Hi!
I am currently working on a project that requires a large amount of
XML-Documents to be stored hierarchically. While the structure of the
repository is already designed and seems to work well (and it confirms
with David's suggestions), I am not that sure about how to store the
XML-Data itself.
My current approach is to have a single node per XML-Document. This node
contains a number of properties which contain Meta-Information regarding
this document and a String-Property containing the XML. This way I can
easily find the document by path, read and write the XML and even search
for meta-data... but I can not that easily search INSIDE the XML (or can
I?). The XML-Data is something like docbook and contains a lot of text
that needs to be searchable.
Do I need to create another sub-node and _import_ the XML, so it is
unfolded inside the JCR in order to be able to search inside the XML? In
this case I would exclude this sub-node (and all descendants) from
versioning and only use the import-structure for searching, because the
XML-String-Property contains the master-XML. Does this make sense?
Any hints would be great!
Cheers,
Dan
