Em Segunda-feira 09 Novembro 2009, às 15:19:12, ext Matthias Dahl escreveu:
> On Monday 09 November 2009 10:11:07 you wrote:
> > 17bf093fa0fb041c2b9a6fec71b90a8630fba1ff is not a tree hash, it's a
> > commit hash.
> 
> I know, sorry for the confusion. Does git even have a real tree hash
> comparable to an svn revision?

Git has a tree hash, but it's not comparable to SVN revision. Git commit 
hashes are comparable to SVN revisions.

If you make a change, commit, then make the opposite change and commit again, 
your tree hash will be the same as it was two commits previously. So a tree 
hash doesn't identify a commit uniquely.

> I never ran across one so far and I just
> grepped through the man pages but there doesn't seem to be any command to
> really get one either.

$ git rev-parse HEAD
c457aebbba18c5431915a664a02264fae96cc848
$ git rev-parse HEAD^{tree}
80cdd6311fe5cec1958adfe9cb40af43401128f8
$ git rev-parse HEAD:
80cdd6311fe5cec1958adfe9cb40af43401128f8
$ git rev-parse HEAD:src
0f054c65231278c5ca59f233b9cb6af8a3b5f144
$ git rev-parse HEAD:src/corelib
e169215d630c845afc91aaa91680098812cc12a7


> > And 7bf093fa0fb041c2b9a6fec71b90a8630fba1ff is not a valid object:
> 
> I was in a hurry but I really should've re-read that mail before sending
>  it, sorry.
> 
> What I meant was the regression must have crept in between those two
>  commits:
> 
> 17bf093fa0fb041c2b9a6fec71b90a8630fba1ff
> 5b4b6b2be7b901ef9a29c37431998034730fa3d3
> 
> But I guess that's not very helpful because it's based on the public
> repository. Just for future reference: What's the best way to narrow the
> search range down for you guys (apart from a proper bisect)?

Actually, what you did is very helpful.

There are 73 commits in that range, including some D-Bus changes I did. 
According to Olivier, he got a backtrace of KMail stuck in D-Bus somewhere, so 
I'm inclined to think it's one of those.

The changes I made are 5c7345809d7f620981f92cc2e93beb14b10504a9 through 
bd9a8091eb8c731569e4972a04f23c9a2f48391c.

My guess is I broke the new KDE notification item library, which is based on D-
Bus but uses bus names to identify clients. I've never liked that D-Bus 
protocol and now would be a good time to fix it properly too.

-- 
Thiago Macieira - thiago.macieira (AT) nokia.com
  Senior Product Manager - Nokia, Qt Development Frameworks
     Sandakerveien 116, NO-0402 Oslo, Norway

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

_______________________________________________
Qt4-preview-feedback mailing list
[email protected]
http://lists.trolltech.com/mailman/listinfo/qt4-preview-feedback

Reply via email to