Ciao Mediii,
on Monday we discussed how to proceed on the 19108 implementation and
the agreement we found was as follows:
1> you (or someone over there 
) committing the version of the code
2> us reviewing and commenting the code you have committed
3> deciding how to proceed (joda vs non-joda as an instance)
Just to recap, since I got distracted by other tasks, what is the
current situation on your side?
Thanks,
Simone
On Mon, May 26, 2008 at 6:03 PM, mediii <mehdi.sidhoum@anonymised.com> wrote:
Hi Simone,
With Martin, we though that it would be safer to leave aside Joda Time for
now because introducing new dependencies in metadata or referencing library
(where a temporal framework should probably lives) is a sensitive issue. We
would like to avoid adding an additional layer of controverse on top of JAXB
in metadata. We see the issue as different than JAXB in that there is no
alternative to JAXB in the JDK, while there is alternatives (admitly
inferiors) to Joda in current JDK.Furthermore Joda is expected to be included in JDK 7, but under another name
(see Date And Time API JSR 310). We would like to wait for its inclusion in
JDK 7 in order to use the Java classes with their appropriate name, and thus
avoid breaking public API during the switch from Joda to JDK 7.For un/marshalling threads we use JAXB and the implementation is based on
jdk6. Currently I have implemented the most important classes of ISO19108
and I have just begun the JUnit tests.Mehdi.
--
-------------------------------------------------------
Eng. Simone Giannecchini
President /CEO GeoSolutions S.A.S.
Via Carignoni 51
55041 Camaiore (LU)
Italy
phone: +39 0584983027
fax: +39 0584983027
mob: +39 333 8128928
-------------------------------------------------------