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)]
31 # Extensions to the String class
33 # TODO make ircify_html() accept an Hash of options, and make riphtml() just
34 # call ircify_html() with stronger purify options.
38 # This method will return a purified version of the receiver, with all HTML
39 # stripped off and some of it converted to IRC formatting
44 # bold and strong -> bold
45 txt.gsub!(/<\/?(?:b|strong)\s*>/, "#{Bold}")
47 # italic, emphasis and underline -> underline
48 txt.gsub!(/<\/?(?:i|em|u)\s*>/, "#{Underline}")
50 ## This would be a nice addition, but the results are horrible
51 ## Maybe make it configurable?
52 # txt.gsub!(/<\/?a( [^>]*)?>/, "#{Reverse}")
54 # Paragraph and br tags are converted to whitespace.
55 txt.gsub!(/<\/?(p|br)\s*\/?\s*>/, ' ')
58 # All other tags are just removed
59 txt.gsub!(/<[^>]+>/, '')
61 # Remove double formatting options, since they only waste bytes
62 txt.gsub!(/#{Bold}(\s*)#{Bold}/, '\1')
63 txt.gsub!(/#{Underline}(\s*)#{Underline}/, '\1')
65 # And finally whitespace is squeezed
68 # Decode entities and strip whitespace
69 return Utils.decode_html_entities(txt).strip!
72 # This method will strip all HTML crud from the receiver
75 self.gsub(/<[^>]+>/, '').gsub(/&/,'&').gsub(/"/,'"').gsub(/</,'<').gsub(/>/,'>').gsub(/&ellip;/,'...').gsub(/'/, "'").gsub("\n",'')
80 # Extensions to the Regexp class, with some common and/or complex regular
85 # A method to build a regexp that matches a list of something separated by
86 # optional commas and/or the word "and", an optionally repeated prefix,
88 def Regexp.new_list(reg, pfx = "")
89 if pfx.kind_of?(String) and pfx.empty?
90 return %r(#{reg}(?:,?(?:\s+and)?\s+#{reg})*)
92 return %r(#{reg}(?:,?(?:\s+and)?(?:\s+#{pfx})?\s+#{reg})*)
98 # We start with some IRC related regular expressions, used to match
99 # Irc::User nicks and Irc::Channel names
101 # For each of them we define three versions of the regular expression:
102 # * a generic one, which should match for any server but may turn out to
103 # match more than a specific server would accept
104 # * an RFC-compliant matcher
105 # * TODO a server-specific one that uses the Irc::Server#supports method to build
106 # a matcher valid for a particular server.
110 CHAN_SAFE = /![A-Z0-9]{5}/
111 CHAN_ANY = /[^\x00\x07\x0A\x0D ,:]/
112 GEN_CHAN = /(?:#{CHAN_FIRST}|#{CHAN_SAFE})#{CHAN_ANY}+/
113 RFC_CHAN = /#{CHAN_FIRST}#{CHAN_ANY}{1,49}|#{CHAN_SAFE}#{CHAN_ANY}{1,44}/
115 CHAN_LIST = Regexp.new_list(GEN_CHAN)
117 # Match "in #channel" or "on #channel" and/or "in private" (optionally
118 # shortened to "in pvt"), returning the channel name or the word 'private'
119 # or 'pvt' as capture
120 IN_CHAN = /#{IN_ON}\s+(#{GEN_CHAN})|(here)|/
121 IN_CHAN_PVT = /#{IN_CHAN}|in\s+(private|pvt)/
123 # As above, but with channel lists
124 IN_CHAN_LIST_SFX = Regexp.new_list(/#{GEN_CHAN}|here/, IN_ON)
125 IN_CHAN_LIST = /#{IN_ON}\s+#{IN_CHAN_LIST_SFX}|anywhere|everywhere/
126 IN_CHAN_LIST_PVT_SFX = Regexp.new_list(/#{GEN_CHAN}|here|private|pvt/, IN_ON)
127 IN_CHAN_LIST_PVT = /#{IN_ON}\s+#{IN_CHAN_LIST_PVT_SFX}|anywhere|everywhere/
129 SPECIAL_CHAR = /[\x5b-\x60\x7b-\x7d]/
130 NICK_FIRST = /#{SPECIAL_CHAR}|[[:alpha:]]/
131 NICK_ANY = /#{SPECIAL_CHAR}|[[:alnum:]]|-/
132 GEN_NICK = /#{NICK_FIRST}#{NICK_ANY}+/
133 RFC_NICK = /#{NICK_FIRST}#{NICK_ANY}{0,8}/
135 # Match a list of nicknames separated by optional commas, whitespace and
136 # optionally the word "and"
137 NICK_LIST = Regexp.new_list(GEN_CHAN)
141 # Next, some general purpose ones
143 HEX_DIGIT = /[0-9A-Fa-f]/
144 HEX_DIGITS = /#{HEX_DIGIT}+/
145 HEX_OCTET = /#{HEX_DIGIT}#{HEX_DIGIT}?/
146 DEC_OCTET = /[01]?\d?\d|2[0-4]\d|25[0-5]/
147 DEC_IP_ADDR = /#{DEC_OCTET}.#{DEC_OCTET}.#{DEC_OCTET}.#{DEC_OCTET}/
148 HEX_IP_ADDR = /#{HEX_OCTET}.#{HEX_OCTET}.#{HEX_OCTET}.#{HEX_OCTET}/
149 IP_ADDR = /#{DEC_IP_ADDR}|#{HEX_IP_ADDR}/
157 class BasicUserMessage
159 # We extend the BasicUserMessage class with a method that parses a string
160 # which is a channel list as matched by IN_CHAN(_LIST) and co. The method
161 # returns an array of channel names, where 'private' or 'pvt' is replaced
162 # by the Symbol :"?", 'here' is replaced by the channel of the message or
163 # by :"?" (depending on whether the message target is the bot or a
164 # Channel), and 'anywhere' and 'everywhere' are replaced by Symbol :*
166 def parse_channel_list(string)
167 return [:*] if [:anywhere, :everywhere].include? string.to_sym
169 /(?:^|,?(?:\s+and)?\s+)(?:in|on\s+)?(#{Regexp::Irc::GEN_CHAN}|here|private|pvt)/