Fix Rails/HttpPositionalArguments offenses

[JoeCohen]

This branch fixes all the Rails/HttpPositionalArguments offenses, including these auxiliary changes:

• Removes mokey-patches that expedited those offenses;
• Enables the Rails/HttpPositionalArguments cop

More details at MO Developers, ruby 2.7.6 and rails 6.1

There is no manual test -- all changes are in test, rather than app

[coveralls]

Coverage remained the same at 93.065% when pulling 9021408 on joe-fix-test-kwargs into 319d80d on master.

[nimmolo]

@JoeCohen If this removes the use of check_for_params and extract_body maybe those methods should be deleted too.

I believe that fixing the kwarg issue for 2.7 will require further changes to functional_test_case methods, since they do not send the args to Rails test_case the way it wants them. Waiting for advice from Jason about that. But that should be a separate PR.

[JoeCohen]

Thanks @nimmolo!
You're right about check_for_params and extract_body. I had it in mind to remove them (as part of removing the old monkey-patches), but then completely forgot to do that.
I'm interested in who you find out about the Ruby 2.7 kwargs warnings.

[JoeCohen]

I'll get to work on the Codeclimate/RuboCop offenses. I'm not sure why there are so many of them.

[nimmolo]

@JoeCohen I think it's that your line length in VS Code is not set to 80? I am happy to open and save the file, that will fix it.

Also -- I believe I solved the kwargs issue once and for all. Pushed it to this branch.

Also -- fixed a rubocop "Prefer index_with to each_with_object in NameControllerTest just now.

[nimmolo]

Hi @JoeCohen -
I was mistaken, Rubocop doesn't do line length.
It's another setting: hit command-, for settings, then search for line length. You'll find it under HTML>Format: Wrap Line Length

Here are the configs to set up rubocop with VS Code, in case you need them

.

[nimmolo]

There's one more kwarg thing to fix, in session_extensions.rb:104. Commit coming soon.

[JoeCohen]

Thank you again! - HTML>Format: Wrap Line Length was set to 120. - I can use the Ruby > Rubocop settings. But I will leave that for another day. - The latest version of RuboCop does some line length autocorrection. We're on an old version. I'll update us after I get done with this PR.

…

On Sat, May 28, 2022 at 4:27 PM andrew nimmo ***@***.***> wrote: Hi @JoeCohen <https://github.com/JoeCohen> - I was mistaken, Rubocop doesn't do line length. It's another setting: hit command-, for settings, then search for line length. You'll find it under HTML>Format: Wrap Line Length Here are the configs to set up rubocop with VS Code, in case you need them [image: Screen Shot 2022-05-28 at 4 15 49 PM] <https://user-images.githubusercontent.com/1948095/170845823-5fa9a999-923e-4979-a6f5-f0f17c6b07d9.png> . — Reply to this email directly, view it on GitHub <#957 (comment)>, or unsubscribe <https://github.com/notifications/unsubscribe-auth/AAALDFBNTCLFEHBSZP7GC23VMKTUJANCNFSM5XHGQSXQ> . You are receiving this because you were mentioned.Message ID: ***@***.***>

[nimmolo]

@JoeCohen I believe this is now good to go. With respect to our calls to http get etc., I believe the "Rails/HttpPositionalArguments" cop is incorrect, as far as I can understand.

See the rails code here, scroll down to the get method etc. It seems to me it expects the args like we're sending them. Maybe it's a case of old rubocop?

[JoeCohen]

Thanks. I'll have to look at it in about an hour. I'm being called to dinner now.

…

On Sat, May 28, 2022 at 5:44 PM andrew nimmo ***@***.***> wrote: @JoeCohen <https://github.com/JoeCohen> I believe this is now good to go. The "Rails/HttpPositionalArguments" cop is incorrect, as far as I can understand, with respect to our calls to http get etc. See the rails code here <https://github.com/rails/rails/blob/aa7c4179ff3b0dba6c14e45cecaf853ac9446d41/actionpack/lib/action_controller/test_case.rb>. It seems to me it expects it like we're serving it. — Reply to this email directly, view it on GitHub <#957 (comment)>, or unsubscribe <https://github.com/notifications/unsubscribe-auth/AAALDFD35LGNSKQGO2TOJATVMK4YBANCNFSM5XHGQSXQ> . You are receiving this because you were mentioned.Message ID: ***@***.***>

[nimmolo]

I'm not at home but thought of another consideration. When a test inherits from functional_test_case, which the tests that inherit from controller_extensions and session_extensions may or may not do, MO intercepts all calls to get etc to do our check_for_unsafe_html. These methods certainly expect get(action, **args) like the test_case classes they inherit from.

[pellaea]

Urk, that was quite a flurry of emails in this thread!! I didn't take the time to read and understand everything. Is there anything that I should be particularly aware of??

…

On Sat, May 28, 2022 at 10:35 PM codeclimate[bot] ***@***.***> wrote: Code Climate has analyzed commit c0329e9 <c0329e9> and detected *5 issues* on this pull request. Here's the issue category breakdown: Category Count Bug Risk 5 View more on Code Climate <https://codeclimate.com/github/MushroomObserver/mushroom-observer/pull/957> . — Reply to this email directly, view it on GitHub <#957 (comment)>, or unsubscribe <https://github.com/notifications/unsubscribe-auth/AAYTNNIOI26ZNS3J73J4HRTVMLJXVANCNFSM5XHGQSXQ> . You are receiving this because you are subscribed to this thread.Message ID: ***@***.***>

[nimmolo]

@pellaea - This PR fixes dozens of HttpPositionalArguments offenses, except for 5 in controller_extensions and session_extensions. I believe the cop is throwing a false positive in those.

The native Rails test_case method get(action, **args) expects a positional argument, plus however many kwargs. The cop complains when we send exactly that, from our own "get"-like (and "post"-like) methods (such as post_without_redirecting and reget). I don't know any way to solve it, but i'm no splat doctor.

[nimmolo]

PS - in case you're reading on email and didn't catch the typo i edited after sending, the native Rails method is of course get(action, **args) not get(action: **args). It is used like get(:show, params)

[JoeCohen]

@pellaea: There's nothing that you should be particularly aware of. We (i.e., 99% Andrew) are on top of things, and all the changes are in test, so they can't break anything in production. @nimmolo: I'd like to wait on merging this, so I will make it a Draft PR to prevent accidental merger. In particular, I want to see if updating RuboCop will fix the false positive HttpPositionalArguments offenses. My plan: - New branch. - Review the default protocol for updating RuboCop - Update our RuboCop gems (rubocop, rubocop_rails, rubocop_performance) - Update RuboCop configuration. - Update rubocop.todo - Push the branch - If everything passes and it's OK with you, merge it to master - Merge master to Fix Rails/HttpPositionalArguments offenses #957 - Rev

…

On Sun, May 29, 2022 at 3:48 AM andrew nimmo ***@***.***> wrote: PS - in case you're reading on email and didn't catch the typo i edited after sending, the native Rails method is of course get(action, **args) not get(action: **args). It is used like get(:show, params) — Reply to this email directly, view it on GitHub <#957 (comment)>, or unsubscribe <https://github.com/notifications/unsubscribe-auth/AAALDFBGUO7PFXFMACKJLMLVMNDRLANCNFSM5XHGQSXQ> . You are receiving this because you were mentioned.Message ID: ***@***.***>

[JoeCohen]

I've resolved the conflicts locally, but am getting 18 errors.
I will push it with errors so that everyone can see what's going on.
I suspect that the changes I had are now incorrect in light of changes @nimmolo made on other branches.

[JoeCohen]

Oh wait. I think I messed up the merger. I will try to fix it, then push if successful.

[JoeCohen]

I'm done for tonight.
I'm fine with merging this.

[pellaea]

Sounds good to me. Great work!

…

On Mon, May 30, 2022 at 12:18 AM Joseph D. Cohen ***@***.***> wrote: I'm done for tonight. I'm fine with merging this. — Reply to this email directly, view it on GitHub <#957 (comment)>, or unsubscribe <https://github.com/notifications/unsubscribe-auth/AAYTNNLIDFR3V2433ZFKPH3VMQ6RTANCNFSM5XHGQSXQ> . You are receiving this because you were mentioned.Message ID: ***@***.***>

[codeclimate]

Code Climate has analyzed commit 9021408 and detected 0 issues on this pull request.

View more on Code Climate.

[nimmolo]

nothing urgent. The 5 stickers are in controller extwnsions and session extensions. I think we can resolve later

…

On May 28, 2022, at 7:52 PM, Jason Hollinger ***@***.***> wrote: Urk, that was quite a flurry of emails in this thread!! I didn't take the time to read and understand everything. Is there anything that I should be particularly aware of?? On Sat, May 28, 2022 at 10:35 PM codeclimate[bot] ***@***.***> wrote: > Code Climate has analyzed commit c0329e9 > <c0329e9> > and detected *5 issues* on this pull request. > > Here's the issue category breakdown: > Category Count > Bug Risk 5 > > View more on Code Climate > <https://codeclimate.com/github/MushroomObserver/mushroom-observer/pull/957> > . > > — > Reply to this email directly, view it on GitHub > <#957 (comment)>, > or unsubscribe > <https://github.com/notifications/unsubscribe-auth/AAYTNNIOI26ZNS3J73J4HRTVMLJXVANCNFSM5XHGQSXQ> > . > You are receiving this because you are subscribed to this thread.Message > ID: ***@***.***> > — Reply to this email directly, view it on GitHub, or unsubscribe. You are receiving this because you were mentioned.