Possible concept leak between Monarch and GO Solr concepts (query for fields that don't exist)

[kltm]

While doing some other poking in the logs, I noticed this coming from what I assume is the monarch webapp/biolink getter:

137.53.238.95 - - [10/Jul/2018:14:50:19 -0700] "GET /solr/select/?facet.limit=25&fl=id%2Cassigned_by%2Csource%2Cbioentity%2Cbioentity_label%2Ctaxon%2Ctaxon_label%2Cqualifier%2Crelation_label%2Cannotation_class%2Cannotation_class_label%2Cobject_taxon%2Cobject_taxon_label%2Cevidence_object%2Cevidence_graph%2Cisa_partof_closure%2Creference%2Cis_redundant_for%2Ctype%2Cevidence%2Cevidence_label%2Cevidence_type%2Cevidence_type_label%2Cevidence_with%2Cevidence_closure%2Cevidence_closure_label%2Cevidence_subset_closure%2Cevidence_subset_closure_label%2Cevidence_type_closure%2Cevidence_type_closure_label&facet=on&fq=document_category%3A%22annotation%22&fq=bioentity%3A%22MGI%3AMGI%3A109522%22&facet.mincount=1&rows=200&wt=json&q=%2A%3A%2A HTTP/1.1" 200 3535 "-" "python-requests/2.19.1"

The issue here would be that I know that we do not have the following fields in GOlr:

• relation_label
• object_taxon
• object_taxon_label
• evidence_object
• evidence_graph
• is_redundant_for

While solr doesn't seem to have a problem at the moment ignoring fields, it does mean that there is a concept leak and that is one of these fields was used outside of fl there would be a problem.

[selewis]

Is this related to this issue? Discrepancies in results for HP and GO (same slim query) #184 where the conflicting fields are publication vs. reference