Hi Günther On Mac, I copy files in the Terminal in order to avoid the ._* metadata files. That worked fine for me with WebDAV and Jackrabbit/Sling so far.
Regarding your question: whether documents are indexed or not depends on your configuration, namely the "textFilterClasses" option of your search index[0]. You can find some more background on available TextExtractors etc. on the Jackrabbit website[1]. Regards Julian [0] http://wiki.apache.org/jackrabbit/Search [1] http://jackrabbit.apache.org/jackrabbit-text-extractors.html On Wed, May 9, 2012 at 6:49 PM, Günther Schmidt <[email protected]> wrote: > Hello, > > I've just managed to upload a local directory to Apache Sling via an > HttpClient I wrote in Java. For those who wonder why I haven't simply used > WebDav for this: I'm using a Mac and every time you copy/move files from > local filesystem to a mounted, non-mac one there are "._*" files added where > Apple will store indespensible, extremely important metadata about the file. > No remedy for this is known to man. > > So anyway, the files and folders are uploaded and I even managed that the > nt:files have the mix:versionable property set. And here's my question: The > jcr:content node which contains the actual file data does not have the > mix:versionable property set, just its parent, the nt:file node. Does this > mean that the actual content of the file is not versioned, do I need to set > it manually? > > Eventually I will study the JCR2.0 spec, just not yet. > > And are the files indexed via Jackrabbit's lucene too? > > Günther
