On Tue, Apr 28, 2009 at 4:20 PM, John McKown <[email protected]> wrote: > > If you have Co:Z installed, you can do this very simply: > > //CONVERT EXEC DTLSPAWN > //STDOUT DD SYSOUT=* > //STDERR DD SYSOUT=* > //STDIN DD * > fromdsn "//DD:INPUT" | tr '\t' '\n' | todsn "//DD:OUTPUT" > /* > //INPUT DD DSN=crappy.file.from.unix,DISP=SHR > //OUTPUT DD DSN=nice.file.output,DISP=(NEW,CATLG), > //... other required DD parameters >
You would need to use the "-b" switch on fromdsn so that record boundaries in the input dataset were ignored. Also, you might want to FTP in a preceding step and receive the file with RECFM=U into a temporary dataset that is passed into the CONVERT step. Kirk Wolf Dovetailed Technologies http://dovetail.com PS> COZBATCH is the new name for DTLSPAWN http://dovetail.com/products/cozbatch.html http://dovetail.com/docs/coz/dsp-ref.html PPS> If you use the Co:Z Launcher rather than FTP, you can launch a script on the Unix box (over SSH) and do everything much simpler: // EXEC PROC=COZPROC, // ARGS='[email protected]' //DOWNLOAD DD DSN=MY.DATASET,DISP=(NEW,CATLG), // DCB=(...),SPACE=(...) //STDIN DD * cp /path/to/unix/file | todsn -l 0x05 //DD:DOWNLOAD // ---------------------------------------------------------------------- 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
