There’s https://issues.apache.org/jira/browse/SOLR-5147
<https://issues.apache.org/jira/browse/SOLR-5147> - but unfortunately it didn’t
make it to the upcoming 5.0 release. It will be in the 5.1 release though.
Using parent/child (block/join) is what I would consider a last resort choice
though, and flattening and working through the query-side use cases thoroughly
with a flattened document/field structure in mind is where I would recommend
starting. Sounds like you’ve already been considering this and determined it’s
not the right fit but I would encourage a serious reconsideration of this.
What use cases are not possible or prohibitive for some reason using a
flattened structure?
Erik
> On Feb 9, 2015, at 7:03 AM, Jens Mayer <mjen...@yahoo.com.INVALID> wrote:
>
> Hey,
> thanks for your answer.Yes I've already checked out the XPathEntityProcessor
> but it's flatten my documents.This is quite good working for the first part
> of my example xml-file, but now i´m looking for a solution of the "timeinfo"
> part.
> I like to keep the nested structure up. So I search for a solution that if I
> do a solr search I'll get a nested structure as result which represent my xml
> structure from the beginning.
>
> Jan Høydahl <jan....@cominvent.com> schrieb am 12:02 Montag, 9.Februar
> 2015:
>
>
> Hi
>
> The most common approach is to flatten the document and map to chosen schema
> fields in Solr.
> You could also store the original XML in a STORED field for the purpose of
> returning to the client.
> Have you checked out XPathEntityProcessor
> (https://cwiki.apache.org/confluence/display/solr/Uploading+Structured+Data+Store+Data+with+the+Data+Import+Handler)?
>
> --
> Jan Høydahl, search solution architect
> Cominvent AS - www.cominvent.com
>
>> 9. feb. 2015 kl. 11.05 skrev Jens Mayer <mjen...@yahoo.com.INVALID>:
>>
>> Hey @all,
>> I have a problem with nested documentsin solr.
>> I like to import avery big XML file which contains a nested structure:
>> <data>
>> <shop>
>> <stores>
>> <code>aa</code>
>> <zipCode>a</zipCode>
>> <city>a</city>
>> <district>a</district>
>> </stores>
>> <timeinfo>
>> <type>open</type>
>> <timeinfo>
>> <from>00:00</from>
>> <to>00:00</to>
>> <weekday>sat</weekday>
>> </timeinfo>
>> </timeinfo>
>> </shop>
>> </data>
>> I like to pass offthe whole nested structure if I search in solr for one of
>> thisdocuments.
>> My search revealedthat the DataImportHandler can't support the import of
>> nestedDocuments.
>> So my ask is, how Ican import nested documents and how I can portray the
>> structure inschema.xml?
>> Thanks in advanced.
>
>
