Merge pull request #2778 from sciencehistory/capybara_puma

Switch capybara back to using default puma webserver

Gemfile

@@ -19,9 +19,6 @@ gem 'rails', '~> 7.2.0'
 # resolution to old versions of sinatra!
 # https://bibwild.wordpress.com/2023/11/09/beware-sinatra-rails-7-1-rack-3-resque-bundler-dependency-resolution/
 gem "rack", ">= 3.0"
-# rackup is normally indirect dependency of Rails. We need to temporarily stick to
-# older rackup, causing problems: https://github.com/rack/rackup/issues/26
-gem "rackup", "< 2.2.0"
 
 gem "sinatra", ">= 4.0"
 

Gemfile.lock

@@ -524,9 +524,8 @@ GEM
       rack (>= 3.0.0)
     rack-test (2.1.0)
       rack (>= 1.3)
-    rackup (2.1.0)
+    rackup (2.2.0)
       rack (>= 3)
-      webrick (~> 1.8)
     rails (7.2.1.1)
       actioncable (= 7.2.1.1)
       actionmailbox (= 7.2.1.1)
@@ -760,7 +759,6 @@ GEM
       addressable (>= 2.8.0)
       crack (>= 0.3.2)
       hashdiff (>= 0.4.0, < 2.0.0)
-    webrick (1.9.0)
     websocket (1.2.11)
     websocket-driver (0.7.6)
       websocket-extensions (>= 0.1.0)
@@ -839,7 +837,6 @@ DEPENDENCIES
   qa (~> 5.2, >= 5.13.0)
   rack (>= 3.0)
   rack-attack (~> 6.6)
-  rackup (< 2.2.0)
   rails (~> 7.2.0)
   rails-controller-testing
   reline (>= 0.2.1)

spec/rails_helper.rb

@@ -29,16 +29,7 @@
 
 require 'scihist_digicoll/solr_wrapper_util'
 
-
-# get puma logs out of console
-# https://github.com/rspec/rspec-rails/issues/1897
-#Capybara.server = :puma, { Silent: true }
-#
-# When running tests, jrochkind is getting a core dump when using puma.
-# Think it has to do with rspec setup of puma to use multiple workers,
-# which is running into a bug. For now, we test with webrick, should be
-# just fine.
-Capybara.server = :webrick
+#Capybara.server = :puma
 
 # Checks for pending migrations and applies them before tests are run.
 # If you are not using ActiveRecord, you can remove these lines.