[BUG]UndefinedObject(Object)>>doesNotUnderstand: #contents

goran.krampe at bluefish.se goran.krampe at bluefish.se
Mon Nov 3 15:19:46 UTC 2003


"Peter van Rooijen" <peter at vanrooijen.com> wrote:
[SNIP]
> > BFAV is a package on SM - it is not in the base image. So Brent has full
> > control over the release cycles of BFAV.
> 
> I understand now. Apparently the bug was not in the image, but in the BFAV,
> and I had just loaded the bug into my image from SM.
> 
> I had no idea the bug was in the BFAV code. How did you know that it was?

Well, since I recognized it from way when it popped up first it was no
real challenge.

> The test behind the link to the fix that you posted (above) doesn't specify
> that it was BFAV code that had the bug (i.e., that the bug was not in the
> image).
> 
> > > While I have your attention, I have a question about the BFAV, too.
> There
> > > seem to be loads of items in the Bug Fixes Archive Viewer that don't
> claim
> > > to be bug fixes, don't even claim to report a bug. What are they doing
> in
> > > the BFAV?
> >
> > BFAV sucks up a lot of stuff, like for example all the ENHs too. So yes,
> > it holds more than just FIXes.
> 
> Why aren't ENHs on SqueakMap? What do they have to do with bugs?

Well, the bug fixes archive was first intended for bug fixes. Then it
has been extended with more XXX tags for ENHancements etc. The
mailinglist has been the medium for maintaining Squeak since "birth of
time".

Larger stuff tends to turn into packages posted on SM - now that we have
SM.

> Ahhh, maybe the BFAV really isn't for bugs and fixes, even though the name
> suggests it. Maybe it is really the place for everything that someone would
> like to see changed in the base image? Maybe it is really the Base Image
> Submissions Tracker???

Typically yes. And it also tends to catch quite a lot of stuff really
intended for packages NOT in the image too. This is just because SM is
new etc - we will solve it eventually. Today we have the "mail to list"
in the changesorters so that is the easiest way to send in a patch.
Eventually we should have a "send to maintainer" or something so that it
gets routed to the correct package maintainer. That is not so trivial
though - but there are ideas.

> Thanks for you guidance, and thanks for your patience.

No problem. Just try to... change the tone a little please (referring to
my other post) ;)
 
> Regards,
> 
> Peter

regards, Göran



More information about the Squeak-dev mailing list