mockexams

A website to practice cbt examination

Urgent

• How do we link questions and their accompaniment with the populate_database.py script?
• Work on login and signup pages
• setup OTP
• setup correct json response for login
  • get from user profile
  • Should user profile contain user records or that can be queried separately when needed.
• We need a way to take care of instructions for particular sections.

Note

• The user auth apis don't support GET requests intentionally becuase we just want information to be posted to it and then it returns the correct response either for login or registration. If we ever need to return a list of all users, we will create a seperate api app for that.

Positons

• Backend Engineer -> Stephen Adesina
• Frontend Engineer -> Abdulla Alaya
• Question gathering team
• Question verification Team

Tech Stack

• Karax
• Django and DRF

Project Structre and Design

• mockexamsapp
• questions_api
• user_auth_api

Features

• User Analytics dashboard
• Admin page(Default django or custom?)
• Quiz page
• No lagging between questions, just show next question staight, no lazy loadings

Database

Getting the database

The database has been created and will soon be populated but it will not be uploaded to github rather a populate_database.py file included in the github will populate the database on your local machine. Make sure not to upload the sqlite databse. It will be later changed to postgresql.

Database schema

The following json Structre will represent the database schema feel free to debate on controversial fields

{
    "question" : {
        "question" : "The question",
        "paragraph" : "for all those comprehension questions",
        "options" : [
            {
                "name" : "option name like option a, b, c etc.",
                "option" : " the answer linked to the option",
                "correct" : "boolean value to know if this is the correct option"
            }
        ]
    },
    "subject" : {
        "name" : "subject name",
        "subjectid" : "specific subject id for each exams",
    },
    "cbt" : {
        "name" : "exam name",
        "description" : "exam's description",
        "image" : "exam's image"
    },
    "exam" : {
        "exam" : cbtfield,
        "examid" : "exam id",
        "subjects" : [
            subjectfield
        ]
    },
    "score" : {
        "name" : "exam name",
        "subject_scores" : [
            {
                "name" : "subject name",
                "score" : "subject score"
            }
        ]
    },
    "record" : {
        "exam" : cbtfield,
        "score" : scorefield,
        "examscore" : "cummulative score of the particular cbt"
    },
    "user" : {
        "id" : "user's id",
        "name" : "The user's full name",
        "email" : "user's email address",
        "image" : "user image url or image base64 encoding",
        "number" : "user's phone number",
        "password" : "user's encoded password",
        "record" : [
            recordfield
            ]
    },
    "superuser" : {
        "user" : userfield,
        "superuserid" : "sudo id"
    }
}

Questions

• question
• option_A
• option_B
• option_C
• option_D
• option_E -> can be null
• correct_ans
• explain
• subject
• year
• acom_type -> can be null
• exam_type
• accompany -> can be null

Accompaniment

• acom_type
• subject
• passage
• diagram

Querying the database

• As discussed, we will have a specific question format

  • 40 english questions
    • 2 passage questions
    • other normal
  • 20 per remaining subjects
    • Some with accompaniments

• We will need to pick questions:

  • with passages
  • with diagrams etc.

• all we need to do is know the number of accompanimet questions we want and then we pick those and we then choose the rest as normal questions. Take the example below.

  • Example 1

    • we need 100 questions -> English = 40, 20 per rem subjects.
    • We need 2 passage questions each with 5 questions. We just go to the accompaniment table and pick two passages and five questions from its questions linked to it in the questions table. And then we pick normal questions from the questions table.
    • We need 2 accompaniment question each having 2 questions each per subject
    • We go to the accompaniment table and pick accompaniments by subject and pick 2 questions from the questions linked to it and pick normal questions from the normal question pool for each subject.

  • If we wanted to pick all the questions linked to an accompaniment, we do that in the same way we have been doing then we calculate how many questions we have left for the subject and pick those from the normal question pool.

Scraping Questions

• The question dirrectory contains past questions to be used for testing the backend.
• The scrape.py script scrapes a particular site for past questions from 2001 to a specified year above 2001.
• The site scrape.py scrapes does not have past questions from about 2013 downwards so i will write another script to scrape past questions from some pdfs with me.
• To use the scrape tool type python3 scrape.py the subjects you want to scrape
• Note you can scrape more than 1 subject at a time.

Structure of the backend

• We are using sqlite for testing and will soon move to postgres.
• The database should store questions such that it is easy to retrieve questions for a particular subject in a particular year
• The database should be structured such that it is easy to pick any amount of random questions from multiple years in a particular subject

Future plans

• Sponsored Ads to make money