Even if the standard is written that way it would be a reasonable extension.
But of course there is not only the return address issue, there is also the whole problem of "stack variables." COBOL does not have the easy distinction of static or not that some languages have. Charles -----Original Message----- From: IBM Mainframe Discussion List [mailto:[email protected]] On Behalf Of McKown, John Sent: Monday, April 25, 2011 7:39 AM To: [email protected] Subject: Recursive PERFORM in COBOL Yes, I know that it is explicitly forbidden. And I understand why: The return address for each PERFORM'ed paragraph is assigned a specific area in the DSA and so there is only one place to put a return address for a given PERFORM'ed paragraph. But I wondering why IBM did it this way. I don't have access to the ANSI COBOL standard. I've been told that many of the things that I consider to be "oddities" in COBOL are not due to IBM's desire, but due to the standard forcing a particular way to do things. But a PERFORM push down stack of return addresses just seems "better" to me. ---------------------------------------------------------------------- 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
