Skip to content

Latest commit

 

History

History
50 lines (32 loc) · 1.54 KB

NOTES.md

File metadata and controls

50 lines (32 loc) · 1.54 KB
Raw

Data API Notes during the Solr9 transition.

On schemas -

We have configuration data used in multiple places:

  • Configuring solr
  • Setting default fields in data api

we cannot add to schema as sent to solr:

Error CREATEing SolrCore 'xgenome_shard1_replica_n1': Unable to create core [xgenome_shard1_replica_n1] Caused by: Invalid field property: x-display-default

Possible solution. Master schema document from which we create our solr and other schemas for various tools. We can start with the solr one and simply augment with x-blah fields that are removed to generate the solr schema. We can also then maintain a single master set of types since we need to be managing the XML anyway.

Query flow

In the dataType router, we flow the request through the following middlewares:

  1. RQLQueryParser. Translate RQL to Solr vformat if req.queryType = 'rql'

  2. Optional SORLQueryParser, commented out. Used for debugging.

  3. DecorateQuery. Add access control clauses to query.

  4. Limiter.

    • Skipped if request call_method != query.
    • Maximum and default limits defined here.

  5. ccc

Request variables:

call_method:

  • query: set if post and content-type is application/x-www-form-urlencoded, or
  • schema: set if incoming request is /schema
  • get: if have url like /genome/83332.12
  • stream
  • post:

ExtractCustomFields

Only for query or stream requests

If we have a fl= in the query, req.fieldSelection set to that

Otherwise, we have a hard coded set of headers (fieldHeader) and fields (fieldSelection) defined for each type