Add new Rails/CreateTableWithTimestamps cop

[wata727]

Hi there,

The other day I made a mistake in forgetting to add timestamps when creating a new table :(
So, I want to prevent such simple mistakes by using RuboCop.

This cop checks the migration for which timestamps are not included when creating a new table. In many cases, timestamps are useful information and should be added.

WDYT?

---

Before submitting the PR make sure the following are checked:

• Wrote good commit messages.
• Commit message starts with [Fix #issue-number] (if the related issue exists).
• Used the same coding conventions as the rest of the project.
• Feature branch is up-to-date with master (if not - rebase it).
• Squashed related commits together.
• Added tests.
• Added an entry to the Changelog if the new code introduces user-observable changes. See changelog entry format.
• All tests(rake spec) are passing.
• The new code doesn't generate RuboCop offenses that are checked by rake internal_investigation.
• The PR relates to only one subject with a clear title
  and description in grammatically correct, complete sentences.
• Updated cop documentation with rake generate_cops_documentation (required only when you've added a new cop or changed the configuration/documentation of an existing cop).

[mikegee]

This seems like a good idea. Here are some questions to consider:

• Should this cop be restricted to db/migrations to avoid false positives?
• Should this cop allow manually declaring updated_at and created_at (because timestamps does not configure a default value, for instance)?
• Should manually declaring only one of created_at or updated_at satisfy this cop?

I'm ok with answering "no" to all of those, just asking ...

[pocke]

Maybe we can use node.parent instead of ancestors.first.

[pocke]

The last begin matcher is not correct. This pattern does not match empty blocks or blocks that have one statement only.
For example, this cop should add offenses for the following code, but it does not add:

create_table :users do |t|
end


create_table :users do |t|
  t.string :name
end

Because the ASTs for this cases don't have begin node.

$ ruby-parse test.rb
(begin
  (block
    (send nil :create_table
      (sym :users))
    (args
      (procarg0 :t)) nil)
  (block
    (send nil :create_table
      (sym :users))
    (args
      (procarg0 :t))
    (send
      (lvar :t) :string
      (sym :name))))

[pocke]

We can use def_node_search instead of child_nodes.any?.
For example:

https://github.com/bbatsov/rubocop/blob/848d34c4818ce7d1a5b1dbd778b49262af8d1182/lib/rubocop/cop/bundler/duplicated_gem.rb#L48

[Drenmi]

This use of a node matcher is a bit gratuitous. You can just use:

node.command?(:create)

on send nodes anywhere. 🙂

[wata727]

@mikegee Thanks for your questions.

Should this cop be restricted to db/migrations to avoid false positives?

Yes. That is a good idea. I will fix config/default.yml for restricting to db/migrates/*.rb

Should this cop allow manually declaring updated_at and created_at (because timestamps does not configure a default value, for instance)?

Yes. If they are declared manually, this cop should not add offenses. I will fix it.

Should manually declaring only one of created_at or updated_at satisfy this cop?

Yes. If a user explicitly declared only one of them, I think we can judge that a user has not forgotten to add timestamps. I will also fix this.