Skip to content

Relational database project for the Run Monash carnival system, featuring schema design, data population, advanced SQL queries, transaction handling, and JSON/MongoDB integration.

Notifications You must be signed in to change notification settings

Nashmia101/run-monash-database

Repository files navigation

run-monash-database

Relational database project for the Run Monash carnival system, featuring schema design, data population, advanced SQL queries, transaction handling, and JSON/MongoDB integration.

Overview

The Run Monash carnival system manages competitors, events, teams, and charities across multiple carnivals.
This project demonstrates my ability to:

  • Design and implement missing schema tables with constraints.
  • Populate the database with realistic test data.
  • Apply DML operations with proper transaction handling.
  • Modify the live database structure while preserving integrity.
  • Write complex SQL queries for reporting.
  • Integrate relational data with JSON and MongoDB.

Features

  • Schema Design (DDL): Implemented Competitor, Entry, and Team tables with keys, constraints, and comments.
  • Data Population: Inserted realistic test data for 15+ competitors, 30+ entries, and 5+ teams, ensuring valid scenarios.
  • DML & Transactions: Created sequences, handled inserts/updates/deletes with transaction safety.
  • Database Modifications: Added new attributes and a charity allocation table with percentage constraints.
  • Advanced SQL Queries: Developed reporting queries (e.g., most popular team names, event records, carnival statistics).
  • JSON & MongoDB: Generated JSON documents from Oracle and inserted them into MongoDB for querying.

Project Deliverables

This repository includes:

  • T1-rm-schema.sql → Schema creation (DDL).
  • T2-rm-insert.sql → Data population with sample competitors, entries, and teams.
  • T3-rm-dm.sql → Data manipulation and transactions.
  • T4-rm-mods.sql → Database modifications (new attributes/tables).
  • T5-rm-select.sql → SQL queries for reports.
  • T6-rm-json.sql → JSON data generation in Oracle SQL.
  • T6-rm-mongo.mongodb.js → MongoDB integration for team documents.
  • FIT2094_Assignment2_Brief.pdf → Project brief.

Results

  • Built and tested a complete Oracle database with enforced constraints.
  • Successfully populated with realistic test data, covering edge cases.
  • Transaction-safe updates and modifications ensured consistency.
  • JSON and MongoDB outputs demonstrated integration with non-relational models.

Author

Nashmia Shakeel

About

Relational database project for the Run Monash carnival system, featuring schema design, data population, advanced SQL queries, transaction handling, and JSON/MongoDB integration.

Resources

Stars

Watchers

Forks

Releases

No releases published

Packages

No packages published