Hi!

> Very thanks for your answer
You're welcome.

>> Doing "find" with large trees is not what I wanted ;-)
>>
> OK, but what in case when I want commit only part files? For example I
> work with new 200 files and I want commit only 150 files, because rest
> of files are not ready to commit ?
Of course you can just say
   find . -iname "*.txt" | xargs fsvs ci -m"..."

I just wanted to remark that FSVS per default takes the directory tree as-is - 
it's not
necessary to say that files were added or removed, like in svn.

>> Hmmm ... I'll take a look. Why do you have this property removed? Did you do
>> that manually?
>>
>>
> No, I did not.
>
> This problems showed when 2 persons commit.
>
> I must describe this case:
>
> For example: We have 2 machine A and B. And machine A commit revision
> 1234, next machine B commit revision 1235, next machine A commit 1236
> without fsvs up. Machine A want commit 1235.When I do 'fsvs up' does not
> update commit 1235. How can I do this? Must I do "fsvs up" before each
> commit?
Sadly, yes.

Due to the multi-URL capability FSVS can't handle mixed working copies yet - 
and that
means that you have to make sure that the whole WC is at the same revision.
That means doing "fsvs up" before a commit, or only using a single WC for 
commit.

The culprit is that FSVS has to store a list of (URL, rev) pairs for each entry 
in the
WC, in order to correctly update in the mixed scenario - and that's simply not 
done yet.


>> This is clearly an internal error when doing the tree walking.
>>
>> You have removed files like usr/local/include/atalk/compat.h ; does
>> usr/local/include/atalk still exist? Does usr/local/include/ include exist?
>>
>>
> I moved files from /usr/local/include/atalk to
> /usr/local/netatalk/include/atalk
Ok, I'll take a look.


Regards,

Phil

-- 
Versioning your /etc, /home or even your whole installation?
             Try fsvs (fsvs.tigris.org)!

------------------------------------------------------
http://fsvs.tigris.org/ds/viewMessage.do?dsForumId=3928&dsMessageId=1294479

To unsubscribe from this discussion, e-mail: 
[[email protected]].

Reply via email to