Fix a redundant crutch call when :update_fields does not contain related field

[skcc321]

what is the issue?

this one https://github.com/toptal/chewy/issues/864

I have an index definition with some crutches inside like this:

class UsersIndex < Chewy::Index
  index_scope User....

  crutch :driver_role do |collection|
      ::User.includes(employees: :group)
        .where(id: collection.map(&:id))
        .group(:id, "groups.kind", "groups.role")
        .pluck(
          :id,
          Arel.sql(
            "jsonb_build_object(
            'driver_role', CASE
                             WHEN (groups.kind = 'carrier' AND groups.role ='ca_driver') THEN TRUE
                             ELSE FALSE
                           END
            )"
          )
        ).to_h
  end

  field :first_name
  field :last_name
  field :driver_role, type: :boolean, value: ->(user, crutch) { crutch.driver_role.dig(user.id, "driver_role") }
  ...
end

when I call

UsersIndex.reset [1], update_fields: [:first_name, :last_name]

it always triggers :driver_role crutch execution even if I don't want to reindex that one by skipping the related field using :update_fields looks like the reason is this line of code https://github.com/toptal/chewy/blob/master/lib/chewy/index/import/bulk_builder.rb#L52 where we don't pass @fields option.

---

• [x] The PR relates to only one subject with a clear title and description in grammatically correct, complete sentences.
• [x] Wrote good commit messages.
• [x] Commit message starts with [Fix #issue-number] (if the related issue exists).
• [x] Feature branch is up-to-date with master (if not - rebase it).
• [x] Squashed related commits together.
• [x] Added tests.
• [x] Added an entry to the changelog if the new code introduces user-observable changes. See changelog entry format for details.

[skcc321]

@konalegi I'm not sure you are the right person (I see you contributed recently as a Toptal org member). Could you please start the CI, please?

[konalegi]

Thanks for the improvement 🚀 Started CI

[skcc321]

Thanks for the improvement 🚀 Started CI

One more time, please. just realized that parent existence can cause extra calls as well after the fix. so I cache data_for result in case of parent existence now.

[skcc321]

@konalegi any objections to getting it merged ( & released)?

[konalegi]

Sorry, I haven't time to review it properly, I'll do it in 1-2 days and most likely release it. One thing that I see, please provide a changelog entry, so that will be easier to generate a release (example)

[konalegi]

@skcc321 sorry for the delay, the person who knows that part very well is now on vacation, so we have to wait until next week to review and release.

[konalegi]

@skcc321 released