twitterscraper-ruby 0.10.0 → 0.15.0

checksums.yaml

@@ -1,7 +1,7 @@
 ---
 SHA256:
-  metadata.gz: c2429cf6172b5f19caede64ac35f5c796a7c8a67e76fff8dd2f08677fb15406b
-  data.tar.gz: 0f32ca6b559a18c4e3aac3205f6503149e372d4d7d1976b1e83db26036d9ff17
+  metadata.gz: a950fb24329aaa1020441e258a8a2144100d732142b6c227bb9b026b8bb73996
+  data.tar.gz: 1f64f31e43189e2ee439f5ef6f6d54bc6ea58895adbed67cb8ddbe91af07681a
 SHA512:
-  metadata.gz: a36ce6c91a363b64b36deeb3abbaaaebb725f3449f280b70be92532497a94dc5915ba449926acfacfc0d852d52471d258d41140a8891e64b6040bf262d0c347f
-  data.tar.gz: a737c7db151190a1493b1a2a92bea304cfcf7512b2ee03fc13c6f25794f5dc727fe548e52cb39eccc2a63261fee0d58fc005920a0e7cd7650d20600e184d79cb
+  metadata.gz: 8573affbc9a5faa05e5e489364bb2ba0da1aa4f12af35445e5de8b1f8c399eb0575cc9f408b2ba96c3d7fd8b2a74b7dd703229053a33c1f8a883856818033cb9
+  data.tar.gz: 2b2b3ad0b2dd9d089a7b6127ed1b0db21e7f4fa5f0c31e6b366d9b5ae444e2244d4200c813b7a3257f43702d2caa9f264515e701602c24f4482a746b89d41328

data/.circleci/config.yml

@@ -0,0 +1,31 @@
+version: 2.1
+orbs:
+  ruby: circleci/ruby@0.1.2
+
+jobs:
+  build:
+    docker:
+      - image: circleci/ruby:2.6.4-stretch-node
+        environment:
+          BUNDLER_VERSION: 2.1.4
+    executor: ruby/default
+    steps:
+      - checkout
+      - run:
+          name: Update bundler
+          command: gem update bundler
+      - run:
+          name: Which bundler?
+          command: bundle -v
+      - restore_cache:
+          keys:
+            - gem-cache-v1-{{ arch }}-{{ .Branch }}-{{ checksum "Gemfile.lock" }}
+            - gem-cache-v1-{{ arch }}-{{ .Branch }}
+            - gem-cache-v1
+      - run: bundle install --path vendor/bundle
+      - run: bundle clean
+      - save_cache:
+          key: gem-cache-v1-{{ arch }}-{{ .Branch }}-{{ checksum "Gemfile.lock" }}
+          paths:
+            - vendor/bundle
+      - run: bundle exec rspec

data/.gitignore

@@ -6,5 +6,5 @@
 /pkg/
 /spec/reports/
 /tmp/
-
+/cache
 /.idea

data/.rspec

@@ -0,0 +1,2 @@
+-fd
+--require spec_helper

data/Gemfile

@@ -5,3 +5,4 @@ gemspec
 
 gem "rake", "~> 12.0"
 gem "minitest", "~> 5.0"
+gem "rspec"

data/Gemfile.lock

@@ -1,19 +1,33 @@
 PATH
   remote: .
   specs:
-    twitterscraper-ruby (0.10.0)
+    twitterscraper-ruby (0.15.0)
       nokogiri
       parallel
 
 GEM
   remote: https://rubygems.org/
   specs:
+    diff-lcs (1.4.4)
     mini_portile2 (2.4.0)
     minitest (5.14.1)
     nokogiri (1.10.10)
       mini_portile2 (~> 2.4.0)
     parallel (1.19.2)
     rake (12.3.3)
+    rspec (3.9.0)
+      rspec-core (~> 3.9.0)
+      rspec-expectations (~> 3.9.0)
+      rspec-mocks (~> 3.9.0)
+    rspec-core (3.9.2)
+      rspec-support (~> 3.9.3)
+    rspec-expectations (3.9.2)
+      diff-lcs (>= 1.2.0, < 2.0)
+      rspec-support (~> 3.9.0)
+    rspec-mocks (3.9.1)
+      diff-lcs (>= 1.2.0, < 2.0)
+      rspec-support (~> 3.9.0)
+    rspec-support (3.9.3)
 
 PLATFORMS
   ruby
@@ -21,6 +35,7 @@ PLATFORMS
 DEPENDENCIES
   minitest (~> 5.0)
   rake (~> 12.0)
+  rspec
   twitterscraper-ruby!
 
 BUNDLED WITH

data/README.md

@@ -1,5 +1,6 @@
 # twitterscraper-ruby
 
+[![Build Status](https://circleci.com/gh/ts-3156/twitterscraper-ruby.svg?style=svg)](https://circleci.com/gh/ts-3156/twitterscraper-ruby)
 [![Gem Version](https://badge.fury.io/rb/twitterscraper-ruby.svg)](https://badge.fury.io/rb/twitterscraper-ruby)
 
 A gem to scrape https://twitter.com/search. This gem is inspired by [taspinar/twitterscraper](https://github.com/taspinar/twitterscraper).
@@ -32,27 +33,39 @@ $ gem install twitterscraper-ruby
 Command-line interface:
 
 ```shell script
-$ twitterscraper --query KEYWORD --start_date 2020-06-01 --end_date 2020-06-30 --lang ja \
-      --limit 100 --threads 10 --proxy --output output.json
+# Returns a collection of relevant tweets matching a specified query.
+$ twitterscraper --type search --query KEYWORD --start_date 2020-06-01 --end_date 2020-06-30 --lang ja \
+      --limit 100 --threads 10 --output tweets.json
+```
+
+```shell script
+# Returns a collection of the most recent tweets posted by the user indicated by the screen_name
+$ twitterscraper --type user --query SCREEN_NAME --limit 100 --output tweets.json
 ```
 
 From Within Ruby:
 
 ```ruby
 require 'twitterscraper'
+client = Twitterscraper::Client.new(cache: true, proxy: true)
+```
+
+```ruby
+# Returns a collection of relevant tweets matching a specified query.
+tweets = client.search(KEYWORD, start_date: '2020-06-01', end_date: '2020-06-30', lang: 'ja', limit: 100, threads: 10)
+```
 
-options = {
-  start_date: '2020-06-01',
-  end_date:   '2020-06-30',
-  lang:       'ja',
-  limit:      100,
-  threads:    10,
-  proxy:      true
-}
+```ruby
+# Returns a collection of the most recent tweets posted by the user indicated by the screen_name
+tweets = client.user_timeline(SCREEN_NAME, limit: 100)
+```
 
-client = Twitterscraper::Client.new
-tweets = client.query_tweets(KEYWORD, options)
 
+## Attributes
+
+### Tweet
+
+```ruby
 tweets.each do |tweet|
   puts tweet.tweet_id
   puts tweet.text
@@ -64,11 +77,6 @@ tweets.each do |tweet|
 end
 ```
 
-
-## Attributes
-
-### Tweet
-
 - screen_name
 - name
 - user_id
@@ -136,15 +144,19 @@ $ cat tweets.json | jq . | less
 | Option | Description | Default |
 | ------------- | ------------- | ------------- |
 | `-h`, `--help` | This option displays a summary of twitterscraper. | |
+| `--type` | Specify a search type. | search |
 | `--query` | Specify a keyword used during the search. | |
-| `--start_date` | Set the date from which twitterscraper-ruby should start scraping for your query. | |
-| `--end_date` | Set the enddate which twitterscraper-ruby should use to stop scraping for your query. | |
+| `--start_date` | Used as "since:yyyy-mm-dd for your query. This means "since the date". | |
+| `--end_date` | Used as "until:yyyy-mm-dd for your query. This means "before the date". | |
 | `--lang` | Retrieve tweets written in a specific language. | |
 | `--limit` | Stop scraping when *at least* the number of tweets indicated with --limit is scraped. | 100 |
+| `--order` | Sort order of the results. | desc |
 | `--threads` | Set the number of threads twitterscraper-ruby should initiate while scraping for your query. | 2 |
-| `--proxy` | Scrape https://twitter.com/search via proxies. | false |
+| `--proxy` | Scrape https://twitter.com/search via proxies. | true |
+| `--cache` | Enable caching. | true |
 | `--format` | The format of the output. | json |
 | `--output` | The name of the output file. | tweets.json |
+| `--verbose` | Print debug messages. | tweets.json |
 
 
 ## Contributing

data/bin/twitterscraper

@@ -7,7 +7,7 @@ begin
   cli.parse
   cli.run
 rescue => e
-  STDERR.puts e.message
+  STDERR.puts e.inspect
   STDERR.puts e.backtrace.join("\n")
   exit 1
 end

data/lib/twitterscraper.rb

@@ -2,6 +2,7 @@ require 'twitterscraper/logger'
 require 'twitterscraper/proxy'
 require 'twitterscraper/http'
 require 'twitterscraper/lang'
+require 'twitterscraper/cache'
 require 'twitterscraper/query'
 require 'twitterscraper/client'
 require 'twitterscraper/tweet'

data/lib/twitterscraper/cache.rb

@@ -0,0 +1,69 @@
+require 'base64'
+require 'digest/md5'
+
+module Twitterscraper
+  class Cache
+    def initialize()
+      @ttl = 3600 # 1 hour
+      @dir = 'cache'
+      Dir.mkdir(@dir) unless File.exist?(@dir)
+    end
+
+    def read(key)
+      key = cache_key(key)
+      file = File.join(@dir, key)
+      entry = Entry.from_json(File.read(file))
+      entry.value if entry.time > Time.now - @ttl
+    rescue Errno::ENOENT => e
+      nil
+    end
+
+    def write(key, value)
+      key = cache_key(key)
+      entry = Entry.new(key, value, Time.now)
+      file = File.join(@dir, key)
+      File.write(file, entry.to_json)
+    end
+
+    def fetch(key, &block)
+      if (value = read(key))
+        value
+      else
+        yield.tap { |v| write(key, v) }
+      end
+    end
+
+    def cache_key(key)
+      value = key.gsub(':', '%3A').gsub('/', '%2F').gsub('?', '%3F').gsub('=', '%3D').gsub('&', '%26')
+      value = Digest::MD5.hexdigest(value) if value.length >= 100
+      value
+    end
+
+    class Entry < Hash
+      attr_reader :key, :value, :time
+
+      def initialize(key, value, time)
+        @key = key
+        @value = value
+        @time = time
+      end
+
+      def attrs
+        {key: @key, value: @value, time: @time}
+      end
+
+      def to_json
+        hash = attrs
+        hash[:value] = Base64.encode64(hash[:value])
+        hash.to_json
+      end
+
+      class << self
+        def from_json(text)
+          json = JSON.parse(text)
+          new(json['key'], Base64.decode64(json['value']), Time.parse(json['time']))
+        end
+      end
+    end
+  end
+end

data/lib/twitterscraper/cli.rb

@@ -16,14 +16,16 @@ module Twitterscraper
       print_version || return if print_version?
 
       query_options = {
+          type: options['type'],
           start_date: options['start_date'],
           end_date: options['end_date'],
           lang: options['lang'],
           limit: options['limit'],
+          daily_limit: options['daily_limit'],
+          order: options['order'],
           threads: options['threads'],
-          proxy: options['proxy']
       }
-      client = Twitterscraper::Client.new
+      client = Twitterscraper::Client.new(cache: options['cache'], proxy: options['proxy'])
       tweets = client.query_tweets(options['query'], query_options)
       export(tweets) unless tweets.empty?
     end
@@ -58,25 +60,36 @@ module Twitterscraper
           'help',
           'v',
           'version',
+          'type:',
           'query:',
           'start_date:',
           'end_date:',
           'lang:',
           'limit:',
+          'daily_limit:',
+          'order:',
           'threads:',
           'output:',
           'format:',
-          'proxy',
+          'cache:',
+          'proxy:',
           'pretty',
           'verbose',
       )
 
+      options['type'] ||= 'search'
+      options['start_date'] = Query::OLDEST_DATE if options['start_date'] == 'oldest'
       options['lang'] ||= ''
       options['limit'] = (options['limit'] || 100).to_i
+      options['daily_limit'] = options['daily_limit'].to_i if options['daily_limit']
       options['threads'] = (options['threads'] || 2).to_i
       options['format'] ||= 'json'
+      options['order'] ||= 'desc'
       options['output'] ||= "tweets.#{options['format']}"
 
+      options['cache'] = options['cache'] != 'false'
+      options['proxy'] = options['proxy'] != 'false'
+
       options
     end
 
@@ -100,7 +113,7 @@ module Twitterscraper
     end
 
     def print_version
-      puts "twitterscraper-#{Twitterscraper::VERSION}"
+      puts "twitterscraper-#{VERSION}"
     end
   end
 end

data/lib/twitterscraper/client.rb

@@ -1,5 +1,18 @@
 module Twitterscraper
   class Client
     include Query
+
+    def initialize(cache: true, proxy: true)
+      @cache = cache
+      @proxy = proxy
+    end
+
+    def cache_enabled?
+      @cache
+    end
+
+    def proxy_enabled?
+      @proxy
+    end
   end
 end

data/lib/twitterscraper/proxy.rb

@@ -17,15 +17,17 @@ module Twitterscraper
           reload
         end
         @cur_index += 1
-        item = @items[@cur_index - 1]
-        Twitterscraper.logger.info("Using proxy #{item}")
-        item
+        @items[@cur_index - 1]
       end
 
       def size
         @items.size
       end
 
+      def empty?
+        @items.empty?
+      end
+
       private
 
       def reload
@@ -51,7 +53,6 @@ module Twitterscraper
         proxies << ip + ':' + port
       end
 
-      Twitterscraper.logger.debug "Fetch #{proxies.size} proxies"
       proxies.shuffle
     rescue => e
       if (retries -= 1) > 0

data/lib/twitterscraper/query.rb

@@ -22,36 +22,41 @@ module Twitterscraper
     RELOAD_URL = 'https://twitter.com/i/search/timeline?f=tweets&vertical=' +
         'default&include_available_features=1&include_entities=1&' +
         'reset_error_state=false&src=typd&max_position=__POS__&q=__QUERY__&l=__LANG__'
-    INIT_URL_USER = 'https://twitter.com/{u}'
-    RELOAD_URL_USER = 'https://twitter.com/i/profiles/show/{u}/timeline/tweets?' +
+    INIT_URL_USER = 'https://twitter.com/__USER__'
+    RELOAD_URL_USER = 'https://twitter.com/i/profiles/show/__USER__/timeline/tweets?' +
         'include_available_features=1&include_entities=1&' +
-        'max_position={pos}&reset_error_state=false'
-
-    def build_query_url(query, lang, pos, from_user = false)
-      # if from_user
-      #   if !pos
-      #     INIT_URL_USER.format(u = query)
-      #   else
-      #     RELOAD_URL_USER.format(u = query, pos = pos)
-      #   end
-      # end
-      if pos
-        RELOAD_URL.sub('__QUERY__', query).sub('__LANG__', lang.to_s).sub('__POS__', pos)
+        'max_position=__POS__&reset_error_state=false'
+
+    def build_query_url(query, lang, from_user, pos)
+      if from_user
+        if pos
+          RELOAD_URL_USER.sub('__USER__', query).sub('__POS__', pos.to_s)
+        else
+          INIT_URL_USER.sub('__USER__', query)
+        end
       else
-        INIT_URL.sub('__QUERY__', query).sub('__LANG__', lang.to_s)
+        if pos
+          RELOAD_URL.sub('__QUERY__', query).sub('__LANG__', lang.to_s).sub('__POS__', pos)
+        else
+          INIT_URL.sub('__QUERY__', query).sub('__LANG__', lang.to_s)
+        end
       end
     end
 
     def get_single_page(url, headers, proxies, timeout = 6, retries = 30)
       return nil if stop_requested?
-      Twitterscraper::Http.get(url, headers, proxies.sample, timeout)
+      unless proxies.empty?
+        proxy = proxies.sample
+        logger.info("Using proxy #{proxy}")
+      end
+      Http.get(url, headers, proxy, timeout)
     rescue => e
       logger.debug "query_single_page: #{e.inspect}"
       if (retries -= 1) > 0
-        logger.info("Retrying... (Attempts left: #{retries - 1})")
+        logger.info "Retrying... (Attempts left: #{retries - 1})"
         retry
       else
-        raise
+        raise Error.new("#{e.inspect} url=#{url}")
       end
     end
 
@@ -70,15 +75,28 @@ module Twitterscraper
       [items_html, json_resp]
     end
 
-    def query_single_page(query, lang, pos, from_user = false, headers: [], proxies: [])
-      logger.info("Querying #{query}")
+    def query_single_page(query, lang, type, pos, headers: [], proxies: [])
+      logger.info "Querying #{query}"
       query = ERB::Util.url_encode(query)
 
-      url = build_query_url(query, lang, pos, from_user)
-      logger.debug("Scraping tweets from #{url}")
+      url = build_query_url(query, lang, type == 'user', pos)
+      http_request = lambda do
+        logger.debug "Scraping tweets from #{url}"
+        get_single_page(url, headers, proxies)
+      end
 
-      response = get_single_page(url, headers, proxies)
-      return [], nil if response.nil?
+      if cache_enabled?
+        client = Cache.new
+        if (response = client.read(url))
+          logger.debug 'Fetching tweets from cache'
+        else
+          response = http_request.call
+          client.write(url, response) unless stop_requested?
+        end
+      else
+        response = http_request.call
+      end
+      return [], nil if response.nil? || response.empty?
 
       html, json_resp = parse_single_page(response, pos.nil?)
 
@@ -90,8 +108,8 @@ module Twitterscraper
 
       if json_resp
         [tweets, json_resp['min_position']]
-      elsif from_user
-        raise NotImplementedError
+      elsif type
+        [tweets, tweets[-1].tweet_id]
       else
         [tweets, "TWEET-#{tweets[-1].tweet_id}-#{tweets[0].tweet_id}"]
       end
@@ -99,33 +117,34 @@ module Twitterscraper
 
     OLDEST_DATE = Date.parse('2006-03-21')
 
-    def validate_options!(query, start_date:, end_date:, lang:, limit:, threads:, proxy:)
+    def validate_options!(queries, type:, start_date:, end_date:, lang:, limit:, threads:)
+      query = queries[0]
       if query.nil? || query == ''
-        raise 'Please specify a search query.'
+        raise Error.new('Please specify a search query.')
       end
 
       if ERB::Util.url_encode(query).length >= 500
-        raise ':query must be a UTF-8, URL-encoded search query of 500 characters maximum, including operators.'
+        raise Error.new(':query must be a UTF-8, URL-encoded search query of 500 characters maximum, including operators.')
       end
 
       if start_date && end_date
         if start_date == end_date
-          raise 'Please specify different values for :start_date and :end_date.'
+          raise Error.new('Please specify different values for :start_date and :end_date.')
         elsif start_date > end_date
-          raise ':start_date must occur before :end_date.'
+          raise Error.new(':start_date must occur before :end_date.')
         end
       end
 
       if start_date
         if start_date < OLDEST_DATE
-          raise ":start_date must be greater than or equal to #{OLDEST_DATE}"
+          raise Error.new(":start_date must be greater than or equal to #{OLDEST_DATE}")
         end
       end
 
       if end_date
         today = Date.today
         if end_date > Date.today
-          raise ":end_date must be less than or equal to today(#{today})"
+          raise Error.new(":end_date must be less than or equal to today(#{today})")
         end
       end
     end
@@ -143,27 +162,32 @@ module Twitterscraper
       end
     end
 
-    def main_loop(query, lang, limit, headers, proxies)
+    def main_loop(query, lang, type, limit, daily_limit, headers, proxies)
       pos = nil
+      daily_tweets = []
 
       while true
-        new_tweets, new_pos = query_single_page(query, lang, pos, headers: headers, proxies: proxies)
+        new_tweets, new_pos = query_single_page(query, lang, type, pos, headers: headers, proxies: proxies)
         unless new_tweets.empty?
+          daily_tweets.concat(new_tweets)
+          daily_tweets.uniq! { |t| t.tweet_id }
+
           @mutex.synchronize {
             @all_tweets.concat(new_tweets)
             @all_tweets.uniq! { |t| t.tweet_id }
           }
         end
-        logger.info("Got #{new_tweets.size} tweets (total #{@all_tweets.size})")
+        logger.info "Got #{new_tweets.size} tweets (total #{@all_tweets.size})"
 
         break unless new_pos
+        break if daily_limit && daily_tweets.size >= daily_limit
         break if @all_tweets.size >= limit
 
         pos = new_pos
       end
 
-      if @all_tweets.size >= limit
-        logger.info("Limit reached #{@all_tweets.size}")
+      if !@stop_requested && @all_tweets.size >= limit
+        logger.warn "The limit you specified has been reached limit=#{limit} tweets=#{@all_tweets.size}"
         @stop_requested = true
       end
     end
@@ -172,37 +196,59 @@ module Twitterscraper
       @stop_requested
     end
 
-    def query_tweets(query, start_date: nil, end_date: nil, lang: '', limit: 100, threads: 2, proxy: false)
+    def query_tweets(query, type: 'search', start_date: nil, end_date: nil, lang: nil, limit: 100, daily_limit: nil, order: 'desc', threads: 2)
       start_date = Date.parse(start_date) if start_date && start_date.is_a?(String)
       end_date = Date.parse(end_date) if end_date && end_date.is_a?(String)
       queries = build_queries(query, start_date, end_date)
-      threads = queries.size if threads > queries.size
-      proxies = proxy ? Twitterscraper::Proxy::Pool.new : []
+      if threads > queries.size
+        logger.warn 'The maximum number of :threads is the number of dates between :start_date and :end_date.'
+        threads = queries.size
+      end
+      if proxy_enabled?
+        proxies = Proxy::Pool.new
+        logger.debug "Fetch #{proxies.size} proxies"
+      else
+        proxies = []
+        logger.debug 'Proxy disabled'
+      end
+      logger.debug "Cache #{cache_enabled? ? 'enabled' : 'disabled'}"
 
-      validate_options!(queries[0], start_date: start_date, end_date: end_date, lang: lang, limit: limit, threads: threads, proxy: proxy)
 
-      logger.info("The number of threads #{threads}")
+      validate_options!(queries, type: type, start_date: start_date, end_date: end_date, lang: lang, limit: limit, threads: threads)
+
+      logger.info "The number of threads #{threads}"
 
       headers = {'User-Agent': USER_AGENT_LIST.sample, 'X-Requested-With': 'XMLHttpRequest'}
-      logger.info("Headers #{headers}")
+      logger.info "Headers #{headers}"
 
       @all_tweets = []
       @mutex = Mutex.new
       @stop_requested = false
 
       if threads > 1
+        Thread.abort_on_exception = true
+        logger.debug "Set 'Thread.abort_on_exception' to true"
+
         Parallel.each(queries, in_threads: threads) do |query|
-          main_loop(query, lang, limit, headers, proxies)
+          main_loop(query, lang, type, limit, daily_limit, headers, proxies)
           raise Parallel::Break if stop_requested?
         end
       else
         queries.each do |query|
-          main_loop(query, lang, limit, headers, proxies)
+          main_loop(query, lang, type, limit, daily_limit, headers, proxies)
           break if stop_requested?
         end
       end
 
-      @all_tweets.sort_by { |tweet| -tweet.created_at.to_i }
+      @all_tweets.sort_by { |tweet| (order == 'desc' ? -1 : 1) * tweet.created_at.to_i }
+    end
+
+    def search(query, start_date: nil, end_date: nil, lang: '', limit: 100, daily_limit: nil, order: 'desc', threads: 2)
+      query_tweets(query, type: 'search', start_date: start_date, end_date: end_date, lang: lang, limit: limit, daily_limit: daily_limit, order: order, threads: threads)
+    end
+
+    def user_timeline(screen_name, limit: 100, order: 'desc')
+      query_tweets(screen_name, type: 'user', start_date: nil, end_date: nil, lang: nil, limit: limit, daily_limit: nil, order: order, threads: 1)
     end
   end
 end

data/lib/twitterscraper/tweet.rb

@@ -43,6 +43,14 @@ module Twitterscraper
     end
 
     class << self
+      def from_json(text)
+        json = JSON.parse(text)
+        json.map do |tweet|
+          tweet['created_at'] = Time.parse(tweet['created_at'])
+          new(tweet)
+        end
+      end
+
       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')]]"))
@@ -51,12 +59,19 @@ module Twitterscraper
       def from_tweets_html(html)
         html.map do |tweet|
           from_tweet_html(tweet)
-        end
+        end.compact
       end
 
       def from_tweet_html(html)
+        screen_name = html.attr('data-screen-name')
+        tweet_id = html.attr('data-tweet-id')&.to_i
+
+        unless html.to_s.include?('js-tweet-text-container')
+          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)
-        tweet_id = html.attr('data-tweet-id').to_i
         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') }
@@ -81,7 +96,7 @@ module Twitterscraper
 
         timestamp = inner_html.xpath("//span[@class[contains(., 'js-short-timestamp')]]").first.attr('data-time').to_i
         new(
-            screen_name: html.attr('data-screen-name'),
+            screen_name: screen_name,
             name: html.attr('data-name'),
             user_id: html.attr('data-user-id').to_i,
             tweet_id: tweet_id,

data/lib/version.rb

@@ -1,3 +1,3 @@
 module Twitterscraper
-  VERSION = '0.10.0'
+  VERSION = '0.15.0'
 end

metadata

@@ -1,14 +1,14 @@
 --- !ruby/object:Gem::Specification
 name: twitterscraper-ruby
 version: !ruby/object:Gem::Version
-  version: 0.10.0
+  version: 0.15.0
 platform: ruby
 authors:
 - ts-3156
 autorequire: 
 bindir: bin
 cert_chain: []
-date: 2020-07-13 00:00:00.000000000 Z
+date: 2020-07-17 00:00:00.000000000 Z
 dependencies:
 - !ruby/object:Gem::Dependency
   name: nokogiri
@@ -46,8 +46,10 @@ executables:
 extensions: []
 extra_rdoc_files: []
 files:
+- ".circleci/config.yml"
 - ".gitignore"
 - ".irbrc"
+- ".rspec"
 - ".ruby-version"
 - ".travis.yml"
 - CODE_OF_CONDUCT.md
@@ -61,6 +63,7 @@ files:
 - bin/twitterscraper
 - lib/twitterscraper-ruby.rb
 - lib/twitterscraper.rb
+- lib/twitterscraper/cache.rb
 - lib/twitterscraper/cli.rb
 - lib/twitterscraper/client.rb
 - lib/twitterscraper/http.rb