Hi Annette,

Here are some guidelines we submit to developers who want to contribute
to the Castor code base:

1- Open a bugzilla item and mark your request as
"enhancement" and make sure you assigned the bug to yourself.

2- Try to be more confident with the code base by learning it, testing
it. You can ask questions for any dark areas. Make sure you understand
the overall consistency.

3- Once you gain sufficient confidence, you can design and code your
contribution. During the coding phase I advise you to follow the 
Exolab Java Conventions (it can be found on the web site).

4- Once you're done, make a diff against the current code base and 
attached it in the bugzilla (diff -u is the preferred patch)

5- Inform us so that we can enter the review process. The
review process can take time but it will be followed for sure. We'll sit
down and make a full review of the code and provide you feedback.

6- Submit a CTF Test case (if possible) for your new feature/patch

7- That's it; the patch if accepted should be in the next release :)

Feel free to ask for more information,

Arnaud

> -----Original Message-----
> From: [EMAIL PROTECTED] [mailto:[EMAIL PROTECTED]]
> Sent: Thursday, December 12, 2002 12:11 PM
> To: Arnaud Blandin; [EMAIL PROTECTED]
> Subject: Re: [castor-dev] Question concerning Unsigned data type
> 
> Hello Arnaud,
> 
> my experience is completely different. I have been working with
several XML
> Schema files where unsigned types of all kinds were used exhaustively,
and
> often in
> places where there was no real reason to restrict the values to
non-negative;
> I
> ascribe this to the habits of some C programmers in the teams
specifying the
> schema.
> 
> So I'd offer as a first step to try with the unsignedInt,short,byte
mapping
> and leave
> unsignedLong for further discussion. Which I open below ...
> Is there anyone whom I can contact to check if my modifications are
ok? You? I
> am a
> newbie to open source programming, but I know the reactions of
teammates when
> you break the code ...
> 
> I understand your argumentation concerning unsignedLong - you do not
want to
> go
> back behind the XML Schema spec. But I would dare mapping unsignedLong
onto
> long despite of the non-literal mapping of the Schema spec for the
following
> reasons:
> - the use of a mapping still cannot solve the problem that you just
cannot
> store
> integer values > Long.MAX properly in Java
> - I think that long values in the range of Long.MAX+1 up to the XML
Schema
> maximum will occur even less frequently than the usage of unsignedLong
at all
> - off-the-shelf usability of Castor is an important argument to
augment the
> audience.
> This comes from my experience again - I find it in most cases
extremely
> difficult to
> convince people to use tools they have not used before, and
argumentation is
> remarkably easier if a tool produces good results without the need for
> reconfiguration.
> 
> Regards, Annette
> 
> --
> Dr. Annette S�ndermann
> Zenturion GmbH
> Postfach 1248,  85730 Ismaning
> Tel.: 089/638 76 706 Fax: 089/638 76 707
> Internet: http://www.zenturion.de

----------------------------------------------------------- 
If you wish to unsubscribe from this mailing, send mail to
[EMAIL PROTECTED] with a subject of:
        unsubscribe castor-dev

Reply via email to