Automatic document processing and data submission pipeline powered by Google Cloud's DocumentAI for the Billion Oyster Project
The current FormParser version provided by Google Cloud is pretrained-form-parser-v2.0-2022-11-10
git clone https://github.com/BillionOysterProjectCommunity/document-processor.git
cd document-processor
python3.11 -m venv .
pip install -r requirements.txt
pip install -e ./Google Cloud
For local ADC configuration follow https://cloud.google.com/docs/authentication/provide-credentials-adc#local-dev
Local SSL
cd theia/cert
openssl req -x509 -nodes -days 365 -newkey rsa:2048 -keyout key.pem -out cert.pem
Project configuration
# theia/config.toml
location = 'us'
project-id = "gcp-project-name"
processor-id = "gcp-project-id"
# Google Drive
master-datasheet-directory-id = "bop-master-sheet-google-drive-directory-id"
# Web
flask-secret-key = "supersecretkey"
iam-file-name = "<document-processor-service-account>.json"
# OAuth
oauth-client-id = "<google-oauth-client-id>.apps.googleusercontent.com"
oauth-client-secret = "oauth-client-secret"
debug = false/true
This project uses Cloud Native Buildpack to build images. To build an image use the pack CLI tool.
pack build theia --path . --builder gcr.io/buildpacks/builder:v1
# Then, run the buildpack.
# Make sure to open 127.0.0.1 in the browser when developing locally
docker run --rm -p 127.0.0.1:8080:8080 theia