Both get my vote. On Thu, Nov 11, 2010 at 7:51 PM, Dennis Lundberg <denn...@apache.org> wrote:
> Hi > > Both proposals are well thought through and are steps in the right > direction. > > On 2010-11-11 19:56, Robert Scholte wrote: > > hi all, > > > > I'm making some nice progress on QDox 2.0 but there are some issues for > > which I'd like to do a proposal. They will be touching the heart of > > QDox, so I'll give everyone a chance to respond. > > > > 1. Change java prerequisite of QDox to Java5 > > Up untill now QDox 1.x this project could be built with java1.4. Main > > reason was to support doclets as some sort of annotation-mechanism for > > those who couldn't make the leap to java5. > > Occording to the Sun/Oracle documentation on Java 1.4.2 [1] is > > considered End Of Life since october 30th 2008 (that's over two years > > ago). And since QDox is used in much more ways than just an > > Annotation solution for pre-java5 stuff, I'd suggest to move to the next > > era and use the java5 QDox 2.x so we can make use of a much stronger > > language-features > > This way we can get rid of duplicate Objects (both an array and a List) > > to keep track of certain models, the code is better readable and > > maintainable. QDox 2.0 would be the perfect moment to make the switch. > > > > 2. (Re)move AbstractQdoxTask (for Ant) and APITestCase(for junit3) > > Main reason to remove them is that these classes don't belong in the > > core-project of QDox, they should be considered projects/ultilities who > > make use of QDox. Their required dependencies kind of spoil the > > pom. They should at least be moved to separate projects (qdox-anttools > > and qdox-junit3tools or something like that) but we might even wonder if > > these projects are still worthwhile to maintain. Especially for junit3, > > since junit4 is the new standard, which uses... annotations :) So no > > need to subclass a TestCase > > I'd suggest to remove them, untill users complain about missing > > them. They can be added in just a few steps afterwards... > > > > > > I've had a look at the streaming parser, but we need to get a new > > feature inside JFlex[2] to have a solid solution. I don't think you can > > vote here, but some extra support for this feature would be nice. > > > > > > regards, > > > > - Robert > > > > [1] http://www.oracle.com/technetwork/java/javase/index-jsp-138567.html > > [2] > > > http://sourceforge.net/tracker/?func=detail&aid=3089140&group_id=14929&atid=364929 > > < > http://sourceforge.net/tracker/?func=detail&aid=3089140&group_id=14929&atid=364929 > > > > > -- > Dennis Lundberg > > --------------------------------------------------------------------- > To unsubscribe from this list, please visit: > > http://xircles.codehaus.org/manage_email > > >
