On 17 Mar 2008 07:15:21 -0700, [EMAIL PROTECTED] (Reda, John) wrote: >There is a difference between a blank and an empty file. The merge >should have worked correctly if the file was empty as opposed to having >a blank line. Let's assume your input had a blank line. There are >multiple ways to get rid of this.
I know that. But there are two jobs that create this file, one that runs a single pass, and another that runs a dual pass and a merge. The design was to pass a single blank line instead of an empty file. I added a new program that processed this output, and it was supposed to abort with an empty input file (actually a blank line) it found a blank line. A few months later the dual pass job created a blank-record output from one pass, but not from the other. I was exploring to see which option has the least impact - changing this merge, or changing my program - I'm leaning towards the latter at the moment. ---------------------------------------------------------------------- For IBM-MAIN subscribe / signoff / archive access instructions, send email to [EMAIL PROTECTED] with the message: GET IBM-MAIN INFO Search the archives at http://bama.ua.edu/archives/ibm-main.html
