On 2020-11-12 21:16, Paul Gilmartin wrote:
On Thu, 12 Nov 2020 21:48:06 +0000, Robert Prins wrote:
For those who don't know, XYplorer <https://www.xyplorer.com/> is a replacement
...
Both can be used on all of your computers.
One of my computers is a Raspberry Pi.
Almost all of your computers.
... auto-convert EBCDIC files. I asked for it (it's kind
of useful to have a quick look at XMIT files containing source, README, or
"$INDEX" like members) and he has hard-coded some defaults:
The conversion is hard-coded to IBM EBCDIC International (codepage 500)
and a record length of 80.
XMIT files of RECVM=VB source contain control information sufficient to
reconstruct line boundaries. Lacking that, it's not only a "quick look" but
also dirty. A PDS member TRANSMITted without the SEQ option is worse:
it's an IEBCOPY PDSU inside a .XMIT envelope.
The option only shows the top of the file, it's never going to be anything line
XMIT Manager or XmitApp.
I suggested the LRECL=80, he himself went for CP 500. I've already suggested
that he "externalizes" the CP, ...
It should be externalized.
We'll see what he decides
For what it's worth, I suggested that the EBCDIC auto-detect is based on long
strings of the ASCII "@" character, the EBCDIC space.
RECFM=V files may lack such long strings of x'40'. Would better be a
preponderance
of octets above x'80'? But there should be an option to override.
If you go for over 0x80, you end up in UTF-8 territory, which is also
auto-detected.
Robert
--
Robert AH Prins
robert(a)prino(d)org
The hitchhiking grandfather - https://prino.neocities.org/indez.html
Some REXX code for use on z/OS - https://prino.neocities.org/zOS/zOS-Tools.html
----------------------------------------------------------------------
For IBM-MAIN subscribe / signoff / archive access instructions,
send email to [email protected] with the message: INFO IBM-MAIN
