Stefan schrieb am Samstag, 12. Dezember 2020 um 11:00:01 UTC+1:

> On Saturday, December 12, 2020 at 10:02:37 AM UTC+1 Tobias Knauss wrote:
>
>> > you don't have to lock that file:
>> > the file is at a new location locally, so all other users don't have 
>> that file (yet). Only once you commit the merge and the moved folder/file, 
>> *then* you can lock the file because then it has a 'connection' to the 
>> repository.
>> Yes, the added file is not yet in the repo, so of course it cannot be 
>> locked. But it is autotically maselected in the lock dialog, which should 
>> not be the case.
>>
>
> the lock dialog does not contact the repository, so it's not possible for 
> the lock dialog to know that the file isn't present (yet) in the repository.
>  
>
Hm, yes, makes sense.


>> > I'm working on a workaround the svn lib problem here when doing a 
>> diff-against-base. However the status will still be reported as normal.
>> > And "added" would be wrong too: the file in relative to it's parent 
>> folder wasn't added at all but stayed the same.
>> "Added" would not be wrong, because the file *was added* in the new 
>> parent folder.
>> If it had *not* been "added", then the consequence must be that there was 
>> no *deleted* file either. But there *is* a delete file, so there also must 
>> be an *added* file.
>>
>
> but since the move information is gone, how would I know that the added 
> and the deleted item were the same before? They could be completely 
> unrelated.
>  
>
I didn't mean that it was the fault of TSVN. It's the SVN lib again.


>> I am thinking about creating a bug report for the SVN lib, but I don't 
>> have any insight to the lib itself, since I don't use it.
>> > so basically there's nothing I can do here.
>> You can. I can contribute to a bug report, but I think you're the one who 
>> needs to create it. Please understand that it's not about offloading work 
>> to you, but if I was the one who starts a conversation with the developers 
>> of the SVN lib, then it basically is a ping-pong with 3 people, and I am in 
>> the middle, although I have the least participation in the whole process.
>> I will assist you in convincing the SVN lib developers if necessary.
>>
>
> good luck with that. You can search the svn dev mailing list for "move 
> handling" and the like and you'll find tons of discussions. Handling 
> renames is done the best way possible. Other implementations would lead to 
> massive performance issues on the server side.
>
> I wll have a look there. I think changing only the reported status should 
be possible; I can hardly imagine that it breaks the whole move handling.

Thank you for your support on this issue!
Best

-- 
You received this message because you are subscribed to the Google Groups 
"TortoiseSVN" group.
To unsubscribe from this group and stop receiving emails from it, send an email 
to [email protected].
To view this discussion on the web visit 
https://groups.google.com/d/msgid/tortoisesvn/947939cd-0a3d-408b-8ae1-38c89a333780n%40googlegroups.com.

Reply via email to