<previous comments in this thread snipped> For any (all) of you who "dislike" the file status "97" - especially anyone involved in a VSE to MVS conversion (where this seems to be a medium-high priority problem), please consider submitting a "Marketing Request" to IBM and reference the existing SHARE requirement:
SSLNGC0413615 Optional (ISO 2002) "0x" file-status code for current "97" (current status of this requirement is "RECOGNIZED") That requirement asks for an enhancement that could be selected by compiler or run-time option to "automatically" turn a "97" into a "0x" file status. (Obviously, run-time would have the advantage of "no recompile required" while "compile-time" would have the advantage of NOT impacting existing programs - without an explicit selection of the feature). * * * To respond to one part of this thread, YES, the Standard (all past and present one) DOES explicitly state that an implementor defined "9x" file status *MUST* be considered "unsuccessful". The way that IBM "gets away" with what they do on a 97, is that the standard does NOT define what must happen on an "unsuccessful" I/O - when no declarative or file status checking is done. * * * * As always, doing a "marketing requirement" does NOT "guaranteed" to do anything more than the existing SHARE requirement does. However, it does tell IBM that this is really impacting existing customers and lets IBM judge how "serious" the impact it. ---------------------------------------------------------------------- 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
