Kanabi Project

Team members:

• Andrew Haley
• Lawrence Gunnell
• Connor Kazmierczak
• Ha Ly
• Sean Mitchell
• Kanra Su
• Alicja Wolak

Startup

There is a launch script that will handle bringing up the containers and performing cleanup on exit. To build and launch the project, type bash launch.bash run.

If you prefer to run the commands manually

sudo docker image rm flask-server:v1          Delete existing (cached) Docker image

sudo docker build -t flask-server:v1 .       Build flask server image (in project folder):

sudo docker-compose up                 Compose and run server

sudo docker-compose up -d               Run in background

ctrl-c                        Stop running server

sudo docker-compose down                Remove built server and PostgreSQL DB containers

If run in background, stop containers with:

sudo docker ps
sudo docker stop <server-container> <db-container>

launch.bash Usage

In project folder:

bash launch.bash clean                                     remove all data and project-specific Docker images
bash launch.bash run                                       build and run the program
bash launch.bash stop                                      stop the program
bash launch.bash rebuild                                   remove all data and rebuild the program
bash launch.bash rebuild-db                                remove only DB data and re-run the program
bash launch.bash test                                      test the program (for a freshly built container)
bash launch.bash backup [path]                             backup current DB to an external file
                                                           if path not provided, save in current directory as <current_date>.sql
bash launch.bash restore <filename>                        restore DB from an external file
bash launch.bash backup-schedule <month|week|day> [path]   schedule regular backups at specified time intervals

Troubleshooting

If you get an error when running sudo docker-compose up indicating that port 5432 (Postgres) is already in use, you need to stop postgresql and try again: sudo service postgresql stop

If you attempt a request and receive a reply regarding unverified SSL certificates, repeat the command with -k provided as an additional argument.

Interacting with the server

You can open a browser and go to https://localhost:443 to connect to the running API. From here, you can hit any of the endpoints specified in the server.py file.

You can also query the API from the command line, using curl:    

/list

List the contents of the intake table:

curl -k https://localhost:443/list?table=intake -b COOKIE_FILE -c COOKIE_FILE

List entries from the intake table, but only the submission date and MRL fields:

curl -k "https://localhost:443/list?table=intake&column=submission_date+mrl" -b COOKIE_FILE -c COOKIE_FILE

List entries from the database where results are filtered based on a JSON-structured query file:

curl -k -X POST https://localhost:443/list -d @resources/test-query-and-1.json -H "Content-Type: application/json" -b COOKIE_FILE -c COOKIE_FILE

 

/load

Post the sample.xlsx file to the /load endpoint:

curl -k -X POST --form "file=@resources/sample.xlsx" https://localhost:443/load?table=intake -b COOKIE_FILE -c COOKIE_FILE

Add a single row, as contained in the sample-row-1.json file:

curl -k -X PUT https://localhost:443/load?table=intake -d @resources/sample-row-1.json -H "Content-Type: application/json" -b COOKIE_FILE -c COOKIE_FILE

 

/update

Update intake table on row 1, for 2 columns, in content-type of JSON:

curl -d '{"row":"1", "receipt_num":200, "phone": 555555}' -H "Content-Type: application/json" -k -X POST https://localhost:443/update -b COOKIE_FILE -c COOKIE_FILE

Update intake table on row 1, for 1 columns, in content-type of x-www-form-urlencoded:

curl -d "row=1&cash_amount=100" -X POST -k https://localhost:443/update -b COOKIE_FILE -c COOKIE_FILE

 

/delete

Delete the row with ID '2' from the intake table:

curl -k "https://localhost:443/delete?table=intake&row=2" -b COOKIE_FILE -c COOKIE_FILE

Delete multiple rows from the intake table:

curl -k "https://localhost:443/delete?table=intake&row=1+2" -b COOKIE_FILE -c COOKIE_FILE

 

/export

Export the intake table as a CSV file named 'intake.csv':

curl https://localhost/export?table=intake -o intake.csv -b COOKIE_FILE -c COOKIE_FILE

 

/restore

Restore a row 1 in the archive table back to the table it was delete from:

curl -k -X PUT "https://localhost:443/restore?row=1" -b COOKIE_FILE -c COOKIE_FILE

 

 

Authentication

/signup

Sign up as a new user:

curl -k -X OPTIONS -H "Origin: GUI_DOMAIN" -H "Content-Type: application/json" --request POST --data '{"email":"EMAIL_ADDRESS","password":"PASSWORD","name":"True"}' https://SERVER:PORT/signup

 

/login

Log in as an existing user:

curl -k -X OPTIONS -H "Origin: GUI_DOMAIN" -H "Content-Type: application/json" --request POST --data '{"email":"EMAIL_ADDRESS","password":"PASSWORD","remember":"True"}' -c cookie.txt https://SERVER:PORT/login

Note that the cookie argument is required for the server to keep track of a user's login status.

In subsequent requests, this named cookie must be provided to the server every time using the -b and -c arguments, like so:

curl -X GET https://localhost/usrhello -k -b a.cookie -c a.cookie

-b uses the named cookie as input; -c saves any updates the server makes to that cookie.

 

/logout

Log out:

curl -k https://SERVER:PORT/logout -c test.cookie -b test.cookie