On 20 December 2017 at 19:40, Phil Smith III <[email protected]> wrote: > I feel silly here, but darned if I can find a definitive list of the EBCDIC > code pages that iconv supports on z/OS. Googled my brains out, and built the > list at the bottom of this note. But I don't know if it's complete.
> Anyone know where this might be documented? I'd start here: https://www-01.ibm.com/software/globalization/cdra/index.html CDRA is, of course, not z/OS specific, but there is little reason to think that z/OS iconv (aka Unicode conversion services) doesn't have available to it all the code pages known to CDRA. It may be the case that not all code pages are installed at a given site, but conversion services is able to do two-step conversions between non-Unicode code pages through Unicode if necessary. And missing code pages, or for that matter custom ones, can be installed as needed. There is also the z/OS Unicode Services User's Guide and Reference (in KC, of course - start here: https://www.ibm.com/support/knowledgecenter/SSLTBW_2.3.0/com.ibm.zos.v2r3.cunu100/toc.htm ), which has two lists in an appendix: https://www.ibm.com/support/knowledgecenter/SSLTBW_2.3.0/com.ibm.zos.v2r3.cunu100/iea3un_Direct_conversions_supported_between_non-Unicode_CCSIDs_.htm and https://www.ibm.com/support/knowledgecenter/SSLTBW_2.3.0/com.ibm.zos.v2r3.cunu100/diruni.htm More information than you need, perhaps, as there are non-EBCDIC, non-Unicode code pages there too. It is also possible to query the system about what code pages are installed. Or just try it and if what you need isn't installed you will get a return/reason code telling you so fairly precisely. I'm not sure how well these return codes are translated to iconv() errno/errno2, but there is a non-iconv C interface too. Tony H. ---------------------------------------------------------------------- For IBM-MAIN subscribe / signoff / archive access instructions, send email to [email protected] with the message: INFO IBM-MAIN
