unicode - cp maps

[Vinay Kapoor]

Hi there,

The ICU home page mentions that:

"ICU provides character set conversion with mapping
tables for a number of important codepages. The
default tables are a subset of IBM's CDRA conversion
table repository"

Could you please point me to the IBM CDRA conversion
table repository. I tried searching for it on the net
and the IBM website but drew a blank.

Best Regards,
Vinay

__________________________________________________
Do You Yahoo!?
Yahoo! - Official partner of 2002 FIFA World Cup
http://fifaworldcup.yahoo.com

[George Rhoten]

It's an internal repository that the public does not have access too.

All Unicode charset mappings from the CDRA repository are in the
http://oss.software.ibm.com/cvs/icu/charset/data/ directory, and those
files from the CDRA start with "ibm-" in the filename. These files are
not the original files, but they are a modified form of the original
files. All of those ibm-* UCM files can be put directly into ICU. The
other UCM files were collected from other platforms, and may require some
modification to add them to ICU.

George Rhoten
IBM Globalization Center of Competency/ICU San Jose, CA, USA




Vinay Kapoor <vinay_...@yahoo.com>
Sent by: icu-chars...@www-124.southbury.usf.ibm.com
05/29/2002 02:11 AM


To: icu-ch...@www-124.southbury.usf.ibm.com
cc:
Subject: unicode - cp maps

_______________________________________________
icu-charsets mailing list
icu-ch...@oss.software.ibm.com
http://oss.software.ibm.com/developerworks/oss/mailman/listinfo/icu-charsets

[Markus Scherer]

Correction: A snapshot of the original repository is available online via
the IBM Developer Connection site:

"The conversion table repository available externally (outside of IBM) via
the web is located at: Developer Connection On-line: Users must register
as a guest to gain access to the catalog. From the catalog select Sample
Code , then scroll down to NLS to find the link to the conversion tables."

The link is http://www.developer.ibm.com/devcon/titlepg.htm


Our ICU collection contains only mappings between Unicode and legacy
codepages, while the CDRA repository also contains direct legacy<->legacy
mapping tables.
The .ucm file format is an extension of the UPMAP file format in CDRA,
with additional information. See
http://oss.software.ibm.com/icu/userguide/conversion-data.html


Best regards,
markus


Markus Scherer IBM GCoC-Unicode/ICU San José, CA
markus....@us.ibm.com (also for SameTime)

[George Rhoten]

Here is what the CDRA has to say about 5026:
This package is a copy of 930 <-> 13488 package. The only difference
between CCSID 930 and 5026 is 930 contains 2490 additional DBCS. Any
changes made to either package must be updated in both.

Here is what the CDRA has to say about 5035:
This package is a copy of 939 <-> 13488 package. The only difference
between CCSID 939 and 5035 is 939 contains 2490 additional DBCS. Any
changes made to either package must be updated in both.

So you could try to use 930 and 939 instead, which seem to be supersets of
5026 and 5035.

George Rhoten
IBM Globalization Center of Competency/ICU San Jose, CA, USA




Vinay Kapoor <vinay_...@yahoo.com>

05/29/2002 07:58 PM


To: George Rhoten/San Jose/IBM@IBMUS
cc:
Subject: Re: unicode - cp maps




Hi George,

Thanks a lot for your input. I have already downloaded
the codepages from that site.

One last question I have on the subject is that the
AS/400 manuals mention CCSIDs 5026 and 5035 as default
CCSIDs for the Japanese markets. These CCSIDs use
codepages 290, 300 and 1027 (charsets 1172 and 370);
the unicode mappings of which I was unable to find in
the ICU CVS repository. Where can I get the unicode
mappings for these codepages? Is there an alternative
I can use?

Thanks and Regards,
Vinay