Private file hosting for you and your friends.

I run a private instance of sr.ht at sr.ht. You can request an invite if you know me personally. Otherwise, here are the setup instructions to run it on your own infrastructure:

Quick overview:

1. Install dependencies
2. Set up dependencies
3. Clone the repository
4. Configure the site
5. Compile static assets
6. Set up SQL
7. Deployment

Install the dependencies

You'll need these things (Arch packages in parenthesis, some from AUR):

• Python 3 (python)
• PostgreSQL (postgresql)
• scss (ruby-sass)
• Flask (python-flask)
• SQLAlchemy (python-sqlalchemy)
• Flask-Login (python-flask-login)
• psycopg2 (python-psycopg2)
• bcrypt (python-bcrypt)

Use the packages your OS provides, or build them from source.

Set up services

I'll leave you to set up PostgreSQL however you please. Prepare a connection string that looks like this when you're done:

postgresql://username:password@hostname:port/database

The connection string I use on localhost is this:

postgresql://postgres@localhost/sr.ht

We need to be able to create/alter/insert/update/delete in the database you give it.

Clone the repository

Find a place you want the code to live.

$ git clone git://github.com/SirCmpwn/sr.ht.git
$ cd sr.ht

Configure the site

$ cp alembic.ini.example alembic.ini
$ cp config.ini.example config.ini

Edit config.ini and alembic.ini to your liking.

Compile static assets

$ make

Run this again whenever you pull the code.

Deployment

What you do from here depends on your site-specific configuration. If you just want to run the site for development, you can source the virtualenv and run

python app.py

To run it in production, you probably want to use gunicorn behind an nginx proxy. There's a sample nginx config in the configs/ directory here, but you'll probably want to tweak it to suit your needs. Here's how you can run gunicorn, put this in your init scripts:

/path/to/sr.ht/bin/gunicorn app:app -b 127.0.0.1:8000

The -b parameter specifies an endpoint to use. You probably want to bind this to localhost and proxy through from nginx. I'd also suggest blocking the port you choose from external access. It's not that gunicorn is bad, it's just that nginx is better.

When running in a production enviornment, run python app.py at least once and then read the SQL stuff below before you let it go for good.

nginx configuration is available in nginx/, modify it to suit your environment. nginx is required to run sr.ht properly.

You can become an admin like so:

$ cd /path/to/sr.ht/
$ source bin/activate
$ python
>>> from srht.database import db
>>> from srht.objects import User
>>> from datetime import datetime
>>> u = User.query.filter(User.username == "your username").first()
>>> u.approved = True # approve yourself
>>> u.approvalDate = datetime.now()
>>> u.admin = True # make yourself an admin
>>> db.commit()

We use alembic for schema migrations between versions. The first time you run the application, the schema will be created. However, you need to tell alembic about it. Run the application at least once, then:

$ cd /path/to/sr.ht/
$ source bin/activate
$ python
>>> from alembic.config import Config
>>> from alembic import command
>>> alembic_cfg = Config("alembic.ini")
>>> command.stamp(alembic_cfg, "head")
>>> exit()

Congrats, you've got a schema in place. Run alembic upgrade head after pulling the code to update your schema to the latest version. Do this before you restart the site.

You can customize the appearance of the site with template overrides. Create a directory called overrides and copy templates from the templates directory into overrides. Modify them as you see fit, they will be used instead of the version from templates.