Url = Struct.new("Url", :channel, :nick, :time, :url)
+TITLE_RE = /<\s*?title\s*?>(.+?)<\s*?\/title\s*?>/im
+LINK_INFO = "[Link Info]"
class UrlPlugin < Plugin
+ 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 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.)")
+ BotConfig.register BotConfigBooleanValue.new('url.first_par',
+ :default => false,
+ :desc => "Also try to get the first paragraph of a web page")
+
def initialize
super
@registry.set_default(Array.new)
end
+
def help(plugin, topic="")
- "urls [<max>=4] => list <max> last urls mentioned in current channel, urls <channel> [<max>=4] => list <max> last urls mentioned in <channel>, urls search <regexp> => search for matching urls, urls search <channel> <regexp>, search for matching urls in channel <channel>"
+ "urls [<max>=4] => list <max> last urls mentioned in current channel, urls search [<max>=4] <regexp> => search for matching urls. In a private message, you must specify the channel to query, eg. urls <channel> [max], urls search <channel> [max] <regexp>"
+ end
+
+ def get_title_from_html(pagedata)
+ return unless TITLE_RE.match(pagedata)
+ $1.ircify_html
end
+
+ def get_title_for_url(uri_str)
+
+ url = uri_str.kind_of?(URI) ? uri_str : URI.parse(uri_str)
+ return if url.scheme !~ /https?/
+
+ title = nil
+ extra = String.new
+
+ begin
+ debug "+ getting #{url.request_uri}"
+ @bot.httputil.get_response(url) { |resp|
+ case resp
+ when Net::HTTPSuccess
+
+ if resp['content-type'] =~ /^text\/|(?:x|ht)ml/
+ # The page is text or HTML, so we can try finding a title and, if
+ # requested, the first par.
+ #
+ # We act differently depending on whether we want the first par or
+ # not: in the first case we download the initial part and the parse
+ # it; in the second case we only download as much as we need to find
+ # the title
+ #
+ if @bot.config['url.first_par']
+ partial = resp.partial_body(@bot.config['http.info_bytes'])
+ title = get_title_from_html(partial)
+ first_par = Utils.ircify_first_html_par(partial, :strip => title)
+ extra << ", #{Bold}text#{Bold}: #{first_par}" unless first_par.empty?
+ return "#{Bold}title#{Bold}: #{title}#{extra}" if title
+ else
+ resp.partial_body(@bot.config['http.info_bytes']) { |part|
+ title = get_title_from_html(part)
+ return "#{Bold}title#{Bold}: #{title}" if title
+ }
+ end
+ # if nothing was found, provide more basic info, as for non-html pages
+ end
+
+ debug resp.to_hash.inspect
+
+ enc = resp['content-encoding']
+
+ extra << ", #{Bold}encoding#{Bold}: #{enc}" if enc
+
+ unless @bot.config['url.titles_only']
+ # content doesn't have title, just display info.
+ size = resp['content-length'].gsub(/(\d)(?=\d{3}+(?:\.|$))(\d{3}\..*)?/,'\1,\2') rescue nil
+ size = size ? ", #{Bold}size#{Bold}: #{size} bytes" : ""
+ return "#{Bold}type#{Bold}: #{resp['content-type']}#{size}#{extra}"
+ end
+ else
+ return "Error getting link (#{resp.code} - #{resp.message})"
+ end
+ }
+ rescue Exception => e
+ error e
+ return "Error connecting to site (#{e.message})"
+ 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+|$)/
- url = Url.new(m.target, m.sourcenick, Time.new, $1)
+ urlstr = $1
list = @registry[m.target]
+
+ if @bot.config['url.display_link_info']
+ Thread.start do
+ debug "Getting title for #{urlstr}..."
+ begin
+ title = get_title_for_url urlstr
+ if title
+ m.reply "#{LINK_INFO} #{title}", :overlong => :truncate
+ debug "Title found!"
+ else
+ debug "Title not found!"
+ end
+ rescue => e
+ debug "Failed: #{e}"
+ end
+ end
+ end
+
+ # check to see if this url is already listed
+ return if list.find {|u| u.url == urlstr }
+
+ url = Url.new(m.target, m.sourcenick, Time.new, urlstr)
debug "#{list.length} urls so far"
- if list.length > 50
+ if list.length > @bot.config['url.max_urls']
list.pop
end
debug "storing url #{url.url}"
end
end
end
- def privmsg(m)
- case m.params
- when nil
- if m.public?
- urls m, m.target
- else
- m.reply "in a private message, you need to specify a channel name for urls"
- end
- when (/^(\d+)$/)
- max = $1.to_i
- if m.public?
- urls m, m.target, max
- else
- m.reply "in a private message, you need to specify a channel name for urls"
- end
- when (/^(#.*?)\s+(\d+)$/)
- channel = $1
- max = $2.to_i
- urls m, channel, max
- when (/^(#.*?)$/)
- channel = $1
- urls m, channel
- when (/^search\s+(#.*?)\s+(.*)$/)
- channel = $1
- string = $2
- search m, channel, string
- when (/^search\s+(.*)$/)
- string = $1
- if m.public?
- search m, m.target, string
- else
- m.reply "in a private message, you need to specify a channel name for urls"
- end
- else
- m.reply "incorrect usage: " + help(m.plugin)
- end
- end
- def urls(m, channel, max=4)
+ def urls(m, params)
+ channel = params[:channel] ? params[:channel] : m.target
+ max = params[:limit].to_i
max = 10 if max > 10
max = 1 if max < 1
list = @registry[channel]
end
end
- def search(m, channel, string, max=4)
+ def search(m, params)
+ channel = params[:channel] ? params[:channel] : m.target
+ max = params[:limit].to_i
+ string = params[:string]
max = 10 if max > 10
max = 1 if max < 1
- regex = Regexp.new(string)
+ regex = Regexp.new(string, Regexp::IGNORECASE)
list = @registry[channel].find_all {|url|
regex.match(url.url) || regex.match(url.nick)
}
end
end
plugin = UrlPlugin.new
-plugin.register("urls")
+plugin.map 'urls search :channel :limit :string', :action => 'search',
+ :defaults => {:limit => 4},
+ :requirements => {:limit => /^\d+$/},
+ :public => false
+plugin.map 'urls search :limit :string', :action => 'search',
+ :defaults => {:limit => 4},
+ :requirements => {:limit => /^\d+$/},
+ :private => false
+plugin.map 'urls :channel :limit', :defaults => {:limit => 4},
+ :requirements => {:limit => /^\d+$/},
+ :public => false
+plugin.map 'urls :limit', :defaults => {:limit => 4},
+ :requirements => {:limit => /^\d+$/},
+ :private => false
projects / user / henk / code / ruby / rbot.git / blobdiff