IridiumIO/pyChart.js

An easy to use, class-based approach to implementing Chart.js into Python projects.

Initially designed as a Django app, it is now self-contained and outputs chart data in JSON, meaning it can easily be used in:

• Django
• Flask
• AJAX/Rest API requests
• Other Python projects

Getting Started

Install with pip

> pip install pychart.js

You will need to have chart.js or chart.min.js ready for use in your HTML document. The following is a drop-in CDN script to use:

2.8.0:  <script src="https://cdnjs.cloudflare.com/ajax/libs/Chart.js/2.8.0/Chart.min.js"></script>
3.9.1:  <script src="https://cdn.jsdelivr.net/npm/[email protected]/dist/chart.min.js"></script>

Basic Usage

1. Chart Element in HTML and JS

As stated above, the output is a JSON object which can be used directly in any template. The following is an example HTML document with a simple chart element. Here, the chart object is going to be passed into the {{ chartJSON | safe }} tag. Note here that it has to be flagged as safe otherwise it will not work.

  
<canvas id="myChart"></canvas>

<script>
    var data = {{ chartJSON | safe }}
    var ctx = document.getElementById("myChart").getContext('2d');
    var myChart = new Chart(ctx, data);
</script>

2. Python Code

The following is a minimal example of a chart you can generate and pass into your HTML using a Django view.

from pychartjs import BaseChart, ChartType, Color                                     

class MyBarGraph(BaseChart):

    type = ChartType.Bar

    class data:
        label = "Numbers"
        data = [12, 19, 3, 17, 10]
        backgroundColor = Color.Green

def homepage(request):

    NewChart = MyBarGraph()
    NewChart.data.label = "My Favourite Numbers"      # can change data after creation
    
    ChartJSON = NewChart.get()

    return render(request=request,
                  template_name='main/home.html',
                  context={"chartJSON": ChartJSON})

3. The Result

Extending With More Features

The above is what's achievable with minimal code but you can do almost anything the JS equivalent can do. Where a feature isn't implemented natively in pyChart.js, you can use a dictionary or list as per normal.

Here's a skeleton of what can be used:

class MyChart(BaseChart):

    type = ChartType.Line

    class labels:
        # ...

    class data: 
        # ...

    class options:
        # ...

    class pluginOptions:
        # ...

Base Variables

Chart Type

Can use the variables in ChartType or can be entered directly

type = ChartType.Line   #...from pychartjs import ChartType
type = 'Bar'

In-line JS

Callbacks to Javascript functions or direct code can be implemented anywhere in the chart class as long as it is a string encapsulated within <<>>

• note: Using this will render the output non-compliant to the JSON standard, and as such it likely will not work with AJAX/REST

callback = "<<myJavascriptFunction>>"
inlineJS = "<<function(value, index, values) {
                      return '$' + value;
                  } >>"

Subclasses

Labels Class

Used to define the labels used for each data item. If it is left blank, labels will be generated automatically from the first data collection.

Can be any of:

• A single list of strings for all labels
• Independent variables for each label
• (Planned) Select from pre-determined lists for common datasets, e.g. Days, Months

class labels:
    grouped = ['Mon', 'Tue', 'Wed']
    # or
    day1 = 'Mon'
    day2 = 'Tue'
    day3 = 'Wed'

Data Class

Used to define data or datasets. If you only have one dataset, this can be defined directly in the class. Otherwise, use subclasses for each dataset. For each subclass, the name of the class is used as the label if one isn't specified.

Can be either of:

• A single dataset, defined directly as variables in the class
• Multiple datasets, each with their own subclass.

Rules:

• Must include a data variable of type list
• Must not have functions/methods. These will not work due to the reference methods used internally. However, you can use in-line operators or call to a function outside the Chart class. You just can't define a function within the dataset class.
• If you don't want a variable to be compiled, prefix it with an underscore, e.g. _color

#One Dataset:
class data: 
    data = [12, 19, 3, 17, 10]
    label = "Fruit Eaten"
    backgroundColor = Color.Palette(Color.Green)
    borderColor = Color.Hex(0xA2E6B1FF)

#Multiple Datasets:
class data: 

    class Apples:
        data = [2, 8, 3, 3, 2]

    class Oranges:
        data = [2, 3, 0, 12, 1]
        label = "Bananas"  # Overrides the generated label 'Oranges'

Options Class

NOTE: Chart.JS 3.0 introduced many significant changes to how the Options are laid out. The examples below work for 2.94 and below but need modification if you're using 3.0 or higher. (For example scales is no longer a list of dictionary items).

Define extended options here. Note however that plugin options get defined under their own heading, not in here (to avoid over-nesting). Options can be defined either as dictionaries or by using the Options.Generic() object. builders for Title, Legend, Legend Labels and Layout are provided for convenience.

Can include:

• Top-level options as variables
• Deeper options as dictionaries or Options.Generic objects
• callbacks or javascript functions can be included if the variable is surrounded by << >> tags

class options: 

    # Object-based
    title = Options.Title("My Fruit Consumption") 
    animation = Options.Generic(duration=1000)

    _labels = Options.Legend_Labels(fonColor=Color.Gray, fullwidth=True)
    legend = Options.Legend(position='bottom', labels=_labels)


    # Dictionary-based
    title = {"text": "My Fruit Consumption", "display": True} 
    animation = {"duration": 1000}
    hover = {"animationDuration": 500 }
    responsiveAnimationDuration = 0

    legend = {
            'position': 'bottom', 
            'labels': {
                'fontColor': Color.Gray, 
                'fullWidth': True
                }
            }

Plugin Options Class

Used to define options for plugins. Could theorectically be included in the above options class, but has been split out here to reduce clutter.

Can include:

• One subclass per plugin (class name = plugin name)
• To disable a plugin, no subclass is required; simply put pluginName = False at the top of the pluginOptions class

class pluginOptions:

    stacked100 = False   # Disables the plugin 'stacked100'

    class colorSchemes: 
        scheme = "brewer.Paired12"
        custom = "<< customColorFunction >>"

Colors

Some rudimentary color functions are provided to make generating charts and graphs easier.

• All colors are returned in a formatted string 'rgba(R, G, B, A)' regardless of input type
  • Color.Hex() accepts a string or a Hex Integer.
  • Color.RGBA() accepts either RGB or RGBA values.
  • Color.HSLA()
  • Color.HSVA()

from pychartjs import Color

color1 = Color.Magenta                  #22 basic colors available
color2 = Color.Hex("#242424")
color3 = Color.Hex(0x242424FF)
color4 = Color.RGBA(35, 22, 225)
color5 = Color.RGBA(35, 22, 255, 1.0)

>>> 'rgba(240, 50, 230, 1.0)'

Color Palettes

Color palettes can be generated using the Color.Palette() function. It returns a list of rgba() formatted colors which can be used directly in the chart.

• BaseColor = Color to use as the generator for the palette. Must be a formatted string as above, which means it can accept any of the Color.X() functions as an input.
• n = Number of colors to generate. Defaults to 5.
• generator = Component to use to generate palette. Can be 'hue', 'saturation', 'lightness' or 'alpha'. Defaults to saturation

p1 = Color.Palette(Color.Red)
p2 = Color.Palette(Color.Hex("#432475"), n=3, generator='lightness')

>>> ['rgba(55, 30, 97, 1.0)', 'rgba(111, 60, 195, 1.0)', 'rgba(183, 157, 224, 1.0)']

Color Gradients

Linear and Radial gradients can be generated using the Color.JSLinearGradient and Color.JSRadialGradient objects. Note that this returns an escaped javascript generator function. Construction can be done in a similar fashion to the javascript way as seen here, or alternatively the whole gradient can be build in the constructor.

• chartContextName = javascript chart context that you use in your HTML. Defaults to 'ctx'.
• x1, y1, x2, y2 (Linear) = start and stop gradient coordinates
• x1, y1, r1, x2, y2, r2 (Radial) = start, stop and radii of the circles to generate the gradient between.
• *colorStops = optional colorstops can be passed through as tuples to condense creation of gradients into one line

Stops can be added by calling the addColorStop() function

#Define and apply color separately

_color = Color.JSLinearGradient('ctx', 0, 0, 1000, 0)                    
_color.addColorStop(0, Color.Blue)
_color.addColorStop(1, Color.Green)

backgroundColor = _color.returnGradient()

#Inline construction of gradient

backgroundColor = Color.JSLinearGradient('ctx2', 0, 0, 0, 1000,
                                          (0, Color.Red),
                                          (1, Color.Blue)
                                          ).returnGradient()

#Radial Gradient

backgroundColor = Color.JSRadialGradient('ctx2', 50, 50, 0, 50, 50, 100
                                          (0, Color.Red),
                                          (1, Color.Blue)
                                          ).returnGradient()

Putting it all together

The following is an example of a complex chart that can be created with many of the above features:

class MyChart(BaseChart):
    
    type = ChartType.Bar
    
    class labels:
        group = ['Mon', 'Tue', 'Wed', 'Thu', 'Fri', 'Sat', 'Sun']
    
    class data:
        
        class apples: 
            data = [2, 8, 11, 7, 2, 4, 3]
            backgroundColor = Color.Palette(Color.Hex('#30EE8090'), 7, 'lightness')
            borderColor = Color.Green
            yAxisID = 'apples'
            
        class totalEnergy: 
            label = "Total Daily Energy Consumption (kJ)"
            type = ChartType.Line
            data = [5665, 5612, 7566, 8763, 5176, 5751, 6546]
            backgroundColor = Color.RGBA(0,0,0,0)
            borderColor = Color.Purple
            yAxisID = 'totalenergy'
    
    class options: 
        
        title = Options.Title("Apples I've eaten compared to total daily energy")
        
        scales = {
            "yAxes": [
                {"id": "apples",
                 "ticks": {
                     "beginAtZero": True,
                     "callback": "<<function(value, index, values) {return value + ' Big Ones';}>>",
                     }
                },
                {"id": "totalenergy",
                 "position": "right",
                 "ticks": {"beginAtZero": True}
                }
            ]
        }

Output

Another example

class NewChart(BaseChart):

    type = ChartType.Line

    class labels:
        Years = list(range(2017, 2023))

    class data:
        class Whales:
            data = [80, 60, 100, 80, 90, 60]
            
            _color = Color.JSLinearGradient('ctx', 0, 0, 1000, 0)
            _color.addColorStop(0, Color.Green)
            _color.addColorStop(1, Color.Purple)
            
            borderColor = _color.returnGradient()
            fill = False
            pointBorderWidth = 10
            pointRadius = 3

        class Bears:
            data = [60, 50, 80, 120, 140, 180]
            borderColor = Color.JSLinearGradient('ctx', 0, 0, 1000, 0,
                                                 (0, Color.Red), 
                                                 (1, Color.Magenta)
                                                 ).returnGradient()
            fill = False
            pointBorderWidth = 10
            pointRadius = 3

        class Dolphins:
            data = [150, 80, 60, 30, 50, 30]
            borderColor = Color.JSLinearGradient('ctx', 0, 0, 1000, 0,
                                                 (0, Color.Yellow), 
                                                 (1, Color.Orange)
                                                 ).returnGradient()
            fill = False
            pointBorderWidth = 10
            pointRadius = 3

    class options:

        title   = Options.Title(text="Wildlife Populations", fontSize=18)
        
        _lables = Options.Legend_Labels(fontColor=Color.Gray, fullWidth=True)
        legend  = Options.Legend(position='Bottom', labels=_lables)
        
        _yAxes = [Options.General(ticks=Options.General(beginAtZero=True, padding=15, max=200))]
        
        scales = Options.General(yAxes=_yAxes)

Output

More Examples

Can be found here