Hi Stephane, It must be that your documents are themselves in XML, right? extract-path normally grabs trees from the persisted document, and so the nodes extracted from an XML document will be XML.
I wonder whether you can add '/text()' to the end of your extract-path expressions in order to force them into something that can be serialized within JSON. That would erase the key names of course. An alternate approach would be to use bulk search (from a client API) and use an output transform to render results of each search result into JSON. (Possible, but I can see why that would not be an appealing solution). If your documents were JSON, I *think* you'd get the results you are expecting. Charles Greer Lead Engineer MarkLogic Corporation ________________________________ From: general-boun...@developer.marklogic.com [general-boun...@developer.marklogic.com] on behalf of stephane.va...@oecd.org [stephane.va...@oecd.org] Sent: Thursday, June 23, 2016 2:19 AM To: general@developer.marklogic.com Subject: [MarkLogic Dev General] format:json && extract-document-data Hi, I am trying to include some document data into my search results, using the following query options: <options xmlns="http://marklogic.com/appservices/search";> <extract-document-data selected="include"> <extract-path>/language-version/ language-version-canonical-model/title</extract-path> <extract-path>/language-version/ language-version-canonical-model/language</extract-path> (…) </extract-document-data> </options> Unfortunately, when I ask for json format (using header Accpet: application/json), the extracted element comes as “stringyfied” xml instead of being converted into json as I would have expected: { "snippet-format": "snippet", "total": 564, "start": 1, "page-length": 10, "selected": "include", "results": [ { "index": 1, "uri": "ENV/CHEM/NANO(2015)22/ANN5/2", "path": "fn:doc(\"ENV/CHEM/NANO(2015)22/ANN5/2\")", (…) "extracted": { "kind": "element", "content": [ "<language>En</language>", "<title>ZINC OXIDE DOSSIERANNEX 5</title>", "<reference>ENV/CHEM/NANO(2015)22/ANN5</reference>", "<classification>2</classification>", "<modificationDate>2015-04-16T00:00:00.000+02:00</modificationDate>", "<subject label_en=\"media\">media</subject>", "<subject label_en=\"fish\">fish</subject>", (…) ] } }, Anything I am doing wrong? Is there some configuration options I could tweak to enforce the conversion of xml to json? Cheers, Stéphane Varin
_______________________________________________ General mailing list General@developer.marklogic.com Manage your subscription at: http://developer.marklogic.com/mailman/listinfo/general
