+#-- vim:sw=2:et\r
+#++\r
+#\r
# RSS feed plugin for RubyBot\r
# (c) 2004 Stanislav Karchebny <berkus@madfire.net>\r
# (c) 2005 Ian Monroe <ian@monroe.nu>\r
# (c) 2005 Mark Kretschmann <markey@web.de>\r
+# (c) 2006 Giuseppe Bilotta <giuseppe.bilotta@gmail.com>\r
+#\r
# Licensed under MIT License.\r
\r
require 'rss/parser'\r
require 'rss/1.0'\r
require 'rss/2.0'\r
require 'rss/dublincore'\r
-begin\r
- # require 'rss/dublincore/2.0'\r
-rescue\r
- warning "Unable to load RSS libraries, RSS plugin functionality crippled"\r
-end\r
+# begin\r
+# require 'rss/dublincore/2.0'\r
+# rescue\r
+# warning "Unable to load RSS libraries, RSS plugin functionality crippled"\r
+# end\r
\r
class ::String\r
def shorten(limit)\r
end\r
\r
def watched_by?(who)\r
- @watchers.include?(who)\r
+ # We need to check bot 'who' itself and the String form, because rss\r
+ # watches added before the new Irc framework represented watchers as\r
+ # Strings whereas they are now Channels.\r
+ #\r
+ @watchers.include?(who) || @watchers.include?(who.to_s) \r
end\r
\r
def add_watch(who)\r
if watched_by?(who)\r
return nil\r
end\r
- @watchers << who unless watched_by?(who)\r
+ # TODO FIXME? should we just store watchers as Strings instead?\r
+ # This should then be @watchers << who.downcase\r
+ @watchers << who\r
return who\r
end\r
\r
def rm_watch(who)\r
+ # See comment to watched_by?\r
+ #\r
@watchers.delete(who)\r
+ @watchers.delete(who.to_s)\r
+ end\r
+\r
+ def to_a\r
+ [@handle,@url,@type,@watchers]\r
end\r
\r
- # def to_ary\r
- # [@handle,@url,@type,@watchers]\r
- # end\r
+ def to_s(watchers=false)\r
+ if watchers\r
+ a = self.to_a.flatten\r
+ else\r
+ a = self.to_a[0,3]\r
+ end\r
+ a.join(" | ")\r
+ end\r
end\r
\r
class RSSFeedsPlugin < Plugin\r
:default => 90, :validate => Proc.new{|v| v > 0 && v < 400},\r
:desc => "How many characters to use of a RSS item text")\r
\r
+ BotConfig.register BotConfigIntegerValue.new('rss.thread_sleep',\r
+ :default => 300, :validate => Proc.new{|v| v > 30},\r
+ :desc => "How many seconds to sleep before checking RSS feeds again")\r
+\r
@@watchThreads = Hash.new\r
@@mutex = Mutex.new\r
\r
rewatch_rss\r
end\r
\r
+ def name\r
+ "rss"\r
+ end\r
+\r
def watchlist\r
@feeds.select { |h, f| f.watched? }\r
end\r
def help(plugin,topic="")\r
case topic\r
when "show"\r
- "rss show #{Bold}handle#{Bold} [#{Bold}limit#{Bold}] : show #{Bold}limit#{Bold} (default: 5, max: 15) entries from rss #{Bold}handle#{Bold}"\r
+ "rss show #{Bold}handle#{Bold} [#{Bold}limit#{Bold}] : show #{Bold}limit#{Bold} (default: 5, max: 15) entries from rss #{Bold}handle#{Bold}; #{Bold}limit#{Bold} can also be in the form a..b, to display a specific range of items"\r
when "list"\r
"rss list [#{Bold}handle#{Bold}] : list all rss feeds (matching #{Bold}handle#{Bold})"\r
when "watched"\r
end\r
end\r
\r
- def report_problem(report, m=nil)\r
- if m\r
+ def report_problem(report, e=nil, m=nil)\r
+ if m && m.respond_to?(:reply)\r
m.reply report\r
else\r
warning report\r
end\r
+ if e\r
+ debug e.inspect\r
+ debug e.backtrace.join("\n") if e.respond_to?(:backtrace)\r
+ end\r
end\r
\r
def show_rss(m, params)\r
handle = params[:handle]\r
- limit = params[:limit].to_i\r
- limit = 15 if limit > 15\r
- limit = 1 if limit <= 0\r
+ lims = params[:limit].to_s.match(/(\d+)(?:..(\d+))?/)\r
+ debug lims.to_a.inspect\r
+ if lims[2]\r
+ ll = [[lims[1].to_i-1,lims[2].to_i-1].min, 0].max\r
+ ul = [[lims[1].to_i-1,lims[2].to_i-1].max, 14].min\r
+ rev = lims[1].to_i > lims[2].to_i\r
+ else\r
+ ll = 0\r
+ ul = [[lims[1].to_i-1, 1].max, 14].min\r
+ rev = false\r
+ end\r
+\r
feed = @feeds.fetch(handle, nil)\r
unless feed\r
m.reply "I don't know any feeds named #{handle}"\r
return\r
end\r
- m.reply("Please wait, querying...")\r
+\r
+ m.reply "lemme fetch it..."\r
title = items = nil\r
@@mutex.synchronize {\r
title, items = fetchRss(feed, m)\r
}\r
return unless items\r
- m.reply("Channel : #{title}")\r
- # TODO: optional by-date sorting if dates present\r
- items[0...limit].reverse.each do |item|\r
- printRssItem(m.replyto,item)\r
+\r
+ # We sort the feeds in freshness order (newer ones first)\r
+ items = freshness_sort(items)\r
+ disp = items[ll..ul]\r
+ disp.reverse! if rev\r
+\r
+ m.reply "Channel : #{title}"\r
+ disp.each do |item|\r
+ printFormattedRss(feed, item, {:places=>[m.replyto],:handle=>nil,:date=>true})\r
end\r
end\r
\r
+ def itemDate(item,ex=nil)\r
+ return item.pubDate if item.respond_to?(:pubDate)\r
+ return item.date if item.respond_to?(:date)\r
+ return ex\r
+ end\r
+\r
+ def freshness_sort(items)\r
+ notime = Time.at(0)\r
+ items.sort { |a, b|\r
+ itemDate(b, notime) || 0 <=> itemDate(a, notime) || 0\r
+ }\r
+ end\r
+\r
def list_rss(m, params)\r
wanted = params[:handle]\r
reply = String.new\r
def add_rss(m, params, force=false)\r
handle = params[:handle]\r
url = params[:url]\r
+ unless url.match(/https?/)\r
+ m.reply "I only deal with feeds from HTTP sources, so I can't use #{url} (maybe you forgot the handle?)"\r
+ return\r
+ end\r
type = params[:type]\r
if @feeds.fetch(handle, nil) && !force\r
m.reply "There is already a feed named #{handle} (URL: #{@feeds[handle].url})"\r
return feed\r
end\r
\r
- def rewatch_rss(m=nil)\r
+ def rewatch_rss(m=nil, params=nil)\r
kill_threads\r
\r
# Read watches from list.\r
private\r
def watchRss(feed, m=nil)\r
if @@watchThreads.has_key?(feed.handle)\r
- report_problem("watcher thread for #{feed.handle} is already running", m)\r
+ report_problem("watcher thread for #{feed.handle} is already running", nil, m)\r
return\r
end\r
@@watchThreads[feed.handle] = Thread.new do\r
- debug 'watchRss thread started.'\r
+ debug "watcher for #{feed} started"\r
oldItems = []\r
firstRun = true\r
+ failures = 0\r
loop do\r
begin\r
- debug 'Fetching rss feed...'\r
+ debug "fetching #{feed}"\r
title = newItems = nil\r
@@mutex.synchronize {\r
title, newItems = fetchRss(feed)\r
}\r
unless newItems\r
- m.reply "no items in feed"\r
- break\r
- end\r
- debug "Checking if new items are available"\r
- if firstRun\r
- debug "First run, we'll see next time"\r
- firstRun = false\r
+ debug "no items in feed #{feed}"\r
+ failures +=1\r
else\r
- otxt = oldItems.map { |item| item.to_s }\r
- dispItems = newItems.reject { |item|\r
- otxt.include?(item.to_s)\r
- }\r
- if dispItems.length > 0\r
- debug "Found #{dispItems.length} new items"\r
- dispItems.each { |item|\r
- debug "showing #{item.title}"\r
- @@mutex.synchronize {\r
- printFormattedRss(feed, item)\r
- }\r
- }\r
+ debug "Checking if new items are available for #{feed}"\r
+ if firstRun\r
+ debug "First run, we'll see next time"\r
+ firstRun = false\r
else\r
- debug "No new items found"\r
+ otxt = oldItems.map { |item| item.to_s }\r
+ dispItems = newItems.reject { |item|\r
+ otxt.include?(item.to_s)\r
+ }\r
+ if dispItems.length > 0\r
+ debug "Found #{dispItems.length} new items in #{feed}"\r
+ # When displaying watched feeds, publish them from older to newer\r
+ dispItems.reverse.each { |item|\r
+ @@mutex.synchronize {\r
+ printFormattedRss(feed, item)\r
+ }\r
+ }\r
+ else\r
+ debug "No new items found in #{feed}"\r
+ end\r
end\r
+ oldItems = newItems.dup\r
end\r
- oldItems = newItems.dup\r
rescue Exception => e\r
- error "IO failed: #{e.inspect}"\r
+ error "Error watching #{feed}: #{e.inspect}"\r
debug e.backtrace.join("\n")\r
+ failures += 1\r
end\r
\r
- seconds = 150 + rand(100)\r
- debug "Thread going to sleep #{seconds} seconds.."\r
+ seconds = @bot.config['rss.thread_sleep'] * (failures + 1)\r
+ seconds += seconds * (rand(100)-50)/100\r
+ debug "watcher for #{feed} going to sleep #{seconds} seconds.."\r
sleep seconds\r
end\r
end\r
end\r
\r
- def printRssItem(loc,item)\r
- if item.kind_of?(RSS::RDF::Item)\r
- @bot.say loc, item.title.chomp.riphtml.shorten(@bot.config['rss.head_max']) + " @ " + item.link\r
- else\r
- desc = String.new\r
- desc << item.pubDate.to_s.chomp + ": " if item.pubDate\r
- desc << item.title.chomp.riphtml.shorten(@bot.config['rss.head_max']) + " :: " if item.title\r
- desc << " @ " + item.link.chomp if item.link\r
- @bot.say loc, desc\r
+ def printFormattedRss(feed, item, opts=nil)\r
+ places = feed.watchers\r
+ handle = "::#{feed.handle}:: "\r
+ date = String.new\r
+ if opts\r
+ places = opts[:places] if opts.key?(:places)\r
+ handle = opts[:handle].to_s if opts.key?(:handle)\r
+ if opts.key?(:date) && opts[:date]\r
+ if item.respond_to?(:pubDate) \r
+ if item.pubDate.class <= Time\r
+ date = item.pubDate.strftime("%Y/%m/%d %H.%M.%S")\r
+ else\r
+ date = item.pubDate.to_s\r
+ end\r
+ elsif item.respond_to?(:date)\r
+ if item.date.class <= Time\r
+ date = item.date.strftime("%Y/%m/%d %H.%M.%S")\r
+ else\r
+ date = item.date.to_s\r
+ end\r
+ else\r
+ date = "(no date)"\r
+ end\r
+ date += " :: "\r
+ end\r
end\r
- end\r
-\r
- def printFormattedRss(feed, item)\r
- feed.watchers.each { |loc|\r
+ title = "#{Bold}#{item.title.chomp.riphtml}#{Bold}" if item.title\r
+ desc = item.description.gsub(/\s+/,' ').strip.riphtml.shorten(@bot.config['rss.text_max']) if item.description\r
+ link = item.link.chomp if item.link\r
+ places.each { |loc|\r
case feed.type\r
when 'blog'\r
- @bot.say loc, "::#{feed.handle}:: #{item.category.content} just blogged at #{item.link}::"\r
- @bot.say loc, "::#{feed.handle}:: #{item.title.chomp.riphtml} - #{item.description.chomp.riphtml.shorten(@bot.config['rss.text_max'])}::"\r
+ @bot.say loc, "#{handle}#{date}#{item.category.content} blogged at #{link}"\r
+ @bot.say loc, "#{handle}#{title} - #{desc}"\r
when 'forum'\r
- @bot.say loc, "::#{feed.handle}:: #{item.pubDate.to_s.chomp+": " if item.pubDate}#{item.title.chomp.riphtml+" :: " if item.title}#{" @ "+item.link.chomp if item.link}"\r
+ @bot.say loc, "#{handle}#{date}#{title}#{' @ ' if item.title && item.link}#{link}"\r
when 'wiki'\r
- @bot.say loc, "::#{feed.handle}:: #{item.title} has been edited by #{item.dc_creator}. #{item.description.split("\n")[0].chomp.riphtml.shorten(@bot.config['rss.text_max'])} #{item.link} ::"\r
+ @bot.say loc, "#{handle}#{date}#{item.title} has been edited by #{item.dc_creator}. #{desc} #{link}"\r
when 'gmame'\r
- @bot.say loc, "::#{feed.handle}:: Message #{item.title} sent by #{item.dc_creator}. #{item.description.split("\n")[0].chomp.riphtml.shorten(@bot.config['rss.text_max'])} ::"\r
+ @bot.say loc, "#{handle}#{date}Message #{title} sent by #{item.dc_creator}. #{desc}"\r
when 'trac'\r
- @bot.say loc, "/---- #{feed.handle} :: #{item.title} :: #{item.link}"\r
- @bot.say loc, "|#{item.description.gsub(/\s+/,' ').strip.riphtml.shorten(@bot.config['rss.text_max'])}"\r
- @bot.say loc, "\\----"\r
+ @bot.say loc, "#{handle}#{date}#{title} @ #{link}"\r
+ unless item.title =~ /^Changeset \[(\d+)\]/\r
+ @bot.say loc, "#{handle}#{date}#{desc}"\r
+ end\r
else\r
- printRssItem(loc,item)\r
+ @bot.say loc, "#{handle}#{date}#{title}#{' @ ' if item.title && item.link}#{link}"\r
end\r
}\r
end\r
def fetchRss(feed, m=nil)\r
begin\r
# Use 60 sec timeout, cause the default is too low\r
- xml = @bot.httputil.get_cached(feed.url,60,60)\r
+ # Do not use get_cached for RSS until we have proper cache handling\r
+ # xml = @bot.httputil.get_cached(feed.url,60,60)\r
+ xml = @bot.httputil.get(feed.url,60,60)\r
rescue URI::InvalidURIError, URI::BadURIError => e\r
- report_problem("invalid rss feed #{feed.url}", m)\r
+ report_problem("invalid rss feed #{feed.url}", e, m)\r
+ return\r
+ rescue => e\r
+ report_problem("error getting #{feed.url}", e, m)\r
return\r
end\r
- debug 'fetched'\r
+ debug "fetched #{feed}"\r
unless xml\r
- report_problem("reading feed #{url} failed", m)\r
+ report_problem("reading feed #{feed} failed", nil, m)\r
return\r
end\r
\r
begin\r
## do validate parse\r
rss = RSS::Parser.parse(xml)\r
- debug 'parsed'\r
+ debug "parsed #{feed}"\r
rescue RSS::InvalidRSSError\r
## do non validate parse for invalid RSS 1.0\r
begin\r
rss = RSS::Parser.parse(xml, false)\r
- rescue RSS::Error\r
- report_problem("parsing rss stream failed, whoops =(", m)\r
+ rescue RSS::Error => e\r
+ report_problem("parsing rss stream failed, whoops =(", e, m)\r
return\r
end\r
- rescue RSS::Error\r
- report_problem("parsing rss stream failed, oioi", m)\r
+ rescue RSS::Error => e\r
+ report_problem("parsing rss stream failed, oioi", e, m)\r
return\r
rescue => e\r
- report_problem("processing error occured, sorry =(", m)\r
- debug e.inspect\r
- debug e.backtrace.join("\n")\r
+ report_problem("processing error occured, sorry =(", e, m)\r
return\r
end\r
items = []\r
if rss.nil?\r
- report_problem("#{feed.url} does not include RSS 1.0 or 0.9x/2.0",m)\r
+ report_problem("#{feed} does not include RSS 1.0 or 0.9x/2.0", nil, m)\r
else\r
begin\r
- rss.output_encoding = "euc-jp"\r
- rescue RSS::UnknownConvertMethod\r
- report_problem("bah! something went wrong =(",m)\r
+ rss.output_encoding = 'UTF-8'\r
+ rescue RSS::UnknownConvertMethod => e\r
+ report_problem("bah! something went wrong =(", e, m)\r
return\r
end\r
rss.channel.title ||= "Unknown"\r
end\r
\r
if items.empty?\r
- report_problem("no items found in the feed, maybe try weed?",m)\r
+ report_problem("no items found in the feed, maybe try weed?", e, m)\r
return\r
end\r
return [title, items]\r
\r
plugin.map 'rss show :handle :limit',\r
:action => 'show_rss',\r
- :requirements => {:limit => /^\d+$/},\r
+ :requirements => {:limit => /^\d+(?:\.\.\d+)?$/},\r
:defaults => {:limit => 5}\r
plugin.map 'rss list :handle',\r
:action => 'list_rss',\r
:action => 'unwatch_rss'\r
plugin.map 'rss rmwatch :handle',\r
:action => 'unwatch_rss'\r
-plugin.map 'rss rewatch :handle',\r
+plugin.map 'rss rewatch',\r
:action => 'rewatch_rss'\r
+\r