Message-ID: <419B9236.2010503@arbash-meinel.com>
Date: Wed, 17 Nov 2004 12:02:30 -0600
From: John A Meinel <john@arbash-meinel.com>
User-Agent: Mozilla Thunderbird 0.9 (Windows/20041103)
MIME-Version: 1.0
To: Mark Stosberg <mark@summersault.com>
Subject: Re: [Gnu-arch-users] Re: darcs vs tla
References: <22800.194.7.240.10.1100508579.squirrel@194.7.240.10>	<419B87EC.8060402@arbash-meinel.com>
	<slrncpn44k.68v.mark@localhost.localdomain>
In-Reply-To: <slrncpn44k.68v.mark@localhost.localdomain>
Cc: gnu-arch-users@gnu.org
Precedence: list
Content-Type: multipart/mixed; boundary="===============1453547160=="
Sender: gnu-arch-users-bounces+migo=homemail.com@gnu.org
Errors-To: gnu-arch-users-bounces+migo=homemail.com@gnu.org
Status: RO
Content-Length: 2808
Lines: 88
This is an OpenPGP/MIME signed message (RFC 2440 and 3156)
--===============1453547160==
Content-Type: multipart/signed; micalg=pgp-sha1;
	protocol="application/pgp-signature";
	boundary="------------enig66BAC7F612B51C64C3EFC0BE"

This is an OpenPGP/MIME signed message (RFC 2440 and 3156)
--------------enig66BAC7F612B51C64C3EFC0BE
Content-Type: text/plain; charset=ISO-8859-1; format=flowed
Content-Transfer-Encoding: 7bit

Mark Stosberg wrote:
> On 2004-11-17, John A Meinel <john@arbash-meinel.com> wrote:
> 
>>I just had a weird experience with the "portability" of darcs. I went to 
>>check out the Carpet repository (just to play around with darcs)
>>http://www.carpetcode.org/get-carpet-darcs.html
>>
>>I then added a line to the readme, and used "darcs diff". It started 
>>scrolling tons of changes by.
>>
>>I think it decided that the line endings were different.
> 
> 
> Sounds likely. 
> 
> I believe some are already working on enhanced line-ending handling for
> darcs. 
> 
>     Mark

I did some web searches and found some comments on "what should we do". 
That the repository should always store LF, and the working dir could be 
CRLF type of thing.

tla punts the decision by saying files are binary, and if you made it as 
CRLF, then it is stored that way, if you create it as LF, it is stored 
that way.

I prefer it because CVS has screwed up lots of files that I've used by 
not doing this. (Checking a CRLF under linux, check it out under 
windows, big mess. Check in a binary file and forget to set the binary 
bit, corrupted).

I would prefer darcs to do this, but I think a lot of people thing the 
CVS way is better.

Whatever, so long as just checking out a tree doesn't cause every file 
to be marked as changed. I find it odd that this exists in the 1.0 
release. Especially because darcs makes a big note about it being 
cross-platform compatible.

John
=:->

--------------enig66BAC7F612B51C64C3EFC0BE
Content-Type: application/pgp-signature; name="signature.asc"
Content-Description: OpenPGP digital signature
Content-Disposition: attachment; filename="signature.asc"

-----BEGIN PGP SIGNATURE-----
Version: GnuPG v1.2.4 (Cygwin)
Comment: Using GnuPG with Thunderbird - http://enigmail.mozdev.org

iD8DBQFBm5I2JdeBCYSNAAMRAmJEAJ9vJ4mNbzMGJCnSQuW9deGfHXrtqwCgn7da
ZE4+1bSGb1uddHFWPXcrcOE=
=SQsh
-----END PGP SIGNATURE-----

--------------enig66BAC7F612B51C64C3EFC0BE--


--===============1453547160==
Content-Type: text/plain; charset="us-ascii"
MIME-Version: 1.0
Content-Transfer-Encoding: 7bit
Content-Disposition: inline

_______________________________________________
Gnu-arch-users mailing list
Gnu-arch-users@gnu.org
http://lists.gnu.org/mailman/listinfo/gnu-arch-users

GNU arch home page:
http://savannah.gnu.org/projects/gnu-arch/
--===============1453547160==--