Selecting data

We now know how to retrieve data from the database, but that isn't much good unless we can get it back again. Let's say we now want to build an account settings page for MyStatus; we've got the user's username stored in a persistent session, but we will retrieve the other profile fields from the database to display in the settings form:

    SELECT * FROM "users"
    WHERE "username" = 'alice';

This query tells Cassandra we want to retrieve the rows where the value for username (the primary key) is alice. The * wildcard simply says we would like all the columns in that row, saving us from having to type them all out. You'll see the rows we requested nicely formatted in the CQL shell as follows:

In other scenarios, we don't need all the columns. When a user tries to log in to MyStatus, we want to retrieve their password and compare it to the one the user provided us with, but we don't care about the email. Avoiding unnecessary columns reduces the amount of data that needs to be transferred between Cassandra and our application, thus making the queries faster. Instead of using the (*) wildcard, we can instead type a list of columns we are interested in:

    SELECT "username", "encrypted_password" FROM "users"
    WHERE "username" = 'alice';

You'll see that in the results, the email column no longer appears since it wasn't in the list of columns that we specified.