+#-- vim:sw=2:et
+#++
+#
+# :title: Url plugin
+
define_structure :Url, :channel, :nick, :time, :url, :info
class ::UrlLinkError < RuntimeError
class UrlPlugin < Plugin
TITLE_RE = /<\s*?title\s*?>(.+?)<\s*?\/title\s*?>/im
LINK_INFO = "[Link Info]"
+ OUR_UNSAFE = Regexp.new("[^#{URI::PATTERN::UNRESERVED}#{URI::PATTERN::RESERVED}%# ]", false, 'N')
- BotConfig.register BotConfigIntegerValue.new('url.max_urls',
+ Config.register Config::IntegerValue.new('url.max_urls',
:default => 100, :validate => Proc.new{|v| v > 0},
:desc => "Maximum number of urls to store. New urls replace oldest ones.")
- BotConfig.register BotConfigBooleanValue.new('url.display_link_info',
- :default => false,
- :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)")
- BotConfig.register BotConfigBooleanValue.new('url.titles_only',
+ Config.register Config::IntegerValue.new('url.display_link_info',
+ :default => 0,
+ :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)")
+ Config.register Config::BooleanValue.new('url.titles_only',
:default => false,
:desc => "Only show info for links that have <title> tags (in other words, don't display info for jpegs, mpegs, etc.)")
- BotConfig.register BotConfigBooleanValue.new('url.first_par',
+ Config.register Config::BooleanValue.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',
+ Config.register Config::BooleanValue.new('url.info_on_list',
:default => false,
:desc => "Show link info when listing/searching for urls")
+ Config.register Config::ArrayValue.new('url.no_info_hosts',
+ :default => ['localhost', '^192\.168\.', '^10\.', '^127\.', '^172\.(1[6-9]|2\d|31)\.'],
+ :on_change => Proc.new { |bot, v| bot.plugins['url'].reset_no_info_hosts },
+ :desc => "A list of regular expressions matching hosts for which no info should be provided")
def initialize
super
@registry.set_default(Array.new)
+ unless @bot.config['url.display_link_info'].kind_of?(Integer)
+ @bot.config.items[:'url.display_link_info'].set_string(@bot.config['url.display_link_info'].to_s)
+ end
+ reset_no_info_hosts
+ end
+
+ def reset_no_info_hosts
+ @no_info_hosts = Regexp.new(@bot.config['url.no_info_hosts'].join('|'), true)
+ debug "no info hosts regexp set to #{@no_info_hosts}"
end
def help(plugin, topic="")
$1.ircify_html
end
- def get_title_for_url(uri_str, nick = nil, channel = nil)
+ def get_title_for_url(uri_str, nick = nil, channel = nil, ircline = nil)
url = uri_str.kind_of?(URI) ? uri_str : URI.parse(uri_str)
return if url.scheme !~ /https?/
+ if url.host =~ @no_info_hosts
+ return "Sorry, info retrieval for #{url.host} is disabled"
+ end
+
logopts = Hash.new
logopts[:nick] = nick if nick
logopts[:channel] = channel if channel
+ logopts[:ircline] = ircline if ircline
title = nil
extra = String.new
if @bot.config['url.first_par']
partial = resp.partial_body(@bot.config['http.info_bytes'])
logopts[:title] = title = get_title_from_html(partial)
+ if url.fragment and not url.fragment.empty?
+ fragreg = /.*?<a\s+[^>]*name=["']?#{url.fragment}["']?.*?>/im
+ partial.sub!(fragreg,'')
+ end
first_par = Utils.ircify_first_html_par(partial, :strip => title)
unless first_par.empty?
logopts[:extra] = first_par
end
end
- def listen(m)
- return unless m.kind_of?(PrivMessage)
- return if m.address?
- # TODO support multiple urls in one line
- if m.message =~ /(f|ht)tps?:\/\//
- if m.message =~ /((f|ht)tps?:\/\/.*?)(?:\s+|$)/
- urlstr = URI.escape $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, m.source.nick, m.channel
- if title
- m.reply "#{LINK_INFO} #{title}", :overlong => :truncate
- debug "Title found!"
- else
- debug "Title not found!"
- end
- rescue => e
- m.reply "Error #{e.message}"
+ def handle_urls(m, urls, display_info=@bot.config['url.display_link_info'])
+ return if urls.empty?
+ debug "found urls #{urls.inspect}"
+ if m.public?
+ list = @registry[m.target]
+ else
+ list = nil
+ end
+ urls_displayed = 0
+ urls.each { |urlstr|
+ debug "working on #{urlstr}"
+ next unless urlstr =~ /^https?:/
+ title = nil
+ debug "display link info: #{display_info}"
+ if display_info > urls_displayed
+ urls_displayed += 1
+ Thread.start do
+ debug "Getting title for #{urlstr}..."
+ begin
+ title = get_title_for_url urlstr, m.source.nick, m.channel, m.message
+ if title
+ m.reply "#{LINK_INFO} #{title}", :overlong => :truncate
+ debug "Title found!"
+ else
+ debug "Title not found!"
end
+ rescue => e
+ m.reply "Error #{e.message}"
end
end
+ end
- # check to see if this url is already listed
- return if list.find {|u| u.url == urlstr }
+ next unless list
- 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
- end
- debug "storing url #{url.url}"
- list.unshift url
- debug "#{list.length} urls now"
- @registry[m.target] = list
+ # check to see if this url is already listed
+ next if list.find {|u| u.url == 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
end
- end
+ debug "storing url #{url.url}"
+ list.unshift url
+ debug "#{list.length} urls now"
+ }
+ @registry[m.target] = list
+ end
+
+ def info(m, params)
+ escaped = URI.escape(params[:urls].to_s, OUR_UNSAFE)
+ urls = URI.extract(escaped)
+ handle_urls(m, urls, params[:urls].length)
+ end
+
+ def listen(m)
+ return unless m.kind_of?(PrivMessage)
+ return if m.address?
+
+ escaped = URI.escape(m.message, OUR_UNSAFE)
+ urls = URI.extract(escaped)
+ handle_urls(m, urls)
end
def reply_urls(opts={})
end
plugin = UrlPlugin.new
+plugin.map 'urls info *urls', :action => 'info'
plugin.map 'urls search :channel :limit :string', :action => 'search',
:defaults => {:limit => 4},
:requirements => {:limit => /^\d+$/},
projects / user / henk / code / ruby / rbot.git / blobdiff