On Tue, 25 Oct 2011 09:37:27 -0500, Shmuel Metz (Seymour J.) wrote: > on 10/24/2011 at 05:56 PM, Paul Gilmartin said: > >>Ummm. Not quite. z/OS Unix System Services uses EBCDIC NL (x'15') >>as the new line indication. > >For EBCDIC files, but I believe that it still uses LF for other >character sets. > I thought my use of "EBCDIC" would be sufficient qualification. I wish there were better support for those "other character sets".
>>Then it violates the specifications of the code pages by >>translating LF(x'0A')<->NL(x'15') > >IBM was caught between a rock and a hard place. None of the available >choices was good. > Were it my choice, I'd define a new code page with LF at 0x15 and NEL at 0x25 to validate the OEMVS311 translation table. I know, there's some quasi-theological specious argument that 0x00 through 0x3F are inviolable. In my estimation, this pales before calling OEMVS311 a mapping between ISO8859-1 and IBM-1047. I wouldn't much care if my 3215 didn't work any more. If I had a 3215. And if I needed it to work, I'd simply translate this putative new character set to IBM-1047 for the 3215, even as EBCDIC is translated to ASCII for my terminal session. -- gil ---------------------------------------------------------------------- For IBM-MAIN subscribe / signoff / archive access instructions, send email to [email protected] with the message: GET IBM-MAIN INFO Search the archives at http://bama.ua.edu/archives/ibm-main.html
