[MSysGit] Windows & executable bit

[MSysGit] Windows & executable bit

[Ferry Huberts]

Hi list,

Our project is (also) using msysgit.
Recently we changed the permissions on a file
(a Linux shell script) by this (gitk) patch:

----------------------------- xxxx/SQC/SQC.sh -----------------------------
old mode 100644
new mode 100755

We did this on a Linux machine.

Now on a Windows machine we pull in the change:
This causes problems when checking out the branch on which the change was
made: after a checkout the status of SQC.sh is always 'modified' and we can't
convince git to reset it. I actually succeeded _once_ but can not remember how
I did it.
The file is _always_ dirty.

Is this a bug?

I have to note that when I re-clone the repo I don't have the problem. So it
seems to be somewhere in the 'update the working copy' code?


Ferry

Re: [msysGit] Windows & executable bit

[Johannes Schindelin]

Hi,

On Tue, 18 Aug 2009, Ferry Huberts wrote:

> after a checkout the status of SQC.sh is always 'modified' and we can't 
> convince git to reset it. I actually succeeded _once_ but can not 
> remember how I did it. The file is _always_ dirty.
> 
> Is this a bug?
> 
> I have to note that when I re-clone the repo I don't have the problem. 
> So it seems to be somewhere in the 'update the working copy' code?

git config trust.fileMode false

Ciao,
Dscho

Re: [msysGit] Windows & executable bit

[Thomas Rast]

Johannes Schindelin wrote:
> 
> git config trust.fileMode false

Isn't that core.filemode ?

-- 
Thomas Rast
trast@{inf,student}.ethz.ch

Re: [msysGit] Windows & executable bit

[Ferry Huberts]

>
> Isn't that core.filemode ?
>

yep. works

however, the man page says that you'll have to configure it in this way when
you're on a broken filesystem like FAT. We're on NTFS however....
NTFS is also broken then?

excuse me for the questions, I'm really a Linux guy, I know very little about
Windows. I'm just the interface towards the community :-)

Re: [msysGit] Windows & executable bit

[Johannes Schindelin]

Hi,

On Tue, 18 Aug 2009, Ferry Huberts wrote:

> > Isn't that core.filemode ?
> 
> yep. works
> 
> however, the man page says that you'll have to configure it in this way 
> when you're on a broken filesystem like FAT. We're on NTFS however.... 
> NTFS is also broken then?

It is more a question of the ACLs not being properly mappable to POSIX 
permissions.  So we cannot trust the file mode.

Ciao,
Dscho

Re: Windows & executable bit

[Johannes Schindelin]

Hi,

On Tue, 18 Aug 2009, Thomas Rast wrote:

> Johannes Schindelin wrote:
> > 
> > git config trust.fileMode false
> 
> Isn't that core.filemode ?

The matching is actually case-insensitive, but I prefer CamelCase here...

Ciao,
Dscho

Re: [msysGit] Windows & executable bit

[Erik Faye-Lund]

On Tue, Aug 18, 2009 at 1:55 PM, Johannes
Schindelin<Johannes.Schindelin@gmx.de> wrote:
>> > git config trust.fileMode false
>>
>> Isn't that core.filemode ?
>
> The matching is actually case-insensitive, but I prefer CamelCase here...

I believe the question was about it being in the trust vs the core section...

-- 
Erik "kusma" Faye-Lund
kusmabite@gmail.com
(+47) 986 59 656

Re: Windows & executable bit

[Johannes Schindelin]

Hi,

On Tue, 18 Aug 2009, Erik Faye-Lund wrote:

> On Tue, Aug 18, 2009 at 1:55 PM, Johannes
> Schindelin<Johannes.Schindelin@gmx.de> wrote:
> >> > git config trust.fileMode false
> >>
> >> Isn't that core.filemode ?
> >
> > The matching is actually case-insensitive, but I prefer CamelCase here...
> 
> I believe the question was about it being in the trust vs the core 
> section...

Bah, you're right.  I let myself fool by the lowercase 'm'.  Sorry.

Ciao,
Dscho

Re: Windows & executable bit

[Johan 't Hart]

Johannes Schindelin schreef:
> Hi,
> Bah, you're right.  I let myself fool by the lowercase 'm'.  Sorry.


Haha. Looking with developer eyes. I know that problem from someone... 
Oh yea, me.