Re: The QSYGETPH API

[Simon Coulter <shc@xxxxxxxxxxxxxxxxx>]

On 27/07/2004, at 1:06 AM, Bruce Vining wrote:

> The Char(*) in this case is indicating that the length of the 
> parameter is
> no longer fixed at 10 bytes.  Per the API documentation if you're 
> using a
> special value for the Password parameter then it's treated as a 
> Char(10)
> and you do not use the Length of password and CCSID of password 
> parameters.
> If you're using a password for the Password parameter then you must 
> provide
> the Length of password (from 1 to 512) and CCSID of password 
> parameters.

So why did they make such an incompatible change? Surely the correct 
behaviour would be to presume a maximum length of 10 and use job CCSID 
if parameters 5 and 6 are not specified. That would stop existing code 
from breaking.

Or were they worried about people activating long/case-sensitive 
passwords and then finding out that their profile handle code no longer 
worked? In which case why not do that at 510 when long password support 
was implemented?

Looking at the documentation for 510 and 520 it looks like this API was 
documented incorrectly. The parameters are in the wrong order and no 
mention of parameters 5 and 6 being required with passwords.

Regards,
Simon Coulter.
--------------------------------------------------------------------
   FlyByNight Software         AS/400 Technical Specialists

   http://www.flybynight.com.au/
   Phone: +61 3 9419 0175   Mobile: +61 0411 091 400        /"\
   Fax:   +61 3 9419 0175                                   \ /
                                                             X
                 ASCII Ribbon campaign against HTML E-Mail  / \
--------------------------------------------------------------------