I've gone through all of those docs quite a bit, but the one point I don't find is how to "interrupt" the flow of the exec and how to either know what the actual statement is or how to map it to the original exec.
I read Chapter 14 many times, but I don't see a way to get between the user and the exec and the language processor. In any case TRACE isn't one of the "Replicable Routines and Exits" that could be "hijacked" to do the trace. And if they were, still, no, one needs to be able to map the original source to what is running and be able to do something with it. Lindy -----Original Message----- From: IBM Mainframe Discussion List [mailto:[EMAIL PROTECTED] On Behalf Of Charles Mills Sent: Saturday, December 30, 2006 8:48 PM To: [email protected] Subject: Re: How to write a full-screen Rexx debugger? Ah! Would it be possible to "replace" TRACE with your own service routine, that would do TSO full-screen presentation of the data, rather than Rexx's line I/O. Interesting question. That would be a useful "product." You can write code that Rexx will call for all I/O including Say. See Chapter 14 of the TSO Rexx reference. I have done this sort of thing; it is not for the faint-hearted. I am *guessing* that Rexx would call it for TRACE I/O. There are special considerations for Say under TSO. As I say, not for the faint-hearted. I would think you could "trap" Rexx debug input and output and display it using your own presentation format. You can also read and write Rexx variables from outside Rexx using IRXEXCOM. It's documented in the same manual. Take a look also at IRXIC. Charles ---------------------------------------------------------------------- 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
