Follow-up from "Backport of Multiple Assignees feature"

The following discussions from !11089 (merged) should be addressed:

  • @smcgivern started a discussion:

    We should split this and assignee_id by issue / MR, so that we don't accept assignee_ids for MRs.

  • @smcgivern started a discussion:

    case selected_users.length

  • @smcgivern started a discussion:

    Same as above, we should only permit by issue / MR.

  • @smcgivern started a discussion:

    This should probably say 'removed assignee' in CE, but it's really not very important.

  • @smcgivern started a discussion:

    - issue.assignees.take(max).each do |assignee|
      # ...
  • @smcgivern started a discussion:

    We should probably move these blocks to their own _sidebar/_assignee partial

  • @smcgivern started a discussion:

    This check if it's an MR is redundant now 😉

  • @smcgivern started a discussion:

    We could do these in a single hash literal, then merge with options[:data].

  • @smcgivern started a discussion: (+1 comment)

    Is this an array or an AR relation here? If it's a relation, we will perform an extra query for no reason. Try issue.assignees.to_a.any?.

  • @smcgivern started a discussion:

    This should also be split into two partials.

  • @smcgivern started a discussion:

    We should make this clearer for CE, because in CE there can only be zero or one assignee.

  • @smcgivern started a discussion:

    a user -> the users?

  • @smcgivern started a discussion:

    We could also use params.tap here?

  • @smcgivern started a discussion: (+1 comment)

    Nice!

  • @smcgivern started a discussion:

    user -> user

Assignee Loading
Time tracking Loading