On Tue, Mar 11, 2008 at 10:18 PM, Johannes Schindelin<Johannes.Schindelin@xxxxxx> wrote:> Hi,>>> On Tue, 11 Mar 2008, Nguyen Thai Ngoc Duy wrote:>> > On Tue, Mar 11, 2008 at 8:20 PM, Johannes Schindelin> > <Johannes.Schindelin@xxxxxx> wrote:> >>> > > On Sun, 2 Mar 2008, Nguyễn Thái Ngọc Duy wrote:> > >> > > > Signed-off-by: Nguyễn Thái Ngọc Duy <pclouds@xxxxxxxxx>> > >> > > I do not like this change. It is IMO completely unnecessary, and> > > might break assumptions.> >> > It could. And I tried my best to make sure it did not break anything.>> That is contradicting. I think that whatever you tried, once you let> rev-parse --git-dir return only absolute paths, you already risked> breakage.>> Besides, I remember that Junio specifically requested that I would _not_> do something like that (this was for --show-cdup, but I think it really> applies here, too).
I might have lost track. Could you provide me the link for reference? Thanks.
> > While it's not really necessary, it would be IMHO a good change as you> > would not have to rely on `cwd` anymore (that would mean whether moving> > cwd by prefix or not would no longer matter).>> I do not see the problem. And therefore I do not see that this patch> solves anything.
It can simplify things (in other patches, such as builtin-config.cchanges). Though without it things just run fine (with a little morecomplication in prefix handling). Recent mails about git-am running insubdirectory (Message-ID:20080301062255.GA27538@xxxxxxxxxxxxxxxxxxxxxxx) made me think therewas room for this.
It is also prerequisite for later changes in the series (IIRC,removing prefix from setup*_gently() needs it).
> Ciao,> Dscho>
-- Duy��.n��������+%������w��{.n��������n�r������&��z�ޗ�zf���h���~����������_��+v���)ߣ�m
