You need to override org.apache.solr.morphlines.solr.LoadSolrBuilder.LoadSolr.doProcess(Record). Now LoadSolrBuilder.LoadSolr.convert(Record) copies all record fields into SolrInputDocuments fields. SolrInputDocument.addChildDocument(SolrInputDocument) nests a doc.
On Fri, Sep 11, 2015 at 11:27 PM, Lewin Joy (TMS) <lewin_...@toyota.com> wrote: > Oh Yes. We are upgrading Cloudera to get solr 4.10 just to get this block > join feature. > But, how do I index a nested document to use for block join for this huge > a dataset? > I could not find anyway to sculpt the morphline file for this use case. > > Thank you for the reply, Mikhail > > -Lewin > > > -----Original Message----- > From: Mikhail Khludnev [mailto:mkhlud...@griddynamics.com] > Sent: Friday, September 11, 2015 2:13 PM > To: solr-user <solr-user@lucene.apache.org> > Subject: Re: Morphline for Indexing Nested Document Structure > > Hello Lewin, > > Block Join support is released in Solr 4.5. > > On Fri, Sep 11, 2015 at 9:05 PM, Lewin Joy (TMS) <lewin_...@toyota.com> > wrote: > > > Hi, > > > > I am having a huge data of about 600 Million documents. > > These documents are relational and I need to maintain the relation in > solr. > > > > So, I am Indexing them as nested documents. It has nested documents > > within nested documents. > > Now, my problem is how to index them. > > > > We are on Cloudera Solr 4.4 and using mapreduce Indexer. > > Can we specify this nested structure in the morphline file? For the > > mapreduce or spark-submit, I need this handled through morphline. > > > > If this can't be done, is there an alternative that I can try? > > > > Thanks, > > Lewin > > > > > > -- > Sincerely yours > Mikhail Khludnev > Principal Engineer, > Grid Dynamics > > <http://www.griddynamics.com> > <mkhlud...@griddynamics.com> > -- Sincerely yours Mikhail Khludnev Principal Engineer, Grid Dynamics <http://www.griddynamics.com> <mkhlud...@griddynamics.com>
