Created
January 25, 2013 18:40
-
-
Save hungtatai/4636786 to your computer and use it in GitHub Desktop.
This file contains bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
# encoding: utf-8 | |
require 'nokogiri' | |
require 'httparty' | |
require 'pp' | |
base = 'http://www.ptt.cc' | |
url = 'http://www.ptt.cc/bbs/joke/index.html' | |
yesterday = Date.today - 1 | |
stop_flag = false | |
topics = [] | |
until stop_flag do | |
res = HTTParty.get(url).to_s.encode('utf-8') | |
doc = Nokogiri::HTML(res) | |
prev_page = doc.search('h2 a')[1]['href'] | |
doc.search('div#prodlist tr').reverse.each do |tr| | |
a = tr.search('td')[-1].children[0] | |
like_count = tr.search('td')[2].content.to_i | |
post_time = tr.search('td')[3].content.to_s.strip.split('/') | |
post_time = Date.new(Date.today.year, post_time[0].to_i, post_time[1].to_i) | |
title = a.content | |
href = base + a['href'] | |
if yesterday < post_time then | |
next | |
elsif yesterday == post_time | |
if like_count > 10 then | |
#topic_content = Nokogiri::HTML(HTTParty.get(href).to_s.force_encoding('utf-8')).search('div#mainContent pre').to_s | |
topic_content = "test" | |
topics.push({ | |
:post_time => post_time, | |
:title => title, | |
:like_count => like_count, | |
:href => href, | |
:topic_content => topic_content | |
}) | |
puts post_time.to_s+"\t"+title+"\t"+like_count.to_s+"\t"+href | |
#puts topic_content | |
end | |
else | |
stop_flag = true | |
break | |
end | |
end | |
url = base + prev_page | |
end | |
require 'rss' | |
require 'time' | |
rss_content = RSS::Maker.make('2.0') do |maker| | |
maker.channel.author = "Honda" | |
maker.channel.title = "Joke "+yesterday.to_s | |
maker.channel.link = base | |
maker.channel.updated = Time.now.to_s | |
maker.channel.description = "desc" | |
topics.each do |topic| | |
item = maker.items.new_item | |
item.title = "[${topic[:like_count]}]"+topic[:title] | |
item.link = topic[:href] | |
item.description = topic[:topic_content] | |
item.date = Time.parse(topic[:post_time].to_s) | |
end | |
end | |
#puts rss_content.encoding.name | |
#puts rss_content.to_s.encode("big5") | |
File.open('joke.rss', 'w') do |file| | |
file.write(rss_content.to_s) | |
end | |
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment