Re: Importing Unicode data to the System i

Hello,

AFIAK, IBM i doesn't support little-endian data. Can't you just load the file into Notepad or something, and use 'Save As' to save it in a big-endian format? Or UTF-8? (Alternately, you could write a program that translates little-endian to big-endian.)

The CCSID for UTF-16 (big endian) is 1200. The CCSID for UTF-8 is 1208. (935 is a Chinese variety of EBCDIC -- not Unicode.)

When you put the data in the IFS, make sure you transfer it with a binary-safe method (such as FTP in binary mode) and change the CCSID on the object (with the CHGATR CL command, for example) to the appropriate CCSID.

After that, you should be able to copy it in with CPYFRMSTMF or CPYFRMIMPF depending on the style of the file and/or the style of the PF.


On 8/31/2010 10:24 PM, sjl wrote:

I have a Unicode text file created by containing double-byte (Simplified)
Chinese data that I need to import to the System i. This file has a
preamble (BOM) of x'FFFE', indicating that it is UTF-16 Low-Endian. How do
I go about getting this data into a physical file with the appropriate
CCSID? (I believe the correct CCSID is 935...)

1) How do I deal with low-endian data?
2) Will l need to copy it to the IFS and read it with an RPG program, using
iconv() to convert the data before writing it to the PF?

Regards,
sjl