Simon Large wrote:
> 2009/2/16 Stefan Küng <tortoisesvn_at_gmail.com>:
>> Tree conflicts:
>> Subversion 1.6 marks tree conflicts. In pre 1.6 clients, such conflicts
>> were simply ignored (shown as 'skipped' in the progress dialog). Now,
>> such conflicts are properly marked as conflicted until the user marks
>> those as resolved.
>> But: Subversion doesn't give much information on how to resolve those
>> conflicts. Resolving is completely up to the user. This will improve in
>> 1.7 together with the sqlite based working copy rewrite, but until then
>> there's not much we can do to help users resolving those conflicts. I've
>> tried to add some info to our docs, but those need a lot more details
>> and maybe step-by-step guides. Simon, could you please have a look at
>> those docs?
>> Here's a description on how the svn devs *want* it to work:
>> https://svn.collab.net/repos/svn/trunk/notes/tree-conflicts/use-cases-resolution.txt
>> But of course, that's not how it works in 1.6 because there's not enough
>> information ready to do all this.
>
> When you say it doesn't all work yet in 1.6 I assume you mean the
> resolution. 1.6 can detect all these tree conflicts OK?
Yes, it detects the conflicts. But it doesn't help much in resolving. So
the resolving is purely manual.
>> Mark Phippard and Steve Elsemore (working on Subclipse) are working on a
>> document which describes several tree conflict situations and how
>> Subclipse tries to help users to resolve them:
>> http://docs.google.com/Doc?id=dd885dbs_30c8x6bvcn
>
> That doc looks broken to me. It is not at all clear that use cases 4,
> 5 and 6 are referring to separate branches. I have clarified our docs
> to show this and will notify Mark.
>
>> But note that TSVN doesn't have enough information ready: Subclipse
>> knows more about a working copy since for example it 'knows' the working
>> copy root (which TSVN doesn't). Also, Subclipse does a lot of guessing,
>> which IMHO is bad, especially if guessing results in < 50% accuracy:
>> that will confuse users even more. That's why TSVN does no guessing and
>> rather leaves the task up to the users.
>> Simon, if you want to test different tree conflict scenarios, you can
>> use the test scripts in /trunk/test/treeconflics to create them.
>
> Out of interest, those scripts all start an instance of svnserve.
> Wouldn't it be a lot easier to use file:// protocol for those
> scripts?
I just used my template script, which had the protocol set to svn://.
But as you can see in those scripts, it's not difficult to change it to
file:// - just comment the svn:// parts an uncomment the file:// parts.
Stefan
--
___
oo // \\ "De Chelonian Mobile"
(_,\/ \_/ \ TortoiseSVN
\ \_/_\_/> The coolest Interface to (Sub)Version Control
/_/ \_\ http://tortoisesvn.net
------------------------------------------------------
http://tortoisesvn.tigris.org/ds/viewMessage.do?dsForumId=757&dsMessageId=1216359
To unsubscribe from this discussion, e-mail: [dev-unsubscribe_at_tortoisesvn.tigris.org].
Received on 2009-02-23 21:34:03 CET