Strange ActiveRecord behaviour when using merge and joins

[iwiznia]

There's a strange behaviour on AR in which using merge on a scope varies the order of the joins depending on how you define the join (string vs symbol).
Here's the testcase:
https://gist.github.com/iwiznia/7547713

I know that in the test there's a table that doesn't exist, it's there only to check the order of the join clauses.
There are 3 tests which should yield the same result, but when you combine joins as strings and joins as symbols or classes, it mangles the order of the joins.
Maybe it has to do with the fact that it's using the join_sources of the Arel::SelectManager class, but I thought that was a valid use case, and doing the same thing, but calling to_sql first solves the error.

What do you think? Is this a bug? I think it is since the three join statements should produce the same sql.

[take]

Hi, I looked into the test code and,

but when you combine joins as strings and joins as symbols or classes, it mangles the order of the joins.

since the following test case passes(which you wrote), isn't the combination which fails is only joins as string, and joins as classes ?

  def test_joins_on_correct_order_works
     assert_equal Post.joins(:comments).merge(Comment.fails).to_sql, 'SELECT "posts".* FROM "posts" INNER JOIN "comments" ON "comments"."post_id" = "posts"."id" INNER JOIN "other_table"'
  end

[iwiznia]

I don't understand your question...There's only one failing test case, the others are there to help pinpoint the problem, since they should be doing outputing the same as the failing test

[rafaelfranca]

This issue has been automatically marked as stale because it has not been commented on for at least
three months.

The resources of the Rails team are limited, and so we are asking for your help.

If you can still reproduce this error on the 4-1-stable, 4-0-stable branches or on master,
please reply with all of the information you have about it in order to keep the issue open.

Thank you for all your contributions.

[rails-bot]

This issue has been automatically closed because of inactivity.

If you can still reproduce this error on the 4-1-stable, 4-0-stable branches or on master,
please reply with all of the information you have about it in order to keep the issue open.

Thank you for all your contributions.

[magikid]

I'm not sure what versions @iwiznia was using but when I run this using ruby 2.3 and rails 4.2.6, all of the tests pass.

[tibbon]

We're running into this problem ourselves. Curious to see if there's any new thoughts on it.