ruby Twitter Webからユーザの発言日时を拾う

Posted 2021-05-16

require 'mechanize'
require 'pp'

user_name = 'innocent_zero'
url = 'https://twitter.com/' + user_name

a = Mechanize.new {|agent|
  agent.user_agent_alias = 'Mac Safari'
}

a.request_headers = {
  'accept-language' => 'ja, ja-JP'
}

page = a.get(url)

xpath = '//*/div/span/a[@class="ProfileTweet-timestamp js-permalink js-nav js-tooltip"]/@title'
regex = %r{(\d+:\d+).+(\d{4,}).(\d+).(\d+).}

pp page.search(xpath).map{|x|
  m = x.value.scan(regex).to_a.flatten
  # may be responses timezone is UTC -0700
  Time.parse("#{m[1]}/#{m[2]}/#{m[3]} #{m[0]} -0700")
}
=begin
[2014-08-15 22:22:00 +0900,
 2014-08-15 22:21:00 +0900,
 2014-08-15 22:20:00 +0900,
 2014-08-15 22:15:00 +0900,
 2014-08-15 22:14:00 +0900,
 2014-08-15 22:13:00 +0900,
 2014-08-15 22:11:00 +0900,
 2014-08-15 22:09:00 +0900,
 2014-08-15 22:06:00 +0900,
 2014-08-15 21:04:00 +0900,
 2014-08-15 18:43:00 +0900,
 2014-08-15 17:17:00 +0900,
 2014-08-15 18:36:00 +0900,
 2014-08-15 18:35:00 +0900,
 2014-08-15 18:33:00 +0900,
 2014-08-15 17:19:00 +0900,
 2014-08-15 16:38:00 +0900,
 2014-08-15 16:35:00 +0900,
 2014-08-15 16:30:00 +0900,
 2014-08-15 15:50:00 +0900]
=end