Some years ago I developed a filter to convert back and forth from EBCDIC to ASCII (for our mainframe email gateway) that used a trip from the source codepage to Unicode and then to the target codepage, so buried in there I have what you're looking for. However it is highly dependent on the format of the translation tables (which I obtained from the IBM Globalization Center of Competency), so is not likely directly usable by anyone else without a lot of setup work. It supports both single and double byte translations. The double byte support only handles Japanese, Chinese (simplified and traditional) and Korean at the moment but since single-byte tables are pretty much all the same format, it can support as many of those translations as you're willing to retrieve tables for. I currently have 47 single-byte EBCDIC, 29 single-byte ASCII, 4 double-byte EBCDIC and 8 double-byte ASCII tables available to the gateway code. Since going to Unicode and then to the target every time for simple single-byte translations isn't the greatest idea from a performance perspective, the gateway code uses the filter to derive single-byte tables on-the-fly the first time the translation is needed and then caches it so that the next time it's a simple Xlate-stage invocation. Anyway, if I can be of any assistance, I'm happy to help. -- bc
On Mon, Aug 30, 2010 at 2:38 PM, Hughes, Jim <[email protected]> wrote: > We have some DB2 systems using 16 bit Unicode. > > > > Would someone show us how to convert 16 bit Unicode to ascii or ebcdic? > > > > Thanks. > > > > ____________________ > > Jim Hughes > > 603-271-5586 > > "It is fun to do the impossible." > > >
