defines a staging execution mode

Author: fxn

.gitignore

@@ -3,3 +3,5 @@
 /tmp
 /rails.git
 config/master.key
+public/assets
+public/cache

config/database.yml

@@ -11,3 +11,7 @@ development:
 test:
   <<: *defaults
   database: rails_contributors_test
+
+staging:
+  <<: *defaults
+  database: rails_contributors_development

config/environments/staging.rb

@@ -0,0 +1 @@
+production.rb

doc/docker.md

@@ -65,6 +65,8 @@ $ docker/rails runner 'p Commit.count'
 $ ...
 ```
 
+The command `docker/server staging` runs the application in `staging` mode. That is a production-like environment (`config/environments/staging.rb` is a symlink to `config/environments/production.rb`), but it uses the development database and serves static files. If there are changes that act differently in `development` and `production`, this execution mode may be useful for checking things up before deployment.
+
 If you modify the name mappings, hard-coded authors, etc., `docker/sync`
 updates the credits and it does so changing as little as possible. The command also accepts an optional argument `all`, which forces the recomputation of all assignments. This is handy when you've changed the heuristics, and it takes less than a full database rebuild, since it does not reimport the commits themselves (which is costly).
 

docker/server

@@ -1,9 +1,10 @@
 #!/bin/bash
 
-if [[ $1 = 'production' ]]; then
-  docker/rails assets:precompile
+if [[ $1 = 'staging' ]]; then
+  docker-compose exec -e RAILS_ENV=staging app \
+    bin/rails assets:precompile
   docker-compose exec -e RAILS_SERVE_STATIC_FILES=1 app \
-    bin/rails server -b 0.0.0.0 -e production
+    bin/rails server -b 0.0.0.0 -e staging
 else
   docker/rails server -b 0.0.0.0
 fi