Note: You can't run DFSORT on zLinux. Cheers, Martin
Martin Packer, zChampion, Principal Systems Investigator, Worldwide Banking Center of Excellence, IBM +44-7802-245-584 email: [email protected] Twitter / Facebook IDs: MartinPacker Blog: https://www.ibm.com/developerworks/mydeveloperworks/blogs/MartinPacker From: John McKown <[email protected]> To: [email protected] Date: 08/06/2014 16:00 Subject: Re: Any need? DFSORT interface to C language Sent by: IBM Mainframe Discussion List <[email protected]> Well, Based on the feedback, this looks like a decent project for me to _try_ to implement. My plans are to first look at the JZOS DfSort classes to see if I can create similar C++ classes. The real effort will be in writing the C code, which will likely be used internally by the C++ classes to minimize the amount of duplicate code. Why implement in both C and C++? Now for the US $60,000 question. My initial coding is being done on my Linux system. Mainly because I like the facilities that I have on it better even that z/OS ISPF. Now the "problem" that I could again use some feedback on. Since I am developing this on Linux, I plan to upload it to a z/OS UNIX subdirectory. If/When it is distributed on the CBT, I plan for it to be like some of my other work: The CBTtape XMIT file will be of a single PDS. This PDS will contain a member which is the PAX unload of the UNIX subdirectory containing the code. I will also make an OBJECT data set PDSE for static linking, as well as a LOADLIB data set PDSE. But I do not intend to make a source PDS. The source will remain in the z/OS UNIX subdirectory. In addition, my C development tends to use make files. So the make file to recompile the source will also be only in the UNIX subdirectory. Now, other than not liking z/OS UNIX, is there any compelling reason that I should create a source PDSE and an JCL PDS for compiling the source in the PDS? Since this will increase my work, and result in duplicate information, it's going to be a really good reason. "I don't like UNIX" won't cut it, sorry <grin/>. I actually, in many ways, prefer doing development in UNIX with make files to the legacy JCL method. As a compromise, if I don't create the PDSes mention, I _might_ create a UNIX shell script & JCL to run it which will create those PDSEs and populate them properly. And put that JCL in the PDS which is initially restored. I'm only putting things into a PDS because that is how the CBTTape accepts contributions. Well, we'll all see if I'm up to this or not. And how long ti takes me, if I succeed. -- There is nothing more pleasant than traveling and meeting new people! Genghis Khan Maranatha! <>< John McKown ---------------------------------------------------------------------- For IBM-MAIN subscribe / signoff / archive access instructions, send email to [email protected] with the message: INFO IBM-MAIN Unless stated otherwise above: IBM United Kingdom Limited - Registered in England and Wales with number 741598. Registered office: PO Box 41, North Harbour, Portsmouth, Hampshire PO6 3AU ---------------------------------------------------------------------- For IBM-MAIN subscribe / signoff / archive access instructions, send email to [email protected] with the message: INFO IBM-MAIN
