In the process of handling a type of web service request, I need to create a series of documents for indexing. They differ by only a couple of field values, but share quite a few others. I would like to clone SolrInputDocument and adjust a couple of fields, index that, lather, rinse, repeat.
However, org.apache.solr.common.SolrInputDocument (branch_3x) does not implement Cloneable, override clone() to make a deep-copy etc. Also observed by looking at the source code is the fact that SolrInputDocument keeps all fields in a LinkedHashMap, and also exposes a Map interface. So, does this sound like a workable idea? I define all my fields in a Map<String, SolrInputField> for the first document, and then tweak and re-use it. E.g.: Collection<SolrInputDocument> docs = new ArrayList<SolrInputDocument>(); Map<String, SolrInputField> fields = ...; //Set up fields for 1st document SolrInputDocument doc = new SolrInputDocument(); doc.putAll(fields); docs.add(doc); //Update values for fields (keys) a and b in fields Map. doc = new SolrInputDocument(); doc.putAll(fields); docs.add(doc); //Update values for fields (keys) a and b in fields Map. doc = new SolrInputDocument(); doc.putAll(fields); docs.add(doc); and so forth. Then: SolrServer solrServer = getSolrServer(); solrServer.add(docs); solrServer.commit(); Map.putAll "Copies all of the mappings from the specified map to this map", so each document will have its own copy of the fields. I will, or course, have to have map values of SolrInputField and instantiate those etc. Perhaps this is not worth the effort and I should be be satisfied repeating the same doc.addField() method calls. Thanks, Jeff -- Jeff Schmidt 535 Consulting j...@535consulting.com (650) 423-1068 http://www.535consulting.com
