Hey Martin,

This is great! One of the biggest things we've tackled in Apache Tika over the 
years is dealing with
Internationalization (same with Nutch) so I understand how important this is. 
Great job committing
the first parts of the framework and I'll look forward to leveraging it as it 
streams in :)

Cheers,
Chris

On Sep 29, 2012, at 3:22 AM, Martin Desruisseaux wrote:

> Hello all
> 
> Yesterday, I committed the first pieces of GeoAPI implementations. My 
> previous commits were only plumbing (utility methods, etc.). So we are 
> getting closer to more interesting stuff.
> 
> The GeoAPI implementation committed yesterday was the most basic type: 
> InternationalString (not yet a very geographic stuff...). The idea is 
> inspired from JSR-150 (Internationalization Service for J2EE) with support 
> for time zone removed (for now). The idea is that we are going to handle 
> potentially big objects that should be the same for every languages. For 
> example the WGS84 Coordinate Reference System is the same object no matter 
> the spoken language, and would be represented in Apache SIS by the same 
> instance for everyone. Nevertheless it still need to provide localized 
> descriptions of itself, especially when used on a server serving web pages to 
> users all around the world. So instead of providing methods like:
> 
>    String getDomainOfValidity();
> 
> it provides:
> 
>    InternationalString getDomainOfValidity();
> 
> where InternationalString is basically a java.lang.CharSequence with a 
> toString(Locale) method. Maybe you noticed that the utility methods committed 
> recently were designed to work on CharSequence rather than String; this was 
> obviously on purpose :-).
> 
> We do not provide the following method:
> 
>    String getDomainOfValidity(Locale);
> 
> because it is sometime too early for knowing the locale. For example we may 
> want to use the domain of validity for building a new metadata object, which 
> needs itself to be internationalizable (the INSPIRE legislation defines a 
> TextGroup for this purpose which works basically in the same way).
> 
> The classes are there:
> 
> https://builds.apache.org/job/sis-trunk/site/apidocs/org/apache/sis/util/type/package-summary.html
> 
> The Javadoc discusses about "GenericName" which are not yet there, but they 
> will be committed probably next week.
> 
>    Martin
> 


++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++
Chris Mattmann, Ph.D.
Senior Computer Scientist
NASA Jet Propulsion Laboratory Pasadena, CA 91109 USA
Office: 171-266B, Mailstop: 171-246
Email: [email protected]
WWW:   http://sunset.usc.edu/~mattmann/
++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++
Adjunct Assistant Professor, Computer Science Department
University of Southern California, Los Angeles, CA 90089 USA
++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++

  • Current state Martin Desruisseaux
    • Re: Current state Mattmann, Chris A (388J)

Reply via email to