For this graph, we created a new table in the database called user_login_counts where we store the:
day;count, the amount of logins of that day.
To populate the table, inside the LoginController::login function, we created a condition to see if the row for that day already exists, and if it does, increment the count by one. If the entrance for that day doesn't already exist, we create it and set the count to 1.
For this graph, we created two tabes in the database, user_session_durations and average_session_durations. Inside user_session_durations, we store the:
session_id;user_id;first_request_time;last_request_time;session_duration.
Inside average_session_durations, we store the:
day;average_duration;session_count.
The way we store values inside the user_session_durations is by creating a Laravel Middleware which we called UserSessionDuration.php which is called every single time a new link is clicked by a user. If there is no entrance for that session_id, we create a new one, setting the values as:
session_id= session id;user_id= user id;first_request_time= current time;last_request_time= current time;session_duration= 0.
If the session_id entrance already exists, we update the session_duration and last_request_time to it's new respective values.
In this case, there is a flow between the two tables. We start by populating the user_session_durations table with all the user sessions, which we then check for abnormal values (such as too short/long session durations) and add the values to the average_session_durations, removing them from the first table.
For this graph, we use a table called finished_topics where we store the values:
topic_id;user_id.
To create the data for the graph, we search for all the rows with the same topic_id and then add them together.
For this graph, we use a table called topic_visits which stores the values:
topic_id;day;count.
To stores values for this table, inside the function TopicController::show, we create a conditional statement to check for an entrance for a certain day, and it if already exists, we increment the count by 1. If the entrance does not exist, we create it and set the count to 1.
For this graph, we set two attributes inside the questions table which are:
correct_answers;incorrect_answers.
Every time a user get a question right or wrong, we update this value. We then simple use those values to create the coordinates to be used in the graph.
For this graph, we created a new table called question_durations where we store the values:
user_id;question_id;first_request_time;duration.
When a user enters a question, a new entrance in the table is created. When the user correctly answers that questions and moves on to the next page, the duration attribute is calculated and stored, which we will then use to display on the graph.