Do SQL statements actually modify the data within the table or overall database?

Hi Evan,

I would like to expand a bit.

SQL generally involves the manipulation of data in a database through commands known as CRUD (Create, Read, Update, Delete). In SQL, these are the INSERT, SELECT, UPDATE, and DELETE commands, respectively.

Most database management systems are by default in auto-commit mode, so the changes are committed to disk when you issue a command that manipulates data in some way (creating, updating, or deleting). If you are simply reading data, then obviously no data will be changed.

In order to prevent changes from committing to disk as they are run, you need to declare BEGIN or BEGIN TRANSACTION at the beginning of your file (or wherever you would like to declare it), and declare COMMIT in the file when you are ready to write the changes to disk. BEGIN turns off auto-commit mode, so that you can make multiple changes in a file before writing them to disk. Basically, you can think of everything between a BEGIN and COMMIT statement as running as one big command all at the same time.

This is especially useful, for example, if you have a huge script that is initializing a database with tons of changes -- using BEGIN works as a safety net in case something disastrous happens in the middle of the script running (like an error or crash). If an error happen in your file before reaching the COMMIT line, then none of the changes are written to disk, so you don't have to deal with a half-finished database monstrosity the next time you come back to it.

The ROLLBACK command can also be used to undo changes that have occurred since the last BEGIN statement was issued.

Happy coding.
Leslie