RE: svn commit: r1355070 - /subversion/trunk/notes/wc-ng/conflict-storage-2.0

> -----Original Message-----
> From: gstein_at_apache.org [mailto:gstein_at_apache.org]
> Sent: donderdag 28 juni 2012 18:41
> To: commits_at_subversion.apache.org
> Subject: svn commit: r1355070 - /subversion/trunk/notes/wc-ng/conflict-
> storage-2.0
>
> Author: gstein
> Date: Thu Jun 28 16:40:30 2012
> New Revision: 1355070
>
> URL: http://svn.apache.org/viewvc?rev=1355070&view=rev
> Log:
> * notes/wc-ng/conflict-storage-2.0: add some comments/questions

Thanks for reviewing!

> Modified:
> subversion/trunk/notes/wc-ng/conflict-storage-2.0
>
> Modified: subversion/trunk/notes/wc-ng/conflict-storage-2.0
> URL: http://svn.apache.org/viewvc/subversion/trunk/notes/wc-ng/conflict-
> storage-2.0?rev=1355070&r1=1355069&r2=1355070&view=diff
> ==========================================================
> ====================
> --- subversion/trunk/notes/wc-ng/conflict-storage-2.0 (original)
> +++ subversion/trunk/notes/wc-ng/conflict-storage-2.0 Thu Jun 28 16:40:30
> 2012
> @@ -159,6 +159,13 @@ independently once this is implemented.
> The --accept options already provide a lot of additional value and the
> WHY model would allow access to the specific sets by URL.
>
> +### gs: there are FOUR values of interest w.r.t property conflicts.
> +### and each of these may be missing to designate the property
> +### does not exist. see 'conflict-storage'.

I followed the definition in svn_wc_conflict_description2_t where this extra value is missing.

Looking at my current test failures (where I switched to three table approach) I really need more information so thanks for confiming.

It appears that the merge code uses a 4-way diff, while the current conflict description used for resolving is modeled as a 3-way diff. So this will need extension too.

> +### gs: the property name needs to be stored. again, see 'conflict-storage'

Conflict-storage documents a nice to have future version of conflicts, including many infrastructure that doesn't exist yet. All the future ideas is why I started the new approach.

But I have a list of properties in my current implementation. See subversion/libsvn_wc/conflicts.h. I will update the property documentation after I get this code working. (Just a few test failures left. So that mostly implies that we need more tests)

> +
> +
> Tree Conflicts
> --------------
>
> @@ -176,3 +183,7 @@ where it applies to the WHY parts.
>
> It would be nice if some other --accept values would be accepted, but that
> is outside the scope of this design.
> +
> +### gs: is LOCAL-STATE defined the same as in 'conflict-storage' ?

Exactly as in the current tree conflict skel. I haven't started implementing this.

> +
> +### gs: what is the definition of INCOMING-ACTION ?

Same: Exactly as in the current tree conflict skel. I haven't started implementing this.

        Bert
Received on 2012-06-28 18:59:31 CEST