X-Git-Url: https://git.netwichtig.de/gitweb/?a=blobdiff_plain;f=data%2Frbot%2Fplugins%2Furl.rb;h=269227a3c468e21d643d83e8f66296be5e6128fb;hb=69db4133c5ccfa41c35b43c67fce1d5ff640bfd5;hp=95ccb5d95dcdc2f8504efc70066cab10d3abf04d;hpb=b11c3c4042b03e36639370002ecf86c44f7ddde4;p=user%2Fhenk%2Fcode%2Fruby%2Frbot.git

diff --git a/data/rbot/plugins/url.rb b/data/rbot/plugins/url.rb
index 95ccb5d9..269227a3 100644
--- a/data/rbot/plugins/url.rb
+++ b/data/rbot/plugins/url.rb
@@ -1,10 +1,12 @@
-require 'uri'
+define_structure :Url, :channel, :nick, :time, :url, :info
 
-Url = Struct.new("Url", :channel, :nick, :time, :url)
-TITLE_RE = /<\s*?title\s*?>(.+?)<\s*?\/title\s*?>/im
-LINK_INFO = "[Link Info]"
+class ::UrlLinkError < RuntimeError
+end
 
 class UrlPlugin < Plugin
+  TITLE_RE = /<\s*?title\s*?>(.+?)<\s*?\/title\s*?>/im
+  LINK_INFO = "[Link Info]"
+
   BotConfig.register BotConfigIntegerValue.new('url.max_urls',
     :default => 100, :validate => Proc.new{|v| v > 0},
     :desc => "Maximum number of urls to store. New urls replace oldest ones.")
@@ -17,6 +19,10 @@ class UrlPlugin < Plugin
   BotConfig.register BotConfigBooleanValue.new('url.first_par',
     :default => false,
     :desc => "Also try to get the first paragraph of a web page")
+  BotConfig.register BotConfigBooleanValue.new('url.info_on_list',
+    :default => false,
+    :desc => "Show link info when listing/searching for urls")
+
 
   def initialize
     super
@@ -32,57 +38,89 @@ 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
 
     begin
-      range = @bot.config['http.info_bytes']
-      response = @bot.httputil.get_response(url, :range => "bytes=0-#{range}")
-      if response.code != "206" && response.code != "200"
-        return "Error getting link (#{response.code} - #{response.message})"
-      end
-      extra = String.new
-
-      if response['content-type'] =~ /^text\//
-
-        body = response.body.slice(0, range)
-        title = String.new
+      debug "+ getting #{url.request_uri}"
+      @bot.httputil.get_response(url) { |resp|
+        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.
+            #
+            # We act differently depending on whether we want the first par or
+            # not: in the first case we download the initial part and the parse
+            # it; in the second case we only download as much as we need to find
+            # the title
+            #
+            if @bot.config['url.first_par']
+              partial = resp.partial_body(@bot.config['http.info_bytes'])
+              logopts[:title] = title = get_title_from_html(partial)
+              first_par = Utils.ircify_first_html_par(partial, :strip => title)
+              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|
+                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
 
-        # since the content is 'text/*' and is small enough to
-        # be a webpage, retrieve the title from the page
-        debug "+ getting #{url.request_uri}"
+          enc = resp['content-encoding']
+          logopts[:extra] = String.new
+          logopts[:extra] << "Content Type: #{resp['content-type']}"
+          if enc
+            logopts[:extra] << ", encoding: #{enc}"
+            extra << ", #{Bold}encoding#{Bold}: #{enc}"
+          end
 
-        # we act differently depending on whether we want the first par or not:
-        # in the first case we download the initial part and the parse it; in the second
-        # case we only download as much as we need to find the title
-        if @bot.config['url.first_par']
-          first_par = Utils.ircify_first_html_par(body)
-          extra << "\n#{LINK_INFO} text: #{first_par}" unless first_par.empty?
-          title = get_title_from_html(body)
-          return "title: #{title}#{extra}" if title
+          unless @bot.config['url.titles_only']
+            # content doesn't have title, just display info.
+            size = resp['content-length'].gsub(/(\d)(?=\d{3}+(?:\.|$))(\d{3}\..*)?/,'\1,\2') rescue nil
+            if size
+              logopts[:extra] << ", size: #{size} bytes"
+              size = ", #{Bold}size#{Bold}: #{size} bytes"
+            end
+            call_event(:url_added, url.to_s, logopts)
+            return "#{Bold}type#{Bold}: #{resp['content-type']}#{size}#{extra}"
+          end
+          call_event(:url_added, url.to_s, logopts)
         else
-          title = get_title_from_html(body)
-          return "title: #{title}" if title
+          raise UrlLinkError, "getting link (#{resp.code} - #{resp.message})"
         end
-
-        # if nothing was found, provide more basic info
-      end
-
-      debug response.to_hash.inspect
-      unless @bot.config['url.titles_only']
-        # content doesn't have title, just display info.
-        size = response['content-length'].gsub(/(\d)(?=\d{3}+(?:\.|$))(\d{3}\..*)?/,'\1,\2') rescue nil
-        size = size ? ", size: #{size} bytes" : ""
-        return "type: #{response['content-type']}#{size}#{extra}"
-      end
+      }
+      return nil
     rescue Exception => e
-      error e.inspect
-      debug e.backtrace.join("\n")
-      return "Error connecting to site (#{e.message})"
+      case e
+      when UrlLinkError
+        raise e
+      else
+        error e
+        raise "connecting to site/processing information (#{e.message})"
+      end
     end
   end
 
@@ -95,11 +133,12 @@ class UrlPlugin < Plugin
         urlstr = $1
         list = @registry[m.target]
 
+        title = nil
         if @bot.config['url.display_link_info']
           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!"
@@ -107,7 +146,7 @@ class UrlPlugin < Plugin
                 debug "Title not found!"
               end
             rescue => e
-              debug "Failed: #{e}"
+              m.reply "Error #{e.message}"
             end
           end
         end
@@ -115,7 +154,7 @@ class UrlPlugin < Plugin
         # check to see if this url is already listed
         return if list.find {|u| u.url == urlstr }
 
-        url = Url.new(m.target, m.sourcenick, Time.new, urlstr)
+        url = Url.new(m.target, m.sourcenick, Time.new, urlstr, title)
         debug "#{list.length} urls so far"
         if list.length > @bot.config['url.max_urls']
           list.pop
@@ -128,6 +167,31 @@ class UrlPlugin < Plugin
     end
   end
 
+  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, 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
+      m.reply disp, :overlong => :truncate
+    end
+  end
+
   def urls(m, params)
     channel = params[:channel] ? params[:channel] : m.target
     max = params[:limit].to_i
@@ -137,9 +201,7 @@ class UrlPlugin < Plugin
     if list.empty?
       m.reply "no urls seen yet for channel #{channel}"
     else
-      list[0..(max-1)].each do |url|
-        m.reply "[#{url.time.strftime('%Y/%m/%d %H:%M:%S')}] <#{url.nick}> #{url.url}"
-      end
+      reply_urls :msg => m, :channel => channel, :list => list, :max => max
     end
   end
 
@@ -151,17 +213,17 @@ 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
-      list[0..(max-1)].each do |url|
-        m.reply "[#{url.time.strftime('%Y/%m/%d %H:%M:%S')}] <#{url.nick}> #{url.url}"
-      end
+      reply_urls :msg => m, :channel => channel, :list => list, :max => max
     end
   end
 end
+
 plugin = UrlPlugin.new
 plugin.map 'urls search :channel :limit :string', :action => 'search',
                           :defaults => {:limit => 4},