On Thu, Aug 24, 2017 at 10:10 AM, Kirk Wolf <[email protected]> wrote:
> Or just use Co:Z SFTP and send the data set directly rather than having to
> copy it to a Unix file.
Then you can use the trim=on/off option to control whether you want
> trailing spaces trimmed.
>
The only reason that I didn't mention Co:Z sftp was, IIRC, it requires
modifying the SSH daemon in addition to just installing the UNIX files &
PDSE libraries. This means that a sysprog must be involved and perhaps even
some change control person to approve a change. Using "fromdsn" is easier
since anyone can install it in they own PDSEs and somewhere in the ${HOME}
directory. Most people can do this without "bothering" (aka "being caught
by") others.
>
> Kirk Wolf
> Dovetailed Technologies
> http://dovetail.com
>
> On Thu, Aug 24, 2017 at 7:46 AM, John McKown <[email protected]
> >
> wrote:
>
> > On Thu, Aug 24, 2017 at 3:14 AM, venkat kulkarni <
> > [email protected]>
> > wrote:
> > > my dataset on Mainframe is FB , So whenever mainframe transfer data
> using
> > > FTP using ascii to AIX it doesnt have trailing spaces.
> > >
> > > But as SFTP cant transfer directly dataset into AIX location, so we
> > > transfer dataset to omvs location using OPUT and then from omvs, we
> > > transfer data to AIX using SFTP.
> > >
> > > So, I think, while transferring data from dataset to omvs it add
> trailing
> > > blanks and which carry forward to unix as well.
> >
> > Technically, the blanks are already there in an FB data set, so OPUT is
> > _preserving_ existing trailing blanks. FTP, with the option
> > NOTRAILINGBLANKS, will remove these existing trailing blanks, whereas
> sftp
> > will not remove them. There are a couple of things you can to. If you are
> > forced to only use standard IBM z/OS UNIX facilities, then you can use
> the
> > "cp" and "sed" commands to transfer the data in the DSN to a UNIX file.
> For
> > example:
> >
> > If your OPUT looks something like:
> >
> > OPUT 'SOME.DATA.SET.NAME' '/u/myid/transfer/unix.file.name.txt' TEXT
> >
> > You could replace it by using the OSHELL command. I am not using BPXBATCH
> > instead of IKJEFT01 to minimize the number of changes for the OP.
> >
> > OSHELL cp "//'SOME.DATA.SET.NAME'
> > "
> > /dev/fd/1 | +
> > sed
> > -E 's/ *$//;' >/u/myid/transfer/unix.file.name.txt
> >
> >
> >
> > Personally, I would install Dovetailed Technologies' "data set pipes"
> from
> > http://dovetail.com/downloads/coz/index.html. This software can be used
> > without any kind of fee or license. You need not even register it. It
> comes
> > with a UNIX command called "fromdsn".
> > Still using OSHELL instead of, and assuming that the proper directories
> > are on the UNIX ${PATH}, you could do:
> >
> > OSHELL fromdsn -K SOME.DATA.SET.NAME >/u/myid/transfer/unix.file.
> name.txt
> >
> >
> > Note that "fromdsn" does not normally prefix the DSN with your TSO id.
> So
> > the DSN is not enclosed in ' characters.
> >
> >
> > > Mainframe to AIX rather then first trasnferring to omvs and then to
> aix.
> >
> >
> > --
> > If you look around the poker table & don't see an obvious sucker, it's
> you.
> >
> > Maranatha! <><
> > John McKown
> >
> > ----------------------------------------------------------------------
> > For IBM-MAIN subscribe / signoff / archive access instructions,
> > send email to [email protected] with the message: INFO IBM-MAIN
> >
>
> ----------------------------------------------------------------------
> For IBM-MAIN subscribe / signoff / archive access instructions,
> send email to [email protected] with the message: INFO IBM-MAIN
>
--
If you look around the poker table & don't see an obvious sucker, it's you.
Maranatha! <><
John McKown
----------------------------------------------------------------------
For IBM-MAIN subscribe / signoff / archive access instructions,
send email to [email protected] with the message: INFO IBM-MAIN
