On Saturday, 01/03/2009 at 05:47 EST, Florian Bilek <florian.bi...@gmail.com> wrote:
> I am using the German/Austrian codepage on my PersCom (273 or 1141 with the > Euro symbol). That means the EBCDIC codes coming from z/VM are presented on > my Workstation according to the codepage 273. Some of the characters used by > z/VM like some of the editing characters or the '@' or the "!" are displayed > differently. Yes, code page 273/1141 moves a lot of of the special characters around. > The codepage is a problem when using the German Umlaute as "ÄÖÜ". I am > wondering also how I can handle C-Programs as the brackets and braces are > incorrectly displayed with xedit. When I enter those characters while > writing a C-program those characters do not match the codepoints the > C-Compiler is expecting. > How can this be addressed? 1. SET INPUT/SET OUTPUT to convert as Kris describes. I don't recommend this as it only leads to confusion when you try to share the file with others. Further, unless you do it for ALL differences between the two code pages, it leads to an abomination that is neither code page 1047 nor 1141, making reliable FTP upload/download impossible without custom TCP/IP translation tables. [While I offer translation table build services, it is only for translations between recognized code pages, not for one-offs. See http://www.vm.ibm.com/devpages/altmarka/tcpxlate.html.] 2. Assuming we're talking about the IBM C/C++ compiler, use the '#pragma filetag' directive to tell the compiler that your source files are encoded with code page 273 or 1141. 3. Change to use code page 924. This is ISO 8859-15 (EBCDIC) and is compatible with code page 1047. (Code page 923 is the ASCII equivalent.) This means that you perform mass global changes to the contents of your files, converting them from code page 1141 to 924. I switched to code page 924 several years ago and except for learning to stop using LOGICAL NOT in my Rexx execs, I didn't have any more trouble. Code page 924 has most of the national characters from Europe and the Americas, though I think one of the Icelandic characters is missing. Go with #2 for now and consider whether there needs to be a strategy within your organization to begin to use a code page that is more international in flavor. Alan Altmark z/VM Development IBM Endicott
