On Wed, 23 May 2012 06:43:53 -0700, Charles Mills wrote:
>I don't understand what I am seeing from Unicode Services translation.
>
>I specify translation from 1047 (Encoding scheme 1100 - EBCDIC, SBCS; Name
>LATIN 1 / OPEN SYSTEM) to 1252 (Encoding scheme 4105 - ASCII, SBCS; Name
>MS-WIN LATIN-1).
>
>As both CCSIDs are SBCS I would expect that any "common" EBCDIC character
>would get translated into a single ASCII byte. But for an input byte of
>X'B0' (logical not in 1047) I am seeing translation to the 2-byte sequence
>C2AC. AC is by my reading correct: it's 1252 logical not. But what the heck
>is that C2 about (C2 is A with an acute accent in 1252).
>
>FWIW technique E, substitution 1A.
>
That appears to be a variable-length encoding, such as UTF-8. For example:
387 $ awk 'BEGIN { printf( "%c", 16*11 ) }' |
iconv -f IBM-1047 -t UTF-8 | od -x
0000000 acc2
(John M. was lately ranting on another forum about the richness of the
POSIX shell command structure compared to TSO. I can only agree.
What would be required to accomplish the same from the TSO "READY"
prompt?)
-- gil
----------------------------------------------------------------------
For IBM-MAIN subscribe / signoff / archive access instructions,
send email to [email protected] with the message: INFO IBM-MAIN
