SQL corner

One of my favorite technologies are relational databases and the SQL language. I like how elleganant a solution of certain programming tasks can be using SQL - mainly graph computation. E.g., computation of neural networks in common procedural programming languages like C# and Java involves three nested loops; in SQL, it's only one loop, and the other program logic is expressed using SELECTs with JOINs. More explanation comes later.

My favorite database system is MySQL 5.0. I intentionally mention the version, because it became my favourite when it allowed to write Stored Procedures.

Through the time I created some handy general-purpose procedures and functions, which ease the developement of other procedures for MySQL.

Various SQL Implementations

Here I publish some results of my efforts in SQL:

• MySQL Stored Procedures

• SQL Implementation of Neural Networks - Perceptron and Backpropagation

• Transport schedule search SQL implementation

MySQL Wishlist - feature requests

Besides that, from time to time, I also write down what do I miss in MySQL. That's quite a long list, but the most important features are:

• Passing of a result set between stored procedures

• Variables should be able to keep a result set

• Java objects in Stored Procedures (sci-fi)

• `ALTER TABLE ... RENAME [**OVERWRITE**] TO`

• Stored Procedure call stack in INFORMATION_SCHEMA

• sprintf() equivalent in SQL (or rather take Java's String::format() as an inspiration)