X-Git-Url: https://git.netwichtig.de/gitweb/?a=blobdiff_plain;f=data%2Frbot%2Fplugins%2Furl.rb;h=269227a3c468e21d643d83e8f66296be5e6128fb;hb=69db4133c5ccfa41c35b43c67fce1d5ff640bfd5;hp=8c064dccb91750d866e111fc87765ac4aaf4635c;hpb=37fb9f96d582d24ffb9bbc14cc75fc9fec52ace3;p=user%2Fhenk%2Fcode%2Fruby%2Frbot.git

diff --git a/data/rbot/plugins/url.rb b/data/rbot/plugins/url.rb
index 8c064dcc..269227a3 100644
--- a/data/rbot/plugins/url.rb
+++ b/data/rbot/plugins/url.rb
@@ -1,4 +1,4 @@
-Url = Struct.new("Url", :channel, :nick, :time, :url, :info)
+define_structure :Url, :channel, :nick, :time, :url, :info
 
 class ::UrlLinkError < RuntimeError
 end
@@ -38,11 +38,15 @@ class UrlPlugin < Plugin
     $1.ircify_html
   end
 
-  def get_title_for_url(uri_str)
+  def get_title_for_url(uri_str, nick = nil, channel = nil)
 
     url = uri_str.kind_of?(URI) ? uri_str : URI.parse(uri_str)
     return if url.scheme !~ /https?/
 
+    logopts = Hash.new
+    logopts[:nick] = nick if nick
+    logopts[:channel] = channel if channel
+
     title = nil
     extra = String.new
 
@@ -52,6 +56,8 @@ class UrlPlugin < Plugin
         case resp
         when Net::HTTPSuccess
 
+          debug resp.to_hash
+
           if resp['content-type'] =~ /^text\/|(?:x|ht)ml/
             # The page is text or HTML, so we can try finding a title and, if
             # requested, the first par.
@@ -63,35 +69,50 @@ class UrlPlugin < Plugin
             #
             if @bot.config['url.first_par']
               partial = resp.partial_body(@bot.config['http.info_bytes'])
-              title = get_title_from_html(partial)
+              logopts[:title] = title = get_title_from_html(partial)
               first_par = Utils.ircify_first_html_par(partial, :strip => title)
-              extra << ", #{Bold}text#{Bold}: #{first_par}" unless first_par.empty?
+              unless first_par.empty?
+                logopts[:extra] = first_par
+                extra << ", #{Bold}text#{Bold}: #{first_par}"
+              end
+              call_event(:url_added, url.to_s, logopts)
               return "#{Bold}title#{Bold}: #{title}#{extra}" if title
             else
               resp.partial_body(@bot.config['http.info_bytes']) { |part|
-                title = get_title_from_html(part)
+                logopts[:title] = title = get_title_from_html(part)
+                call_event(:url_added, url.to_s, logopts)
                 return "#{Bold}title#{Bold}: #{title}" if title
               }
             end
           # if nothing was found, provide more basic info, as for non-html pages
+          else
+            resp.no_cache = true
           end
 
-          debug resp.to_hash.inspect
-
           enc = resp['content-encoding']
-
-          extra << ", #{Bold}encoding#{Bold}: #{enc}" if enc
+          logopts[:extra] = String.new
+          logopts[:extra] << "Content Type: #{resp['content-type']}"
+          if enc
+            logopts[:extra] << ", encoding: #{enc}"
+            extra << ", #{Bold}encoding#{Bold}: #{enc}"
+          end
 
           unless @bot.config['url.titles_only']
             # content doesn't have title, just display info.
             size = resp['content-length'].gsub(/(\d)(?=\d{3}+(?:\.|$))(\d{3}\..*)?/,'\1,\2') rescue nil
-            size = size ? ", #{Bold}size#{Bold}: #{size} bytes" : ""
+            if size
+              logopts[:extra] << ", size: #{size} bytes"
+              size = ", #{Bold}size#{Bold}: #{size} bytes"
+            end
+            call_event(:url_added, url.to_s, logopts)
             return "#{Bold}type#{Bold}: #{resp['content-type']}#{size}#{extra}"
           end
+          call_event(:url_added, url.to_s, logopts)
         else
           raise UrlLinkError, "getting link (#{resp.code} - #{resp.message})"
         end
       }
+      return nil
     rescue Exception => e
       case e
       when UrlLinkError
@@ -117,7 +138,7 @@ class UrlPlugin < Plugin
           Thread.start do
             debug "Getting title for #{urlstr}..."
             begin
-              title = get_title_for_url urlstr
+              title = get_title_for_url urlstr, m.source.nick, m.channel
               if title
                 m.reply "#{LINK_INFO} #{title}", :overlong => :truncate
                 debug "Title found!"
@@ -146,13 +167,24 @@ class UrlPlugin < Plugin
     end
   end
 
-  def reply_urls(m, list, max)
+  def reply_urls(opts={})
+    list = opts[:list]
+    max = opts[:max]
+    channel = opts[:channel]
+    m = opts[:msg]
+    return unless list and max and m
     list[0..(max-1)].each do |url|
       disp = "[#{url.time.strftime('%Y/%m/%d %H:%M:%S')}] <#{url.nick}> #{url.url}"
       if @bot.config['url.info_on_list']
-        title = url.info || get_title_for_url(url.url) rescue nil
-        if title and not url.info
-          url.info = title
+        title = url.info || get_title_for_url(url.url, url.nick, channel) rescue nil
+        # If the url info was missing and we now have some, try to upgrade it
+        if channel and title and not url.info
+          ll = @registry[channel]
+          debug ll
+          if el = ll.find { |u| u.url == url.url }
+            el.info = title
+            @registry[channel] = ll
+          end
         end
         disp << " --> #{title}" if title
       end
@@ -169,7 +201,7 @@ class UrlPlugin < Plugin
     if list.empty?
       m.reply "no urls seen yet for channel #{channel}"
     else
-      reply_urls(m, list, max)
+      reply_urls :msg => m, :channel => channel, :list => list, :max => max
     end
   end
 
@@ -181,12 +213,13 @@ class UrlPlugin < Plugin
     max = 1 if max < 1
     regex = Regexp.new(string, Regexp::IGNORECASE)
     list = @registry[channel].find_all {|url|
-      regex.match(url.url) || regex.match(url.nick)
+      regex.match(url.url) || regex.match(url.nick) ||
+        (@bot.config['url.info_on_list'] && regex.match(url.info))
     }
     if list.empty?
       m.reply "no matches for channel #{channel}"
     else
-      reply_urls(m, list, max)
+      reply_urls :msg => m, :channel => channel, :list => list, :max => max
     end
   end
 end