4 # :title: Standard classes extensions
6 # Author:: Giuseppe "Oblomov" Bilotta <giuseppe.bilotta@gmail.com>
7 # Copyright:: (C) 2006,2007 Giuseppe Bilotta
10 # This file collects extensions to standard Ruby classes and to some core rbot
11 # classes to be used by the various plugins
13 # Please note that global symbols have to be prefixed by :: because this plugin
14 # will be read into an anonymous module
17 # Extensions to the Array class
21 # This method returns a random element from the array, or nil if the array is
25 return nil if self.empty?
26 self[rand(self.length)]
30 # Extensions for the Numeric classes
34 # This method forces a real number to be not more than a given positive
35 # number or not less than a given positive number, or between two any given
38 def clip(left,right=0)
39 raise ArgumentError unless left.kind_of?(Numeric) and right.kind_of?(Numeric)
48 # Extensions to the String class
50 # TODO make ircify_html() accept an Hash of options, and make riphtml() just
51 # call ircify_html() with stronger purify options.
55 # This method will return a purified version of the receiver, with all HTML
56 # stripped off and some of it converted to IRC formatting
58 def ircify_html(opts={})
62 txt.gsub!(/<script(?:\s+[^>]*)?>.*?<\/script>/im, "")
65 txt.gsub!(/<style(?:\s+[^>]*)?>.*?<\/style>/im, "")
67 # bold and strong -> bold
68 txt.gsub!(/<\/?(?:b|strong)(?:\s+[^>]*)?>/im, "#{Bold}")
70 # italic, emphasis and underline -> underline
71 txt.gsub!(/<\/?(?:i|em|u)(?:\s+[^>]*)?>/im, "#{Underline}")
73 ## This would be a nice addition, but the results are horrible
74 ## Maybe make it configurable?
75 # txt.gsub!(/<\/?a( [^>]*)?>/, "#{Reverse}")
76 case val = opts[:a_href]
77 when Reverse, Bold, Underline
78 txt.gsub!(/<(?:\/a\s*|a (?:[^>]*\s+)?href\s*=\s*(?:[^>]*\s*)?)>/, val)
80 # Not good for nested links, but the best we can do without something like hpricot
81 txt.gsub!(/<a (?:[^>]*\s+)?href\s*=\s*(?:([^"'>][^\s>]*)\s+|"((?:[^"]|\\")*)"|'((?:[^']|\\')*)')(?:[^>]*\s+)?>(.*?)<\/a>/) { |match|
83 debug [$1, $2, $3, $4].inspect
89 warn "unknown :a_href option #{val} passed to ircify_html" if val
92 # Paragraph and br tags are converted to whitespace
93 txt.gsub!(/<\/?(p|br)(?:\s+[^>]*)?\s*\/?\s*>/i, ' ')
97 # Superscripts and subscripts are turned into ^{...} and _{...}
98 # where the {} are omitted for single characters
99 txt.gsub!(/<sup>(.*?)<\/sup>/, '^{\1}')
100 txt.gsub!(/<sub>(.*?)<\/sub>/, '_{\1}')
101 txt.gsub!(/(^|_)\{(.)\}/, '\1\2')
103 # All other tags are just removed
104 txt.gsub!(/<[^>]+>/, '')
106 # Convert HTML entities. We do it now to be able to handle stuff
108 txt = Utils.decode_html_entities(txt)
110 # Remove double formatting options, since they only waste bytes
111 txt.gsub!(/#{Bold}(\s*)#{Bold}/, '\1')
112 txt.gsub!(/#{Underline}(\s*)#{Underline}/, '\1')
114 # Simplify whitespace that appears on both sides of a formatting option
115 txt.gsub!(/\s+(#{Bold}|#{Underline})\s+/, ' \1')
116 txt.sub!(/\s+(#{Bold}|#{Underline})\z/, '\1')
117 txt.sub!(/\A(#{Bold}|#{Underline})\s+/, '\1')
119 # And finally whitespace is squeezed
120 txt.gsub!(/\s+/, ' ')
122 # Decode entities and strip whitespace
126 # As above, but modify the receiver
128 def ircify_html!(opts={})
130 replace self.ircify_html(opts)
131 return self unless self.hash == old_hash
134 # This method will strip all HTML crud from the receiver
137 self.gsub(/<[^>]+>/, '').gsub(/&/,'&').gsub(/"/,'"').gsub(/</,'<').gsub(/>/,'>').gsub(/&ellip;/,'...').gsub(/'/, "'").gsub("\n",'')
142 # Extensions to the Regexp class, with some common and/or complex regular
147 # A method to build a regexp that matches a list of something separated by
148 # optional commas and/or the word "and", an optionally repeated prefix,
150 def Regexp.new_list(reg, pfx = "")
151 if pfx.kind_of?(String) and pfx.empty?
152 return %r(#{reg}(?:,?(?:\s+and)?\s+#{reg})*)
154 return %r(#{reg}(?:,?(?:\s+and)?(?:\s+#{pfx})?\s+#{reg})*)
161 # Match a list of channel anmes separated by optional commas, whitespace
162 # and optionally the word "and"
163 CHAN_LIST = Regexp.new_list(GEN_CHAN)
165 # Match "in #channel" or "on #channel" and/or "in private" (optionally
166 # shortened to "in pvt"), returning the channel name or the word 'private'
167 # or 'pvt' as capture
168 IN_CHAN = /#{IN_ON}\s+(#{GEN_CHAN})|(here)|/
169 IN_CHAN_PVT = /#{IN_CHAN}|in\s+(private|pvt)/
171 # As above, but with channel lists
172 IN_CHAN_LIST_SFX = Regexp.new_list(/#{GEN_CHAN}|here/, IN_ON)
173 IN_CHAN_LIST = /#{IN_ON}\s+#{IN_CHAN_LIST_SFX}|anywhere|everywhere/
174 IN_CHAN_LIST_PVT_SFX = Regexp.new_list(/#{GEN_CHAN}|here|private|pvt/, IN_ON)
175 IN_CHAN_LIST_PVT = /#{IN_ON}\s+#{IN_CHAN_LIST_PVT_SFX}|anywhere|everywhere/
177 # Match a list of nicknames separated by optional commas, whitespace and
178 # optionally the word "and"
179 NICK_LIST = Regexp.new_list(GEN_NICK)
189 class BasicUserMessage
191 # We extend the BasicUserMessage class with a method that parses a string
192 # which is a channel list as matched by IN_CHAN(_LIST) and co. The method
193 # returns an array of channel names, where 'private' or 'pvt' is replaced
194 # by the Symbol :"?", 'here' is replaced by the channel of the message or
195 # by :"?" (depending on whether the message target is the bot or a
196 # Channel), and 'anywhere' and 'everywhere' are replaced by Symbol :*
198 def parse_channel_list(string)
199 return [:*] if [:anywhere, :everywhere].include? string.to_sym
201 /(?:^|,?(?:\s+and)?\s+)(?:in|on\s+)?(#{Regexp::Irc::GEN_CHAN}|here|private|pvt)/