Create `rails public_data:commit` task, to share public data dumps

I'm starting to port over the functionality that was previously just,
me running `yarn db:export:public-data` in `impress-2020` and
committing it to Git LFS every time.

My immediate motivation is that the `impress-2020` git repository is
getting weirdly large?? Idk how these 40MB files have blown up to a
solid 16GB of Git LFS data (we don't have THAT many!!!), but I guess
there's something about Git LFS's architecture and disk usage that I'm
not understanding.

So, let's move to a simpler system in which we don't bind the public
data to the codebase, but instead just regularly dump it in production
and make it available for download.

This change adds the `rails public_data:commit` task, which when run in
production will make the latest available at
`https://impress.openneo.net/public-data/latest.sql.gz`, and will also
store a running log of previous dumps, viewable at
`https://impress.openneo.net/public-data/`.

Things left to do:
1. Create a `rails public_data:pull` task, to download `latest.sql.gz`
   and import it into the local development database.
2. Set up a cron job to dump this out regularly, idk maybe weekly? That
   will grow, but not very fast (about 2GB per year), and we can add
   logic to rotate out old ones if it starts to grow too far. (If we
   wanted to get really intricate, we could do like, daily for the past
   week, then weekly for the past 3 months, then monthly for the past
   year, idk. There must be tools that do this!)

.gitignore

@@ -8,6 +8,8 @@ tmp/**/*
 /app/assets/builds/*
 !/app/assets/builds/.keep
 
+/public/public-data
+
 /node_modules
 
 .pnp.*

Gemfile

@@ -79,6 +79,9 @@ gem "stackprof", "~> 0.2.25"
 gem "sentry-ruby", "~> 5.12"
 gem "sentry-rails", "~> 5.12"
 
+# For tasks that use shell commands.
+gem "shell", "~> 0.8.1"
+
 # For automated testing.
 group :test do
 	gem 'sqlite3', '~> 1.7'

Gemfile.lock

@@ -139,6 +139,7 @@ GEM
       railties (>= 3.2)
     drb (2.2.0)
       ruby2_keywords
+    e2mmap (0.1.0)
     erubi (1.12.0)
     execjs (2.9.1)
     falcon (0.43.0)
@@ -316,6 +317,9 @@ GEM
       sentry-ruby (~> 5.16.1)
     sentry-ruby (5.16.1)
       concurrent-ruby (~> 1.0, >= 1.0.2)
+    shell (0.8.1)
+      e2mmap
+      sync
     sprockets (4.2.1)
       concurrent-ruby (~> 1.0)
       rack (>= 2.2.4, < 4)
@@ -327,6 +331,7 @@ GEM
       mini_portile2 (~> 2.8.0)
     stackprof (0.2.26)
     stringio (3.1.0)
+    sync (0.5.0)
     temple (0.10.3)
     terser (1.2.0)
       execjs (>= 0.3.0, < 3)
@@ -385,6 +390,7 @@ DEPENDENCIES
   sass-rails (~> 6.0)
   sentry-rails (~> 5.12)
   sentry-ruby (~> 5.12)
+  shell (~> 0.8.1)
   sprockets (~> 4.2)
   sqlite3 (~> 1.7)
   stackprof (~> 0.2.25)

config/environments/development.rb

@@ -112,4 +112,8 @@ Rails.application.configure do
   # we keep this in a long-term location instead!)
   config.neopets_media_archive_root = Rails.root / "tmp" /
     "neopets_media_archive" / "development"
+
+  # When developing the `public_data:commit` command, save to the local `tmp`
+  # folder. (In production, we keep this in a long-term location instead!)
+  config.public_data_root = Rails.root / "tmp" / "public_data"
 end

config/environments/production.rb

@@ -130,4 +130,8 @@ Rails.application.configure do
   # Save the Neopets Media Archive in `/var/lib/neopets-media-archive`, a
   # long-term storage location.
   config.neopets_media_archive_root = "/var/lib/neopets-media-archive"
+
+  # Save our public data exports in `public/public-data`. (This should be
+  # symlinked to a shared folder persisted across all versions.)
+  config.public_data_root = Rails.root / "public" / "public-data"
 end

deploy/deploy.yml

@@ -51,6 +51,12 @@
           - "--exclude=.git"
           - "--filter=':- .gitignore'"
 
+    - name: Link the public-data folder to the shared public-data folder
+      file:
+        src: "{{ remote_project_root }}/shared/public-data"
+        dest: "{{ remote_app_root }}/public/public-data"
+        state: link
+
     - name: Configure Bundler to run in deployment mode
       command:
         chdir: "{{ remote_app_root }}"

deploy/files/sites-available/impress.conf

@@ -33,6 +33,10 @@ server {
     add_header  ETag "";
   }
 
+  location /public-data/ {
+    autoindex on;
+  }
+
   # On status 503, return the maintenance page. (We'll trigger this ourselves
   # in the @app location, if $maintenance is on.)
   error_page 503 /maintenance.html;

deploy/setup.yml

@@ -275,6 +275,13 @@
         - Reload systemctl
         - Restart impress
 
+    - name: Create the shared public-data folder
+      become_user: impress
+      file:
+        dest: /srv/impress/shared/public-data
+        mode: "755"
+        state: directory
+
     - name: Create service file for impress
       copy:
         src: files/impress.service

lib/tasks/public_data.rake

@@ -0,0 +1,67 @@
+require "open3"
+
+desc "Tools to save and import DTI's public modeling data"
+namespace :public_data do
+	desc "Save the local database's public data to a local file"
+	task :commit, [:name] => :environment do |_, args|
+		if Rails.env.development?
+			puts "NOTE: The `public_data:commit` task is primarily meant to be " +
+				"run in production, to create public data files we can copy to our " +
+				"development machines via `public_data:pull`. I'll still run it " +
+				"locally and save to #{Rails.configuration.public_data_root}, though!"
+		end
+
+		config = ApplicationRecord.connection_db_config.configuration_hash
+
+		# Generate a filename from the current time, and the option name argument
+		# provided to the command (e.g. `rails public_data:commit[scheduled]`).
+		timestamp = Time.now.utc.iso8601.gsub(':', '_')
+		name = args.fetch(:name, "manual")
+		filename = "#{timestamp}-#{name}.sql.gz"
+		dest_path = Rails.configuration.public_data_root / filename
+
+		args = []
+
+		# The connection details for our database!
+		args << "--host=#{config[:host]}" if config[:host]
+		args << "--user=#{config[:username]}" if config[:username]
+		args << "--password=#{config[:password]}" if config[:password]
+
+		# Don't lock the database to do it!
+		args << "--single-transaction"
+
+		# Dump the public data tables from the primary database.
+		args << config.fetch(:database)
+		args += %w(species colors zones) # manual constants
+		args += %w(alt_styles items parents_swf_assets pet_states pet_types
+		           swf_assets) # from modeling
+
+		# Set up a shell, and register the commands we need.
+		Shell.def_system_command("mysqldump")
+		Shell.def_system_command("gzip")
+		sh = Shell.new
+
+		# Ensure the output directory exists.
+		dest_path.dirname.mkpath
+
+		# Run mysqldump, pipe it into gzip, and output to the destination file.
+		sh.mysqldump(*args) | sh.gzip("-c") > dest_path.to_s
+		puts "Saved dump to #{dest_path}"
+
+		# Link this latest dump as `latest.sql.gz`.
+		latest_path = Rails.configuration.public_data_root / "latest.sql.gz"
+		File.unlink(latest_path) if File.exist?(latest_path)
+		File.symlink(dest_path, latest_path)
+		puts "Linked dump to #{latest_path}"
+	end
+
+	desc "Pull and import the latest public data from production (dev only)"
+	task :pull do
+		unless Rails.env.development?
+			raise "Can only pull public data in development mode! This helps us " +
+				"ensure we won't overwrite the production database accidentally."
+		end
+
+		raise NotImplementedError, "TODO!"
+	end
+end