Re: CCSID Conversion -- MIDRANGE-L


On 16/03/2005, at 6:18 AM, Brescia, Richard wrote:

Is there an easy way on the Iseries (V5R2) to convert English characters to Japanese either through PF, LF, or PRTF using the CCSID keyword?

CCSID conversion does not do language translation. Notice the different use of words: CCSIDs are for 'conversion' not 'translation'. Their purpose is to ensure that a given character retains its meaning when moving data between different code pages and character sets. For example, a US dollar sign will be treated as a Pound Sterling symbol if viewed using a UK code page without conversion. Tagging the original data with CCSID 37 (US) allows the dollar sign to stay as a dollar sign when viewed with CCSID 285 (UK). They do not convert text in one language into another.

Or is there a conversion API with an associate translation table I could use?

The conversion APIs are not translation APIs. All they do is use either a CCSID or a conversion table to ensure characters maintain integrity in different code pages and character sets.

I read something that if your system is has the CCSID set to 65535 a conversion is not possible. Is this true?

Having the system value QCCSID set to 65535 or the job CCSID set to 65535 means you are telling the system you don't want conversion to occur and that data should be read as hexadecimal values. Running with QCCSID set to 65535 is dumb--even though that's how most sites run.

Each job has two values governing CCSID conversion:
        o Job CCSID
        o Job Default CCSID

Run DSPJOB option 2 and go to the 3rd page. Put the cursor on the two CCSID value and press Help. Read.

In some cases the system will ignore a job CCSID of 65535 and use the default job CCSID instead.

            My goal is to print an item label in Japanese or English
depending on which user and printer is being used.

Then you must handle the translation yourself. You must externalise the text you want to print on the label. Message files are a good method for doing this but you could use database files. You then retrieve the appropriate text according to the user and printer. The usual way is to have one message file with English text, duplicate it and translate manually into Japanese, then at run time set your library list appropriately so the correct message file is found. If you use an externally described printer file then you can get the system to retrieve the message text for you. Investigate the MSGCON keyword.

You should also read the information IBM provides about internationalising your application. Start at the Globalisation section of the Information Centre.

Regards,
Simon Coulter.
--------------------------------------------------------------------
   FlyByNight Software         AS/400 Technical Specialists

   http://www.flybynight.com.au/
   Phone: +61 3 9419 0175   Mobile: +61 0411 091 400        /"\
   Fax:   +61 3 9419 0175                                   \ /
                                                             X
                 ASCII Ribbon campaign against HTML E-Mail  / \
--------------------------------------------------------------------