dsmiley commented on PR #3163: URL: https://github.com/apache/solr/pull/3163#issuecomment-2701775120
> So what is the point of changing from a LinkedHashMap to LinkedHashMap in a SMO? A LinkedHashMap doesn't implement NamedList. An SMO does. Thus a Solr 10 server could start passing maps and MapWriter things (e.g. SolrParams) to the response in situations that a SolrJ 9 is still anticipating reading a NamedList. With SOM; it's both things in one. The choice of LinkedHashMap vs ArrayList would be a regression for cases where the client code already expected a Map; was assuming O(1) performance of the map if lots of stuff is in it. Because NamedList/Map is nestable, at the time of unmarshalling we have no idea what the actual code looking at the object is casting it as. In a fantasy world if there was just exactly one thing to unmarshall, then there would be maybe readMap vs readNamedList and we wouldn't have concerns that this PR tries to solve. That said, let's not actually change SMO to be based on a LinkedHashMap; at least not soon. I proposed only reading maps as SMO for a more narrow circumstance (9.x talking to 10.x) that would be somewhat temporary / limited, thus ameliorating a performance risk. -- This is an automated message from the Apache Git Service. To respond to the message, please log on to GitHub and use the URL above to go to the specific comment. To unsubscribe, e-mail: issues-unsubscr...@solr.apache.org For queries about this service, please contact Infrastructure at: us...@infra.apache.org --------------------------------------------------------------------- To unsubscribe, e-mail: issues-unsubscr...@solr.apache.org For additional commands, e-mail: issues-h...@solr.apache.org
