Large file transfer for the DCC

Quickstart

Initialize the database

Initialize the database based on the last migrations:

python manage.py db upgrade

Load the database with a new server and server-token

python manage.py authorize_server "my-server-id" "My Server Name" "my-server-token"

Start up the development server

Start up the development server in one console:

python manage.py runserver --port=8000

Generate a transfer code

Run the following curl command in the console using a valid server token (configurable in config.py) to obtain a transfer code valid for 24 hours

curl -H "X-Server-Token: my-server-token" -H "Content-Type: application/json" -X POST -d '{"user":"your-user-id","name":"your-user-name","email":"your-user-email","duration":7}' http://localhost:8000/transfers/

Upload a file:

For example, create a random 10MB file:

cat /dev/random | head -c 10000000 > test.bin

Open a browser to localhost:8000 , then log in using your transfer code. Add a sample and drop your file to upload the file in chunks of 1MB:

Setting up the environment

Create and active a new virtual environment:

$ virtualenv -p python2.7 .virtualenv
$ source .virtualenv/bin/active

Install the dependencies:

$ pip install -r requirements.txt

Performing database migrations

After modifying the schema, create a new migration script:

$ python manage.py db migrate

Upgrade to the new version:

$ python manage.py db upgrade

Downgrade to the previous version:

$ python manage.py db downgrade

*Note: for OperationalErrors with SQLite when dropping or adding columns, please modify the migration script to use the following workaround:

with op.batch_alter_table("some_table") as batch_op:
    # Add a column
    batch_op.add_column(Column('foo', Integer))
    # Drop a column
    batch_op.drop_column('bar')