On Fri, 4 Jun 2021 18:03:04 +0000, Farley, Peter x23353 wrote:

>Variable and U format will always be issues.  
>
No.  Why?

> ... Converting them on the PC to unlabeled AWSTAPE has similar issues.
>
No.  Why?  Labels is just bytes.  The format is documented and irrelevant.

We once had an offline, in-house  "Tape Replication System", hardware
and software supplied by an overseas vendor.  At some point the decision
was made to move the system, physically and logically, to an out-of-state
contractor and supply electronic images rather than physical tapes.

I was tasked with replicating the internal data format.  (hot AWSTAPE;
should have been.  I was not part of the specification process and would
nor have bee  aware of AWSTAPE in thee day.)

I reverse-engineered the vendor's data format from their source code
and wrote a Rexx program to generate the vendor's format from our
master tapes, mounted overriding to RECFM=U,LABEL=BLP.

Worked readily.  My code didn't need to understand the formats of
labels, BDWs, or RDWs.  Bytes is bytes.  One wrinkle was that Rexx
in the day didn't handle RECFM=U -- I needed to add a REPRO step
to convert U to VB.

>BUT I believe that IFF the original tape has standard header and trailer 
>files, then AWSTAPE is realistic to use.  Then the AWSTAPE file can be binary 
>transferred to the MF and processed there with the mainframe AWSTAPE utility 
>(or is it a HET utility? I don’t remember now).
>
>The key step is to capture the binary data with no translation from degrading 
>9-track.  Figuring out how to successfully use it can come after that step.
>
>PC utilities like HXD (HexEdit) can view binary EBCDIC files with ease so you 
>know what you are dealing with after you capture the data.

-- gil

----------------------------------------------------------------------
For IBM-MAIN subscribe / signoff / archive access instructions,
send email to [email protected] with the message: INFO IBM-MAIN

Reply via email to