Create Interactive Charts Using Plotly.js, Part 2: Line Charts

Create Interactive Charts Using Plotly.js, Part 2: Line Charts

In the Getting Started With Plotly.js tutorial of this series, you were presented with a quick start guide that briefly covered all the features, bundles and chart types available in the library. If you have not already read that tutorial, I would suggest that you go over it at least once to get a broad idea of the Plotly.js library.

In this tutorial, you will learn how to create line charts in Plotly. I have also written another series in the past about a lightweight library called Chart.js that can be used to create canvas-based charts in JavaScript. In one tutorial that is titled Getting Started With Chart.js: Line and Bar Charts, I covered the process of creating line charts using Chart.js. People who have read that tutorial will probably remember that you have to set the type attribute to line to create a line chart in Chart.js.

With Plotly.js, you cannot set the type attribute to line to create a line chart. You will have to set the type attribute to scatter and then set the mode attribute to "lines", "lines+markers", or "lines+markers+text". Keeping this in mind, let's get started with the tutorial and create some awesome line charts.

Creating a Basic Line Chart

In this section, we will plot two different scatter traces in the form of line charts using Plotly. The code is very similar to the one we used in the last tutorial to create a line chart using a single trace. Here is the actual code:

var lineDiv = document.getElementById('line-chart');

var traceA = {
  x: [1, 5, 13, 24, 35, 46, 60],
  y: [80, 40, 70, 65, 15, 75, 49],
  type: 'scatter'
};

var traceB = {
  x: [4, 9, 17, 21, 31, 42, 56],
  y: [64, 81, 3, 49, 25, 17, 26],
  type: 'scatter'
};

var data = [traceA, traceB];

var layout = {
  title:'A Line Chart in Plotly'
};

Plotly.plot( lineDiv, data, layout );

Styling the Chart Lines Using Different Attributes

At this point, everything in our chart has been created using the default options. The rest of the sections in the tutorial will show you how to use all the attributes available for the scatter type trace to create custom line charts. The options to change the appearance of a plotted line are all available under the line key.

The color of a line can be specified using the color key. You can also control the width of a line by using the width attribute. The width is specified in pixels, and its default value is 2. 

The shape of the line between different points that are being plotted can be specified using the shape attribute. The shape is linear by default, but you can also set it to spline, vh, hv, hvh, or vhv. When shape is set to linear, there are no bends in the line that connects two consecutive points. In the case of vh, hv, hvh, and vhv, the lines are never drawn at an angle. They are either horizontal or vertical, with a 90-degree bend that can occur at the first point, second point, mid point, or both the points, depending on the specified shape value. 

The last option is to set the shape to spline. In this case, the line actually becomes a curve, with no sharp bends. The smoothness of this curve can be set with the help of another attribute called smoothing. This attribute can accept any value between 0 and 1.3 inclusive. Setting it to zero will result in straight lines like the linear value. Here is some code that uses all these attributes to plot five different lines on a chart:

var lineDiv = document.getElementById('line-chart');

var traceLinear = {
  x: [1, 5, 13, 24, 35, 46, 60],
  y: [180, 160, 170, 165, 175, 175, 149],
  type: 'scatter',
  name: 'Linear Shape',
  line: {
    shape: 'linear',
    color: 'rgb(255, 157, 98)'
  }
};

var traceSpline = {
  x: [1, 5, 13, 24, 35, 46, 60],
  y: [150, 130, 140, 135, 145, 145, 119],
  type: 'scatter',
  name: 'Spline Shape',
  line: {
    shape: 'spline',
    color: 'rgb(157, 255, 98)'
  }
};

var traceVH = {
  x: [1, 5, 13, 24, 35, 46, 60],
  y: [120, 100, 110, 105, 115, 115, 89],
  type: 'scatter',
  name: 'VH Shape',
  line: {
    shape: 'vh',
    color: 'rgb(157, 98, 255)'
  }
};

var traceHVH = {
  x: [1, 5, 13, 24, 35, 46, 60],
  y: [90, 70, 80, 75, 85, 85, 59],
  type: 'scatter',
  name: 'HVH Shape',
  line: {
    shape: 'hvh',
    color: 'rgb(98, 157, 255)'
  }
};

var traceSplineB = {
  x: [1, 5, 13, 24, 35, 46, 60],
  y: [60, 40, 50, 45, 55, 55, 29],
  type: 'scatter',
  name: 'Spline Shape (1.3)',
  line: {
    shape: 'spline',
    smoothing: 1.3,
    color: 'rgb(255, 98, 157)'
  }
};

var data = [traceLinear, traceSpline, traceVH, traceHVH, traceSplineB];

var layout = {
  title:'Different Shapes of a Line Chart',
  yaxis: {
    rangemode: 'tozero'
  }
};

Plotly.plot( lineDiv, data, layout );

The following line chart shows the difference between different values of the shape attribute. I have used the name parameter to assign different names to each line so that you can clearly see the shape created by each value.

One more parameter that you can use while drawing the plot lines is the dash parameter. You can set this parameter to a string value to set a dash style for your lines. Valid values for this attribute are: solid, dot, dash, longdash, dashdot, and longdashdot. 

Another option is to provide the dash length yourself using a list of pixel values like "4px, 4px, 10px". The following demo uses different values of the dash property to create unique line styles.

Setting Fill Colors and Markers

The area under a plotted line in a graph remains transparent by default, but you can fill it with a color of your choice if you wish to. This can be achieved by using the fill parameter. It is set to none by default, but other valid values include tozeroy, tozerox, tonexty, tonextx, toself, and tonext. 

The value tozeroy will fill all the area starting from the line trace to y=0. Similarly, the value tozerox will fill all the area starting from the line trace to x=0. The tonexty and tonextx values fill all the area between endpoints of the current trace and the trace before it. When no other traces are left, these values will act like tozeroy and tozerox respectively. You can use these two values to create stacked graphs. 

As the name suggests, the value toitself connects the endpoints of the trace to each other, forming a closed shape. The last value, tonext, fills all the available space between two traces only when one trace is completely enclosing the other one.

By default, the area specified by the value of the fill parameter is filled using a semi-transparent variant of the line color, marker color, or marker line color, depending on availability. However, you can specify your own color to fill that area using the fillcolor parameter.

The dots that represent the plotted points on the graph are marked using circles by default. Plotly.js also provides a lot of other marker symbols for you to choose from. There are about 44 different marker symbols, and almost all of them are available in four different versions. You can find a list of all these symbols in the documentation. The markers also have opacity, size, and color parameters that let you control the opacity, size, and color of these symbols. The default value of opacity is 1, and the default value of size is 6.

Here is a little code that uses all the parameters from this section to create line charts with filled areas and non-circular markers:

var lineDiv = document.getElementById('line-chart');

var traceA = {
  x: [1, 8, 13, 24, 35, 46, 60],
  y: [180, 150, 210, 165, 175, 185, 220],
  type: 'scatter',
  name: 'marker: circle',
  fill: 'tonexty',
  marker: {
    symbol: 'circle',
    size: 10
  }
};

var traceB = {
  x: [1, 8, 13, 24, 35, 46, 60],
  y: [130, 120, 160, 135, 145, 145, 119],
  type: 'scatter',
  name: 'marker: diamond',
  fill: 'tonexty',
  marker: {
    symbol: 'diamond',
    size: 10
  }
};

var traceC = {
  x: [1, 8, 13, 24, 35, 46, 60],
  y: [110, 100, 105, 100, 115, 85, 95],
  type: 'scatter',
  name: 'marker: pentagon',
  fill: 'tonexty',
  marker: {
    symbol: 'pentagon',
    size: 12
  }
};

var traceD = {
  x: [1, 8, 13, 24, 35, 46, 60],
  y: [70, 80, 60, 85, 75, 50, 59],
  type: 'scatter',
  name: 'marker: star',
  fill: 'tonexty',
  marker: {
    symbol: 'star',
    size: 12
  }
};

var traceE = {
  x: [1, 8, 13, 24, 35, 46, 60],
  y: [30, 20, 30, 15, 55, 15, 30],
  type: 'scatter',
  name: 'dash: 4px, 4px, 10px',
  fill: 'tonexty',
  marker: {
    symbol: 'bowtie',
    size: 12
  }
};

var data = [traceE, traceD, traceC, traceB, traceA];

var layout = {
  title:'Dash values for a Line Chart',
  xaxis: {
    rangemode: 'tozero'
  },
  yaxis: {
    rangemode: 'tozero'
  }
};

Plotly.plot( lineDiv, data, layout );

When filling the area between different traces with a given color, you need to keep in mind that all these parameters fill the trace area based on the order in which the traces are supplied to the plot() function, and not the order in which the traces were declared. 

In our case, traceE is the first trace, and there is no trace before it. This means that the value tonexty effectively becomes tozeroy this time. We have passed traceD after traceE, so all the points between these two traces will be filled with the color of traceD. If we had instead passed traceA first, then the fill area would have extended all the way from the top to y=0. That might not be a desirable outcome in some cases. So you should always keep the order in which the traces are passed in mind.

Conclusion

In this tutorial, you learned how to create line charts using Plotly.js. Different sections of the tutorial discussed a variety of customization options, like setting the shape, color, and width of the lines. 

You also learned about different marker symbols and the fill option provided by Plotly.js that can be used to further customize a line chart. I have only covered some of the major attributes here, but there are a few others, like showlegend and hoverinfo, that you should probably know about. You can read about all these attributes under the scatter reference section on the Plotly website.

In the next tutorial of the series, you will learn how to create bar charts in Plotly. If you have any questions or tips related to this tutorial, feel free to share them in the comments.