Using the searchservice in EWS

I'm currently working with the SearchService in the LiveLink EWS and have a question about the location of the nodes. When i iterate through my searchresults, the location of the nodes is specified like using the DataIDs from the DTree-table in the database like this: 2000 45043 12248587 etc etc

I want to avoid having to do multiple calls to the EWS in order to get a string-representation of the path to the node. Is there a way to get the full path without having to call getNode(...) of the DocumentManagementService?

My other question is hopefully easier to answer. When i create a SingleSearchRequest, i see that there's a property which specifies which is the first result to retrieve and a property which specifies how many results to retrieve. Is it necessary to specify these properties or are they only used to help optimize the usage of the EWS?

Find more posts tagged with

Comments

Greg_Griffiths

Andre,

Not OOTB, I've implemented an additional WebService for Search for a project recently as they wanted this and Hit Highlighting returned in the data. Get in touch and I can discuss this in more detail.

Andre_Sovde

I've sent you an e-mail. I've included a specification document regarding the service. The document will be sent to the companies using this solution in order to implement clients.

André

Kyle_Swidrovich

The DocumentManagement service exposes a method called "getNodes" which allows you to pass in a list of IDs in a single call and receive an array of Node objects from which you could retrieve the names you need.

public Node[] getNodes(int[] IDs)

Get node objects.

Parameters:: IDs - List of IDs of nodes to retrieve.
Returns:: Returns an array of nodes. If any node cannot be found, a null node will be in the corresponding position in the array. Throws fault exception with fault codes: DocMan.NodeRetrievalError.

If you convert the OTLocation value into an array of IDs, this could help here.
However, if all nodes returned reside in a similar area of Content Server, some optimization could be made here by deduping IDs for all results in the result page so that only unique IDs would be used in your getNodes call. This would result in less DB calls to retrieve information about the same nodes.

As for your second question, these parameters should always be provided, as it clearly outlines the page of results you are requesting.

Andre_Sovde

Quoted Kyle Swidrovich (kswidrov) on 2010/07/14 16:32:

Get node objects.

Parameters:: IDs - List of IDs of nodes to retrieve.
Returns:: Returns an array of nodes. If any node cannot be found, a null node will be in the corresponding position in the array. Throws fault exception with fault codes: DocMan.NodeRetrievalError.

Ahhh, of course!! I'd forgotten about that method. Thanks. i think that should do the trick:)