Material
SQL Data Manipulation Language
Objective: The research question is how to transform raw data to adapt to a database relational schema and load data to populate it.
Theory background:
- The notion of consistency and a consistent database
- Extraction, loading and transformation (ETL) processes [PDF]
- Database (consistent) state
ToDo
- Using the SQL data definition language, create a script to insert some example tuples in the database. Remember to respect a specific order according to how you defined foreign keys. Otherwise, you will have errors. Query your schema if necessary, as seen in Activity 1.
- Propose an algorithm in pseudo-code for transforming raw data into your schema. Input is raw data, and output is an SQL script with the insertion commands.
- Implement and test your algorithm.
- Run the resulting script on the DBMS for populating your database schema.
Deliverables:
- SQL script for inserting tuples in the relational database implemented in Activity 1 06/12/2023
- Program transforming raw data into instances (tuples) of your relational schema -13/12/2023
- Test on the created database 13/12/2023