A voter validation tool for petition gathering. This uses a raw txt voter file with a specified schema, as mentioned in update_voters.py. Voters are stored in a database, and fuzzy search can be used to search petition signers and mark them as valid. The voter file itself is stored as MVF_YYYY_MM_DD.tsv.
This tool can allow grassroots campaigns to run low-cost petition validation software on Heroku to validate petition signatures against a voter file, without having to pay for proprietary software. It has features like fuzzy search and a simple search ranking algorithm using trigram similarity and exact text match, with some speed optimizations.
A lot of this was developed for the San Francisco voter file, and is only designed to work with a single jurisdiction's voter file. Typical users for this kind of tool would be campaign staff and volunteers looking to validate signatures to qualify a ballot measure.
DISCLAIMER: It is illegal to use signatures on an initiative petition in California for a purpose other than qualification of the proposed measure for the ballot. Petitions cannot be used to create or add to mailing lists or similar lists for any purpose, including fundraising or requests for support. Any such misuse constitutes a crime. Handle your data carefully, and only keep it for as long as absolutely necessary for validation purposes.
The steps below should get you set up on an Arch Linux computer. If you have issues, contact the author.
If you don't already have git:
sudo apt-get install git
To pull down the repository:
git clone https://github.com/lakshbhasin/VoterValidation.git
Get the Heroku command line tools and login.
yaourt -S heroku-cli
heroku login
To add the production Heroku remote:
git remote add heroku https://git.heroku.com/sf-voter-validation.git
And you should be all set up to push to production (if you have been given push access).
Note: The following instructions assume you have Python 3.8.2+, which is referred to as just "python". If "python" points to "python2" on your computer, please take note of this key difference!
Install postgresql and dev headers from your package manager, and configure it. If you don't already have them, also grab the the python3 dev headers and virtualenv.
On Arch Linux:
pacman -S postgresql python redis flake8
To configure PostgreSQL to work with the local version of this website (see backend/settings.py for the latest authentication info), follow these steps:
sudo -u postgres psql template1
# The following lines are run in psql, where "template1=#" is the prompt
template1=# CREATE USER voter_validation_test WITH PASSWORD 'voter_validation_test_pass';
template1=# CREATE DATABASE voter_validation_test_db;
template1=# GRANT ALL PRIVILEGES ON DATABASE voter_validation_test_db TO voter_validation_test;
template1=# \connect voter_validation_test_db
You are now connected to database "voter_validation_test_db" as user "postgres".
voter_validation_test_db=# CREATE EXTENSION pg_trgm;
voter_validation_test_db=# \q
The last few lines connect to the DB so we can install the Postgres trigram
similarity extension, which supports fuzzy search. You can also run \dx in
psql after installing the extension, to check it is installed.
On Arch Linux, you will have to run chmod og+X /home /home/$USER and
follow the instructions on https://wiki.archlinux.org/index.php/PostgreSQL#Installing_PostgreSQL
(through starting the service with systemctl) before trying the above commands.
Install all of the Python requirements inside a virtualenv, in the root directory of the project:
python3.8 -m pip install --user virtualenv
python3.8 -m virtualenv venv
source venv/bin/activate
pip install -r requirements.txt
Note that, in every new terminal window where you want to run the server, you will first have to source the virtualenv (as above). It might be useful to create an alias that "cd"s into the repo directory and sources the virtualenv, to make development easier.
If any environment variables can't be found, you can source the .env file and export all of its variables:
set -a
source .env
set +a
A helpful alias for development and opening new terminal windows (change ~/Documents/code/VoterValidation to your repo location):
alias voterval='cd ~/Documents/code/VoterValidation && set -a && source .env && set +a && source venv/bin/activate'
You then need to migrate any DB changes as follows:
python manage.py migrate
Next, create a superuser for your local copy:
python manage.py createsuperuser
# Follow the prompts to set up your local account
# Do this for as many users as you need for testing
Create UserProfiles for each of your users as well:
python manage.py create_user <username>
To run locally:
heroku local
The web server will run at localhost:5000.
Run flake8 to lint and correct style issues.
You can use the update_voters script to update the voter file in an atomic
transaction, with confirmation required before finalizing changes.
python manage.py update_voters <mvf_tsv_url> [--dry_run]
Where mvf_tsv_url is a URL holding the Master Voter File as a TSV.
This can happen after a PostgreSQL update on Arch Linux. The error for this looks like:
could not connect to server: Connection refused
Is the server running on host "localhost" (::1) and accepting
TCP/IP connections on port 5432?
To resolve this, follow the update instructions in
https://wiki.archlinux.org/index.php/PostgreSQL#Quick_guide. Once the database
has been upgraded, run sudo systemctl start postgresql.service and sudo systemctl enable postgresql.service. Then rerun heroku local.