NOTE: Keep in mind that this project is a work in progress. Things may change drastically as the project stabilises.
Aggregates the sentiment analysis score of news articles from various news sources and exposes these scores in an API.
- Python (pip)
- AWS CLI and AWS SAM CLI.
- Node 10+ (npm)
- Docker
# Installing AWS SAM CLI and AWS CLI
pip install awscli aws-sam-cli
# Running a local version of DynamoDB (which is the main datastore of this service)
docker run -p 8000:8000 -d amazon/dynamodb-local
# Dependency install
npm install
# Copy .env.sample to .env
cp .env.sample .env
The .env file contains all you need to configure this application, so feel free to edit that. Locally, the app loads the .env file from the root project directory, but not when doing a single build (as is the case in ./.ci/deploy.sh).
Probably the most important thing that you can edit is where the app can access your local DynamoDB instance:
...
NODESCRAPE_LOCAL_ENDPOINT=http://192.168.99.100:8000 # or http://localhost:8000 if you're lucky enough to have a UNIX-based machine
...
npm start
GET /news-scores - returns a list of news sources along with their average sentiment score.
GET /calculate - (this should be a PUT, but made it a GET for easy debugging for now) triggers the news sentiment calculation and stores the result in DynamoDB, which is exposed through GET /news-scores .
See https://github.com/verzac/news-neutrality-scraper/blob/master/.travis.yml. All commits to master are pushed to production immediately by Travis CI. Please make sure you've linted/tested/built the project before deploying.
Contributions are currently locked for now while I get the project up and running. However, feel free to shoot me a message or open an issue should you want to contribute! Feedbacks are always appreciated :)