tweet.rb in twitterscraper-ruby-0.17.0

- old
+ new

@@ -4,10 +4,11 @@
   class Tweet
     KEYS = [
         :screen_name,
         :name,
         :user_id,
+        :profile_image_url,
         :tweet_id,
         :text,
         :links,
         :hashtags,
         :image_urls,
@@ -49,10 +50,15 @@
           tweet['created_at'] = Time.parse(tweet['created_at'])
           new(tweet)
         end
       end
 
+      # .js-stream-item
+      #   .js-stream-tweet{data: {screen-name:, tweet-id:}}
+      #     .stream-item-header
+      #     .js-tweet-text-container
+      #     .stream-item-footer
       def from_html(text)
         html = Nokogiri::HTML(text)
         from_tweets_html(html.xpath("//li[@class[contains(., 'js-stream-item')]]/div[@class[contains(., 'js-stream-tweet')]]"))
       end
 
@@ -70,10 +76,12 @@
           Twitterscraper.logger.warn "html doesn't include div.js-tweet-text-container url=https://twitter.com/#{screen_name}/status/#{tweet_id}"
           return nil
         end
 
         inner_html = Nokogiri::HTML(html.inner_html)
+
+        profile_image_url = inner_html.xpath("//img[@class[contains(., 'js-action-profile-avatar')]]").first.attr('src').gsub(/_bigger/, '')
         text = inner_html.xpath("//div[@class[contains(., 'js-tweet-text-container')]]/p[@class[contains(., 'js-tweet-text')]]").first.text
         links = inner_html.xpath("//a[@class[contains(., 'twitter-timeline-link')]]").map { |elem| elem.attr('data-expanded-url') }.select { |link| link && !link.include?('pic.twitter') }
         image_urls = inner_html.xpath("//div[@class[contains(., 'AdaptiveMedia-photoContainer')]]").map { |elem| elem.attr('data-image-url') }
         video_url = inner_html.xpath("//div[@class[contains(., 'PlayableMedia-container')]]/a").map { |elem| elem.attr('href') }[0]
         has_media = !image_urls.empty? || (video_url && !video_url.empty?)
@@ -97,9 +105,10 @@
         timestamp = inner_html.xpath("//span[@class[contains(., 'js-short-timestamp')]]").first.attr('data-time').to_i
         new(
             screen_name: screen_name,
             name: html.attr('data-name'),
             user_id: html.attr('data-user-id').to_i,
+            profile_image_url: profile_image_url,
             tweet_id: tweet_id,
             text: text,
             links: links,
             hashtags: text.scan(/#\w+/).map { |tag| tag.delete_prefix('#') },
             image_urls: image_urls,