|
From: Tony W. <wuy...@gm...> - 2006-12-29 15:15:14
|
Hi George, Thanks. My comments inline :) On 12/29/06, George Rhoten <gr...@us...> wrote: > Since there is no data item called ucdata, I presume you're talking about > ucadata. exactly > It should be noted that using data from one version of ICU in > another version of ICU (both ICU4C and ICU4J) may not work very well. I= f > you really want to use data from a specific version of ICU, it's just > better to use that version of ICU. It's not uncommon for the data to > change with a lock in step change in the code. > Is there any potential risk for icu4j3.4 with 3.2's ucadata besides mentioned by Mark? I think there must be some users like me who have this requirement. We want to enjoy the new feature in high version of icu but we have to be compatible with older unicode data. I have a naive thoughts that it will be a good feature for icu to switch its ucadata freely to be compatible with any unicode version. Do you have any concern about that? > If you decide to ignore my advice, you should at least run the tests with > the new data. > Good suggestion, I'll try. > George Rhoten > IBM Globalization Center of Competency/ICU San Jos=E9, CA, USA > http://www.icu-project.org/ > http://icu.sourceforge.net/ > > > > "Tony Wu" <wuy...@gm...> > Sent by: icu...@li... > 12/28/2006 02:12 AM > Please respond to > ICU support mailing list <icu...@li...> > > > To > "ICU support mailing list" <icu...@li...> > cc > > Subject > Re: [icu-support] [icu_support]how to customize the data for icu4j? > > > > > > > Hi Mark, > Thanks for your remindering. Actually, I just want to replace the > ucdata in 3.4 with that in 3.2 :) > > On 12/28/06, Mark Davis <mar...@ic...> wrote: > > I want to reinforce what Ram said. Changing the fundamental Unicode > > properties is very strongly discouraged. ICU and other programs are > built on > > the basis of those properties, and changing them just opens the way for > > strange, intermittent bugs, and incompatibilities with other programs. > > > > Mark > > > > > > On 12/27/06, Ram Viswanadha <ram...@gm...> wrote: > > > > > > Hi Tony, > > > > > > > > > On 12/27/06, Tony Wu <wuy...@gm... > wrote: > > > > > > > > I have another question here. I noticed that the data files were > > > > combined to a .dat file in the version 3.4 other than previous. > > > > > > > > > I do not understand this statement. > > > > > > > I wonder whether my change in the > > source/data/unidata/UnicodeData.txt > > > > will affect the output data? That is, if I change a defination of > > > > 0x0000 in the .txt file, would it be changed in the distribution? > > > > > > > > > Please be informed that it is not recommended that you change the > > definitions of code points in UnicodeData.txt file. If you changed > > definitions in UnicodeData.txt and then ran gendtjar.pl, then yes the > > changes are taken transferred to data files generated and hence > icudata.jar. > > If you change the data then please make sure to change the manifest.stu= b > > files to reflect that fact that you have modified icu4j.jar. > > > -- > > > > > > Best Regards, > > > Ram Viswanadha > > > ICU > > > > ------------------------------------------------------------------------- > Take Surveys. Earn Cash. Influence the Future of IT > Join SourceForge.net's Techsay panel and you'll get the chance to share y= our > opinions on IT & business topics through brief surveys - and earn cash > http://www.techsay.com/default.php?page=3Djoin.php&p=3Dsourceforge&CID=3D= DEVDEV > _______________________________________________ > icu-support mailing list - icu...@li... > To Un/Subscribe: https://lists.sourceforge.net/lists/listinfo/icu-support > --=20 Tony Wu China Software Development Lab, IBM |