Sergey A. Glukhov

I have fixed the pull request - the last commit was about this issue, but didn't either caused or fixed it

@shatalov-boris I would go with a format listed in official docs as an example: https://lucene.apache.org/solr/guide/6_6/updating-parts-of-documents.html#UpdatingPartsofDocuments-Example e.g. `Post.atomic_update(post1.id => { tags: { set: ['tag 1'] }, title: 'A new title' })`

UPD: That relates to atomic updates only

submitted a ticket to rsolr team https://github.com/rsolr/rsolr/issues/200

@heaven seem like this one got stuck?

I would make this configurable. I can imagine tasks when you want to index objects without persisting them (In my case I was indexing objects which aren't ActiveRecord siblings)

a lot of type changes here, hard to review. The question is - is this really needed? Perhaps importing old types in the new managed schema is better option?

I like the docker setup btw 👍 perhaps its worth making a separate PR for it

@l8nite any chance this can have specs?

is there away to have a spec for this? it's kind of hard to understand the changes made here