Final project for the Data Engineer @ DataScientest, jan2023 Bootcamp training
Authors S. Delahaies, A. Boutier, I. Noui, H. Elmi Ali
Travel planning: design, implement, deploy a travel planning web application
A brief overview of the project can be found in the post Travel Planning Web App.
# download the repo and enter the folder
git clone https://github.com/sdelahaies/travelplanner-full.git
cd travelplanner-full
docker compose up -d
# change ownership for the datalake folder if not already owned
sudo chown $USER:$USER datalake
# download the index_processed.json
cd datalake
wget --load-cookies /tmp/cookies.txt "https://docs.google.com/uc?export=download&confirm=$(wget --quiet --save-cookies /tmp/cookies.txt --keep-session-cookies --no-check-certificate 'https://docs.google.com/uc?export=download&id=1wPf55Bl9ObDQlxeN4kJDKqiYVZFRQrFT' -O- | sed -rn 's/.*confirm=([0-9A-Za-z_]+).*/\1\n/p')&id=1wPf55Bl9ObDQlxeN4kJDKqiYVZFRQrFT" -O index_processed.json && rm -rf /tmp/cookies.txt
The next step consists in initializing the databases. Go to airflow UI at http://localhost:8080 with the following credentials:
- login: airflow
- password: airflow
and run the dags importer_datatourism_to_db and importer_user_to_db to import the data into mongoDB.
Go to mongo-express at http://localhost:8111 and check that the itinerary DB exists and that it contains the two collections point_of_interest and user.
You can then access the web app at http://localhost:8181/login with the credentials:
- login: travel
- password: planner
and test the planner and auto-planner modes.
Note: the search bar is using the geolocation API adresse.data.gouv.fr it is therefore limited to France, the same goes for DATAtourisme data... Feel free to plug any worldwide geolocation API together with opentripmap to enable global travel planning.
