Class: AkaneBigquery::CLI

Inherits:
Thor
  • Object
show all
Defined in:
lib/akane-bigquery/cli.rb

Instance Method Summary collapse

Instance Method Details

#init ⇒ Object 



16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
 
# File 'lib/akane-bigquery/cli.rb', line 16

def init
  # check dataset existence
  dataset =  client.execute(
    api_method: api.datasets.get,
    parameters: {
      'projectId' => config['project_id'],
      'datasetId' => config['dataset_id'],
    }
  )

  if dataset.error?
    if dataset.error_message =~ /^Not Found:/i
      puts "Creating dataset #{config['dataset_id']} ..."
      dataset = client.execute(
        api_method: api.datasets.insert,
        parameters: {
          'projectId' => config['project_id'],
        },
        body_object: {
          'datasetReference' => {
            'datasetId' => config['dataset_id'],
          },
          'description' => 'akane',
        }
      )

      raise dataset.error_message if dataset.error?
    else
      raise dataset.error_message
    end
  end

  schemas = AkaneBigquery::Schema::SCHEMA

  schemas.each do |table_id, schema|
    table = client.execute(
      api_method: api.tables.get,
      parameters: {
        'projectId' => config['project_id'],
        'datasetId' => config['dataset_id'],
        'tableId' => table_id,
      },
    )

    if table.error?
      if table.error_message =~ /^Not Found:/i
        puts "Creating table #{table_id} ..."
        table = client.execute(
          api_method: api.tables.insert,
          parameters: {
            'projectId' => config['project_id'],
            'datasetId' => config['dataset_id'],
          },
          body_object: {
            'tableReference' => {
              'projectId' => config['project_id'],
              'datasetId' => config['dataset_id'],
              'tableId' => table_id,
            },
            'friendlyName' => table_id,
            'schema' => schema,
          }
        )
        raise table.error_message if table.error?
      else
        raise table.error_message
      end
    end

  end
end

#prepare(source, prefix) ⇒ Object 



91
92
93
94
95
96
97
98
99
100
101
102
103
104
105
106
107
108
109
110
111
112
113
114
115
116
117
118
119
120
121
122
123
124
125
126
127
128
129
130
131
132
133
134
135
136
137
138
139
140
141
142
143
144
145
146
147
148
149
150
151
152
153
154
155
156
157
158
159
160
161
162
163
164
165
166
167
168
169
170
171
172
173
174
175
176
177
178
179
180
181
182
183
184
 
# File 'lib/akane-bigquery/cli.rb', line 91

def prepare(source, prefix)
  limit = 524288000 # 500MBytes

  count = -1
  bytes = 0

  new_io = lambda do
    bytes = 0
    count += 1
    path = File.join(prefix, "tweets.#{count.to_s.rjust(4,'0')}.txt")
    puts "=> Using #{path}"
    File.open(path, 'w')
  end
  io = new_io.call

  months = options[:months] && options[:months].split(/,/)
  before = options[:before] && Time.parse(options[:before])

  userdirs = Dir.entries(File.join(source, "users"))
  userdirs.each_with_index do |user_dirname, index|
    next if user_dirname == "." || user_dirname == ".."
    puts " * #{user_dirname} (#{index.succ}/#{userdirs.size}, #{((index.succ/userdirs.size.to_f)*100).to_i}%)"

    userdir = File.join(source, "users", user_dirname)

    tweet_filepaths = if options[:months]
                        months.map { |_| File.join(userdir, "tweets.#{_}.txt") }
                      else
                        Dir[File.join(userdir, 'tweets.*.txt')]
                      end
    tweet_filepaths.each do |file|
      begin
        File.open(file, 'r') do |tweets_io|
          tweets_io.each_line do |line|
            json = line.chomp

            tweet  = Oj.load(json)

            created_at = Time.parse(tweet['created_at'.freeze])
            next if before && before <= created_at

            new_json = {
              'json'.freeze => json,
              'id_str'.freeze => tweet['id_str'.freeze],
              'id'.freeze => tweet['id'.freeze],
              'text'.freeze => tweet['text'.freeze],
              'lang'.freeze => tweet['lang'.freeze],
              'source'.freeze => tweet['source'.freeze],
              'in_reply_to_status_id'.freeze => tweet['in_reply_to_status_id'.freeze],
              'in_reply_to_status_id_str'.freeze => tweet['in_reply_to_status_id_str'.freeze],
              'in_reply_to_user_id'.freeze => tweet['in_reply_to_user_id'.freeze],
              'in_reply_to_user_id_str'.freeze => tweet['in_reply_to_user_id_str'.freeze],
              'in_reply_to_screen_name'.freeze => tweet['in_reply_to_screen_name'.freeze],
              'user'.freeze => {
                'id_str'.freeze => tweet['user'.freeze]['id_str'.freeze],
                'id'.freeze => tweet['user'.freeze]['id'.freeze],
                'name'.freeze => tweet['user'.freeze]['name'.freeze],
                'screen_name'.freeze => tweet['user'.freeze]['screen_name'.freeze],
                'protected'.freeze => tweet['user'.freeze]['protected'.freeze],
              },
              'created_at'.freeze => created_at.to_i
            }

            if tweet['coordinates'.freeze]
              new_json['coordinates_longitude'.freeze] = tweet['coordinates'.freeze]['coordinates'.freeze][0]
              new_json['coordinates_latitude'.freeze] = tweet['coordinates'.freeze]['coordinates'.freeze][1]
            end

            if tweet['place'.freeze]
              place = tweet['place'.freeze]
              new_json['place'.freeze] = {
                'id'.freeze => place['id'.freeze],
                'country'.freeze => place['country'.freeze],
                'country_code'.freeze => place['country_code'.freeze],
                'name'.freeze => place['name'.freeze],
                'full_name'.freeze => place['full_name'.freeze],
                'place_type'.freeze => place['place_type'.freeze],
                'url'.freeze => place['url'.freeze],
              }
            end

            new_json_str = Oj.dump(new_json)
            io.puts new_json_str
            bytes += new_json_str.size + 1
            io = new_io.call if limit <= bytes
          end
        end
      rescue Errno::ENOENT
      end

    end

  end
end