tweetlr 0.0.10 → 0.1.0

data/README.md

@@ -21,6 +21,7 @@ tumblr_username: YOUR_TUMBLR_EMAIL
 tumblr_password: YOUR_TUMBLR_PW
 update_period: 300 #check for updates every 300 secs = 5 minutes
 shouts: 'says' # will be concatenated after the username, before the message: @mr_x says: awesome things on a photo!
+loglevel: 0 # 0: debug, 1: info (default), 2: warn, 3: error, 5: fatal
 whitelist: #twitter accounts in that list will have their tweets published immediately. post from others will be saved as drafts
   - whitey_mc_whitelist
   - sven_kr

data/Rakefile

@@ -7,7 +7,7 @@ require 'rake/testtask'
 
 spec = Gem::Specification.new do |s|
   s.name = 'tweetlr'
-  s.version = '0.0.10'
+  s.version = '0.1.0'
   s.has_rdoc = true
   s.extra_rdoc_files = ['README.md', 'LICENSE']
   s.summary = %{tweetlr crawls twitter for a given term, extracts photos out of the collected tweets' short urls and posts the images to tumblr.}

data/bin/tweetlr

@@ -8,26 +8,26 @@ require_relative '../lib/tweetlr.rb'
 
 begin
   config_file = File.join( Dir.pwd,  'config', 'tweetlr.yml')
-  @log_file = File.join( Dir.pwd, 'tweetlrd.log')
   CONFIG = YAML.load_file(config_file)
   TERM = CONFIG['search_term']
   USER = CONFIG['tumblr_username']
   PW   = CONFIG['tumblr_password']
   TIMESTAMP = CONFIG['twitter_timestamp']
   UPDATE_PERIOD = CONFIG['update_period']
-  @tweetlr = Tweetlr.new(USER, PW, nil, TIMESTAMP, TERM, config_file)
+  LOGLEVEL = CONFIG['loglevel'] || Logger::INFO
+  @tweetlr = Tweetlr.new(USER, PW, config_file, {:since_id => TIMESTAMP, :terms => TERM, :loglevel => LOGLEVEL})
 rescue SystemCallError
   $stderr.puts "Ooops - looks like there is no ./config/tweetlr.yml found. I'm affraid tweetlr won't work properly until you introduced that configuration file."
   exit(1)
 end
 
 Daemons.run_proc('tweetlr', :dir_mode => :script, :dir => './', :backtrace => true, :log_output => true) do
-  #@log = Logger.new(@log_file)
-  puts "#{Time.now} - starting tweetlr daemon..." 
-  puts "creating a new tweetlr instance using this config: #{CONFIG.inspect}"   
+  @log = Logger.new(STDOUT)
+  @log.info "starting tweetlr daemon..." 
+  @log.info "creating a new tweetlr instance using this config: #{CONFIG.inspect}"   
   EventMachine::run { 
     EventMachine::add_periodic_timer( UPDATE_PERIOD ) {
-     puts "#{Time.now} - starting tweetlr crawl..."
+     @log.info "starting tweetlr crawl..."
      response = @tweetlr.lazy_search_twitter
      if response
        tweets = response['results']
@@ -35,20 +35,20 @@ Daemons.run_proc('tweetlr', :dir_mode => :script, :dir => './', :backtrace => tr
        tweets.each do |tweet|
            tumblr_post = @tweetlr.generate_tumblr_photo_post tweet
            if tumblr_post.nil? ||  tumblr_post[:source].nil?
-              puts "could not get image source: tweet: #{tweet} --- tumblr post: #{tumblr_post.inspect}"
-         else
+              @log.warn "could not get image source: tweet: #{tweet} --- tumblr post: #{tumblr_post.inspect}"
+           else
              #@log.debug tumblr_post
              #@log.debug @tweetlr.post_to_tumblr tumblr_post
              #puts "tumblr post: #{tumblr_post}"
              res = @tweetlr.post_to_tumblr tumblr_post
-             puts "tumblr response: #{res.header_str} #{res.body_str}" unless res.response_code == 201
+             @log.warn "tumblr response: #{res.header_str} #{res.body_str}" unless res.response_code == 201
            end
          end
        end
      else
-       puts "#{Time.now} - twitter serach returned no response. Hail the whale!"
+       @log.error "twitter serach returned no response. hail the failwhale!"
      end
-     puts "#{Time.now} - finished tweetlr crawl."
+     @log.info "finished tweetlr crawl."
      }
    }
 

data/lib/tweetlr.rb

@@ -9,51 +9,68 @@ class Tweetlr
   LOCATION_START_INDICATOR = 'Location: '
   LOCATION_STOP_INDICATOR  = "\r\n"
   
-  def initialize(email, password, cookie=nil, since_id=nil, terms=nil, config_file) #TODO use a hash or sth more elegant here...
-    @log = Logger.new(File.join( Dir.pwd, 'tweetlr.log'))
+  def initialize(email, password, config_file, args={:cookie => nil, :since_id=>nil, :terms=>nil, :loglevel=>Logger::INFO})
+    @log = Logger.new(STDOUT)
+    @log.level = args[:loglevel] if (Logger::DEBUG..Logger::UNKNOWN).to_a.index(args[:loglevel])
+    @log.debug "log level set to #{@log.level}"
     config = YAML.load_file(config_file)
+    @email = email
+    @password = password
+    @since_id = args[:since_id]
+    @search_term = args[:terms]
+    @cookie = args[:cookie]
     @results_per_page = config['results_per_page']
     @result_type = config['result_type']
     @api_endpoint_twitter = config['api_endpoint_twitter']
     @api_endpoint_tumblr = config['api_endpoint_tumblr']
     @whitelist = config['whitelist']
     @shouts = config['shouts']
-    @since_id = since_id
-    @search_term = terms
     @whitelist.each {|entry| entry.downcase!}
-    @email = email
-    @password = password
-    @refresh_url = "#{@api_endpoint_twitter}?ors=#{terms}&since_id=#{since_id}&rpp=#{@results_per_page}&result_type=#{@result_type}" if (since_id && terms)
-    if !cookie
+    @refresh_url = "#{@api_endpoint_twitter}?ors=#{@search_term}&since_id=#{@since_id}&rpp=#{@results_per_page}&result_type=#{@result_type}" if (@since_id && @search_term)
+    if !@cookie
       response = Curl::Easy.http_post(
         "#{@api_endpoint_tumblr}/login",
         :body => {
           :email => @email,
-          :password => password
+          :password => @password
         }
       )
-      @log.debug("initial login response: #{response}")
+      @log.debug("initial login response header: #{response.header_str}") if response
       @cookie = response.headers['Set-Cookie']
-      @log.debug("--------login cookie via new login: #{@cookie.inspect}")
+      @log.debug("login cookie via new login: #{@cookie.inspect}")
     else
-      @cookie = cookie
-      @log.debug("--------login cookie via argument: #{@cookie.inspect}")
+      @cookie = args[:cookie]
+      @log.debug("login cookie via argument: #{@cookie.inspect}")
     end
     
   end
-  #post a tumblr photo entry. required arguments are :type, :date, :source, :caption, :state
+  #post a tumblr photo entry. required arguments are :type, :date, :source, :caption, :state. optional argument: :tags 
   def post_to_tumblr(options={})
+    tries = 3
     if options[:type] && options[:date] && options[:source] && options[:caption] && options[:state]
-      response = Curl::Easy.http_post("#{@api_endpoint_tumblr}/api/write", 
-      Curl::PostField.content('generator', GENERATOR),
-      Curl::PostField.content('email', @email), 
-      Curl::PostField.content('password', @password),
-      Curl::PostField.content('type', options[:type]),
-      Curl::PostField.content('date', options[:date]),
-      Curl::PostField.content('source', options[:source]),
-      Curl::PostField.content('caption', options[:caption]),
-      Curl::PostField.content('state', options[:state])
-      )
+      tags = options[:tags]
+      begin
+        response = Curl::Easy.http_post("#{@api_endpoint_tumblr}/api/write", 
+        Curl::PostField.content('generator', GENERATOR),
+        Curl::PostField.content('email', @email), 
+        Curl::PostField.content('password', @password),
+        Curl::PostField.content('type', options[:type]),
+        Curl::PostField.content('date', options[:date]),
+        Curl::PostField.content('source', options[:source]),
+        Curl::PostField.content('caption', options[:caption]),
+        Curl::PostField.content('state', options[:state]),
+        Curl::PostField.content('tags', tags)
+        )
+      rescue Curl::Err => err
+        @log.error "Failure in Curl call: #{err}"
+        tries -= 1
+        sleep 3
+        if tries > 0
+            retry
+        else
+            response = nil
+        end
+      end
     end
     response
   end
@@ -63,13 +80,13 @@ class Tweetlr
     tumblr_post = nil
     message = tweet['text']
     if !retweet? message
-      #@log.debug "tweet: #{tweet}"
-      #puts "tweet: #{tweet}"
+      @log.debug "tweet: #{tweet}"
       tumblr_post = {}
       tumblr_post[:type] = 'photo'
       tumblr_post[:date] = tweet['created_at']
       tumblr_post[:source] = extract_image_url tweet
       user = tweet['from_user']
+      tumblr_post[:tags] = user
       tweet_id = tweet['id']
       if @whitelist.member? user.downcase
         state = 'published'
@@ -78,7 +95,7 @@ class Tweetlr
       end
       tumblr_post[:state] = state
       shouts = " #{@shouts}" if @shouts
-      tumblr_post[:caption] = %?<a href="http://twitter.com/#{user}/statuses/#{tweet_id}" alt="#{user}">@#{user}</a>#{shouts}: #{tweet['text']}? #TODO make this a bigger matter of yml configuration
+      tumblr_post[:caption] = %?<a href="http://twitter.com/#{user}/statuses/#{tweet_id}" alt="tweet">@#{user}</a>#{shouts}: #{tweet['text']}? #TODO make this a bigger matter of yml configuration
     end
     tumblr_post
   end
@@ -98,13 +115,11 @@ class Tweetlr
     @refresh_url = "#{@api_endpoint_twitter}#{@response['refresh_url']}" unless (@response.nil? || @response['refresh_url'].nil? || @response['refresh_url'].empty?)
     if @refresh_url
      #FIXME persist the refresh url - server restart would be a pain elsewise
-     #@log.info "lazy search using '#{@refresh_url}'"
      search_url = "#{@refresh_url}&result_type=#{@result_type}&rpp=#{@results_per_page}"
-     puts "lazy search using '#{search_url}'" #workaround to get refresh url logged w/ the Daemons gem
+     @log.info "lazy search using '#{search_url}'" #workaround to get refresh url logged w/ the Daemons gem
      @response = http_get search_url
     else
-      #@log.debug "regular search using '#{term}'"
-      puts "regular search using '#{@search_term}'"
+      @log.debug "regular search using '#{@search_term}'"
       @response = search_twitter()
     end
   end
@@ -128,12 +143,16 @@ class Tweetlr
     end
     url
   end
-  
+  #find the image's url for a lockerz link
+  def image_url_lockerz(link_url)
+    response = http_get "http://api.plixi.com/api/tpapi.svc/json/metadatafromurl?details=false&url=#{link_url}"
+    response['BigImageUrl'] if response
+  end
+  #find the image's url for an twitter shortened link
   def image_url_tco(link_url)
     service_url = link_url_redirect link_url
     find_image_url service_url
   end
-
   #find the image's url for an instagram link
   def image_url_instagram(link_url)
     link_url['instagram.com'] = 'instagr.am' if link_url.index 'instagram.com' #instagram's oembed does not work for .com links
@@ -149,7 +168,11 @@ class Tweetlr
     #if short url fails, try long url
     #response = HTTParty.get "http://picplz.com/api/v2/pic.json?longurl_ids=#{id}"
     #extract url
-    response['value']['pics'].first['pic_files']['640r']['img_url'] if response
+    if response && response['value'] && response['value']['pics'] && response['value']['pics'].first && response['value']['pics'].first['pic_files'] && response['value']['pics'].first['pic_files']['640r']
+      response['value']['pics'].first['pic_files']['640r']['img_url'] 
+    else
+      nil
+    end
   end
   #find the image's url for a twitpic link
   def image_url_twitpic(link_url)
@@ -202,18 +225,37 @@ class Tweetlr
   
   #convenience method for curl http get calls
   def http_get(request)
+    tries = 3
     begin
       res = Curl::Easy.http_get(request)
       JSON.parse res.body_str
     rescue Curl::Err::ConnectionFailedError => err
-      #@log.error "Connection failed: #{err}"
-      puts "Connection failed: #{err}"
-      nil
+      @log.error "Connection failed: #{err}"
+      tries -= 1
+      sleep 3
+      if tries > 0
+          retry
+      else
+          nil
+      end
+    rescue Curl::Err::RecvError => err
+      @log.error "Failure when receiving data from the peer: #{err}"
+      tries -= 1
+      sleep 3
+      if tries > 0
+          retry
+      else
+          nil
+      end
+    rescue Curl::Err => err
+      @log.error "Failure in Curl call: #{err}"
+      tries -= 1
+      sleep 3
+      if tries > 0
+          retry
+      else
+          nil
+      end
     end
-  end
-  
-end
-
-
-
-
+  end  
+end

metadata

@@ -2,7 +2,7 @@
 name: tweetlr
 version: !ruby/object:Gem::Version 
   prerelease: 
-  version: 0.0.10
+  version: 0.1.0
 platform: ruby
 authors: 
 - Sven Kraeuter
@@ -10,7 +10,7 @@ autorequire:
 bindir: bin
 cert_chain: []
 
-date: 2011-05-22 00:00:00 Z
+date: 2011-06-02 00:00:00 Z
 dependencies: 
 - !ruby/object:Gem::Dependency 
   name: daemons