QuickStart Development
Nancy Palmquist
nlp at vss3.com
Wed Jun 15 11:52:12 PDT 2016
On 6/15/2016 11:33 AM, Kenneth Brody via Filepro-list wrote:
> On 6/15/2016 11:11 AM, Kenneth Brody via Filepro-list wrote:
>> On 6/15/2016 10:54 AM, Nancy Palmquist via Filepro-list wrote:
>> [...]
>>> A footnote to remember is that while every effort is made to make
>>> filepro
>>> and Quickstart filepro run exactly the same on the same tables it is
>>> not
>>> always the case. It is critical that you compile the tables with the
>>> associated automatic table indicated. I have seen differences in
>>> how they
>>> throw error messages and like issues over the years. So it is
>>> important to
>>> test your software in both environments. QS does not have a
>>> debugger level
>>> so you will have to get very friendly with logtext to track your flow.
>> [...]
>>
>> If you have a dummy field which is defined in automatic processing,
>> but not
>> in the auto processing to be used for the output, and you use that dummy
>> field for multi-level break subtotals[*], then you will get different
>> output
>> in rreport than in dreport. I don't recall if there is a problem with a
>> dummy field not defined in automatic, but defined in the auto processing
>> used for the output. (rclerk-vs-dclerk is not affected.)
> [...]
>
> I forgot to (re)mention that Nancy had already said -- that this issue
> is only if you don't compile the prc with the correct automatic
> processing.
>
Ken,
I recall a weird error that it would increment a counter by 5 instead of
by 1 when the automatic table was not right when the process was
tokenized. It was in the MOS and I am not sure what caused it but I
could always tell when that happened that was my mistake. I would expect
it was with normal dummy variables, since most of that programming was
too old for declared variables. I have seen many cases where it did not
seem to matter but with the complexity of that programming there had to
be something connected in a strange way. But that was more strange than
a defined variable difference.
But these little tidbits always made my days more interesting, back in
the day. I suggest making a menu for development that has all your
setups so you can compile your processing quickly, before distribution.
rcabe filename -y auto2 -c test test1 input oldput etc
I kept a menu for just this and listed all the special token setups so I
could just hit it and it would redo all the tables correctly, in case, I
forgot and loaded a table to make changes without the right auto table.
SUGGESTION - would have been nice to have a way to attach the auto table
with the output table without all that fuss. I use declared variables
for most things these days so the automatic table is less important than
it once was, but it still has it's purpose.
Nancy
--
Nancy Palmquist MOS & filePro Training Available
Virtual Software Systems Web Based Training and Consulting
PHONE: (412) 835-9417 Web site: http://www.vss3.com
More information about the Filepro-list
mailing list