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
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}")
77 # Paragraph and br tags are converted to whitespace
78 txt.gsub!(/<\/?(p|br)(?:\s+[^>]*)?\s*\/?\s*>/i, ' ')
82 # Superscripts and subscripts are turned into ^{...} and _{...}
83 # where the {} are omitted for single characters
84 txt.gsub!(/<sup>(.*?)<\/sup>/, '^{\1}')
85 txt.gsub!(/<sub>(.*?)<\/sub>/, '_{\1}')
86 txt.gsub!(/(^|_)\{(.)\}/, '\1\2')
88 # All other tags are just removed
89 txt.gsub!(/<[^>]+>/, '')
91 # Convert HTML entities. We do it now to be able to handle stuff
93 txt = Utils.decode_html_entities(txt)
95 # Remove double formatting options, since they only waste bytes
96 txt.gsub!(/#{Bold}(\s*)#{Bold}/, '\1')
97 txt.gsub!(/#{Underline}(\s*)#{Underline}/, '\1')
99 # Simplify whitespace that appears on both sides of a formatting option
100 txt.gsub!(/\s+(#{Bold}|#{Underline})\s+/, ' \1')
101 txt.sub!(/\s+(#{Bold}|#{Underline})\z/, '\1')
102 txt.sub!(/\A(#{Bold}|#{Underline})\s+/, '\1')
104 # And finally whitespace is squeezed
105 txt.gsub!(/\s+/, ' ')
107 # Decode entities and strip whitespace
111 # As above, but modify the receiver
115 replace self.ircify_html
116 return self unless self.hash == old_hash
119 # This method will strip all HTML crud from the receiver
122 self.gsub(/<[^>]+>/, '').gsub(/&/,'&').gsub(/"/,'"').gsub(/</,'<').gsub(/>/,'>').gsub(/&ellip;/,'...').gsub(/'/, "'").gsub("\n",'')
127 # Extensions to the Regexp class, with some common and/or complex regular
132 # A method to build a regexp that matches a list of something separated by
133 # optional commas and/or the word "and", an optionally repeated prefix,
135 def Regexp.new_list(reg, pfx = "")
136 if pfx.kind_of?(String) and pfx.empty?
137 return %r(#{reg}(?:,?(?:\s+and)?\s+#{reg})*)
139 return %r(#{reg}(?:,?(?:\s+and)?(?:\s+#{pfx})?\s+#{reg})*)
146 # Match a list of channel anmes separated by optional commas, whitespace
147 # and optionally the word "and"
148 CHAN_LIST = Regexp.new_list(GEN_CHAN)
150 # Match "in #channel" or "on #channel" and/or "in private" (optionally
151 # shortened to "in pvt"), returning the channel name or the word 'private'
152 # or 'pvt' as capture
153 IN_CHAN = /#{IN_ON}\s+(#{GEN_CHAN})|(here)|/
154 IN_CHAN_PVT = /#{IN_CHAN}|in\s+(private|pvt)/
156 # As above, but with channel lists
157 IN_CHAN_LIST_SFX = Regexp.new_list(/#{GEN_CHAN}|here/, IN_ON)
158 IN_CHAN_LIST = /#{IN_ON}\s+#{IN_CHAN_LIST_SFX}|anywhere|everywhere/
159 IN_CHAN_LIST_PVT_SFX = Regexp.new_list(/#{GEN_CHAN}|here|private|pvt/, IN_ON)
160 IN_CHAN_LIST_PVT = /#{IN_ON}\s+#{IN_CHAN_LIST_PVT_SFX}|anywhere|everywhere/
162 # Match a list of nicknames separated by optional commas, whitespace and
163 # optionally the word "and"
164 NICK_LIST = Regexp.new_list(GEN_NICK)
174 class BasicUserMessage
176 # We extend the BasicUserMessage class with a method that parses a string
177 # which is a channel list as matched by IN_CHAN(_LIST) and co. The method
178 # returns an array of channel names, where 'private' or 'pvt' is replaced
179 # by the Symbol :"?", 'here' is replaced by the channel of the message or
180 # by :"?" (depending on whether the message target is the bot or a
181 # Channel), and 'anywhere' and 'everywhere' are replaced by Symbol :*
183 def parse_channel_list(string)
184 return [:*] if [:anywhere, :everywhere].include? string.to_sym
186 /(?:^|,?(?:\s+and)?\s+)(?:in|on\s+)?(#{Regexp::Irc::GEN_CHAN}|here|private|pvt)/