I4GL Reports to files

Path: emory!mathcs.emory.edu!walt From: walt@mathcs.emory.edu (Walt Hultgren) Newsgroups: comp.databases.informix Message-ID: <8400@emory.mathcs.emory.edu> Date: 17 Jan 92 20:52:42 GMT Sender: walt@mathcs.emory.edu Reply-To: walt@rmy.emory.edu Organization: Emory University I'd like to know if an old problem (bug?) with sending I4GL reports to output files has been fixed. Then: Some time ago (~2 yrs) I wrote a 4GL report program that had to generate several different reports for each account number in a table. For each account, the summary report had to *precede* the detail report. The scheme I used was to send the summary and detail reports to a disk file. When each report had finished for the current account, I would cat the summary file, then the detail file. After cat'ing the files, I would rm them. This process was done inside a foreach loop that selected the accounts from the table. This worked fine except that the program bombed out when the number of accounts reported on one run was high. My guess at the time was that somehow a new file descriptor was used each time through the "foreach account" loop even though the previous report had been "finished." Eventually, the program would hit the limit for open descriptors for one process. Under that assumption, I developed a work-around using a shell script wrapper and a couple of extra programs. However, this process runs more slowly than a single program execution, and there's more code to maintain. I don't recall whether I called Informix Tech Support to see if this problem was a known bug. I may not have. Deadlines were tight, and I wasn't finding ISI Support too responsive in those days. Now: I was recently making some changes to the program, so I thought I'd test the full program run without the wrapper. It now appears to work correctly, but I don't now if the problem has been fixed or if I just haven't triggered it yet in my testing. So... My questions are: Does this problem sound familiar to anyone? Was this ever a known bug? If so, is it definitely fixed? All of this has been with the SE engine under SunOS. Our current versions are I4GL .03F, and SunOS 4.1. Any evidence, rumors and/or war stories would be greatly appreciated. Thanks, Walt. -- Walt Hultgren Internet: walt@rmy.emory.edu (IP 128.140.8.1) Emory University UUCP: {...,gatech,rutgers,uunet}!emory!rmy!walt 954 Gatewood Road, NE BITNET: walt@EMORY Atlanta, GA 30329 USA Voice: +1 404 727 0648