add sinatra, json pull, etc

Dockerfile

@@ -163,5 +163,5 @@ chmod 744 /etc/cron.d/dockerhub-stats
 
 # ports and volumes
 #WORKDIR /opt/docker-readme-sync
-EXPOSE 3000 8083 8086
+EXPOSE 3000 8083 8086 4567
 VOLUME /config

app/Gemfile

@@ -7,3 +7,4 @@ gem 'httparty', '~> 0.14.0'
 gem 'influxdb', '~> 0.3.12'
 
 gem 'awesome_print'
+gem 'sinatra', '~> 2.0', '>= 2.0.1'

app/Gemfile.lock

@@ -10,7 +10,17 @@ GEM
       json
     json (2.0.2)
     multi_xml (0.5.5)
+    mustermann (1.0.2)
+    rack (2.0.4)
+    rack-protection (2.0.1)
+      rack
     rake (11.3.0)
+    sinatra (2.0.1)
+      mustermann (~> 1.0)
+      rack (~> 2.0)
+      rack-protection (= 2.0.1)
+      tilt (~> 2.0)
+    tilt (2.0.8)
 
 PLATFORMS
   ruby
@@ -20,9 +30,10 @@ DEPENDENCIES
   httparty (~> 0.14.0)
   influxdb (~> 0.3.12)
   rake (~> 11.3)
+  sinatra (~> 2.0, >= 2.0.1)
 
 RUBY VERSION
    ruby 2.3.1p112
 
 BUNDLED WITH
-   1.13.6
+   1.16.1

app/stats.rb

@@ -0,0 +1,31 @@
+require "awesome_print"
+require "influxdb"
+require "sinatra"
+
+def get_data
+  port = 8086
+  port = ENV['INFLUXDB_PORT'].to_i unless ENV['INFLUXDB_PORT'].nil?
+
+  influxdb = InfluxDB::Client.new database: "dockerhub_stats", port: port
+  query = 'select max(pull_count) as count from dockerhub_stats where time > now() - 1h group by repo,arch'
+  res = influxdb.query(query)
+
+  data = []
+  res.each do |repo|
+    repo_name = repo['tags']['repo']
+    count = repo['values'][0]['count']
+    arch = repo['tags']['arch']
+    data << { name: repo_name, count: count, arch: arch }
+  end
+
+  data
+end
+
+set :logging, true
+set :dump_errors, true
+set :bind, '0.0.0.0'
+
+get '/stats' do
+  content_type :json
+  get_data.to_json
+end

app/stats_pull.rb

@@ -14,15 +14,21 @@ def processs_org(org)
 
   res = HTTParty.get(
     "https://hub.docker.com/v2/repositories/#{org}/",
-    query: { page_size: 100 })
-  data = JSON.parse(res.body)
+    query: { page_size: 20 })
 
   pull_counts = {}
-  while data.include?('next') && data['next'].nil? == false
-    url = data['next']
-    res = HTTParty.get(url)
+  continue = true
+  while continue
     data = JSON.parse(res.body)
     pull_counts.merge!(process_counts(data, arch))
+
+    if data.include?('next') && data['next'].nil? == false
+      url = data['next']
+      res = HTTParty.get(url)
+    else
+      continue = false
+    end
+
   end
 
   pull_counts
@@ -40,7 +46,10 @@ def process_counts(data, arch)
 end
 
 def insert_counts(data)
-  influxdb = InfluxDB::Client.new database: "dockerhub_stats", port: 18086
+  port = 8086
+  port = ENV['INFLUXDB_PORT'].to_i unless ENV['INFLUXDB_PORT'].nil?
+
+  influxdb = InfluxDB::Client.new database: "dockerhub_stats", port: port
   data.each do |repo, repo_data|
     influx_data = {
       values: { pull_count: repo_data[:pull_count] },
@@ -54,10 +63,12 @@ end
 def run
   orgs = nil
 
+  # pull orgs from env var
   unless ENV['DOCKERHUB_ORGS'].nil?
     orgs = ENV['DOCKERHUB_ORGS'].split(',')
   end
 
+  # pull orgs from settings file if not specified in env var
   if orgs.nil?
     config_file = File.dirname(__FILE__) + "/config/settings.yml"
     config = YAML.load_file(config_file)
@@ -65,9 +76,6 @@ def run
     orgs = config["orgs"]
   end
 
-  ap orgs
-  exit
-
   counts = {}
   orgs.each do |org|
     counts.merge!(processs_org(org))

root/etc/services.d/sinatra/run

@@ -0,0 +1,4 @@
+#!/usr/bin/with-contenv bash
+
+cd /opt/dockerhub-stats/
+ruby stats.rb