hi, @all: would be great if some jackrabbit users could help us to understand how things are supposed to work with jackrabbit.
@lars: for the question about userland cascading the versioning. i did not hear of anybody doing that. it would be sad if we can't use the content repository, but you might want to ask on the symfony-cmf-devs list about this. cheers,david Am 05.03.2013 16:49, schrieb Lars Janssen: > Hi David, > > Thanks for the link. From what I can gather, using just 'version' or > 'copy' in the CND format implies OnParventVersion=version or copy. I > *think* I want copy, as it appears to fully cascade the version rather > than just creating a reference to the parent history. > > However, I'm not sure if I should apply it to specific properties, or to > children. For example, I'm now using: > > [phpcr:versionCascade] > nt:unstructured > - foo_c (String) copy > + * copy > > After running the code this PR (but with the above node definition): > https://github.com/phpcr/phpcr-api-tests/pull/85 > > Here is how my repository looks: > https://gist.github.com/fazy/5091056 > > There is evidence of versions of the "foo_c" parameter being captured; > if you search the Gist for "foo_c" there are different values at > different version stages. However, I'm not sure how to get to these > values. I have some test code that just > calls getVersionHistory('/tests_version_base/versioned/version_child') > and then loops through the results... unfortunately none of them > contains a 'foo_c' parameter at all. > > I'm wondering if anyone has had any experience of using cascading > versions at all (especially if it's an open source project where I can > see the code). > > Otherwise, as a temporary solution I might need to move the cascade > logic up into my application. i.e. when versioning a page node, iterate > through all child nodes and version those too. It means potentially > duplicating what's already possible in Jackrabbit, but on the other hand > if my application needs very specific rules (e.g. how to handle a > 'reference block' if I add that in future), then it might make sense to > control it there. Again, has anyone tried doing this? -- Liip AG // Agile Web Development // T +41 26 422 25 11 CH-1700 Fribourg // PGP 0xA581808B // www.liip.ch
