Adding CSS & JS and Overriding the Page-Load Template

Dash apps are rendered in the web browser with CSS and JavaScript.
On page load, Dash serves a small HTML template that includes references to
the CSS and JavaScript that are required to render the app.
This chapter covers everything that you need to know about configuring
this HTML file and about including external CSS and JavaScript in Dash
apps.

Adding Your Own CSS and JavaScript to Dash Apps

Dash supports adding custom CSS or JavaScript in your apps.

Create a folder named assets in the root of your app directory
and include your CSS and JavaScript
files in that folder. Dash automatically serves all the files that
are included in this folder. By default, the URL to request the assets
is /assets, but you can customize this with the assets_url_path argument
to dash.Dash.

Example: Including Local CSS and JavaScript

We’ll create several files: app.jl, a folder named assets, and
three files in that folder:

 

app.jl

This example has not been ported to Julia yet - showing the Python version instead.

Visit the old docs site for Julia at: https://community.plotly.com/c/dash/julia/20

from dash import Dash, html

app = Dash(__name__)

app.layout = html.Div([
    html.Div(
        className="app-header",
        children=[
            html.Div('Plotly Dash', className="app-header--title")
        ]
    ),
    html.Div(
        children=html.Div([
            html.H1('Overview'),
            html.Div('''
                This is an example of a simple Dash app with
                local, customized CSS.
            ''')
        ])
    )
])

if __name__ == '__main__':
    app.run(debug=True)

 

typography.css

body {
    font-family: sans-serif;
}
h1, h2, h3, h4, h5, h6 {
    color: hotpink
}

 

header.css

.app-header {
    height: 60px;
    line-height: 60px;
    border-bottom: thin lightgrey solid;
}

.app-header .app-header--title {
    font-size: 22px;
    padding-left: 5px;
}

 

custom-script.js

alert('If you see this alert, then your custom JavaScript script has run!')

When you run your app, it should look something like this:

Preview of the local-css example Dash app with the CSS styling applied

Embedding Images in Your Dash Apps

In addition to CSS and JavaScript files, you can include images in
the assets folder. An example of the folder structure:

- app.py
- assets/
    |-- image.png

 

In your app.jl file you can use the relative path to that image:

This example has not been ported to Julia yet - showing the Python version instead.

Visit the old docs site for Julia at: https://community.plotly.com/c/dash/julia/20

from dash import Dash, html

app = Dash(__name__)

app.layout = html.Div([
    html.Img(src='/assets/image.png')
])

if __name__ == '__main__':
    app.run(debug=True)

If placing images inside the assets folder isn’t an option, then you can
also embed images “inline” with base64 encoding:

This example has not been ported to Julia yet - showing the Python version instead.

Visit the old docs site for Julia at: https://community.plotly.com/c/dash/julia/20

import base64
from dash import Dash, html

app = Dash(__name__)

image_filename = 'my-image.png'

def b64_image(image_filename):
    with open(image_filename, 'rb') as f:
        image = f.read()
    return 'data:image/png;base64,' + base64.b64encode(image).decode('utf-8')

app.layout = html.Img(src=b64_image(image_filename))

Changing the Favicon

It is possible to override the default favicon by adding
a file named favicon.ico to your assets folder. Changes to
this file will implement cache-busting automatically.

- app.py
- assets/
    |-- favicon.ico

Adding External CSS/JavaScript

Example:

This example has not been ported to Julia yet - showing the Python version instead.

Visit the old docs site for Julia at: https://community.plotly.com/c/dash/julia/20

from dash import Dash, html


# external JavaScript files
external_scripts = [
    'https://www.google-analytics.com/analytics.js',
    {'src': 'https://cdn.polyfill.io/v2/polyfill.min.js'},
    {
        'src': 'https://cdnjs.cloudflare.com/ajax/libs/lodash.js/4.17.10/lodash.core.js',
        'integrity': 'sha256-Qqd/EfdABZUcAxjOkMi8eGEivtdTkh3b65xCZL4qAQA=',
        'crossorigin': 'anonymous'
    }
]

# external CSS stylesheets
external_stylesheets = [
    'https://codepen.io/chriddyp/pen/bWLwgP.css',
    {
        'href': 'https://stackpath.bootstrapcdn.com/bootstrap/4.1.3/css/bootstrap.min.css',
        'rel': 'stylesheet',
        'integrity': 'sha384-MCw98/SFnGE8fJT3GXwEOngsV7Zt27NXFoaoApmYm81iuXoPkFOJwJ8ERdknLPMO',
        'crossorigin': 'anonymous'
    }
]


app = Dash(__name__,
                external_scripts=external_scripts,
                external_stylesheets=external_stylesheets)

app.layout = html.Div()

if __name__ == '__main__':
    app.run(debug=True)

Controlling the Plotly.js Version Used by dcc_graph

The dcc_graph component leverages the Plotly.js library to render visualizations.

You can override the Plotly.js version by placing a Plotly.js bundle in the assets directory.

This technique can be used to: * take advantage of new features in a version of Plotly.js that is more recent than the one that is included in the currently installed version of Dash or Dash Design Kit. * take advantage of more desirable behavior of a version of Plotly.js that is less recent than the one that is included in the currently installed version of Dash or Dash Design Kit. We strive to make Plotly.js releases completely backwards-compatible, so you shouldn’t have to do this very often. * use a Plotly-distributed Plotly.js partial bundle or a custom-built Plotly.js bundle which only includes the subset of Plotly.js features that your Dash app uses. Partial bundles are smaller than the full Plotly.js bundles that come with the dcc_graph component and can therefore improve your app’s loading time.

Rendering LaTeX Inside dcc_graph Figures

To use the built-in Plotly.js capability of rendering LaTeX inside figure labels, the external_script functionality described above can be used: add external_scripts=["https://cdnjs.cloudflare.com/ajax/libs/mathjax/2.7.5/MathJax.js?config=TeX-MML-AM_CHTML" ] to the app = dash.Dash() call.

Customizing Dash’s Document or Browser Tab Title

The document title is the name of the web page that appears in your
web browser’s tab.

By default, it is Dash.

Update the Document Title Dynamically Based on the URL or Tab

To set the document title dynamically, you can use a clientside callback
that updates the document.title as a side effect. The example below
sets the document.title based off of the currently selected tab.

This example has not been ported to Julia yet - showing the Python version instead.

Visit the old docs site for Julia at: https://community.plotly.com/c/dash/julia/20

from dash import Dash, html, dcc, Input, Output, clientside_callback

app = Dash(__name__)

app.layout = html.Div([
    html.Div(id='blank-output'),
    dcc.Tabs(id='tabs-example', value='tab-1', children=[
        dcc.Tab(label='Tab one', value='tab-1'),
        dcc.Tab(label='Tab two', value='tab-2'),
    ]),
])


clientside_callback(
    """
    function(tab_value) {
        if (tab_value === 'tab-1') {
            document.title = 'Tab 1'
        } else if (tab_value === 'tab-2') {
            document.title = 'Tab 2'
        }
    }
    """,
    Output('blank-output', 'children'),
    Input('tabs-example', 'value')
)

if __name__ == '__main__':
    app.run(debug=True)

Updating the page based off of the URL would be similar: the input of the
callback would be the pathname property of dcc_location. See the
URLs and Multi Page Apps chapter for
a dcc_location example.

Customizing or Removing Dash’s “Updating…” Message

When a callback is running, Dash updates the document title
(the title that appears in your browser tab) with the “Updating…”
message.

Customize this message with the update_title= property:


Customizing Dash’s HTML Index Template

Dash’s UI is generated dynamically with Dash’s React.js front-end. So,
on page load, Dash serves a very small HTML template string that includes
the CSS and JavaScript that is necessary to render the page and some simple
HTML meta tags.

This simple HTML string is customizable. You might want to customize this
string if you wanted to:
- Customize the way that your CSS or JavaScript is included in the page.
For example, if you wanted to include remote scripts or if you wanted to
include the CSS before the Dash component CSS
- Include custom meta tags in your app. Note that meta tags can also be
added with the meta_tags argument (example below).
- Include a custom version of dash-renderer, by instantiating the
DashRenderer JavaScript class yourself. You can add request hooks this way, by providing
a hooks config object as in the example below.


Customizing dash-renderer with Request Hooks

This example has not been ported to Julia yet - showing the Python version instead.

Visit the old docs site for Julia at: https://community.plotly.com/c/dash/julia/20

from dash import Dash, html

external_stylesheets = ['https://codepen.io/chriddyp/pen/bWLwgP.css']

app = Dash(__name__, external_stylesheets=external_stylesheets)

app.renderer = 'var renderer = new DashRenderer();'

app.layout = html.Div('Simple Dash App')

if __name__ == '__main__':
    app.run(debug=True)

When you provide your own DashRenderer, you can also pass in a hooks object that holds request_pre and request_post functions. These request hooks will be fired
before and after Dash makes a request to its backend. Here’s an example:

This example has not been ported to Julia yet - showing the Python version instead.

Visit the old docs site for Julia at: https://community.plotly.com/c/dash/julia/20

from dash import Dash, html

external_stylesheets = ['https://codepen.io/chriddyp/pen/bWLwgP.css']

app = Dash(__name__, external_stylesheets=external_stylesheets)

app.renderer = '''
var renderer = new DashRenderer({
    request_pre: (payload) => {
        // print out payload parameter
        console.log(payload);
    },
    request_post: (payload, response) => {
        // print out payload and response parameter
        console.log(payload);
        console.log(response);
    }
})
'''

app.layout = html.Div('Simple Dash App')

if __name__ == '__main__':
    app.run(debug=True)

Notice the request_pre function takes the payload of the request being sent as its argument, and the request_post function takes both the payload and the response of the server
as arguments. These can be altered in our function, allowing you to modify the response and request objects that Dash sends to the server. In the example above, the request_pre
function is fired before each server call, and in the case of this example, it will console.log() the request parameter. The request_post function will fire after each server
call, and in our example will also print out the response parameter.


Customizing Meta Tags

Not sure what meta tags are?
Check out this tutorial on meta tags and why you might want to use them.

Dash adds some meta tags to your app by default:

Overriding Default Tags

To override or add custom meta tags to your app,
you can specify meta tags directly in the Dash constructor:

Adding Additional Tags

You can also add additional tags. Here’s an example of adding description
and robots meta tags.

This example has not been ported to Julia yet - showing the Python version instead.

Visit the old docs site for Julia at: https://community.plotly.com/c/dash/julia/20

from dash import Dash, html

app = Dash(meta_tags=[
    # A description of the app, used by e.g.
    # search engines when displaying search results.
    {
        'name': 'description',
        'content': 'My description'
    },
    # To request that crawlers not index a page
    {
        'name': 'robots',
        'content': 'noindex'
    }
])

app.layout = html.Div('Simple Dash App')

if __name__ == '__main__':
    app.run(debug=True)

If you inspect the source of your app, you’ll see the meta tags.
In this example, there are the two custom tags we added, along with
the three default meta tags.

Dash App with Custom Meta Tags


Serving Dash’s Component Libraries Locally or from a CDN

Changed in Dash 1.0.0 - now serve_locally defaults to True,
previously it defaulted to False

Dash’s component libraries, like dash_core_components and dash_html_components,
are bundled with JavaScript and CSS files. Dash automatically checks with
component libraries are being used in your app and will automatically
serve these files in order to render the app.

By default, Dash serves the JavaScript and CSS resources from the
local files on the server where Dash is running. This is the more flexible
and robust option: in some cases, such as firewalled or airgapped
environments, it is the only option. It also avoids some hard-to-debug
problems like packages that have not been published to NPM or CDN downtime,
and the unlikely but possible scenario of the CDN being hacked. And of
course, component developers will want the local version while changing the
code, so when dev bundles are requested (such as with debug=True) we
always serve locally.

However, for performance-critical apps served beyond an intranet, online
CDNs can often deliver these files much faster than loading the resources
from the file system, and will reduce the load on the Dash server.

from dash import Dash

app = Dash(__name__, serve_locally=False)

This will load the bundles from the https://unpkg.com/ CDN, which is a community-maintained project that serves JavaScript bundles from NPM. We don’t maintain it, so we cannot guarantee or attest to its uptime, performance, security, or long term availability.

Also note that we don’t publish the dev bundles to unpkg, so when running the app locally with python app.py, the local JavaScript files will be served. When the app is deployed with gunicorn, it’ll switch to the CDN.


Sample Dash CSS Stylesheet

Currently, Dash does not include styles by default.

To get started with Dash styles, we recommend starting with this CSS stylesheet hosted
on Codepen.

To include this stylesheet in your app, copy and paste it into a file
in your assets folder. You can view the raw CSS source here:
https://codepen.io/chriddyp/pen/bWLwgP.css.

Here is an embedded version of this stylesheet.

<iframe><iframe>


Syntax Highlighting With Markdown

Both dash-table and
dash-core-components
support Markdown formatting, which you can use to specify
syntax highlighting for inline code.

Highlighting is handled by
highlight.js. By default, only
certain languages are recognized, and there is only one color
scheme available. However, you can override this by downloading a
custom highlight.js package. To do this, visit
https://highlightjs.org/download/,
and in the Custom package section, select all the
languages that you require, download your package, and place the
resulting highlight.min.js file into the assets folder. The package should also come with a styles/ directory;
to use a different color scheme, copy the corresponding
stylesheet into your app’s assets folder.