dreport and debug

Richard Kreiss rkreiss at gccconsulting.net
Thu Apr 11 07:44:07 PDT 2013


I know there has been a recent discussion of dreport and debug.

I recently had to rewrite an output process which creates a CSV file. The only changes which I had to make was to replace fixed dates used in the processing with variables so that one date entered would be used to calculate all of the other dates which are required.

My problem has nothing to do with these new date variables. However, when I run the output, upon reaching certain records the processing stops. To track the specific record on which the processing stops. I have added some show statements, which gives me the transaction number and the date. In addition, I have show statements which show me which subroutines have run. This was to allow me to find where the problem was more quickly.

I originally hardcoded in debug on when that specific transaction number was reached. My problem is that when I manually cycled through the program, everything ran as programmed and when done the next record for processing came up.

There are 147,000 records to be processed. So I added a variable at @once processing to get the transaction number to turn on the debug process. So far, I have only found one place where I need to correct the processing as I am getting a negative dollar value (credit).

I have manually looked at 14 records with the debugger, when I run the output again, the processing continues through those records which previously caused the program to hang up. This is the same whether I run dreport or rreport. The only differences is rreport runs a lot faster than dreport.

I really don't understand why my output processing would hang up on a particular record and when I manually go through the processing it completes and moves the next record to be process. One thing that I have done is to hold down the enter key with the debugger running and watching the countdown of the records left to be processed, so far only once have I noticed the program looping without the getnext been executed. This was with the negative number, and I know I need to correct that.

There are still 51,000 records which need to be processed and I need to find a common thread as to why the processing stops, but after processing the records with the debugger, on the next run those records process without a problem.

This is more of a rant than a question, unless someone has some suggestions as to what I might look for.

Yes, I'm aware that without the processing table it is difficult to analyze what might be causing the problem. There are some single lookups to other files to get information and there is also a lookup with a getnext to accumulate a total. I know I'll eventually get the whole outfit to run without stopping, but is taking a long time to debug this program.


Richard Kreiss
GCC Consulting

Office: 410-653-2813




More information about the Filepro-list mailing list