Vincent van Ravesteijn - TNW wrote:
Tommaso,
Would you please post this kind of patches before
committing. I thought we have been very clear by now.
Don't be surprised when I start reverting your
commits when I don't like them.
Well, on this particular issue, I think we discussed
it last week Vincent.
Well, we discussed about adding a isInternal() function, but I would
rather see a patch beforehand than that I have to start fixing things
afterwards.
I won't repeat myself what was wrong/missing or not completely along our
style guidelines, but there were a few issues.
Ok, maybe I'm being a bit pricky, but it's just much more effective to
comment on a patch before committing.
Well, in this case I told Tommaso about the TocBackend modification so i
understand that he felt that as an indication that it could commit
directly. IOW, this is probably my fault, not Tommaso's.
Maybe we want to think about how to solve the TocBackend problem. Now,
every buffer gets a toc_backend. So, I'm not sure whether we want this
also for the internal ones and then add the buffer.isInternal() calls
into TocBackend... It feels wrong. And well, the outliner thing is still
completely broken.
So, what's your proposal? Maybe that toc_backend_ becomes a pointer that
is not instantiated if the buffer is internal? That would be a solution
indeed. In any case, let's please try to focus on the solution instead
of the problem.
Abdel.
