add opensearch rss results to dht collection (due to text = snippet)

which is used to differentiate meta from full data
- make sure check for dht is not dependant on number of collection entries

source/net/yacy/cora/federate/opensearch/OpenSearchConnector.java

@@ -113,10 +113,10 @@ public class OpenSearchConnector extends AbstractFederateSearchConnector impleme
                                 }
                                 doc.setField(CollectionSchema.publisher_t.getSolrFieldName(), item.getCopyright());
 
-                                // TODO: we likely got only a search related snippet (take is as text content)
-                                // we need a way to differentiate metadata from full crawl data in the index (would be also good for rwi transferred/received metadata)
-                                // or considere to add this to snippet cache, without adding text_t
                                 doc.setField(CollectionSchema.text_t.getSolrFieldName(), item.getDescriptions());
+                                // we likely got only a search related snippet (take is as text content)
+                                // add collection "dht" which is used to differentiate metadata from full crawl data in the index
+                                doc.setField(CollectionSchema.collection_sxt.getSolrFieldName(), "dht");
 
                                 if (item.getLat() != 0.0 && item.getLon() != 0.0) {
                                     doc.setField(CollectionSchema.coordinate_p.getSolrFieldName(), item.getLat() + "," + item.getLon());

source/net/yacy/search/index/Fulltext.java

@@ -352,9 +352,9 @@ public final class Fulltext {
                         putDocument(getDefaultConfiguration().metadata2solr(entry));
                     } else {
                         Collection<Object> collections = doc.getFieldValues(CollectionSchema.collection_sxt.getSolrFieldName());
-                        for (Object s: collections) {
-                            if (!"dht".equals(s)) return;
-                        }
+                        // collection dht is used to identify metadata from full crawled documents (if "dht" exists don't overwrite rich crawldata with metadata
+                        if (!collections.contains("dht")) return;
+                        
                         // passed all checks, overwrite document
                         putDocument(getDefaultConfiguration().metadata2solr(entry));
                     }