Ever since DB2 version 8 DB2 has supported UNICODE character sets as most of us consider as the same as ASCII. And god forbid such a horrible character set on our beloved EBCDIC mainframe. One of the consequences was that the tables of the DB2 catalog all are in UNICODE and now supports tables and columns with very long names. This is just not the full story I realised recently. You can also use all sorts of strange characters in your names including spaces.

Using this new discovery I can name my columns in the SELECT part in a completely new way. Here is a very simple example you can try for yourself:

SELECT current date as "Today is"
FROM SYSIBM.SYSDUMMY1

Using the above example in SPUFI or QMF will provide you with a very nice heading including both upper and lowercase letters and even a space between Today and is. It is just so nice. All the places in DB2 where you can specify a name you can put the name in quotes and use all sorts of characters. The challenge is we are used to DB2 converting our names into uppercase when we do what we are used to do. But this behaviour comes to an end when we start naming objects in quotes. Then DB2 suddenly becomes case sensitive and it becomes extremely important to specify your name in exactly the same case as the name you want to reference.

It might happen that on your installation you will have to use apostrophe (') instead of quote (") to acheive the above described functionality. I have though not yet seen a DB2 installation where they use quotes around character constants.