> Sometimes allowing the programmer to control the severity of selected messages by specific message code.
Good idea, hadn't thought of that. Thanks. Maybe we can turn eights into fours as well as turning fours into zeroes. > if this capability is provided in PARM (one plausible implementation) it readily exceeds the 100-character JCL ANYTHING implemented in PARM quickly hits the 100-character limit! We've beat that one to death here <g>. No, I've got a parameter file and it would be an additional command or parameter there. PARM= gets passed to the invoked compilers as-is. Also, I've thought a little more about what Ray said. True, the software can't ignore certain conditions, but this is not about ignoring any conditions. Nothing in the "functional" behavior of the software will change. A fatal error will still be a fatal error, even if you choose to specify ignoring the message. But potentially - if that's what the customer specifies - a fatal error could cause termination with a zero return code. And thank you Chris. Good thoughts. I appreciate your input. I THINK at this point (inspired by the C/C++ PARM= option) I'm going to opt for SUPPRESS as the keyword. I think I'm going to print the message anyway but "suppress" the return code escalation - the nature of the product is such that we just don't under any circumstances end up with an overly chatty list file, so suppressing the message is unnecessary and potentially counter-productive - what the customer wants is just that precious zero return code - he doesn't even look at the list unless there is a problem. The option to increase/decrease generally the severity of messages is a cool idea but is probably overkill at this time. Thanks again all. Charles -----Original Message----- From: IBM Mainframe Discussion List [mailto:[EMAIL PROTECTED] On Behalf Of Paul Gilmartin Sent: Friday, March 10, 2006 7:59 AM To: [email protected] Subject: Re: Any products that let you ignore certain messages? In a recent note, Ray Mullins said: > Date: Fri, 10 Mar 2006 07:08:05 -0800 > > Many C/C++ compilers have a #pragma option (varies from compiler to > compiler, of course) that can suppress (and then unsuppress) messages. I > Sometimes allowing the programmer to control the severity of selected messages by specific message code. > > -----Original Message----- > > [mailto:[log in to unmask]] On Behalf Of Charles Mills > > Sent: Thursday March 09 2006 15:45 > > > > I guess the HL assembler ACONTROL FLAG(NOxxxx) is the closest > > example I can think of. > > > > I'm thinking of a syntax something like IGNORE > > MSG(XYZ0123E,XYZ0456W,XYZ0789W) > > You might even wish, according to the customers' coding standards ,to be able to increase the severity of messages, treating a "nnnI" or a "nnnW" as if it were a "nnnE". Alas, if this capability is provided in PARM (one plausible implementaion) it readily exceeds the 100-character JCL limit. ---------------------------------------------------------------------- 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
