Stephen Colebourne wrote:
Once you start getting into a list like this you must consider the IBM ICU
project, which tackles these kind of issues. (note, I haven't used ICU).
ICU tackles a lot of important functionality, mainly related to
extended Unicode support (compared to the Java RT). Most notably
it provides character normalization, as well as localized
collation, generalized number parsing and formatting, script
dependent glyph shaping and some other character/script/text
related stuff.
It also has all kind of calendars, localized date and time parsing
and formatting, extensions for dealing with holidays (very useful),
and a somewhat extended support for currencies (also compared to
Java RT).
Some of the functionality, in particular BIDI support, was adopted by
Java 1.4.
A look at the ICU4J jar size should make clear that proper i18n is
no small undertaking.
There's still more to i18n, especially for interactive programs
which get significant text input from the user. This requires
internationalized input methods, something which is rather
nontrivial to get right. There's also hyphenation, word inflection,
spell check, punctuation and a few other text/script specific stuff.
Other, hardly explored i18n features are writing modes and dialog
component arrangement (imagine a drop down box in a lr-tb writing
mode), cultural dependent use of colors, patterns and pictograms,
and the gadzillion of small differences in national standards and
customs for expressing measurements, city names, legalese,
lies+threats, formulas and whatnot.
Regards
J.Pietschmann
---------------------------------------------------------------------
To unsubscribe, e-mail: [EMAIL PROTECTED]
For additional commands, e-mail: [EMAIL PROTECTED]
