Re: Is user space cleared with QSPPUTSP?

[CRPence <CRPbottle@xxxxxxxxx>]

On 20 Feb 2013 12:19, Glenn Gundermann wrote:
> I have three AFPDS spooled files that I'm merging into one using
> QSPGETSP and QSPPUTSP for each spooled file.
>
> is it possible to add data to a user space using QSPGETSP and then
> use QSPPUTSP just once?

  Best I can tell, the Get Spooled File Data (QSPGETSP) API does not 
accept a start position, so there would be no capability to /append/ the 
results of consecutive requests.  Seems each request to the API will 
have the output start writing from the beginning of the UsrSpc.

> With the way I'm doing it today, I'm wondering if I need to clear the
> user space between spooled files.
> I'm not clearing it now and everything looks great to me but I'm
> wondering if I'm doing it correctly. I don't want it to have
> intermittent problems or to be broken in the future. I don't see
> anything in the Information Centre.

  An API that writes to a User Space will write whatever is relevant to 
the request; i.e. what is requested for output.  Regardless of whether 
the existing space is cleared by the API is moot, because any data not 
overwritten was not returned by the API and thus should not be accessed 
except by an error in the navigation of the returned data, which would 
be a defect in the user program; or the returned data could falsely 
direct the user code to some residuals, which would be a system API 
defect.  However I believe the issue of residual data is not an issue 
for at least the length of data returned for most APIs that write to a 
User Space, because I believe that most in fact will either clear at 
least up to that point, or will write every byte up to that point to 
ensure no residual data would remain in /gaps/ that the output-format 
would not overwrite.  Thus I expect, that is why there is no problem as 
currently coded, and probably why not clearing would never become an issue.

> Lastly, does anyone have code to change the page #s using a user
> space?

  Would that be a question of the spooled data?  If so, that would 
required a scan\replace of the data, which I expect requires knowing the 
layout of the data in a particular spool file for the best effect.?