Add Spring, spring-watcher-listen, and Listen

[nimmolo]

Another development-only tool that i'm finding helpful.

Spring is a Rails-maintained application preloader. "It speeds up development by keeping your application running in the background, so you don't need to boot it every time you run a test, rake task or migration."

Listen allows Spring to work more efficiently by sending it notifications of directory changes. It can also be used independently, but this PR does not implement it outside of Spring.

spring-watcher-listen connects the two.

Versions of all three locked to work with Rails 5.2 and each other.

Spring (v4) is, bundled and active by default in Rails 6 (this PR is for Spring v3, the last version that works with Rails 5). It's back to inactive by default in Rails 7 (because "computers are faster" they say), but it's still bundled with Rails 7 currently.

If anyone objects for any reason, no problem. I'm open to using it local-only.

[coveralls]

Coverage remained the same at 92.893% when pulling e4cd6f7 on nimmo-add-spring-listen into 9fc48e7 on master.

[JoeCohen]

I'd love it to have some time shaved off test runs. But I'm seeing some issues:

1. Will this configuration run on the M1? (Nathan has an M1.)
   See Spring taking minutes to run on MacOS with M1 Chip rails/spring#635 (suggesting upgrading to listen >= 3.3.0; but our Gemfile says gem "listen", ">= 3.0.5", "< 3.2)
2. Should the following change be made in config/environments/test.rb:
   config.cache_classes = true false
   See https://github.com/rails/spring#enable-reloading
3. I think it's not working for me. In particular: after a test run spring is not running:

vagrant@ubuntu-focal /vagrant/mushroom-observer (nimmo-add-spring-listen)
$ grep "config.cache_classes" config/environments/test.rb
  # config.cache_classes = true
  config.cache_classes = false
vagrant@ubuntu-focal /vagrant/mushroom-observer (nimmo-add-spring-listen)
vagrant@ubuntu-focal /vagrant/mushroom-observer (nimmo-add-spring-listen)
$ rails t
...
vagrant@ubuntu-focal /vagrant/mushroom-observer (nimmo-add-spring-listen)
$ spring status
Spring is not running.

Compare to https://github.com/rails/spring#usage. (I'll try to see if I'm doing something wrong.)

4. When I start a test run, I get the following warnings. I don't understand their significance. (And maybe they're related to the fact that Spring isn't running after the tests.):

vagrant@ubuntu-focal /vagrant/mushroom-observer (nimmo-add-spring-listen)
$ rails t
/vagrant/mushroom-observer/app/classes/map_collapsible.rb:49: warning: already initialized constant CollapsibleCollectionOfMappableObjects::PRECISION
/vagrant/mushroom-observer/app/classes/map_collapsible.rb:49: warning: previous definition of PRECISION was here
/vagrant/mushroom-observer/app/classes/map_collapsible.rb:70: warning: already initialized constant CollapsibleCollectionOfMappableObjects::MAX_PRECISION
/vagrant/mushroom-observer/app/classes/map_collapsible.rb:70: warning: previous definition of MAX_PRECISION was here
/vagrant/mushroom-observer/app/classes/map_collapsible.rb:71: warning: already initialized constant CollapsibleCollectionOfMappableObjects::MIN_PRECISION
/vagrant/mushroom-observer/app/classes/map_collapsible.rb:71: warning: previous definition of MIN_PRECISION was here
Started with run options --seed 40014

[codeclimate]

Code Climate has analyzed commit e4cd6f7 and detected 0 issues on this pull request.

View more on Code Climate.

[pellaea]

The MIN_MAX_PRECISION are constants I probably added incorrectly to the map stuff. I bet there's a better way of doing it. So you can probably add that to the "uncovering hidden bugs" column.

…

On Tue, Jan 18, 2022 at 5:32 PM codeclimate[bot] ***@***.***> wrote: Code Climate has analyzed commit e4cd6f7 <e4cd6f7> and detected *0 issues* on this pull request. View more on Code Climate <https://codeclimate.com/repos/58cfc16d1035280287001d93/pull/824>. — Reply to this email directly, view it on GitHub <#824 (comment)>, or unsubscribe <https://github.com/notifications/unsubscribe-auth/AAYTNNJY2STFV5K5IMZFARLUWXS67ANCNFSM5MGFY23A> . Triage notifications on the go with GitHub Mobile for iOS <https://apps.apple.com/app/apple-store/id1477376905?ct=notification-email&mt=8&pt=524675> or Android <https://play.google.com/store/apps/details?id=com.github.android&referrer=utm_campaign%3Dnotification-email%26utm_medium%3Demail%26utm_source%3Dgithub>. You are receiving this because you are subscribed to this thread.Message ID: ***@***.***>

[mo-nathan]

I've merged this into the mo-docker branch and everything seems to be working. Like Joe, I see not speed improvements running the tests, but I also see no downside. I'm fine merging it into master.

[JoeCohen]

@mo-nathan: Can you run tests in the mo-docker branch?

[JoeCohen]

That was a dumb question. You said you were running them.

[JoeCohen]

More important question:
When you run the tests, are you getting the "already initialized constant" warnings? See #824 (comment)
If so, what if anything, should we do about that before merging?
I just switched to master, and am getting the same warnings there.

[JoeCohen]

It's just me; the "already initialized constant" warnings are just on my desktop, not on my laptop.

[JoeCohen]

In order to fix this, I had to rebuild my vm, starting from cloning the repo.
Perhaps my problem was due to my following the directions to "springify executables". https://github.com/rails/spring#setup
I'll double-check that tomorrow if I have time.
But right now I'll merge this.

[JoeCohen]

Using Spring in test environment

1. The "already initialized constant" warnings were due to my changing config.cache_classes to false in config/environments/test.rb. You have to do that in order to run Spring in that environment. https://github.com/rails/spring#enable-reloading
2. I won't be using Spring for the tests. Although Spring did improve "user" and "sys" times, it didn't change "real" time.

Without Spring:

$ time rails t
...
Finished in 198.49247s
...
real	3m24.552s
user	2m9.653s
sys	0m42.491s

With Spring:

$ time spring rails t
Running via Spring preloader in process 15117
...
Finished in 195.71937s
real	3m19.578s
user	0m0.251s
sys	0m0.028s

$ time spring rails t
Running via Spring preloader in process 20955
...
Finished in 212.87733s
...
real	3m36.885s
user	0m0.245s
sys	0m0.037s

[nimmolo]

Hi Joe - sorry to take awhile to get back to these questions, I couldn't remember how i had it set up. I realize it's merged...

• I'm realizing i didn't run the bundle exec spring binstub --all setup on this branch, so it's unlikely that Spring is active by default on master now
• When you tested: Did you set it up to use the bin/ prefix before the rails or rake command, or did you set up an .envrc file?

[JoeCohen]

Neither.
I manually modified config/environments/test.rb.
Then ran time spring rails t.