Hi, I've logged this as a spec bug.
You are right. The intended behavior is that the connected non-versionable (CNV) subtree of a versionable V should be updated or not according to the merge test result of V. In other words,V and its CNV subtree must be treated as a single unit for purposes of merge (where the CNV subtree is the subtree accessible from V without traversing any verisonable nodes). The current pseudo-code in the spec, and consequently the Jackrabbit implementation of merge doesn't make any sense. We will fix it in the spec, at which point it will become a Jackrabbit bug. Does someone want to file a bug against Jackrabbit? I guess the only consolation I can offer is that you guys win a prize for being among the first to seriously try out the merge functionality! Thanks for reporting the issue. I'm always glad when people read the spec :-) Cheers, Peeter On 7/27/07, Peeter Piegaze <[EMAIL PROTECTED]> wrote: > Hi Tim, Jose, > > Thanks for pointing this out. I'll take a look at the issue and get > back to you guys soon. > > Cheers, > Peeter > > On 7/18/07, Jose Alvarez <[EMAIL PROTECTED]> wrote: > > Tim Kuchlein <kuchlein <at> gmail.com> writes: > > > > > > > > Hello > > > > > > I am currently having the same issue with merging two workspaces and > > > I noticed that no one has responded to this issue. I can't see how > > > the algorithm presented in JSR-170 (8.2.10.1) or the newly released > > > JSR-283 could possibly implement the textual description (8.2.10). > > > Nowhere in the algorithm is the merge result of the nearest > > > versionable ancestor taken into consideration. Specifically the > > > version of doleave(n) that dofail(n, v') calls needs to be altered to > > > read: > > > > > > doLeave(n) > > > for each versionable child node c of n domerge(c). > > > > > > rather than what is currently has: > > > > > > doLeave(n) > > > for each child node c of n domerge(c). > > > > > > thanks in advance > > > tim > > > > > > > Hi, > > > > I have found the same issue in my current project. > > This issue make me think that I don't understand correctly the JCR > > specification about how to merge nodes. > > > > Can someone explain that? > > > > Thanks in advance > > > > > > > > > > > -- > Peeter Piegaze > > Day Software > Suite 331 > 67 Mowat Avenue > Toronto Ontario M6K 3E3 > Canada > > office +1 416 987 5720 > mobile +1 647 205 2403 > fax +1 866 719 3988 > -- Peeter Piegaze Day Software Suite 331 67 Mowat Avenue Toronto Ontario M6K 3E3 Canada office +1 416 987 5720 mobile +1 647 205 2403 fax +1 866 719 3988
