2

I want to upload a csv-file with complex numbers into a bokeh web app. The solution of DuCorey works perfectly for integer and floats, see: Upload a CSV file and read it in Bokeh Web app But when I change the values_type from Int to Complex in models.py and insert a complex number into the csv-file, it doesn't work anymore. I do not want to plot the imported array as DuCorey did (and it would not work in his example). Just import the values and f.e. print them.

I changed the following three files:

models.py

from bokeh.core.properties import List, String, Dict, Int, Complex, Float
from bokeh.models import LayoutDOM

class FileInput(LayoutDOM):
    __implementation__ = 'static/js/extensions_file_input.coffee'
    __javascript__ = './input_widget/static/js/papaparse.js'

    value = String(help="""
    Selected input file.
    """)

    file_name = String(help="""
    Name of the input file.
    """)

    accept = String(help="""
    Character string of accepted file types for the input. This should be
    written like normal html.
    """)

    data = List(Dict(keys_type=String, values_type=Complex))

csv-file

x,y
1,2
3,3
3,5+1j
10,25

and main.py

from bokeh.core.properties import List, String, Dict, Int
from bokeh.models import LayoutDOM

from bokeh.layouts import column
from bokeh.models import Button, ColumnDataSource
from bokeh.io import curdoc
from bokeh.plotting import Figure

import pandas as pd

from models import FileInput

button_input = FileInput(id="fileSelect",
                         accept=".csv")

def change_plot_data(attr, old, new):
    new_df = pd.DataFrame(new)
    print(new_df)

button_input.on_change('data', change_plot_data)

layout = column(button_input)
curdoc().add_root(layout)

I get the following error message: DeserializationError('Complex expected Complex, got 5+1j').

Thanks for all advices!

astair
  • 21
  • 3

1 Answers1

0

The approach used in the other example uses a JavaScript library to do the parsing of the file, and send it back. AFAIK this will not work in this instance for two reasons:

  • the papaparse JS library does not, itself, know anything about parsing complex numbers
  • I'm not actually sure sending comlex values from JS back to Python is workable. AFAIK you are the first person to want to try it since 2013

I think to make this work your best bet is to simply send the entire contents of the raw file back to the Bokeh server (as one big String) and then use Pandas to parse the file from the string.

bigreddot
  • 33,642
  • 5
  • 69
  • 122
  • Thanks! Do you have or know about an example how to get the entire content of the file as a String into the Bokeh server? I'm really new to it... – astair Jun 13 '19 at 09:46
  • In your custom extension you could add a `raw_data = String(...)` property. When you set that in JS, it will automatically be sent to the Python side, and you can have a callback `my_model.on_change('raw_data',...)` that takes the value and passes it to pandas. I don't know how to read the local file, but evidently the [Papaparse](https://www.papaparse.com/) library used in the example you linked does, so you could study its source code. – bigreddot Jun 13 '19 at 15:51
  • Studying the code of papaparse was a good idea! In this case reading the documentation was enough. Simply setting the papaparse option `dynamicTyping: false` and change `values_type=String` in models.py do it. Valeu – astair Jun 15 '19 at 09:30