----------------------------------------<snip>-------------------------------
I would hope that I would never do that, either. However, for code that
uses an interface that has been unchanged for 45 years with a 100-byte
limit, I'm not sure I'd be quite that hard on someone whose code copied
the data into a 100-byte buffer using the supplied length.
To turn it on its head: I wouldn't play you-bet-your-system with this.
The code in most vendor products isn't as well vetted as your own
internal code, and is a lot harder to examine.
Guns have safeties for a reason; so should a change like this. Whether
it's a new interface (PARMX) or an LE setting that must be explicitly
enabled, there needs to be some informed consent. I can't imagine IBM
being willing to even consider it otherwise -- else a (poorly written)
program that's been happily running for decades could crater production,
and nobody wants that.
------------------------------------<unsnip>------------------------------------
Granted that guns have safeties (most of them, anyway). But if you have
to worry about whether it's on or off, you're already doing something
very wrong.
There's no excuse for knowing what you're doing and PLANNING AHEAD. I've
always coded my programs with the assumption that the parm field might
be as long as 255 bytes, the max that can be described in a single byte.
Rick
--
If you're not the lead dog, the view never changes.
----------------------------------------------------------------------
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
