mcs/class/Mono.C5/UserGuideExamples/Anagrams.cs

   1 /*
   2  Copyright (c) 2003-2006 Niels Kokholm and Peter Sestoft
   3  Permission is hereby granted, free of charge, to any person obtaining a copy
   4  of this software and associated documentation files (the "Software"), to deal
   5  in the Software without restriction, including without limitation the rights
   6  to use, copy, modify, merge, publish, distribute, sublicense, and/or sell
   7  copies of the Software, and to permit persons to whom the Software is
   8  furnished to do so, subject to the following conditions:
   9
  10  The above copyright notice and this permission notice shall be included in
  11  all copies or substantial portions of the Software.
  12
  13  THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR
  14  IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY,
  15  FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL THE
  16  AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER
  17  LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM,
  18  OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE
  19  SOFTWARE.
  20 */
  21
  22 // C5 example: anagrams 2004-08-08, 2004-11-16
  23
  24 // Compile with
  25 //   csc /r:C5.dll Anagrams.cs
  26
  27 using System;
  28 using System.IO;                        // StreamReader, TextReader
  29 using System.Text;                      // Encoding
  30 using System.Text.RegularExpressions;   // Regex
  31 using C5;
  32 using SCG = System.Collections.Generic;
  33
  34 namespace Anagrams
  35 {
  36   class MyTest
  37   {
  38     public static void Main(String[] args)
  39     {
  40       Console.OutputEncoding = Encoding.GetEncoding("iso-8859-1");
  41       SCG.IEnumerable<String> ss;
  42       if (args.Length == 2)
  43         ss = ReadFileWords(args[0], int.Parse(args[1]));
  44       else
  45         ss = args;
  46       // foreach (String s in FirstAnagramOnly(ss))
  47       //   Console.WriteLine(s);
  48       //   Console.WriteLine("===");
  49       Timer t = new Timer();
  50       SCG.IEnumerable<SCG.IEnumerable<String>> classes = AnagramClasses(ss);
  51       int count = 0;
  52       foreach (SCG.IEnumerable<String> anagramClass in classes)
  53       {
  54         count++;
  55         // foreach (String s in anagramClass)
  56         //   Console.Write(s + " ");
  57         // Console.WriteLine();
  58       }
  59       Console.WriteLine("{0} non-trivial anagram classes", count);
  60       Console.WriteLine(t.Check());
  61     }
  62
  63     // Read words at most n words from a file
  64
  65     public static SCG.IEnumerable<String> ReadFileWords(String filename, int n)
  66     {
  67       Regex delim = new Regex("[^a-zæøåA-ZÆØÅ0-9-]+");
  68       Encoding enc = Encoding.GetEncoding("iso-8859-1");
  69       using (TextReader rd = new StreamReader(filename, enc))
  70       {
  71         for (String line = rd.ReadLine(); line != null; line = rd.ReadLine())
  72         {
  73           foreach (String s in delim.Split(line))
  74             if (s != "")
  75               yield return s.ToLower();
  76           if (--n == 0)
  77             yield break;
  78         }
  79       }
  80     }
  81
  82     // From an anagram point of view, a word is just a bag of
  83     // characters.  So an anagram class is represented as TreeBag<char>
  84     // which permits fast equality comparison -- we shall use them as
  85     // elements of hash sets or keys in hash maps.
  86
  87     public static TreeBag<char> AnagramClass(String s)
  88     {
  89       TreeBag<char> anagram = new TreeBag<char>(Comparer<char>.Default, EqualityComparer<char>.Default);
  90       foreach (char c in s)
  91         anagram.Add(c);
  92       return anagram;
  93     }
  94
  95     // Given a sequence of strings, return only the first member of each
  96     // anagram class.
  97
  98     public static SCG.IEnumerable<String> FirstAnagramOnly(SCG.IEnumerable<String> ss)
  99     {
 100       SCG.IEqualityComparer<TreeBag<char>> tbh
 101         = UnsequencedCollectionEqualityComparer<TreeBag<char>, char>.Default;
 102       HashSet<TreeBag<char>> anagrams = new HashSet<TreeBag<char>>(tbh);
 103       foreach (String s in ss)
 104       {
 105         TreeBag<char> anagram = AnagramClass(s);
 106         if (!anagrams.Contains(anagram))
 107         {
 108           anagrams.Add(anagram);
 109           yield return s;
 110         }
 111       }
 112     }
 113
 114     // Given a sequence of strings, return all non-trivial anagram
 115     // classes.  Should use a *sequenced* equalityComparer on a TreeBag<char>,
 116     // obviously: after all, characters can be sorted by ASCII code.  On
 117     // 347 000 distinct Danish words this takes 70 cpu seconds, 180 MB
 118     // memory, and 263 wall-clock seconds (due to swapping).
 119
 120     // Using a TreeBag<char> and a sequenced equalityComparer takes 82 cpu seconds
 121     // and 180 MB RAM to find the 26,058 anagram classes among 347,000
 122     // distinct words.
 123
 124     // Using an unsequenced equalityComparer on TreeBag<char> or HashBag<char>
 125     // makes it criminally slow: at least 1200 cpu seconds.  This must
 126     // be because many bags get the same hash code, so that there are
 127     // many collisions.  But exactly how the unsequenced equalityComparer works is
 128     // not clear ... or is it because unsequenced equality is slow?
 129
 130     public static SCG.IEnumerable<SCG.IEnumerable<String>> AnagramClasses(SCG.IEnumerable<String> ss)
 131     {
 132       bool unseq = true;
 133       IDictionary<TreeBag<char>, TreeSet<String>> classes;
 134       if (unseq)
 135       {
 136         SCG.IEqualityComparer<TreeBag<char>> unsequencedTreeBagEqualityComparer
 137     = UnsequencedCollectionEqualityComparer<TreeBag<char>, char>.Default;
 138         classes = new HashDictionary<TreeBag<char>, TreeSet<String>>(unsequencedTreeBagEqualityComparer);
 139       }
 140       else
 141       {
 142         SCG.IEqualityComparer<TreeBag<char>> sequencedTreeBagEqualityComparer
 143     = SequencedCollectionEqualityComparer<TreeBag<char>, char>.Default;
 144         classes = new HashDictionary<TreeBag<char>, TreeSet<String>>(sequencedTreeBagEqualityComparer);
 145       }
 146       foreach (String s in ss)
 147       {
 148         TreeBag<char> anagram = AnagramClass(s);
 149         TreeSet<String> anagramClass;
 150         if (!classes.Find(anagram, out anagramClass))
 151           classes[anagram] = anagramClass = new TreeSet<String>();
 152         anagramClass.Add(s);
 153       }
 154       foreach (TreeSet<String> anagramClass in classes.Values)
 155         if (anagramClass.Count > 1)
 156           yield return anagramClass;
 157     }
 158   }
 159
 160 // Crude timing utility ----------------------------------------
 161
 162   public class Timer
 163   {
 164     private DateTime start;
 165
 166     public Timer()
 167     {
 168       start = DateTime.Now;
 169     }
 170
 171     public double Check()
 172     {
 173       TimeSpan dur = DateTime.Now - start;
 174       return dur.TotalSeconds;
 175     }
 176   }
 177 }