xml metadata

[gokceng]

hi, I'was looking at the xml extractors. I need something different. My nodes are not type of xml. My nodes have some custom properties and some of them are xml.  For example,

var dummy = <rootNode>
<node1>value1</node1>
<node2>value2</node2>
<node3>value3</node3>
</rootNode>

document.properties["my:dummy"] = dummy;

Let's say I have 1000 nodes that have that structure. Is there a way to find if there is a node that has <nodeX>valueX</nodeX> ?
Thanks in advance..

[mrogers]

Either you extract the xml values and put them into some sort of queryable structure that you can search against or you will need to use full text search for your text data.

[gokceng]

thanks for your reply. can you give an example for "queryable structure" or "using full text search"? I mean how can I get a queryable strcuture or make Alfresco usable for full text search.

[invictus9]

thanks for your reply. can you give an example for "queryable structure" or "using full text search"? I mean how can I get a queryable strcuture or make Alfresco usable for full text search.

Full text search refers to the thing that Google does – words are indexed, in context, and you can construct queries on those words. The way full text search works in Alfresco removes the XML context, though, much like HTML is stripped off web pages on Google searches.

Queryable structure implies a structure that the Lucene search engine knows about and can build indexes for, and can be efficiently searched. The content type and its metadata form the queryable structure. So, if your document has a set of metadata properties, node1, node2, etc, you can push the document content out to the document properties. They will then be indexed and made searchable. In effect, you are telling the search engine that "value1" is a significant piece of data, but only in the context of "node1" meaning.

[gokceng]

thank you for your replies invictus9 and mrogers. I will look at the points you mentioned.