A website for serving media super fast, by SirCmpwn and jdiez, and several other contributors.
What is this? It's a website you can upload images, audio, and video to, and receive a link to share it with your friends. This readme documents contributor guidelines and installation instructions. For information on the official MediaCrush instance, see https://mediacru.sh/about
Support us on Gittip? https://www.gittip.com/mediacrush/
If you aren't looking to contribute, but just want to do some cool stuff with the site, you might be interested in our developer documentation, which documents our API and a few other nice things.
See CONTRIBUTING.md. To get started, join our our IRC channel (#mediacrush on irc.freenode.net) to listen in on dev chatter. We can help you sort out your ideas and we'll work with you directly to fine tune your pull requests.
Here's a quick overview of installation:
- Install Python 2, virtualenv, redis, ffmpeg, tidy, jhead, node.js, and optipng.
- Clone the MediaCrush git repository.
- Activate the virtualenv.
- Install pip requirements.
- Install coffeescript.
- Configure MediaCrush.
- Start the services and you're done!
Your milage may vary, be prepared to deal with unforeseen complications.
Here it is again, in more detail.
Install the requirements
Our servers and our dev machines both run Arch Linux, which makes getting updated packages a little easier. We need to
install a few things: sudo pacman -S redis imagemagick python2 python-virtualenv nodejs. You also
need to install ffmpeg-full from the AUR. Feel free to modify the PKGBUILD a little bit to suit your enviornment -
you probably don't need x11grab, for example. If you aren't on Arch Linux, you should be able to use your distribution
packages, with the exception of ffmpeg, which you must compile yourself. Make sure you enable libtheora, libvorbis,
libx264, libfdk_aac, and libvpx when you configure it.
Optional dependencies:
- jpegtran for JPG support - via extra/libjpeg-turbo
- optipng for PNG support
- tidyhtml for SVG support
- xcftools for XCF support
- otfinfo for subtitle support - via extra/texlive-bin
Clone the repository
git clone http://github.com/MediaCrush/MediaCrush && cd MediaCrush
Create a virtual environment
Note: you'll need to use Python 2. If Python 3 is your default python interpreter (python --version), add
--python=python2 to the virtualenv command.
virtualenv . --no-site-packages
Activate the virtualenv
source bin/activate
Install pip requirements
pip install -r requirements.txt
Install coffeescript
npm install -g coffee-script
Configure MediaCrush
cp config.ini.sample config.ini
Review config.ini and change any details you like. The default place to store uploaded files is ./storage,
which you'll need to create (mkdir storage) and set the storage_folder variable in the config to an absolute path to this folder.
Start the services
You'll want to make sure Redis is running at this point. It's probably best to set it up to run when you boot
up the server (systemctl enable redis.service on Arch).
MediaCrush requires the daemon and the website to be running concurently to work correctly. The website is
app.py, and the daemon is celery. The daemon is responsible for handling media processing. Run the
daemon, then the website:
celery worker -A mediacrush
python app.py
This runs the site in debug mode. If you want to run this on a production server, you'll probably want to run it with gunicorn, and probably behind an nginx proxy like we do.
gunicorn -w 4 app:app
To run the unit tests, simply execute python tests.py.
Note: do not execute the test script on a live instance - it clears the storage and database.
Updating a MediaCrush instance isn't pretty. We don't have a great mechanism in place for handling breaking changes. However, we will be posting to the mediacrush@librelist.com mailing list whenever we push noteworthy changes. Send an email to that address to subscribe to the list. Anyone who runs a third-party MediaCrush instance should be on that list. Feel free to send any questions related to maintaining your instance as well, but be sure to browse the archives first.