Ben:
For comparison, CVS's default _is_ to make backup copies of the original,
pre-merged file as is being mentioned as a possibility if, apparently, only
a --paranoid flag is used. So the cute Carl Sagan "users have been using CVS
for years" doesn't stack up, hehe; well, unless you're arguing FOR file
backups. Personally, I wish this were also the default case in Subversion,
it would make my development team happier about the recent switch as well.
The backup is of the working copy file that was there before the merge, and
is saved as ".#<filename>.<revision number working copy of file was at
before merge>".
[root(2)@ironclad cvs]# pwd
/tmp/cvs
[root(2)@ironclad cvs]# cvs -d /tmp/cvs/repos init
[root(2)@ironclad cvs]# mkdir /tmp/cvs/repos/module
[root(2)@ironclad cvs]# cvs -d /tmp/cvs/repos co -d work1 module
cvs checkout: Updating work1
[root(2)@ironclad cvs]# cvs -d /tmp/cvs/repos co -d work2 module
cvs checkout: Updating work2
[root(2)@ironclad cvs]# cd work1
[root(2)@ironclad work1]# echo -ne "1\n2\n3\n4\n5\n6\n7\n" > test
[root(2)@ironclad work1]# cvs add test; cvs commit -m ""
cvs add: scheduling file `test' for addition
cvs add: use 'cvs commit' to add this file permanently
cvs commit: Examining .
RCS file: /tmp/cvs/repos/module/test,v
done
Checking in test;
/tmp/cvs/repos/module/test,v <-- test
initial revision: 1.1
done
[root(2)@ironclad work1]# cd ../work2
[root(2)@ironclad work2]# cvs up
cvs update: Updating .
U test
[root(2)@ironclad work2]# echo -ne "8\n" >> test
[root(2)@ironclad work2]# cvs commit -m ""
cvs commit: Examining .
Checking in test;
/tmp/cvs/repos/module/test,v <-- test
new revision: 1.2; previous revision: 1.1
done
[root(2)@ironclad work2]# cd ../work1
[root(2)@ironclad work1]# echo -ne "1\n3\n4\n5\n6\n7\n" > test
[root(2)@ironclad work1]# cvs up
cvs update: Updating .
RCS file: /tmp/cvs/repos/module/test,v
retrieving revision 1.1
retrieving revision 1.2
Merging differences between 1.1 and 1.2 into test
M test
[root(2)@ironclad work1]# ls -la
total 20
drwxr-xr-x 3 root root 4096 Aug 18 12:22 .
drwxr-xr-x 5 root root 4096 Aug 18 12:20 ..
drwxr-xr-x 2 root root 4096 Aug 18 12:22 CVS
-rw-r--r-- 1 root root 14 Aug 18 12:22 test
-rw-r--r-- 1 root root 12 Aug 18 12:22 .#test.1.1
[root(2)@ironclad work1]# cat .#test.1.1
1
3
4
5
6
7
[root(2)@ironclad work1]#
Sincerely,
Jay Freeman (saurik)
saurik@saurik.com
----- Original Message -----
From: "Ben Collins-Sussman" <sussman@collab.net>
To: "Wadsworth, Eric (Contractor)" <wadswore@fhu.disa.mil>
Cc: <kfogel@collab.net>; <james-tigris@jrv.org>;
<users@subversion.tigris.org>; <dev@subversion.tigris.org>
Sent: Monday, August 18, 2003 10:40 AM
Subject: Re: Disabling automatic conflict resolution?
...
> Or put on your Carl Sagan voice: "*billions* of users have been using
> CVS for years, and have had no problems with this..."
>
> Seriously, your scenario reminds me of the kind of PHB reactions I
> sometimes see when trying to explain concurrent versioning -- "what?
> no locking? it will never work!". These days, I just point them to
> the beginning of chapter 2 in the svnbook.
---------------------------------------------------------------------
To unsubscribe, e-mail: dev-unsubscribe@subversion.tigris.org
For additional commands, e-mail: dev-help@subversion.tigris.org
Received on Mon Aug 18 19:42:02 2003