-require 'net/http'
require 'uri'
-require 'cgi'
Url = Struct.new("Url", :channel, :nick, :time, :url)
TITLE_RE = /<\s*?title\s*?>(.+?)<\s*?\/title\s*?>/im
-UNESCAPE_TABLE = {
- 'raquo' => '>>',
- 'quot' => '"',
- 'micro' => 'u',
- 'copy' => '(c)',
- 'trade' => '(tm)',
- 'reg' => '(R)',
- '#174' => '(R)',
- '#8220' => '"',
- '#8221' => '"',
- '#8212' => '--',
- '#39' => '\'',
-=begin
- # extras codes, for future use...
- 'zwnj' => '‌',
- 'aring' => '\xe5',
- 'gt' => '>',
- 'yen' => '\xa5',
- 'ograve' => '\xf2',
- 'Chi' => 'Χ',
- 'bull' => '•',
- 'Egrave' => '\xc8',
- 'Ntilde' => '\xd1',
- 'upsih' => 'ϒ',
- 'Yacute' => '\xdd',
- 'asymp' => '≈',
- 'radic' => '√',
- 'otimes' => '⊗',
- 'nabla' => '∇',
- 'aelig' => '\xe6',
- 'oelig' => 'œ',
- 'equiv' => '≡',
- 'Psi' => 'Ψ',
- 'auml' => '\xe4',
- 'circ' => 'ˆ',
- 'Acirc' => '\xc2',
- 'Epsilon' => 'Ε',
- 'Yuml' => 'Ÿ',
- 'Eta' => 'Η',
- 'lt' => '<',
- 'Icirc' => '\xce',
- 'Upsilon' => 'Υ',
- 'ndash' => '–',
- 'there4' => '∴',
- 'Prime' => '″',
- 'prime' => '′',
- 'psi' => 'ψ',
- 'Kappa' => 'Κ',
- 'rsaquo' => '›',
- 'Tau' => 'Τ',
- 'darr' => '↓',
- 'ocirc' => '\xf4',
- 'lrm' => '‎',
- 'zwj' => '‍',
- 'cedil' => '\xb8',
- 'Ecirc' => '\xca',
- 'not' => '\xac',
- 'amp' => '&',
- 'AElig' => '\xc6',
- 'oslash' => '\xf8',
- 'acute' => '\xb4',
- 'lceil' => '⌈',
- 'laquo' => '\xab',
- 'shy' => '\xad',
- 'rdquo' => '”',
- 'ge' => '≥',
- 'Igrave' => '\xcc',
- 'Ograve' => '\xd2',
- 'euro' => '€',
- 'dArr' => '⇓',
- 'sdot' => '⋅',
- 'nbsp' => '\xa0',
- 'lfloor' => '⌊',
- 'lArr' => '⇐',
- 'Auml' => '\xc4',
- 'larr' => '←',
- 'Atilde' => '\xc3',
- 'Otilde' => '\xd5',
- 'szlig' => '\xdf',
- 'clubs' => '♣',
- 'diams' => '♦',
- 'agrave' => '\xe0',
- 'Ocirc' => '\xd4',
- 'Iota' => 'Ι',
- 'Theta' => 'Θ',
- 'Pi' => 'Π',
- 'OElig' => 'Œ',
- 'Scaron' => 'Š',
- 'frac14' => '\xbc',
- 'egrave' => '\xe8',
- 'sub' => '⊂',
- 'iexcl' => '\xa1',
- 'frac12' => '\xbd',
- 'sbquo' => '‚',
- 'ordf' => '\xaa',
- 'sum' => '∑',
- 'prop' => '∝',
- 'Uuml' => '\xdc',
- 'ntilde' => '\xf1',
- 'sup' => '⊃',
- 'theta' => 'θ',
- 'prod' => '∏',
- 'nsub' => '⊄',
- 'hArr' => '⇔',
- 'rlm' => '‏',
- 'THORN' => '\xde',
- 'infin' => '∞',
- 'yuml' => '\xff',
- 'Mu' => 'Μ',
- 'le' => '≤',
- 'Eacute' => '\xc9',
- 'thinsp' => ' ',
- 'ecirc' => '\xea',
- 'bdquo' => '„',
- 'Sigma' => 'Σ',
- 'fnof' => 'ƒ',
- 'Aring' => '\xc5',
- 'tilde' => '˜',
- 'frac34' => '\xbe',
- 'emsp' => ' ',
- 'mdash' => '—',
- 'uarr' => '↑',
- 'permil' => '‰',
- 'Ugrave' => '\xd9',
- 'rarr' => '→',
- 'Agrave' => '\xc0',
- 'chi' => 'χ',
- 'forall' => '∀',
- 'eth' => '\xf0',
- 'rceil' => '⌉',
- 'iuml' => '\xef',
- 'gamma' => 'γ',
- 'lambda' => 'λ',
- 'harr' => '↔',
- 'rang' => '〉',
- 'xi' => 'ξ',
- 'dagger' => '†',
- 'divide' => '\xf7',
- 'Ouml' => '\xd6',
- 'image' => 'ℑ',
- 'alefsym' => 'ℵ',
- 'igrave' => '\xec',
- 'otilde' => '\xf5',
- 'Oacute' => '\xd3',
- 'sube' => '⊆',
- 'alpha' => 'α',
- 'frasl' => '⁄',
- 'ETH' => '\xd0',
- 'lowast' => '∗',
- 'Nu' => 'Ν',
- 'plusmn' => '\xb1',
- 'Euml' => '\xcb',
- 'real' => 'ℜ',
- 'sup1' => '\xb9',
- 'sup2' => '\xb2',
- 'sup3' => '\xb3',
- 'Oslash' => '\xd8',
- 'Aacute' => '\xc1',
- 'cent' => '\xa2',
- 'oline' => '‾',
- 'Beta' => 'Β',
- 'perp' => '⊥',
- 'Delta' => 'Δ',
- 'loz' => '◊',
- 'pi' => 'π',
- 'iota' => 'ι',
- 'empty' => '∅',
- 'euml' => '\xeb',
- 'brvbar' => '\xa6',
- 'iacute' => '\xed',
- 'para' => '\xb6',
- 'micro' => '\xb5',
- 'cup' => '∪',
- 'weierp' => '℘',
- 'uuml' => '\xfc',
- 'part' => '∂',
- 'icirc' => '\xee',
- 'delta' => 'δ',
- 'omicron' => 'ο',
- 'upsilon' => 'υ',
- 'Iuml' => '\xcf',
- 'Lambda' => 'Λ',
- 'Xi' => 'Ξ',
- 'kappa' => 'κ',
- 'ccedil' => '\xe7',
- 'Ucirc' => '\xdb',
- 'cap' => '∩',
- 'mu' => 'μ',
- 'scaron' => 'š',
- 'lsquo' => '‘',
- 'isin' => '∈',
- 'Zeta' => 'Ζ',
- 'supe' => '⊇',
- 'deg' => '\xb0',
- 'and' => '∧',
- 'tau' => 'τ',
- 'pound' => '\xa3',
- 'hellip' => '…',
- 'curren' => '\xa4',
- 'int' => '∫',
- 'ucirc' => '\xfb',
- 'rfloor' => '⌋',
- 'ensp' => ' ',
- 'crarr' => '↵',
- 'ugrave' => '\xf9',
- 'notin' => '∉',
- 'exist' => '∃',
- 'uArr' => '⇑',
- 'cong' => '≅',
- 'Dagger' => '‡',
- 'oplus' => '⊕',
- 'times' => '\xd7',
- 'atilde' => '\xe3',
- 'piv' => 'ϖ',
- 'ni' => '∋',
- 'Phi' => 'Φ',
- 'lsaquo' => '‹',
- 'Uacute' => '\xda',
- 'Omicron' => 'Ο',
- 'ang' => '∠',
- 'ne' => '≠',
- 'iquest' => '\xbf',
- 'eta' => 'η',
- 'yacute' => '\xfd',
- 'Rho' => 'Ρ',
- 'uacute' => '\xfa',
- 'Alpha' => 'Α',
- 'zeta' => 'ζ',
- 'Omega' => 'Ω',
- 'nu' => 'ν',
- 'sim' => '∼',
- 'sect' => '\xa7',
- 'phi' => 'φ',
- 'sigmaf' => 'ς',
- 'macr' => '\xaf',
- 'minus' => '−',
- 'Ccedil' => '\xc7',
- 'ordm' => '\xba',
- 'epsilon' => 'ε',
- 'beta' => 'β',
- 'rArr' => '⇒',
- 'rho' => 'ρ',
- 'aacute' => '\xe1',
- 'eacute' => '\xe9',
- 'omega' => 'ω',
- 'middot' => '\xb7',
- 'Gamma' => 'Γ',
- 'Iacute' => '\xcd',
- 'lang' => '〈',
- 'spades' => '♠',
- 'rsquo' => '’',
- 'uml' => '\xa8',
- 'thorn' => '\xfe',
- 'ouml' => '\xf6',
- 'thetasym' => 'ϑ',
- 'or' => '∨',
- 'raquo' => '\xbb',
- 'acirc' => '\xe2',
- 'ldquo' => '“',
- 'hearts' => '♥',
- 'sigma' => 'σ',
- 'oacute' => '\xf3',
-=end
-}
-
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 => true,
+ :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.)")
+
def initialize
super
@registry.set_default(Array.new)
"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 unescape_title(htmldata)
- # first pass -- let CGI try to attack it...
- htmldata = CGI::unescapeHTML htmldata
-
- # second pass -- destroy the remaining bits...
- htmldata.gsub(/(&(.+?);)/) {
- symbol = $2
-
- # remove the 0-paddng from unicode integers
- if symbol =~ /#(.+)/
- symbol = "##{$1.to_i.to_s}"
- end
-
- # output the symbol's irc-translated character, or a * if it's unknown
- UNESCAPE_TABLE[symbol] || '*'
- }
- end
-
def get_title_from_html(pagedata)
return unless TITLE_RE.match(pagedata)
title = $1.strip.gsub(/\s*\n+\s*/, " ")
- title = unescape_title title
+ title = Utils.decode_html_entities title
title = title[0..255] if title.length > 255
"[Link Info] title: #{title}"
end
- def get_title_for_url(uri_str, depth=10)
+ def read_data_from_response(response, amount)
+
+ amount_read = 0
+ chunks = []
+
+ response.read_body do |chunk| # read body now
+
+ amount_read += chunk.length
+
+ if amount_read > amount
+ amount_of_overflow = amount_read - amount
+ chunk = chunk[0...-amount_of_overflow]
+ end
+
+ chunks << chunk
+
+ break if amount_read >= amount
+
+ end
+
+ chunks.join('')
+
+ end
+
+ def get_title_for_url(uri_str, depth=@bot.config['http.max_redir'])
# This god-awful mess is what the ruby http library has reduced me to.
# Python's HTTP lib is so much nicer. :~(
-
+
if depth == 0
raise "Error: Maximum redirects hit."
end
-
- puts "+ Getting #{uri_str}"
- url = URI.parse(uri_str)
+
+ debug "+ Getting #{uri_str.to_s}"
+ url = uri_str.kind_of?(URI) ? uri_str : URI.parse(uri_str)
return if url.scheme !~ /https?/
-
- puts "+ connecting to #{url.host}:#{url.port}"
- http = @bot.httputil.get_proxy(url)
- title = http.start do |http|
- url.path = '/' if url.path == ''
- head = http.request_head(url.path)
- case head
- when Net::HTTPRedirection then
- # call self recursively if this is a redirect
- redirect_to = head['location']
- puts "+ redirect location: #{redirect_to}"
- url = URI.join url.to_s, redirect_to
- puts "+ whee, redirecting to #{url.to_s}!"
- title = get_title_for_url(url.to_s, depth-1)
- when Net::HTTPSuccess then
- if head['content-type'] =~ /^text\// and (not head['content-length'] or head['content-length'].to_i < 400000)
- # since the content is 'text/*' and is small enough to
- # be a webpage, retrieve the title from the page
- puts "+ getting #{url.request_uri}"
- response = http.request_get(url.request_uri)
- return get_title_from_html(response.body)
+
+ title = nil
+
+ debug "+ connecting to #{url.host}:#{url.port}"
+ http = @bot.httputil.get_proxy(url)
+ http.start { |http|
+
+ http.request_get(url.request_uri(), @bot.httputil.headers) { |response|
+
+ case response
+ when Net::HTTPRedirection
+ # call self recursively if this is a redirect
+ redirect_to = response['location'] || '/'
+ debug "+ redirect location: #{redirect_to.inspect}"
+ url = URI.join(url.to_s, redirect_to)
+ debug "+ whee, redirecting to #{url.to_s}!"
+ return get_title_for_url(url, depth-1)
+ when Net::HTTPSuccess
+ if response['content-type'] =~ /^text\//
+ # since the content is 'text/*' and is small enough to
+ # be a webpage, retrieve the title from the page
+ debug "+ getting #{url.request_uri}"
+ # was 5*10^4 ... seems to much to me ... 4k should be enough for everybody ;)
+ data = read_data_from_response(response, 4096)
+ return get_title_from_html(data)
+ else
+ 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')
+ size = size ? ", size: #{size} bytes" : ""
+ return "[Link Info] type: #{response['content-type']}#{size}"
+ end
+ end
else
- # content doesn't have title, just display info.
- size = head['content-length'].gsub(/(\d)(?=\d{3}+(?:\.|$))(\d{3}\..*)?/,'\1,\2')
- #lastmod = head['last-modified']
- return "[Link Info] type: #{head['content-type']}#{size ? ", size: #{size} bytes" : ""}"
- end
- when Net::HTTPClientError then
- return "[Link Info] Error getting link (#{head.code} - #{head.message})"
- when Net::HTTPServerError then
- return "[Link Info] Error getting link (#{head.code} - #{head.message})"
- end
- end
+ return "[Link Info] Error getting link (#{response.code} - #{response.message})"
+ end # end of "case response"
+
+ } # end of request block
+ } # end of http start block
+
+ return title
+
rescue SocketError => e
return "[Link Info] Error connecting to site (#{e.message})"
end
if @bot.config['url.display_link_info']
debug "Getting title for #{urlstr}..."
+ begin
title = get_title_for_url urlstr
if title
m.reply title
debug "Title found!"
else
debug "Title not found!"
- end
+ end
+ rescue => e
+ debug "Failed: #{e}"
+ 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 > @bot.config['url.max_urls']