Re: japi and protected field declaring classes

[Mark Wielaard]

Hi Stuart,

On Tue, 2005-10-04 at 13:22 -0400, Stuart Ballard wrote:
> It appears that Sun have a field numBands in java.awt.image that's
> exactly like Implementation 2 above (declared in both
> ComponentSampleModel and SampleModel), except that it's protected.
> (someone fixed classpath to conform to this, gmane isn't helping me
> find out who actually did it right now though).

You are looking for the following thread by the two Toms:
http://lists.gnu.org/archive/html/classpath-patches/2005-10/msg00007.html

> If it really doesn't matter, I'm sorry to the person who made the
> awt.image fix based on this misinformation out of japi... :(

Changing things just because japi doesn't give a full green bar is wrong
to start with anyway. Japi can contain bugs. And it doesn't tell you
anything about the why. In the end the why is much more important. As is
getting real programs to work with our library and you do that by
studying what they actually need our library to provide, not just by
going through a list of names and constants.

I do hope someone packages japitools for Debian so that it can be used
easily by the packagers. Running japi on two versions of a library
written in java will be a real help to actually see if an upgrade might
break applications depending on it. It was one of the complaints the
packagers had about a lot of libraries out there. Lots of libraries
actually break binary or source compatibility between minor releases.
Japi could really help people not to make such mistakes between
releases.

Cheers,

Mark

signature.asc
Description: This is a digitally signed message part