FreeTDS Development Group

[Michael Peppler <mpeppler AT bluewin.ch>]

On Wed, 2003-07-30 at 16:58, Lowden, James K wrote:
> > From: James Vanns [mailto:jimv AT canterbury.ac.uk]
> > Sent: July 30, 2003 9:13 AM
> > 
> > I had a friend create a table on his M$ SQL server 2000 with just one
> > row and as many columns as data types. He filled each field with a
> > different value corresponding to data type. E.g.
> > 
> > bigint, bit, char, datetime, float, money, nvarchar, unique
> > 1111119,1,c,30/04/2003 13:00:19,23.45, 123.45, hello, 
> > {F44A-etc-...blah}
> > 
> > Note the Unique identifier in the above example is not the 
> > real value I
> > just couldn't be bothered to write it out in full!
> > 
> > Anyway using the code snippet below works great except when trying to
> > convert and print the unique ID. What should I be doing to allow all
> > data types to be printed successfully as with tsql and sqsh?
> 
> Welcome, James.  
> 
> sqsh doesn't understand the uniqueidentifier datatype.  tsql does.  It's not
> a blob; it's a structure.  I don't know if it can be converted to an image
> or varbinary type.

It is interesting that MS-SQL defines a new datatype for this. For
Sybase it's just a string with or without the '-':


[23] mplap.master.1> select newid(), newid(1)
[23] mplap.master.2> go
                                                                              
  
 ------------------------------------
------------------------------------
 fff916c37f284b629fe7a5b2bcc45fb0    
e7bdc109-6de2-40fb-9d61-e883d82d5202

(newid() is part of the new features of ASE 12.5.0.3).

Michael