bobik 0.0.1

data/Gemfile

@@ -0,0 +1,3 @@
+source "https://rubygems.org"
+
+gemspec

data/Gemfile.lock

@@ -0,0 +1,20 @@
+PATH
+  remote: .
+  specs:
+    bobik (0.0.1)
+      httparty
+
+GEM
+  remote: https://rubygems.org/
+  specs:
+    httparty (0.8.3)
+      multi_json (~> 1.0)
+      multi_xml
+    multi_json (1.3.6)
+    multi_xml (0.5.1)
+
+PLATFORMS
+  ruby
+
+DEPENDENCIES
+  bobik!

data/README.md

@@ -0,0 +1,6 @@
+bobik_ruby_gem
+==============
+
+Bobik SDK for Ruby
+
+TODO: provide documentation on how to install and use the SDK

data/lib/bobik.rb

@@ -0,0 +1,2 @@
+require_relative "bobik/error"
+require_relative "bobik/client"

data/lib/bobik/client.rb

@@ -0,0 +1,73 @@
+require 'json'
+require 'httparty'
+
+module Bobik
+  class Client
+    include HTTParty
+    base_uri 'https://usebobik.com/api/v1'
+
+    def initialize(opts)
+      @auth_token = opts[:auth_token] || raise(Error.new("'auth_token' was not provided"))
+      @timeout_ms = opts[:timeout_ms] || 30000
+      @log = opts[:logger] || (defined?(Rails.logger) && Rails.logger)
+    end
+    
+    
+    def scrape(request, block_until_done, &block)
+      request = Marshal.load(Marshal.dump(request))
+      request[:auth_token] = @auth_token
+
+      job_response = self.class.post('/jobs.json', :body => request)
+      raise Error.new(job_response['errors'].join("\n")) if job_response['errors']
+      job_id = job_response['job']
+
+      Thread.abort_on_exception = true
+      t = Thread.new do
+        wait_until_finished(job_id, &block)
+      end
+      t.join if block_until_done
+      true
+    end
+    
+  private
+    
+    # Blocks until the job is finished or timeout is reached.
+    # When done, yields results to the optional block.
+    # Exceptions thrown: Timeout::Error, Errno::ECONNRESET, Errno::ECONNREFUSED
+    def wait_until_finished(job_id, &block)
+      log("Waiting for job #{job_id} to finish")
+      results = nil
+      errors = nil
+      Timeout::timeout(@timeout_ms.to_f/1000) do
+        while true
+          job_response = get_job_data(job_id, false)
+          progress = job_response['progress']
+          log("Job #{job_id} progress: #{progress*100}%")
+          if progress == 1
+            job_response = get_job_data(job_id, true)
+            results = job_response['results']
+            errors = job_response['errors']
+            break
+          end
+          sleep(job_response['estimated_time_left_ms'].to_f/1000)
+        end
+      end
+      block.call(results, errors)
+    end
+    
+    
+    def get_job_data(job_id, with_results)
+      job_response = self.class.get('/jobs.json', :body => {
+        auth_token: @auth_token,
+        no_results: !with_results,
+        job:        job_id
+      })
+    end
+
+
+    def log(msg)
+      return unless @log
+      @log.debug(msg)
+    end
+  end
+end

data/lib/bobik/error.rb

@@ -0,0 +1,3 @@
+module Bobik
+  class Error < StandardError; end
+end

metadata

@@ -0,0 +1,66 @@
+--- !ruby/object:Gem::Specification
+name: bobik
+version: !ruby/object:Gem::Version
+  version: 0.0.1
+  prerelease: 
+platform: ruby
+authors:
+- Eugene Mirkin
+autorequire: 
+bindir: bin
+cert_chain: []
+date: 2012-07-05 00:00:00.000000000 Z
+dependencies:
+- !ruby/object:Gem::Dependency
+  name: httparty
+  requirement: !ruby/object:Gem::Requirement
+    none: false
+    requirements:
+    - - ! '>='
+      - !ruby/object:Gem::Version
+        version: '0'
+  type: :runtime
+  prerelease: false
+  version_requirements: !ruby/object:Gem::Requirement
+    none: false
+    requirements:
+    - - ! '>='
+      - !ruby/object:Gem::Version
+        version: '0'
+description: Bobik is a web service for scraping the web in real-time
+email: support@usebobik.com
+executables: []
+extensions: []
+extra_rdoc_files: []
+files:
+- lib/bobik/client.rb
+- lib/bobik/error.rb
+- lib/bobik.rb
+- Gemfile
+- Gemfile.lock
+- README.md
+homepage: https://github.com/emirkin/bobik_ruby_gem
+licenses: []
+post_install_message: Bobik says, "Happy scraping!"
+rdoc_options: []
+require_paths:
+- lib
+required_ruby_version: !ruby/object:Gem::Requirement
+  none: false
+  requirements:
+  - - ! '>='
+    - !ruby/object:Gem::Version
+      version: '0'
+required_rubygems_version: !ruby/object:Gem::Requirement
+  none: false
+  requirements:
+  - - ! '>='
+    - !ruby/object:Gem::Version
+      version: '0'
+requirements: []
+rubyforge_project: 
+rubygems_version: 1.8.24
+signing_key: 
+specification_version: 3
+summary: A web service for scraping the web in real-time
+test_files: []