1.html')
repointing the webbrowser object in excel using .Navigate to the local plotly.html. Banner pops up with
".... restricted this file from showing active content that could access your computer"
clicking on the banner, I run into this error:
The same HTML can be opened in a browser.
Is there any way to show interactive plots in excel?
Finally, I have managed to bring the interactive plot to excel after a discussion from Microsoft QnA and Web Browser Control & Specifying the IE Version
To insert a Microsoft webpage to excel you have to change the compatibility Flag in the registry editor
Computer\HKEY_LOCAL_MACHINE\SOFTWARE\Microsoft\Office\ClickToRun\REGISTRY\MACHINE\Software\Wow6432Node\Microsoft\Office\16.0\Common\COM Compatibility{8856F961-340A-11D0-A96B-00C04FD705A2}
Change the DWord 0 instead of 400
Now you can insert the web browser object to excel, Step by step details are here
Edit the HTML File generated from plotly manually by adding a tag for Using the X-UA-Compatible HTML Meta Tag
Originally generated HTML file from plotly looks like this
<html>
<head><meta charset="utf-8" /></head>
<body>
Modified HTML with browser compatibility
<html>
<head>
<meta charset="utf-8" />
<meta http-equiv="X-UA-Compatible" content="IE=edge" />
</head>
<body>
After this, I can able to view the interactive plot in excel also able to do the interactions same as a web browser
Macro used:
Sub Button4_Click()
ActiveSheet.WebBrowser1.Navigate "file:///C:/Users/vignesh.rajendran/Desktop/test5.html"
End Sub
As mentioned by #jerlich, Excel blocks javascript. You should try the workaround they linked if you want full interactivity.
If you want at least some degree of controllability or interactivity, try using xlwings. With excel buttons, you can still have some communication between Excel and Python (including reading data and sending graphs).
The limitations are:
Only being able to use data entries and buttons, instead of the default plotly interactive features. Many could be replicated, but it would be more work.
It will only work on a computer you can set up your python script on (however, it looks like xlwings pro allows you to store your program inside the file)
It seems you will need to pass plotly graphs by saving and then adding the figure, because directly passing them requires xlwings pro. Direct passing without pro is possible with MatPlotLib.
Plotly guide to making interactive(ish) graphs
xlwings docs on MatPlotLib and Plotly graphs
Interactive plots require javascript to work. Excel, for security reasons, blocks that javascript. You can put a static image easily into excel.
The challenge of including javascript in excel has been addressed in this question: How can I use JavaScript within an Excel macro?
I like your question! And I'd wish I could give you a better answer, but It seems that the only way you can achieve anything remotely resembling an interactive plotly plot would be to use pyxll and follow the steps outlined under Charts and plotting / Plotly including a plot function like this:
from pyxll import xl_func, plot
import plotly.express as px
#xl_func
def plotly_plot():
# Get some sample data from plotly.express
df = px.data.gapminder()
# Create a scatter plot figure
fig = px.scatter(df.query("year==2007"),
x="gdpPercap", y="lifeExp",
size="pop", color="continent",
log_x=True, size_max=60)
# Show the figure in Excel using pyxll.plot
plot(fig)
This will produce the following plot:
Alas, this will not be a fully interactive plotly plot like we all know and love, since it's also stated on the very same page that:
The plot that you see in Excel is exported as an image so any
interactive elements will not be available. To make a semi-interactive
plot you can add arguments to your function to control how the plot is
done and when those arguments are changed the plot will be redrawn.
But as far as I know this is as close as you'll get to achieving what you're seeking in your question. If you're not limited to Excel, but somehow limited to the realm of Microsoft, one of the commenters mentioned that you can unleash a fully interactive plotly plot in PowerBI. If that is an option, you should take a closer look at Is it possible to use R Plotly library in R Script Visual of Power BI?. This approach uses R though...
Related
If an EXCEL chart is pasted into a PowerPoint presentation as an EXCEL object, it is possible to hover the mouse over the line and see it's value. My boss likes this feature. I don't like having to do charts in EXCEL.
It is possible to get the same effect with bokeh or plotly, but as far as I am aware, that either relies on a stand-alone html file or a server instance.
Is it possible to paste a bokeh chart into a PowerPoint presentation preserving the feature that when you hover your mouse over a point, the value (and some other information) pops up?
Or is there another python solution that would allow this feature in PowerPoint (ppt is essential) while also allowing plots to be generated via code?
An available option is to create powerpoints with Plotly: https://towardsdatascience.com/embed-interactive-plots-in-your-slides-with-plotly-fde92a5865a and https://evidencen.com/how-to-embed-plotly-graphs-in-powerpoint/
Also it seems this question has been asked before and you can hyperlink the HTML source in the .pptx file: Output of Plotly in PowerPoint
yes. it's a hover tool. here is an example
So, I'm learning how to use matplotlib and stumbled into MPLD3 as the most used way to get these plots in a webpage. However, MPLD3 doesn't support tick formatting and that's something critical for the project I'm in right now. I'd like to know if there is another way to add a matplotlib graph to a webpage while keeping the tick formatting and also having tooltips to display data on hover.
Thanks!
For a web application I would consider bokeh. You can output html with interactive charts or run another backend server. If you go to the ipython notebook tutorial section 10 has demos for embedding.
Alternatively you can just save images from matplotlib and use them as static assets in your page.
I have a Flask web app that uses Bokeh to deliver interactive charts. My end goal is to export whatever the current Bokeh view is to Matplotlib (so that I can create a printable pdf file after that). This would include how the current axes look like after the user zooms and pans. Is there a way to export that data so that I can create those Matplotlib charts behind the scenes? (Printing the page directly or printing to pdf results in low-quality and blurred charts.)
Thanks!
No, currently there is no way to export bokeh to matplotlib. Actually you can do it otherway. You can create matplotlib plot, save, and after that you can export matplotlib to bokeh. I think this is the best option. Eventually you can export bokeh plot as png but it still would not solve problem with quality
I am using bokeh to plot my math functions created with python/numpy.
I would like to use sliders as shown in
http://docs.bokeh.org/en/latest/docs/server_gallery/sliders_server.html
Once I create the html file with the plot, I would like to select different values on the sliders which modify the plot and then read back the chosen values in into python to use it for other manipulations.
What is the best way to read the chosen value on the slider from the html file back into python ?
I saw pyquery could be useful, but I cannot really figure that out.
Any suggestions would be appreciated based on above scenario.
There are two slider examples in the bokeh repo, where the slider is connected back to python via the bokeh server.
Sliders App
Taylor server
If this isn't what you were after, can you elaborate a little more?
A static HTML file and the state of the slider is inside a web browser and never reflected back to the HTML file. What you should be doing is to use bokeh-server - answered via bokeh-google group
I have created a plot in R using rbokeh and I want to convert it to HTML/Javascript in order to embed it inside a web page. I'm currently able to achieve this with mpld3 on Python (and I know that bokeh on python does it too) but I want to be able to do this with rbokeh in R. So basically I'm looking for something similar to Python's mpld3.fig_to_html(), e.g:
fig, ax = plt.subplots()
ax.p1 = plt.bar(...)
html = mpld3.fig_to_html(fig) # <- converts the plot to html/javascript!
print html # prints out the html/javascript code as text
Or using Python's bokeh module: embed.autoload_static()
Can rbokeh plots be converted to html/JavaScript code?
Many thanks in advance.
It appears that, to date, saveWidget is the closest to what I want, so thanks hrbrmstr. I wanted to avoid having to read and write from the disk every time a graph is exported to html but it's not possible yet.