6 define_structure :Url, :channel, :nick, :time, :url, :info
8 class UrlPlugin < Plugin
9 LINK_INFO = "[Link Info]"
10 OUR_UNSAFE = Regexp.new("[^#{URI::PATTERN::UNRESERVED}#{URI::PATTERN::RESERVED}%# ]", false, 'N')
12 Config.register Config::IntegerValue.new('url.max_urls',
13 :default => 100, :validate => Proc.new{|v| v > 0},
14 :desc => "Maximum number of urls to store. New urls replace oldest ones.")
15 Config.register Config::IntegerValue.new('url.display_link_info',
17 :desc => "Get the title of links pasted to the channel and display it (also tells if the link is broken or the site is down). Do it for at most this many links per line (set to 0 to disable)")
18 Config.register Config::BooleanValue.new('url.titles_only',
20 :desc => "Only show info for links that have <title> tags (in other words, don't display info for jpegs, mpegs, etc.)")
21 Config.register Config::BooleanValue.new('url.first_par',
23 :desc => "Also try to get the first paragraph of a web page")
24 Config.register Config::BooleanValue.new('url.info_on_list',
26 :desc => "Show link info when listing/searching for urls")
27 Config.register Config::ArrayValue.new('url.no_info_hosts',
28 :default => ['localhost', '^192\.168\.', '^10\.', '^127\.', '^172\.(1[6-9]|2\d|31)\.'],
29 :on_change => Proc.new { |bot, v| bot.plugins['url'].reset_no_info_hosts },
30 :desc => "A list of regular expressions matching hosts for which no info should be provided")
31 Config.register Config::ArrayValue.new('url.only_on_channels',
32 :desc => "Show link info only on these channels",
37 @registry.set_default(Array.new)
38 unless @bot.config['url.display_link_info'].kind_of?(Integer)
39 @bot.config.items[:'url.display_link_info'].set_string(@bot.config['url.display_link_info'].to_s)
44 def reset_no_info_hosts
45 @no_info_hosts = Regexp.new(@bot.config['url.no_info_hosts'].join('|'), true)
46 debug "no info hosts regexp set to #{@no_info_hosts}"
49 def help(plugin, topic="")
50 "url info <url> => display link info for <url> (set url.display_link_info > 0 if you want the bot to do it automatically when someone writes an url), urls [<max>=4] => list <max> last urls mentioned in current channel, urls search [<max>=4] <regexp> => search for matching urls. In a private message, you must specify the channel to query, eg. urls <channel> [max], urls search <channel> [max] <regexp>"
53 def get_title_from_html(pagedata)
54 return pagedata.ircify_html_title
57 def get_title_for_url(uri_str, opts = {})
59 url = uri_str.kind_of?(URI) ? uri_str : URI.parse(uri_str)
60 return if url.scheme !~ /https?/
62 # also check the ip, the canonical name and the aliases
64 checks = TCPSocket.gethostbyname(url.host)
67 return "Unable to retrieve info for #{url.host}: #{e.message}"
73 unless checks.grep(@no_info_hosts).empty?
74 return "Sorry, info retrieval for #{url.host} (#{checks.first}) is disabled"
83 debug "+ getting info for #{url.request_uri}"
84 info = @bot.filter(:htmlinfo, url)
88 logopts[:title] = title = info[:title]
91 logopts[:extra] = info[:content]
92 extra << "#{Bold}text#{Bold}: #{info[:content]}" if @bot.config['url.first_par']
94 logopts[:extra] = String.new
95 logopts[:extra] << "Content Type: #{resp['content-type']}"
96 extra << "#{Bold}type#{Bold}: #{resp['content-type']}" unless title
97 if enc = resp['content-encoding']
98 logopts[:extra] << ", encoding: #{enc}"
99 extra << "#{Bold}encoding#{Bold}: #{enc}" if @bot.config['url.first_par'] or not title
102 size = resp['content-length'].first.gsub(/(\d)(?=\d{3}+(?:\.|$))(\d{3}\..*)?/,'\1,\2') rescue nil
104 logopts[:extra] << ", size: #{size} bytes"
105 extra << "#{Bold}size#{Bold}: #{size} bytes" if @bot.config['url.first_par'] or not title
108 rescue Exception => e
114 raise "connecting to site/processing information (#{e.message})"
118 call_event(:url_added, url.to_s, logopts)
120 extra.unshift("#{Bold}title#{Bold}: #{title}")
122 return extra.join(", ") if title or not @bot.config['url.titles_only']
125 def handle_urls(m, urls, display_info=@bot.config['url.display_link_info'])
126 unless (channels = @bot.config['url.only_on_channels']).empty?
127 return unless channels.map { |c| c.downcase }.include?(m.channel.downcase)
130 return if urls.empty?
131 debug "found urls #{urls.inspect}"
132 list = m.public? ? @registry[m.target] : nil
133 debug "display link info: #{display_info}"
135 urls.each do |urlstr|
136 debug "working on #{urlstr}"
137 next unless urlstr =~ /^https?:\/\/./
139 debug "Getting title for #{urlstr}..."
142 title = get_title_for_url(urlstr,
143 :nick => m.source.nick,
144 :channel => m.channel,
145 :ircline => m.message)
146 debug "Title #{title ? '' : 'not '} found"
147 reply = "#{LINK_INFO} #{title}" if title
150 # we might get a 404 because of trailing punctuation, so we try again
151 # with the last character stripped. this might generate invalid URIs
152 # (e.g. because "some.url" gets chopped to some.url%2, so catch that too
153 if e.message =~ /\(404 - Not Found\)/i or e.kind_of?(URI::InvalidURIError)
154 # chop off last character, and retry if we still have enough string to
155 # look like a minimal URL
156 retry if urlstr.chop! and urlstr =~ /^https?:\/\/./
158 reply = "Error #{e.message}"
161 if display_info > urls_displayed
163 m.reply reply, :overlong => :truncate, :to => :public,
164 :nick => (m.address? ? :auto : false)
171 # check to see if this url is already listed
172 next if list.find {|u| u.url == urlstr }
174 url = Url.new(m.target, m.sourcenick, Time.new, urlstr, title)
175 debug "#{list.length} urls so far"
176 list.pop if list.length > @bot.config['url.max_urls']
177 debug "storing url #{url.url}"
179 debug "#{list.length} urls now"
181 @registry[m.target] = list
185 escaped = URI.escape(params[:urls].to_s, OUR_UNSAFE)
186 urls = URI.extract(escaped)
187 Thread.new { handle_urls(m, urls, params[:urls].length) }
193 escaped = URI.escape(m.message, OUR_UNSAFE)
194 urls = URI.extract(escaped, ['http', 'https'])
195 return if urls.empty?
196 Thread.new { handle_urls(m, urls) }
199 def reply_urls(opts={})
202 channel = opts[:channel]
204 return unless list and max and m
205 list[0..(max-1)].each do |url|
206 disp = "[#{url.time.strftime('%Y/%m/%d %H:%M:%S')}] <#{url.nick}> #{url.url}"
207 if @bot.config['url.info_on_list']
209 get_title_for_url(url.url,
210 :nick => url.nick, :channel => channel) rescue nil
211 # If the url info was missing and we now have some, try to upgrade it
212 if channel and title and not url.info
213 ll = @registry[channel]
215 if el = ll.find { |u| u.url == url.url }
217 @registry[channel] = ll
220 disp << " --> #{title}" if title
222 m.reply disp, :overlong => :truncate
227 channel = params[:channel] ? params[:channel] : m.target
228 max = params[:limit].to_i
231 list = @registry[channel]
233 m.reply "no urls seen yet for channel #{channel}"
235 reply_urls :msg => m, :channel => channel, :list => list, :max => max
239 def search(m, params)
240 channel = params[:channel] ? params[:channel] : m.target
241 max = params[:limit].to_i
242 string = params[:string]
245 regex = Regexp.new(string, Regexp::IGNORECASE)
246 list = @registry[channel].find_all {|url|
247 regex.match(url.url) || regex.match(url.nick) ||
248 (@bot.config['url.info_on_list'] && regex.match(url.info))
251 m.reply "no matches for channel #{channel}"
253 reply_urls :msg => m, :channel => channel, :list => list, :max => max
258 plugin = UrlPlugin.new
259 plugin.map 'urls info *urls', :action => 'info'
260 plugin.map 'url info *urls', :action => 'info'
261 plugin.map 'urls search :channel :limit :string', :action => 'search',
262 :defaults => {:limit => 4},
263 :requirements => {:limit => /^\d+$/},
265 plugin.map 'urls search :limit :string', :action => 'search',
266 :defaults => {:limit => 4},
267 :requirements => {:limit => /^\d+$/},
269 plugin.map 'urls :channel :limit', :defaults => {:limit => 4},
270 :requirements => {:limit => /^\d+$/},
272 plugin.map 'urls :limit', :defaults => {:limit => 4},
273 :requirements => {:limit => /^\d+$/},