toolkit/components/url-classifier/src/nsUrlClassifierUtils.h

   1 /* ***** BEGIN LICENSE BLOCK *****
   2  * Version: MPL 1.1/GPL 2.0/LGPL 2.1
   3  *
   4  * The contents of this file are subject to the Mozilla Public License Version
   5  * 1.1 (the "License"); you may not use this file except in compliance with
   6  * the License. You may obtain a copy of the License at
   7  * http://www.mozilla.org/MPL/
   8  *
   9  * Software distributed under the License is distributed on an "AS IS" basis,
  10  * WITHOUT WARRANTY OF ANY KIND, either express or implied. See the License
  11  * for the specific language governing rights and limitations under the
  12  * License.
  13  *
  14  * The Original Code is Url Classifier code
  15  *
  16  * The Initial Developer of the Original Code is
  17  * Google Inc.
  18  * Portions created by the Initial Developer are Copyright (C) 2007
  19  * the Initial Developer. All Rights Reserved.
  20  *
  21  * Contributor(s):
  22  *
  23  * Alternatively, the contents of this file may be used under the terms of
  24  * either the GNU General Public License Version 2 or later (the "GPL"), or
  25  * the GNU Lesser General Public License Version 2.1 or later (the "LGPL"),
  26  * in which case the provisions of the GPL or the LGPL are applicable instead
  27  * of those above. If you wish to allow use of your version of this file only
  28  * under the terms of either the GPL or the LGPL, and not to allow others to
  29  * use your version of this file under the terms of the MPL, indicate your
  30  * decision by deleting the provisions above and replace them with the notice
  31  * and other provisions required by the GPL or the LGPL. If you do not delete
  32  * the provisions above, a recipient may use your version of this file under
  33  * the terms of any one of the MPL, the GPL or the LGPL.
  34  *
  35  * ***** END LICENSE BLOCK ***** */
  36
  37 #ifndef nsUrlClassifierUtils_h_
  38 #define nsUrlClassifierUtils_h_
  39
  40 #include "nsAutoPtr.h"
  41 #include "nsIUrlClassifierUtils.h"
  42 #include "nsTArray.h"
  43 #include "nsDataHashtable.h"
  44
  45 class nsUrlClassifierUtils : public nsIUrlClassifierUtils
  46 {
  47 private:
  48   /**
  49    * A fast, bit-vector map for ascii characters.
  50    *
  51    * Internally stores 256 bits in an array of 8 ints.
  52    * Does quick bit-flicking to lookup needed characters.
  53    */
  54   class Charmap
  55   {
  56   public:
  57     Charmap(PRUint32 b0, PRUint32 b1, PRUint32 b2, PRUint32 b3,
  58             PRUint32 b4, PRUint32 b5, PRUint32 b6, PRUint32 b7)
  59     {
  60       mMap[0] = b0; mMap[1] = b1; mMap[2] = b2; mMap[3] = b3;
  61       mMap[4] = b4; mMap[5] = b5; mMap[6] = b6; mMap[7] = b7;
  62     }
  63
  64     /**
  65      * Do a quick lookup to see if the letter is in the map.
  66      */
  67     PRBool Contains(unsigned char c) const
  68     {
  69       return mMap[c >> 5] & (1 << (c & 31));
  70     }
  71
  72   private:
  73     // Store the 256 bits in an 8 byte array.
  74     PRUint32 mMap[8];
  75   };
  76
  77
  78 public:
  79   nsUrlClassifierUtils();
  80   ~nsUrlClassifierUtils() {}
  81
  82   NS_DECL_ISUPPORTS
  83   NS_DECL_NSIURLCLASSIFIERUTILS
  84
  85   nsresult Init();
  86
  87   nsresult CanonicalizeHostname(const nsACString & hostname,
  88                                 nsACString & _retval);
  89   nsresult CanonicalizePath(const nsACString & url, nsACString & _retval);
  90
  91   // This function will encode all "special" characters in typical url encoding,
  92   // that is %hh where h is a valid hex digit.  The characters which are encoded
  93   // by this function are any ascii characters under 32(control characters and
  94   // space), 37(%), and anything 127 or above (special characters).  Url is the
  95   // string to encode, ret is the encoded string.  Function returns true if
  96   // ret != url.
  97   PRBool SpecialEncode(const nsACString & url,
  98                        PRBool foldSlashes,
  99                        nsACString & _retval);
 100
 101   void ParseIPAddress(const nsACString & host, nsACString & _retval);
 102   void CanonicalNum(const nsACString & num,
 103                     PRUint32 bytes,
 104                     PRBool allowOctal,
 105                     nsACString & _retval);
 106
 107   // Convert an urlsafe base64 string to a normal base64 string.
 108   // This method will leave an already-normal base64 string alone.
 109   static void UnUrlsafeBase64(nsACString & str);
 110
 111   // Takes an urlsafe-base64 encoded client key and gives back binary
 112   // key data
 113   static nsresult DecodeClientKey(const nsACString & clientKey,
 114                                   nsACString & _retval);
 115 private:
 116   // Disallow copy constructor
 117   nsUrlClassifierUtils(const nsUrlClassifierUtils&);
 118
 119   // Function to tell if we should encode a character.
 120   PRBool ShouldURLEscape(const unsigned char c) const;
 121
 122   void CleanupHostname(const nsACString & host, nsACString & _retval);
 123
 124   nsAutoPtr<Charmap> mEscapeCharmap;
 125 };
 126
 127 // An MRU list of fragments.  This is used by the DB service to
 128 // keep a set of known-clean fragments that don't need a database
 129 // lookup.
 130 class nsUrlClassifierFragmentSet
 131 {
 132 public:
 133   nsUrlClassifierFragmentSet() : mFirst(nsnull), mLast(nsnull), mCapacity(16) {}
 134
 135   PRBool Init(PRUint32 maxEntries) {
 136     mCapacity = maxEntries;
 137     if (!mEntryStorage.SetCapacity(mCapacity))
 138       return PR_FALSE;
 139
 140     if (!mEntries.Init())
 141       return PR_FALSE;
 142
 143     return PR_TRUE;
 144   }
 145
 146   PRBool Put(const nsACString &fragment) {
 147     Entry *entry;
 148     if (mEntries.Get(fragment, &entry)) {
 149       // Remove this entry from the list, we'll add it back
 150       // to the front.
 151       UnlinkEntry(entry);
 152     } else {
 153       if (mEntryStorage.Length() < mEntryStorage.Capacity()) {
 154         entry = mEntryStorage.AppendElement();
 155         if (!entry)
 156           return PR_FALSE;
 157       } else {
 158         // Reuse the oldest entry.
 159         entry = mLast;
 160         UnlinkEntry(entry);
 161         mEntries.Remove(entry->mFragment);
 162       }
 163       entry->mFragment = fragment;
 164       mEntries.Put(fragment, entry);
 165     }
 166
 167     // Add the entry to the front of the list
 168     entry->mPrev = nsnull;
 169     entry->mNext = mFirst;
 170     mFirst = entry;
 171     if (!mLast) {
 172       mLast = entry;
 173     }
 174
 175     return PR_TRUE;
 176   }
 177
 178   PRBool Has(const nsACString &fragment) {
 179     return mEntries.Get(fragment, nsnull);
 180   }
 181
 182   void Clear() {
 183     mFirst = mLast = nsnull;
 184     mEntries.Clear();
 185     mEntryStorage.Clear();
 186     mEntryStorage.SetCapacity(mCapacity);
 187   }
 188
 189 private:
 190   // One entry in the set.  We maintain a doubly-linked list, with
 191   // the most recently used entry at the front.
 192   class Entry {
 193   public:
 194     Entry() : mNext(nsnull), mPrev(nsnull) {};
 195     ~Entry() { }
 196
 197     Entry *mNext;
 198     Entry *mPrev;
 199     nsCString mFragment;
 200   };
 201
 202   void UnlinkEntry(Entry *entry)
 203   {
 204     if (entry->mPrev)
 205       entry->mPrev->mNext = entry->mNext;
 206     else
 207       mFirst = entry->mNext;
 208
 209     if (entry->mNext)
 210       entry->mNext->mPrev = entry->mPrev;
 211     else
 212       mLast = entry->mPrev;
 213
 214     entry->mPrev = entry->mNext = nsnull;
 215   }
 216
 217   // The newest entry in the cache.
 218   Entry *mFirst;
 219   // The oldest entry in the cache.
 220   Entry *mLast;
 221
 222   // Max entries in the cache.
 223   PRUint32 mCapacity;
 224
 225   // Storage for the entries in this set.
 226   nsTArray<Entry> mEntryStorage;
 227
 228   // Entry lookup by fragment.
 229   nsDataHashtable<nsCStringHashKey, Entry*> mEntries;
 230 };
 231
 232 #endif // nsUrlClassifierUtils_h_