> -----Original Message----- > From: IBM Mainframe Discussion List > [mailto:[EMAIL PROTECTED] On Behalf Of Farley, Peter x23353 > Sent: Thursday, September 22, 2005 9:08 AM > To: [email protected] > Subject: Re: HOD/HACL running on z/OS? > > > Thanks, I will look at that. However, I'm not really looking > for something > tied to a webserver, if it can be avoided. All I need is the > ability to > programmatically act as a 3270 terminal and record the > outputs that I get > from the applications that I run. And no, I don't really > want to write VTAM > LU code either. BTDT, much more complicated than my minimalist needs. > > Peter
If I understand correctly, what you basically want is the ability to capture 3270 screens for later comparison. You would like to do this capture on the z/OS system if at all possible. And the solution needs to be zero, or at least very low, cost. (Do I ever understand that last one!). From other posts, I would guess that you'd also like to capture all the 3270 inputs. The eventual goal is to be able to capture 3270 inputs and outputs so that the inputs could be replayed later and the new outputs compared against the old outputs. (regression testing of code changes?). The following is just "off the top of my head" (like my hair <grin>). You mentioned that you don't want to write a VTAM "capture" type routine due to its complexity. How about a TCP/IP "capture" type routine? I think you'd write something like a "transparent proxy". What your routine would do is LISTEN on a particular IP port. The end-user would connect to this port. Upon connection, your routine would attempt to connect to the normal TN3270E port. Your routine would simple "route" (or "copy") the IP packets between the TN3270E server on the mainframe and the TN3270E client on the other end. You shouldn't need to actually inspect or change them. As it was routing the packets, it could record them to a file on the mainframe (with an indicator as to which direction the packet was flowing). When you wanted to "replay" a session, you would write another program to connect to the TN3270E port on the mainframe, read the previously recorded file, and send that information to the TN3270E server. You could then compare the response from the application to the response recorded in the file. The main problem that I can see is that there are many possible TN3270E sequences which would result in the same 3270 screen display. So I guess the "comparison" would require actually creating a 3270 "buffer" from the file contents and the response, then compare the buffers. This instead of directly comparing the TN3270E data streams, byte for byte. There is still a matter of error recovery or propagation if either socket is abnormally terminated. But I think this is easier to program than the LU error recovery / propagation. I suggest this simply because, to me, this TCPIP program would be easier to code than the equivalent VTAM program. ---------------------------------------------------------------------- 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
