On Sun, 16 May 2021 10:26:25 -0500, Paul Gilmartin wrote:
>    ...
>https://www.lbdsoftware.com/XMITIP-Guide.pdf
>    Note that all data on the mainframe is stored in the EBCDIC character set
>    and is translated to the ASCII character set during the transmission.
>    Any data that should not be translated should be attached in Binary format.
>
(More):
o Note that since EBCDIC is a 256-character set and ASCII is 128,
  there is inevitable loss of information.
o It doesn't specify *which* EBCDIC character set (there are many, not "the".)

>Possibly:  iconv -f IBM-420 -t UTF-8
>Then transmit as FORMAT BINary.
>
>You might yet be frustrated because IBM mainframes (I believe)
>store Arabic, etc. text backwards.


On Sun, 16 May 2021 09:06:10 -0500, Joel C. Ewing wrote:

>Terminology Clarification:   UTF-8 is of course just one representation
>of the Unicode codeset, not the codeset itself.  UTF-8 has the advantage
>that characters in standard ASCII charset have identical byte
>representation in UTF-8, which makes possible some use of UTF-8 even in
>email headers.
> 
Use of encoded non-ASCII characters in headers requires considerable
gyrations.  See section 2 of RFC 2047.  The IBM-MAIN LISTSERV is not
RFC 2047 savvy.

-- gil

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

Reply via email to