API to scrap and store images from Wikipedia
# Create a Google Cloud Project
https://cloud.google.com/appengine/docs/standard/python3/building-app/creating-gcp-project
# Connect SDK To Google to access DataBase
$ gcloud auth application-default login
# Create an isolated Python environment in a directory external to your project and activate it:
$ python3 -m venv env
$ source env/bin/activate
# Install dependencies
$ pip install -r requirements.txt
# Run Server (http://localhost:8080)
python main.py
# Create a Google Cloud Project
https://cloud.google.com/appengine/docs/standard/python3/building-app/creating-gcp-project
# Connect SDK To Google
$ gcloud auth application-default login
# Deploy App
$ gcloud app deploy
# Go To Url
$ gcloud app browse
- GET /scrap/
- GET /geturls/