Radically simplified static file serving for Python web apps
With a couple of lines of config WhiteNoise allows your web app to serve its own static files, making it a self-contained unit that can be deployed anywhere without relying on nginx, Amazon S3 or any other external service. (Especially useful on Heroku, OpenShift and other PaaS providers.)
It’s designed to work nicely with a CDN for high-traffic sites so you don’t have to sacrifice performance to benefit from simplicity.
WhiteNoise works with any WSGI-compatible app but has some special auto-configuration features for Django.
WhiteNoise takes care of best-practices for you, for instance:
- Serving compressed content (gzip and Brotli formats, handling Accept-Encoding and Vary headers correctly)
- Setting far-future cache headers on content which won’t change
Worried that serving static files with Python is horribly inefficient? Still think you should be using Amazon S3? Have a look at the Infrequently Asked Questions below.
QuickStart for Django apps¶
settings.py file and add WhiteNoise to the
list, above all other middleware apart from Django’s SecurityMiddleware:
MIDDLEWARE = [ # 'django.middleware.security.SecurityMiddleware', 'whitenoise.middleware.WhiteNoiseMiddleware', # ... ]
That’s it, you’re ready to go.
Want forever-cacheable files and compression support? Just add this to your
STATICFILES_STORAGE = 'whitenoise.storage.CompressedManifestStaticFilesStorage'
For more details, including on setting up CloudFront and other CDNs see the Using WhiteNoise with Django guide.
QuickStart for other WSGI apps¶
To enable WhiteNoise you need to wrap your existing WSGI application in a WhiteNoise instance and tell it where to find your static files. For example:
from whitenoise import WhiteNoise from my_project import MyWSGIApp application = MyWSGIApp() application = WhiteNoise(application, root='/path/to/static/files') application.add_files('/path/to/more/static/files', prefix='more-files/')
And that’s it, you’re ready to go. For more details see the full documentation.
Using WhiteNoise with Flask¶
WhiteNoise was not specifically written with Flask in mind, but as Flask uses the standard WSGI protocol it is easy to integrate with WhiteNoise (see the Using WhiteNoise with Flask guide).
WhiteNoise works with any WSGI-compatible application and is tested on Python 2.7, 3.3 – 3.6 and PyPy, on both Linux and Windows.
Django WhiteNoiseMiddlware is tested with Django versions 1.8 — 1.11
WhiteNoise owes its initial popularity to the nice things that some of Django and pip’s core developers said about it:
It’s now being used by thousands of projects, including some high-profile sites such as mozilla.org.
Issues & Contributing¶
Infrequently Asked Questions¶
Isn’t serving static files from Python horribly inefficient?¶
The short answer to this is that if you care about performance and efficiency then you should be using WhiteNoise behind a CDN like CloudFront. If you’re doing that then, because of the caching headers WhiteNoise sends, the vast majority of static requests will be served directly by the CDN without touching your application, so it really doesn’t make much difference how efficient WhiteNoise is.
That said, WhiteNoise is pretty efficient. Because it only has to serve a fixed set of
files it does all the work of finding files and determining the correct headers
upfront on initialization. Requests can then be served with little more than a
dictionary lookup to find the appropriate response. Also, when used with
gunicorn (and most other WSGI servers) the actual business of pushing the file
down the network interface is handled by the kernel’s very efficient
sendfile syscall, not by Python.
Shouldn’t I be pushing my static files to S3 using something like Django-Storages?¶
No, you shouldn’t. The main problem with this approach is that Amazon S3 cannot
currently selectively serve compressed content to your users. Compression
(using either the venerable gzip or the more modern brotli algorithms) can make
in order to do this correctly the server needs to examine the
Accept-Encoding header of the request to determine which compression
formats are supported, and return an appropriate
Vary header so that
intermediate caches know to do the same. This is exactly what WhiteNoise does,
but Amazon S3 currently provides no means of doing this.
The second problem with a push-based approach to handling static files is that it adds complexity and fragility to your deployment process: extra libraries specific to your storage backend, extra configuration and authentication keys, and extra tasks that must be run at specific points in the deployment in order for everything to work. With the CDN-as-caching-proxy approach that WhiteNoise takes there are just two bits of configuration: your application needs the URL of the CDN, and the CDN needs the URL of your application. Everything else is just standard HTTP semantics. This makes your deployments simpler, your life easier, and you happier.
What’s the point in WhiteNoise when I can do the same thing in a few lines of Apache/nginx config?¶
There are two answers here. One is that WhiteNoise is designed to work in
situations were Apache, nginx and the like aren’t easily available. But more
importantly, it’s easy to underestimate what’s involved in serving static files
correctly. Does your few lines of nginx config distinguish between files which
might change and files which will never change and set the cache headers
appropriately? Did you add the right CORS headers so that your fonts load
correctly when served via a CDN? Did you turn on the special nginx setting
which allows it to send gzipped content in response to an
which for some reason CloudFront still uses? Did you install the extension which
allows you to serve pre-compressed brotli-encoded content to modern browsers?
None of this is rocket science, but it’s fiddly and annoying and WhiteNoise takes care of all it for you.