On 2018-09-06, at 07:13:54, Sankaranarayanan, Vignesh wrote: > > On mainframe, a dataset has £ (the pound symbol) in multiple records. > I'm transferring to a RHEL box with lzopts="mode=text". > The £ shows up as a $. > > I ran the COZ job with the following in the first //SFTPIN input: > export COZ_LOG="T,Translator=F" > > I see these relevant lines in the log: > ZosSettingsÝI¨: Transfer options: > clientcp=IBM-1047,mode=text,servercp=ISO8859-1,trim > TranslatorÝF¨: -> Translator(IBM-1047, ISO8859-1, , 0, 0) > > This is the output for 'locale' in the target RHEL machine: > -sh-4.2$ locale > LANG=en_GB.UTF-8 > LC_CTYPE="en_GB.UTF-8" > My first guess would be to use UTF-8 ratner than ISO8859-1 in the Translator(), since that's what Locale tells you.
How does Co:Z deal with MBDATACONN? Why do I see incorrectly rendered characters in "ZosSettingsÝI¨:"? I hate EBCDIC! -- gil ---------------------------------------------------------------------- For IBM-MAIN subscribe / signoff / archive access instructions, send email to [email protected] with the message: INFO IBM-MAIN
