Re: how to hard code @ sign in RPGLE program with out having to worry about CCSID

J.Beckeringh@xxxxxxxxxxxxxxxxxxxxxxxxxx wrote:

<<SNIP>> Let's see what the manual has to say:
'When converting between character and UCS-2 data,
the CCSID of the character data is assumed to be
the mixed-byte CCSID related to the job CCSID.'
Aha, I don't understand that either. What on earth
is 'the mixed-byte CCSID related to the job CCSID'?.

So, indeed I tested. I created a program with
CCSID 37 (well, actually DSPPGM says that the program
has 65535 and the module has 37) that converted
u'005B' ([) and u'005D' (]) to char. In a job with
CCSID 37 that became x'BA' and x'BB' ([] in CCSID 37);
in a job with CCSID 500 it became x'4A' and
x'5A' ([] in CCSID 500).

Apparently the unicode data is converted to the job CCSID.

With a SBCS [Single Byte Character Set] EBCDIC job CCSID, for any DBCS [Double Byte Character Set] processing, there is an associated /mixed/ or /mixed-byte/ CCSID according to the job language\CCSID settings. The term mixed-data indicates that both DBCS and SBCS can be intermixed into the same stream; e.g. in a character string variable, or in database & display file fields having data type of /Open/ [DDS O data type] which defines the same capability. The data in the stream can either /shift in/ to, or /shift out/ of DBCS, where the EBCDIC Control Codes at the hex 0x0E and 0x0F code points represent the two /shift/ characters which serve as the indicators of transition between the DBCS & SBCS data. With UCS2 there are no shift characters for DBCS, as every character is already two bytes [see UTF-16 CCSID 1200 however, versus UCS2-L1 13488]. So if for example the UCS2 data being converted includes Japanese Katakana characters, then only a mixed-byte CCSID would be capable of representing that data; i.e. the CCSID 37 does not have those characters, but the CCSID 5026 would.

QTQ_DEFAULT_CCSID environment variable can set the CCSIDs.
Some links for some related information:
http://publib.boulder.ibm.com/infocenter/iseries/v5r4/topic/nls/rbagsucs2.htm
http://publib.boulder.ibm.com/infocenter/iseries/v5r4/topic/nls/rbagsunicodeandprior.htm
http://publib.boulder.ibm.com/infocenter/iseries/v5r4/topic/nls/rbagsutf16.htm
http://www.ibm.com/systems/i/software/globalization/
http://www.ibm.com/systems/i/software/globalization/unicodeinfo.html
http://publib.boulder.ibm.com/infocenter/systems/topic/nls/rbagslngidsdefaultccsids.htm
http://www.ibm.com/systems/i/software/globalization/get_related.html
http://publib.boulder.ibm.com/infocenter/iseries/v5r4/topic/apis/CDRGRDC.htm

Regards, Chuck