[CSVObjects-users] Contributions and ideas

[Benoit O. <bo...@ze...>]

Hello !

I've integrated your CSV parser library into one of my project. 
Everything went well and I really appreciated the Hibernate-like way of 
defining the mapping from CSV to Java :-)

I don't know if it's normal but I had to add the xdoclet jars in the lib 
directory in order to build the project. I also had to modify the 
definition of the xdoclet taskdef in build.xml for ant to
recognize the task (my build is launched from the command line). If you 
want to integrate these modifications, just let me know.

Now, let's move on to serious things :-). As I did not want to embed my 
mapping and formatter configuration files into the CSVObjects jar 
(because it forced me to maintain my files in two points
and to re-build the CSVObjects jar each time I wanted to make a 
modification on those files), I made some modifications into your code 
to be able to (but not forced to) dynamically specify those two
files. If you're interested in integrating these modifications, I can 
send you a patch. Just let me know !

Last thing I wanted to speak about with you : I saw you used the 
singleton pattern in your implementation. This is fine as long as you 
have only one mapping and formatter configuration. For now, this is also 
fine for me but I know that, sooner or later, I will have to deal with 
multiple concurrent mapping and formatter configurations. Do you have 
ideas on how to deal with this use case ?

Thanks again for this nice and handy project !


Regards,


Benoit.
-- 
Benoit Orihuela - ben...@ze... - GPG-PGP : C94615BE
Architecte logiciel, Zenexity - http://www.zenexity.fr
Mobile : +33 (0)6 87 42 71 07 - Tél : +33 (0)1 44 89 47 66
Zenexity, 101-103 boulevard Mac Donald 75019 Paris