Get an anonimized database

anonymize_database.sh

# Run as: ./anonymize_database.sh my-database-production (database name)

set -e

# check if pganonymize is installed
if ! command -v pganonymize &> /dev/null
then
    echo "pganonymize could not be found"
    exit 1
fi

# check if the pganonymize.yml file exists
if [ ! -f pganonymize.yml ]; then
    echo "pganonymize.yml does not exist"
    exit 1
fi

# Get the database name from the first command line argument
DB_NAME=$1

if [ -z "$DB_NAME" ]; then
    echo "No database name provided"
    exit 1
fi

# dump the original database
pg_dump -Fc $DB_NAME > /tmp/$DB_NAME.dump

# create a new database for the anonymized data
createdb $DB_NAME-anonimized

# restore the original dump into the anonymized database
pg_restore -d $DB_NAME-anonimized /tmp/$DB_NAME.dump

# delete the dump file
rm /tmp/$DB_NAME.dump

# anonymize the anonymized database
pganonymize --schema=pganonymize.yml --dbname=$DB_NAME-anonimized

# dump the anonymized database
pg_dump -Fc $DB_NAME-anonimized > /tmp/$DB_NAME-anonimized.dump

# delete the anonymized database
dropdb $DB_NAME-anonimized

# /tmp/$DB_NAME-anonimized.dump must be deleted after the file was downloaded

pganonymize.yml

tables:
 - user_user:
    primary_key: id
    chunk_size: 5000
    fields:
     - first_name:
        provider:
          name: fake.first_name
     - last_name:
        provider:
          name: set
          value: "Bar"
     - email:
        provider:
          name: md5
        append: "@localhost.com"

truncate:
  - django_session

view.py

import io, os

def plug_cleaning_into_stream(stream, filename):
    try:
        closer = getattr(stream, 'close')
        def new_closer():
            closer()
            # removes the file when finishes
            os.remove(filename)

        setattr(stream, 'close', new_closer)
    except:
        raise

# path_to_file is /tmp/$DB_NAME-anonimized.dump
def send_file(request, path_to_file):
    # Call the anonymize_database.sh inside python and use the DB_NAME from the settings as the argument
    with io.open(path_to_file, 'rb') as ready_file:
        plug_cleaning_into_stream(ready_file, path_to_file)
        response = HttpResponse(ready_file.read(), content_type='application/force-download')
        return response