1 define_structure :Url, :channel, :nick, :time, :url, :info
3 class ::UrlLinkError < RuntimeError
6 class UrlPlugin < Plugin
7 TITLE_RE = /<\s*?title\s*?>(.+?)<\s*?\/title\s*?>/im
8 LINK_INFO = "[Link Info]"
10 BotConfig.register BotConfigIntegerValue.new('url.max_urls',
11 :default => 100, :validate => Proc.new{|v| v > 0},
12 :desc => "Maximum number of urls to store. New urls replace oldest ones.")
13 BotConfig.register BotConfigBooleanValue.new('url.display_link_info',
15 :desc => "Get the title of any links pasted to the channel and display it (also tells if the link is broken or the site is down)")
16 BotConfig.register BotConfigBooleanValue.new('url.titles_only',
18 :desc => "Only show info for links that have <title> tags (in other words, don't display info for jpegs, mpegs, etc.)")
19 BotConfig.register BotConfigBooleanValue.new('url.first_par',
21 :desc => "Also try to get the first paragraph of a web page")
22 BotConfig.register BotConfigBooleanValue.new('url.info_on_list',
24 :desc => "Show link info when listing/searching for urls")
29 @registry.set_default(Array.new)
32 def help(plugin, topic="")
33 "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>"
36 def get_title_from_html(pagedata)
37 return unless TITLE_RE.match(pagedata)
41 def get_title_for_url(uri_str, nick = nil, channel = nil)
43 url = uri_str.kind_of?(URI) ? uri_str : URI.parse(uri_str)
44 return if url.scheme !~ /https?/
47 logopts[:nick] = nick if nick
48 logopts[:channel] = channel if channel
54 debug "+ getting #{url.request_uri}"
55 @bot.httputil.get_response(url) { |resp|
61 if resp['content-type'] =~ /^text\/|(?:x|ht)ml/
62 # The page is text or HTML, so we can try finding a title and, if
63 # requested, the first par.
65 # We act differently depending on whether we want the first par or
66 # not: in the first case we download the initial part and the parse
67 # it; in the second case we only download as much as we need to find
70 if @bot.config['url.first_par']
71 partial = resp.partial_body(@bot.config['http.info_bytes'])
72 logopts[:title] = title = get_title_from_html(partial)
73 if url.fragment and not url.fragment.empty?
74 fragreg = /.*?<a\s+[^>]*name=["']?#{url.fragment}["']?.*?>/im
75 partial.sub!(fragreg,'')
77 first_par = Utils.ircify_first_html_par(partial, :strip => title)
78 unless first_par.empty?
79 logopts[:extra] = first_par
80 extra << ", #{Bold}text#{Bold}: #{first_par}"
82 call_event(:url_added, url.to_s, logopts)
83 return "#{Bold}title#{Bold}: #{title}#{extra}" if title
85 resp.partial_body(@bot.config['http.info_bytes']) { |part|
86 logopts[:title] = title = get_title_from_html(part)
87 call_event(:url_added, url.to_s, logopts)
88 return "#{Bold}title#{Bold}: #{title}" if title
91 # if nothing was found, provide more basic info, as for non-html pages
96 enc = resp['content-encoding']
97 logopts[:extra] = String.new
98 logopts[:extra] << "Content Type: #{resp['content-type']}"
100 logopts[:extra] << ", encoding: #{enc}"
101 extra << ", #{Bold}encoding#{Bold}: #{enc}"
104 unless @bot.config['url.titles_only']
105 # content doesn't have title, just display info.
106 size = resp['content-length'].gsub(/(\d)(?=\d{3}+(?:\.|$))(\d{3}\..*)?/,'\1,\2') rescue nil
108 logopts[:extra] << ", size: #{size} bytes"
109 size = ", #{Bold}size#{Bold}: #{size} bytes"
111 call_event(:url_added, url.to_s, logopts)
112 return "#{Bold}type#{Bold}: #{resp['content-type']}#{size}#{extra}"
114 call_event(:url_added, url.to_s, logopts)
116 raise UrlLinkError, "getting link (#{resp.code} - #{resp.message})"
120 rescue Exception => e
126 raise "connecting to site/processing information (#{e.message})"
132 return unless m.kind_of?(PrivMessage)
134 urls = URI.extract(m.message)
135 return if urls.empty?
136 debug "found urls #{urls.inspect}"
137 list = @registry[m.target]
139 debug "working on #{urlstr}"
140 next unless urlstr =~ /^https?:/
142 if @bot.config['url.display_link_info']
144 debug "Getting title for #{urlstr}..."
146 title = get_title_for_url urlstr, m.source.nick, m.channel
148 m.reply "#{LINK_INFO} #{title}", :overlong => :truncate
151 debug "Title not found!"
154 m.reply "Error #{e.message}"
159 # check to see if this url is already listed
160 next if list.find {|u| u.url == urlstr }
162 url = Url.new(m.target, m.sourcenick, Time.new, urlstr, title)
163 debug "#{list.length} urls so far"
164 if list.length > @bot.config['url.max_urls']
167 debug "storing url #{url.url}"
169 debug "#{list.length} urls now"
171 @registry[m.target] = list
174 def reply_urls(opts={})
177 channel = opts[:channel]
179 return unless list and max and m
180 list[0..(max-1)].each do |url|
181 disp = "[#{url.time.strftime('%Y/%m/%d %H:%M:%S')}] <#{url.nick}> #{url.url}"
182 if @bot.config['url.info_on_list']
183 title = url.info || get_title_for_url(url.url, url.nick, channel) rescue nil
184 # If the url info was missing and we now have some, try to upgrade it
185 if channel and title and not url.info
186 ll = @registry[channel]
188 if el = ll.find { |u| u.url == url.url }
190 @registry[channel] = ll
193 disp << " --> #{title}" if title
195 m.reply disp, :overlong => :truncate
200 channel = params[:channel] ? params[:channel] : m.target
201 max = params[:limit].to_i
204 list = @registry[channel]
206 m.reply "no urls seen yet for channel #{channel}"
208 reply_urls :msg => m, :channel => channel, :list => list, :max => max
212 def search(m, params)
213 channel = params[:channel] ? params[:channel] : m.target
214 max = params[:limit].to_i
215 string = params[:string]
218 regex = Regexp.new(string, Regexp::IGNORECASE)
219 list = @registry[channel].find_all {|url|
220 regex.match(url.url) || regex.match(url.nick) ||
221 (@bot.config['url.info_on_list'] && regex.match(url.info))
224 m.reply "no matches for channel #{channel}"
226 reply_urls :msg => m, :channel => channel, :list => list, :max => max
231 plugin = UrlPlugin.new
232 plugin.map 'urls search :channel :limit :string', :action => 'search',
233 :defaults => {:limit => 4},
234 :requirements => {:limit => /^\d+$/},
236 plugin.map 'urls search :limit :string', :action => 'search',
237 :defaults => {:limit => 4},
238 :requirements => {:limit => /^\d+$/},
240 plugin.map 'urls :channel :limit', :defaults => {:limit => 4},
241 :requirements => {:limit => /^\d+$/},
243 plugin.map 'urls :limit', :defaults => {:limit => 4},
244 :requirements => {:limit => /^\d+$/},