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',
: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 BotConfigIntegerValue.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)")
BotConfig.register BotConfigBooleanValue.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.)")
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
end
def help(plugin, topic="")
$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
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.
#
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)
+ 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)
- 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
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 = $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
- 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}"
+
+ escaped = URI.escape(m.message, OUR_UNSAFE)
+ urls = URI.extract(escaped)
+ return if urls.empty?
+ debug "found urls #{urls.inspect}"
+ list = @registry[m.target]
+ urls_displayed = 0
+ urls.each { |urlstr|
+ debug "working on #{urlstr}"
+ next unless urlstr =~ /^https?:/
+ title = nil
+ debug "display link info: #{@bot.config['url.display_link_info']}"
+ if @bot.config['url.display_link_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
+ 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 }
+ # 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
- debug "storing url #{url.url}"
- list.unshift url
- debug "#{list.length} urls now"
- @registry[m.target] = 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
- end
+ debug "storing url #{url.url}"
+ list.unshift url
+ debug "#{list.length} urls now"
+ }
+ @registry[m.target] = list
end
def reply_urls(opts={})
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
+ 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]