X-Git-Url: https://git.netwichtig.de/gitweb/?a=blobdiff_plain;f=lib%2Frbot%2Firc.rb;h=c2c1e82fdd518c30889a1eed9cad0830857ea9de;hb=c705ba5a89cd7b5c19677f4950c9784828ffc5c6;hp=6d46f75fc3500a47730bb30c0d3382378699f900;hpb=82b553fe5a5ad6406045229d5929c6328da6ecc2;p=user%2Fhenk%2Fcode%2Fruby%2Frbot.git

diff --git a/lib/rbot/irc.rb b/lib/rbot/irc.rb
index 6d46f75f..c2c1e82f 100644
--- a/lib/rbot/irc.rb
+++ b/lib/rbot/irc.rb
@@ -475,6 +475,103 @@ class ArrayOf < Array
 end
 
 
+# We extend the Regexp class with an Irc module which will contain some
+# Irc-specific regexps
+#
+class Regexp
+
+  # We start with some general-purpose ones which will be used in the
+  # Irc module too, but are useful regardless
+  DIGITS = /\d+/
+  HEX_DIGIT = /[0-9A-Fa-f]/
+  HEX_DIGITS = /#{HEX_DIGIT}+/
+  HEX_OCTET = /#{HEX_DIGIT}#{HEX_DIGIT}?/
+  DEC_OCTET = /[01]?\d?\d|2[0-4]\d|25[0-5]/
+  DEC_IP_ADDR = /#{DEC_OCTET}.#{DEC_OCTET}.#{DEC_OCTET}.#{DEC_OCTET}/
+  HEX_IP_ADDR = /#{HEX_OCTET}.#{HEX_OCTET}.#{HEX_OCTET}.#{HEX_OCTET}/
+  IP_ADDR = /#{DEC_IP_ADDR}|#{HEX_IP_ADDR}/
+
+  # IPv6, from Resolv::IPv6, without the \A..\z anchors
+  HEX_16BIT = /#{HEX_DIGIT}{1,4}/
+  IP6_8Hex = /(?:#{HEX_16BIT}:){7}#{HEX_16BIT}/
+  IP6_CompressedHex = /((?:#{HEX_16BIT}(?::#{HEX_16BIT})*)?)::((?:#{HEX_16BIT}(?::#{HEX_16BIT})*)?)/
+  IP6_6Hex4Dec = /((?:#{HEX_16BIT}:){6,6})#{DEC_IP_ADDR}/
+  IP6_CompressedHex4Dec = /((?:#{HEX_16BIT}(?::#{HEX_16BIT})*)?)::((?:#{HEX_16BIT}:)*)#{DEC_IP_ADDR}/
+  IP6_ADDR = /(?:#{IP6_8Hex})|(?:#{IP6_CompressedHex})|(?:#{IP6_6Hex4Dec})|(?:#{IP6_CompressedHex4Dec})/
+
+  # We start with some IRC related regular expressions, used to match
+  # Irc::User nicks and users and Irc::Channel names
+  #
+  # For each of them we define two versions of the regular expression:
+  #  * a generic one, which should match for any server but may turn out to
+  #    match more than a specific server would accept
+  #  * an RFC-compliant matcher
+  #
+  module Irc
+
+    # Channel-name-matching regexps
+    CHAN_FIRST = /[#&+]/
+    CHAN_SAFE = /![A-Z0-9]{5}/
+    CHAN_ANY = /[^\x00\x07\x0A\x0D ,:]/
+    GEN_CHAN = /(?:#{CHAN_FIRST}|#{CHAN_SAFE})#{CHAN_ANY}+/
+    RFC_CHAN = /#{CHAN_FIRST}#{CHAN_ANY}{1,49}|#{CHAN_SAFE}#{CHAN_ANY}{1,44}/
+
+    # Nick-matching regexps
+    SPECIAL_CHAR = /[\x5b-\x60\x7b-\x7d]/
+    NICK_FIRST = /#{SPECIAL_CHAR}|[[:alpha:]]/
+    NICK_ANY = /#{SPECIAL_CHAR}|[[:alnum:]]|-/
+    GEN_NICK = /#{NICK_FIRST}#{NICK_ANY}+/
+    RFC_NICK = /#{NICK_FIRST}#{NICK_ANY}{0,8}/
+
+    USER_CHAR = /[^\x00\x0a\x0d @]/
+    GEN_USER = /#{USER_CHAR}+/
+
+    # Host-matching regexps
+    HOSTNAME_COMPONENT = /[[:alnum:]](?:[[:alnum:]]|-)*[[:alnum:]]*/
+    HOSTNAME = /#{HOSTNAME_COMPONENT}(?:\.#{HOSTNAME_COMPONENT})*/
+    HOSTADDR = /#{IP_ADDR}|#{IP6_ADDR}/
+
+    GEN_HOST = /#{HOSTNAME}|#{HOSTADDR}/
+
+    # # FreeNode network replaces the host of affiliated users with
+    # # 'virtual hosts' 
+    # # FIXME we need the true syntax to match it properly ...
+    # PDPC_HOST_PART = /[0-9A-Za-z.-]+/
+    # PDPC_HOST = /#{PDPC_HOST_PART}(?:\/#{PDPC_HOST_PART})+/
+
+    # # NOTE: the final optional and non-greedy dot is needed because some
+    # # servers (e.g. FreeNode) send the hostname of the services as "services."
+    # # which is not RFC compliant, but sadly done.
+    # GEN_HOST_EXT = /#{PDPC_HOST}|#{GEN_HOST}\.??/ 
+
+    # Sadly, different networks have different, RFC-breaking ways of cloaking
+    # the actualy host address: see above for an example to handle FreeNode.
+    # Another example would be Azzurra, wich also inserts a "=" in the
+    # cloacked host. So let's just not care about this and go with the simplest
+    # thing:
+    GEN_HOST_EXT = /\S+/
+
+    # User-matching Regexp
+    GEN_USER_ID = /(#{GEN_NICK})(?:(?:!(#{GEN_USER}))?@(#{GEN_HOST_EXT}))?/
+
+    # Things such has the BIP proxy send invalid nicks in a complete netmask,
+    # so we want to match this, rather: this matches either a compliant nick
+    # or a a string with a very generic nick, a very generic hostname after an
+    # @ sign, and an optional user after a !
+    BANG_AT = /#{GEN_NICK}|\S+?(?:!\S+?)?@\S+?/
+
+    # # For Netmask, we want to allow wildcards * and ? in the nick
+    # # (they are already allowed in the user and host part
+    # GEN_NICK_MASK = /(?:#{NICK_FIRST}|[?*])?(?:#{NICK_ANY}|[?*])+/
+
+    # # Netmask-matching Regexp
+    # GEN_MASK = /(#{GEN_NICK_MASK})(?:(?:!(#{GEN_USER}))?@(#{GEN_HOST_EXT}))?/
+
+  end
+
+end
+
+
 module Irc
 
 
@@ -517,7 +614,9 @@ module Irc
       # Now we can see if the given string _str_ is an actual Netmask
       if str.respond_to?(:to_str)
         case str.to_str
-        when /^(?:(\S+?)(?:!(\S+)@(?:(\S+))?)?)?$/
+          # We match a pretty generic string, to work around non-compliant
+          # servers
+        when /^(?:(\S+?)(?:(?:!(\S+?))?@(\S+))?)?$/
           # We do assignment using our internal methods
           self.nick = $1
           self.user = $2
@@ -1331,8 +1430,8 @@ module Irc
           :typec => nil, # Type C: needs a parameter when set
           :typed => nil  # Type D: must not have a parameter
         },
-        :channellen => 200,
-        :chantypes => "#&",
+        :channellen => 50,
+        :chantypes => "#&!+",
         :excepts => nil,
         :idchan => {},
         :invex => nil,
@@ -1370,6 +1469,7 @@ module Irc
     def clear
       reset_lists
       reset_capabilities
+      @hostname = @version = @usermodes = @chanmodes = nil
     end
 
     # This method is used to parse a 004 RPL_MY_INFO line