Turning on/of indexing by defining a new type

[torghal]

For some documents in our repository the content needs to be indexed for searching, for other documents this indexing should be turned of.  To do this we created a new document type:

<type name="cm:contentNoIndexing">
         <title>Content</title>
         <parent>cm:cmobject</parent>
         <archive>true</archive>
         <properties>
            <property name="cm:contentNoIndexing">
               <type>d:content</type>
               <mandatory>false</mandatory>
               <!— Index content in the background –>
               <index enabled="false">
                  <atomic>false</atomic>
                  <stored>false</stored>
                  <tokenised>false</tokenised>
               </index>
            </property>
         </properties>
      </type>‍‍‍‍‍‍‍‍‍‍‍‍‍‍‍‍‍

With the webservice client we can create documents of this type. To retrieve the documents we have 2 problems:

1) the documents of this type are not displayed in de web client.

2) when we try to retrieve the content of a document of this type by

content = getContentService().read(new Predicate(new Reference[]{new Reference(STORE, dmsId, null)}, STORE, null), PROP_CONTENT_NO_INDEXING);‍

and

ContentUtils.getContentAsInputStream(content[0]) ‍

we get:

The node's content is missing:
   node: workspace://SpacesStore/449bd2ae-2bf4-11dd-acee-fd7ebfbe7924
   reader: null 
Please contact your system administrator.

Can anybody solve these problems or is there a better way to handle the switching between content indexing and non-indexing.

Thx

[jbarmash]

your approach makes sense.    What do you mean that documents of this type are not displayed in the web client?  You mean you add a new document of that type, but then you can't see it?  That is definitely strange.  

I am not sure if there are issues with SOAP based web services, so I can't comment on that.

[torghal]

We added the SOAP interface to create documents of this type.  The documents are in the repository because we can retreive them with the soap interface.  The documents just don't show up in the web client.

[derek]

Hi,
The web client is looking for cm:content and cm:folder types and their descendents only - it is a file/folder browser.  You can extend your cm:contentNoIndexing from the cm:content type and override the cm:content property.  Then the web client will recognise it as being a file.
If you want to force indexing of content into the background, you can set the following in custom-repository.properties:

lucene.maxAtomicTransformationTime=0‍

[plecras]

Hi,
The web client is looking for cm:content and cm:folder types and their descendents only - it is a file/folder browser.  You can extend your cm:contentNoIndexing from the cm:content type and override the cm:content property.  Then the web client will recognise it as being a file.

Hi,

Can you tell me how to override the <index> property please ?

Because when I try to use the <overrides> I get an error.

Is there a particular tag for this operation ?

Thank you.