From: Otto Stolz (Otto.Stolz@uni-konstanz.de)
Date: Mon May 12 2003 - 05:28:12 EDT
John Cowan wrote:
> I know that the point of codepage 1047 is that it swaps NEL and LF so
> that mainframe Unix files see the same line-end character as mainframe
> MVS and VM files.
No. Rather:
> But what does the rest of the page look like? Is it
> exactly CP 037, or is it a bit different?
Codepage 1047 has the brackets and braces (and some other
syntactical marks, as listed by Doug Ewell) on those code
points as expected by IBM compilers and other IBM software.
Prior to the announcement of CECP 1047, no IBM-defined
I/O interface code agreed with the de-facto software code,
used by IBM. It was really a horrible mess.
Under <http://www.rz.uni-konstanz.de/Antivirus/tests/EBCDIC/A523/>,
you'll find scans of the presentation I gave, in 1991,
on this topic (and related topics). On page 11, I have asked
IBM to formally recognize their de-facto software code, and
on pages 23 and 24, I have given a table of this code
(the core of what is now CECP 1047).
I've also put a CECP 1047 code table under
<http://www.rz.uni-konstanz.de/Antivirus/tests/EBCDIC/CECP1047-1.png>.
Best wishes,
Otto Stolz
This archive was generated by hypermail 2.1.5 : Mon May 12 2003 - 06:24:10 EDT
