Archives of the TeradataForum

Message Posted: Fri, 18 Feb 2011 @ 11:14:10 GMT

	<Prev	Next>		<<First	<Prev	Next>	Last>>

Hi Guys,

We've been considering moving away from Upper Case Table names and, instead, start to using Mixed Case.

So for example, rather than 'PARCELINTRANSIT' (or even 'PARCEL_IN_TRANSIT'), we'd have 'ParcelInTransit'

Now, apart form the slightly unfortunate lower case 'l' looking the same as an upper case 'I', can anyone thing of a good reason for not making this change in standards ?

I've done some simple tests and I can't see any particular problem. Teradata 'sees' them the same as far as I can tell.

For example,

     CREATE SET TABLE mytinytable ,NO FALLBACK ,
          NO BEFORE JOURNAL,
          NO AFTER JOURNAL,
          CHECKSUM = DEFAULT
          (
           dummy BYTEINT CHECK ( dummy =  1  ) NOT NULL DEFAULT 1 ,
           mydatecol1 DATE FORMAT 'yyyy-mm-dd',
           mycharcol1 CHAR(2) CHARACTER SET LATIN NOT CASESPECIFIC)
     PRIMARY INDEX ( dummy );

cannot co-exist with Table MYTINYTABLE, as I get a 3803 ('MYTINYTABLE' already exists) message.

Also, Selecting from the Table returns rows whether I use :

     SEL * FROM MYTINYTABLE;
     SEL * FROM mytinytable;
     SEL * FROM MYTINYTABLE;

On top of that, Inserts work regardless of CaSe.

So, are we making a BIG mistake moving from Upper Case to Mixed case, or does it just come down to personal choice, would you say ?

Regards

David Clough
Database Developer
Database Design Group

	<Prev	Next>		<<First	<Prev	Next>	Last>>