utils: support hex HTML entities

[user/henk/code/ruby/rbot.git] / lib / rbot / core / utils / wordlist.rb

diff --git a/lib/rbot/core/utils/wordlist.rb b/lib/rbot/core/utils/wordlist.rb
index 23e141ac3b32497941b09f5fb5a7471e992ce176..81d7d775ef329f4058531cdb8444365a2d87a0a4 100755 (executable)
--- a/lib/rbot/core/utils/wordlist.rb
+++ b/lib/rbot/core/utils/wordlist.rb
@@ -11,13 +11,13 @@ module ::Irc
 class Bot
 class Wordlist
   def self.wordlist_base
-    @@wordlist_base ||= File.join(Utils.bot.botclass, 'wordlists')
+    @@wordlist_base ||= Utils.bot.path 'wordlists'
   end
 
-  def self.get(path, options={})
+  def self.get(where, options={})
     opts = { :spaces => false }.merge(options)
 
-    wordlist_path = File.join(wordlist_base, path)
+    wordlist_path = File.join(wordlist_base, where)
     raise "wordlist not found: #{wordlist_path}" unless File.exist?(wordlist_path)
 
     # Location is a directory -> combine all lists beneath it
@@ -33,12 +33,34 @@ class Wordlist
       File.readlines(wordlist_path)
     end
 
-    wordlist.map! { |l| l.strip }
+    # wordlists are assumed to be UTF-8, but we need to strip the BOM, if present
+    wordlist.map! { |l| l.sub("\xef\xbb\xbf",'').strip }
     wordlist.reject do |word|
       word =~ /\s/ && !opts[:spaces] ||
       word.empty?
     end
   end
+
+  # Return an array with the list of available wordlists.
+  # Available options:
+  # pattern:: pattern that should be matched by the wordlist filename
+  def self.list(options={})
+    pattern = options[:pattern] || "**"
+    # refuse patterns that contain ../
+    return [] if pattern =~ /\.\.\//
+    striplen = self.wordlist_base.length+1
+    Dir.glob(File.join(self.wordlist_base, pattern)).map { |name|
+      name[striplen..-1]
+    }
+  end
+
+  def self.exist?(path)
+    fn = path.to_s
+    # refuse to check outside of the wordlist base directory
+    return false if fn =~ /\.\.\//
+    File.exist?(File.join(self.wordlist_base, fn))
+  end
+
+end
 end
 end
-end
\ No newline at end of file