[
https://issues.apache.org/jira/browse/SOLR-16079?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=17585741#comment-17585741
]
Jan Høydahl commented on SOLR-16079:
------------------------------------
So we'll have three clients to choose from:
* HttpSolrClient (apache http deps)
* Http2SolrClient (jetty deps)
* CloudSolrClient (zk deps)
Now, should solrj-core (or whatever we call it) depend on any of these three,
thus pulling in lots those extra jars, or should SolrCore be slim and require
users to add another solrj-jetty or solrj-apachehttp to pull in either jetty or
apache dependencies? Ideally I'd love to see a slim core with a simple
SolrClient based on JDK11 http client that could solve 80% of the use cases...
Lacking that JDK client we could of course choose to pull in jetty dependencies
in "core"?
> New solrj-apachehttp module
> ---------------------------
>
> Key: SOLR-16079
> URL: https://issues.apache.org/jira/browse/SOLR-16079
> Project: Solr
> Issue Type: Sub-task
> Components: SolrJ
> Reporter: Jan Høydahl
> Priority: Major
>
> As the HttpSolrClient based on apache http client will be deprecated in 9.0,
> then the new {{solrj-core}} module does not need to add those in its
> build.gradle. Instead we can make an "empty" module that pulls in those
> dependencies, so that users wanting to use the old client must add that
> dependency. Then in 10.0 we'll remove both the old clients and this jar.
--
This message was sent by Atlassian Jira
(v8.20.10#820010)
---------------------------------------------------------------------
To unsubscribe, e-mail: issues-unsubscr...@solr.apache.org
For additional commands, e-mail: issues-h...@solr.apache.org
