student-dashboard-django

student dashboard based on django framework

Environment configuration

There is some environment configuration and addtional files needed for this app to run. You can put this in a file called .env for testing.

(Sample .env)

# Any logs that django outputs
DJANGO_LOG_LEVEL=DEBUG
# Any other logs
ROOT_LOG_LEVEL=INFO

# MySQL Configuration
MYSQL_DATABASE=student_dashboard
MYSQL_USER=student_dashboard_user
MYSQL_PASSWORD=student_dashboard_pw
MYSQL_HOST=student_dashboard_mysql
MYSQL_PORT=3306
MYSQL_ROOT_PASSWORD=student_dashboard_root_pw

# UDW configuration
UDW_ENDPOINT=
UDW_USER=
UDW_PASSWORD=
UDW_PORT=
UDW_DATABASE=

# Canvas Configuration
CANVAS_COURSE_IDS=
CANVAS_USER=

# Paths to special login credentials (uses SAML)
GOOGLE_APPLICATION_CREDENTIALS=/secrets/bq_cred.json

# SAML Configuration
# Need to set this to false to turn it off for dev, but default is true
STUDENT_DASHBOARD_SAML=false

# Where your secrets are stored
SAML2_FILES_BASE=/secrets/saml/

# Base for Django to your accounts path
DJANGO_SAML2_URL_BASE=http://localhost:5001/accounts/

# ACS Redirect
DJANGO_ACS_DEFAULT_REDIRECT=http://localhost:5001/

# Login Redirect
LOGIN_REDIRECT_URL=http://localhost:5001/

The bq_cred.json is service account for Big Query, it neesd to be supplied and put into the secrets directory and setup in the environment.

The dashboard/saml directory needs to contain 4 files for SAML configuration. These are currently hard-coded in settings.py though the path comes from the environment SAML2_FILES_BASE.

remote-metadata.xml 
student-dashboard-saml.key
student-dashboard-saml.pem

Docker commands for deploying the app

1. Tear down running application and db instances: docker-compose down

2. Build the application: docker-compose build

3. Run the application in a detached mode: docker-compose up -d

4. Place all of your secrets in a directory and copy it into the docker with docker cp ~/secrets student_dashboard:/secrets

5. Initialize the MySQL database with mysql/init.sql: http://localhost:5001/load_data

Load user, file, file access data into database

Use the following URL patterns to load data into databases:

** For testing now just use /testloader, this has all the URL's ready to load **

1. update_with_udw_user: load user enrollment information from Unizin Data Warehouse
2. update_with_udw_file: load file info from Unizin Data Warehouse
3. update_with_udw_access: load file access information from Canvas Live Event records hosted on Unizin Data Platform
4. update_assignment: loading assignment info in a course
5. update_groups: groups that assignment belongs has weight/point information associated with assignment group
6. submission: assignment submission information of students in the course
7. weight_consideration: load information about weights (as boolean) considered in grading of an assignment.

Query the MySQL database within the container:

docker exec -t -i student_dashboard_mysql /bin/bash

1. login as mysql -u <user> -p
2. use student_dashboard

Clean outdated docker images

The docker images will take up more disk spaces as time goes on, you can delete those outdated docker images by using the following command:

docker rmi $(docker images -q -f "dangling=true")