[cvsnt] Re: static file

Arthur Barrett arthur.barrett at march-hare.com
Tue Oct 25 10:21:58 BST 2005


Olaf,

Thanks for your contributions to the discussion about static files.

The definition of a static file (-ks) is one the one that Tony has
stated "never considered modified".  

The static feature is in stable builds and documented as such and to
change the behaviour would break existing use-cases (ok ok I know we
have done that in the past - but we do try not to).

If you see value in another file type defined as "will not committed
unless forced" (maybe called "noncommit files") then we would be happy
to review your patches to CVSNT and the help files.

Regards,


Arthur


-----Original Message-----
From: cvsnt-bounces at cvsnt.org [mailto:cvsnt-bounces at cvsnt.org] On Behalf
Of Olaf Groeger
Sent: Tuesday, October 25, 2005 7:28 AM
To: cvsnt at cvsnt.org
Subject: [cvsnt] Re: static file


Tony Hoyle wrote:

> Olaf Groeger wrote:
>> to be a clever way to implement, i think it is a bad idea, because 
>> all frontends mark a static file now as "not modified" regardless of 
>> it real status. This means that if i change a static file somewhere 
>> in my sandbox i
> 
> That *is* its real status.  A static file is never considered modified

> - that's the whole point.
> 

No, Tony, the status is modified. Modified in the sandbox, because i
changed the file. But the frontends consider it as uptodate=*not*
modified. So if i activate the flat mode in WinCVS to see modified
files, the modified static file is not shown and the message from WinCVS
for me is that my sandbox is
*identical* to the repository, which is obviously wrong because i
changed a file. This means that the static file like you define
it("never considered
modified") undermines the mutual trust between cvs and the developer,
because cvs reports an incorrect status about the sandbox. cvs reports
that the sandbox is untouched, but it in fact a file is different.

My proposition was to change the definition of a static file to "will
not committed unless forced".

Olaf

_______________________________________________
cvsnt mailing list
cvsnt at cvsnt.org http://www.cvsnt.org/cgi-bin/mailman/listinfo/cvsnt



More information about the cvsnt mailing list