pyvisacharts

pyvisacharts

This package wraps @visa/charts web components for use in Python and jupyter notebooks, leveraging the widget-cookiecutter Python package. You can find pyvisacharts on pypi, installation steps provided below.

---

Installation Steps

• Using pip:

  $ pip install pyvisacharts
  

• or conda:

  $ conda install -c conda-forge pyvisacharts
  

---

Components with Ready status in this bundle

• @visa/bar-chart
• @visa/clustered-bar-chart
• @visa/stacked-bar-chart
• @visa/line-chart
• @visa/pie-chart
• @visa/scatter-plot
• @visa/heat-map
• @visa/circle-packing
• @visa/parallel-plot
• @visa/dumbbell-plot
• @visa/world-map
• @visa/alluvial-diagram

# Use VCC as Python functions

Step 1: Install:

$ pip install pyvisacharts

Step 2: Use component as any other Python function

import pyvisacharts as vcc
import pandas as pd

bar_chart_data = pd.read_json("https://github.com/visa/visa-chart-components/tree/main/packages/charts-pythondocs/demo_data/bar_chart_data.json")
line_chart_data = pd.read_json("https://github.com/visa/visa-chart-components/tree/main/packages/charts-pythondocs/demo_data/line_chart_data.json")

vcc.BarChart(
    accessibility={
        "purpose": "Demonstration of a bar chart built with VCC and minimal properties provided.",
        "statisticalNotes": "This chart is using dummy data."
    },
    data=bar_chart_data,
    ordinalAccessor="item",
    valueAccessor="value"
)

vcc.LineChart(
    accessibility={
        "purpose": "Demonstration of a line chart built with VCC and minimal properties provided.",
        "statisticalNotes": "This chart is using dummy data."
    },
    data=line_chart_data, # a pandas data frame
    ordinalAccessor="date",
    valueAccessor="value",
    seriesAccessor="category",
    config={
        "hoverOpacity": 0.25
    }
)

See our VCC Demo Notebook for more examples.

---

Development Steps

To the python widget locally, you will need to follow the below installation and build steps to symlink the necessary packages across the monorepo.

    $ yarn
    $ yarn dev --i
    $ yarn dev --b
    $ yarn dev --ipy
    $ yarn dev --spy (spins up a local jupyter notebook)
    or
    $ yarn dev --lpy (spins up a local jupyter lab)

After running these commands, the js lib @visa/charts will by symlink'd and a jupyter notebook will be spun up locally for development and testing work. If you update the js build and/or python code you will likely need to restart/refresh the juptyer notebook to see development changes reflected.

In addition to the core project team, special thanks to Luis Chaves Rodriguez (@visa) for his assistance in development of pyvisacharts.