modules/libjar/nsZipArchive.h

   1 /* -*- Mode: C++; tab-width: 2; indent-tabs-mode: nil; c-basic-offset: 2 -*- */
   2 /* ***** BEGIN LICENSE BLOCK *****
   3  * Version: MPL 1.1/GPL 2.0/LGPL 2.1
   4  *
   5  * The contents of this file are subject to the Mozilla Public License Version
   6  * 1.1 (the "License"); you may not use this file except in compliance with
   7  * the License. You may obtain a copy of the License at
   8  * http://www.mozilla.org/MPL/
   9  *
  10  * Software distributed under the License is distributed on an "AS IS" basis,
  11  * WITHOUT WARRANTY OF ANY KIND, either express or implied. See the License
  12  * for the specific language governing rights and limitations under the
  13  * License.
  14  *
  15  * The Original Code is Mozilla Communicator client code, released
  16  * March 31, 1998.
  17  *
  18  * The Initial Developer of the Original Code is
  19  * Netscape Communications Corporation.
  20  * Portions created by the Initial Developer are Copyright (C) 1998-1999
  21  * the Initial Developer. All Rights Reserved.
  22  *
  23  * Contributor(s):
  24  *   Daniel Veditz <dveditz@netscape.com>
  25  *   Samir Gehani <sgehani@netscape.com>
  26  *   Mitch Stoltz <mstoltz@netscape.com>
  27  *
  28  * Alternatively, the contents of this file may be used under the terms of
  29  * either the GNU General Public License Version 2 or later (the "GPL"), or
  30  * the GNU Lesser General Public License Version 2.1 or later (the "LGPL"),
  31  * in which case the provisions of the GPL or the LGPL are applicable instead
  32  * of those above. If you wish to allow use of your version of this file only
  33  * under the terms of either the GPL or the LGPL, and not to allow others to
  34  * use your version of this file under the terms of the MPL, indicate your
  35  * decision by deleting the provisions above and replace them with the notice
  36  * and other provisions required by the GPL or the LGPL. If you do not delete
  37  * the provisions above, a recipient may use your version of this file under
  38  * the terms of any one of the MPL, the GPL or the LGPL.
  39  *
  40  * ***** END LICENSE BLOCK ***** */
  41
  42 #ifndef nsZipArchive_h_
  43 #define nsZipArchive_h_
  44
  45 #define ZIP_MAGIC     0x5A49505FL   /* "ZIP_" */
  46 #define ZIPFIND_MAGIC 0x5A495046L   /* "ZIPF" */
  47 #define ZIP_TABSIZE   256
  48 // Keep this odd. The -1 is significant.
  49 #define ZIP_BUFLEN    (4 * 1024 - 1)
  50
  51 #ifdef STANDALONE
  52 #define nsZipArchive nsZipArchiveStandalone
  53
  54 #define ZIP_Seek(fd,p,m) (fseek((fd),(p),(m))==0)
  55
  56 #else
  57
  58 #define PL_ARENA_CONST_ALIGN_MASK 7
  59 #include "plarena.h"
  60 #define ZIP_Seek(fd,p,m) (PR_Seek((fd),((PROffset32)p),(m))==((PROffset32)p))
  61
  62 #endif
  63
  64 #include "zlib.h"
  65
  66 class nsZipFind;
  67 class nsZipReadState;
  68 class nsZipItemMetadata;
  69
  70 #ifndef STANDALONE
  71 struct PRFileDesc;
  72 #endif
  73
  74 /**
  75  * This file defines some of the basic structures used by libjar to
  76  * read Zip files. It makes use of zlib in order to do the decompression.
  77  *
  78  * A few notes on the classes/structs:
  79  * nsZipArchive   represents a single Zip file, and maintains an index
  80  *                of all the items in the file.
  81  * nsZipItem      represents a single item (file) in the Zip archive.
  82  * nsZipFind      represents the metadata involved in doing a search,
  83  *                and current state of the iteration of found objects.
  84  *
  85  * There is a lot of #ifdef STANDALONE here, and that is so that these
  86  * basic structures can be reused in a standalone static
  87  * library. In order for the code to be reused, these structures
  88  * should never use anything from XPCOM, including such obvious things
  89  * as NS_ASSERTION(). Instead, use the basic NSPR equivalents.
  90  *
  91  * There is one key difference in the way that this code behaves in
  92  * STANDALONE mode. The nsZipArchive owns a single file descriptor and
  93  * that is used to read the ZIP file index, and for 'Test' and 'Extract'.
  94  * Since there is only one nsZipArchive per file, you can only Test/Extract
  95  * one file at a time from the Zip file.
  96  * 'MT''safe' reading from the zipfile is performed through JARInputStream,
  97  * which maintains its own file descriptor, allowing for multiple reads
  98  * concurrently from the same zip file.
  99  */
 100
 101 /**
 102  * nsZipItem -- a helper struct for nsZipArchive
 103  *
 104  * each nsZipItem represents one file in the archive and all the
 105  * information needed to manipulate it.
 106  */
 107 struct nsZipItem
 108 {
 109   nsZipItem*  next;
 110
 111   PRUint32    headerOffset;
 112   PRUint32    dataOffset;
 113   PRUint32    size;             /* size in original file */
 114   PRUint32    realsize;         /* inflated size */
 115   PRUint32    crc32;
 116
 117   /*
 118    * Keep small items together, to avoid overhead.
 119    */
 120   PRUint16     time;
 121   PRUint16     date;
 122   PRUint16     mode;
 123   PRUint8      compression;
 124   PRPackedBool hasDataOffset : 1;
 125   PRPackedBool isDirectory : 1;
 126   PRPackedBool isSynthetic : 1;  /* whether item is an actual zip entry or was
 127                                     generated as part of a real entry's path,
 128                                     e.g. foo/ in a zip containing only foo/a.txt
 129                                     and no foo/ entry is synthetic */
 130 #if defined(XP_UNIX) || defined(XP_BEOS)
 131   PRPackedBool isSymlink : 1;
 132 #endif
 133
 134   char        name[1]; // actually, bigger than 1
 135 };
 136
 137 /**
 138  * nsZipArchive -- a class for reading the PKZIP file format.
 139  *
 140  */
 141 class nsZipArchive
 142 {
 143   friend class nsZipFind;
 144
 145 public:
 146 #ifdef STANDALONE
 147   /** cookie used to validate supposed objects passed from C code */
 148   const PRInt32 kMagic;
 149 #endif
 150
 151   /** constructing does not open the archive. See OpenArchive() */
 152   nsZipArchive();
 153
 154   /** destructing the object closes the archive */
 155   ~nsZipArchive();
 156
 157   /**
 158    * OpenArchive
 159    *
 160    * It's an error to call this more than once on the same nsZipArchive
 161    * object. If we were allowed to use exceptions this would have been
 162    * part of the constructor
 163    *
 164    * @param   fd            File descriptor of file to open
 165    * @return  status code
 166    */
 167   nsresult OpenArchive(PRFileDesc* fd);
 168
 169   /**
 170    * Test the integrity of items in this archive by running
 171    * a CRC check after extracting each item into a memory
 172    * buffer.  If an entry name is supplied only the
 173    * specified item is tested.  Else, if null is supplied
 174    * then all the items in the archive are tested.
 175    *
 176    * @return  status code
 177    */
 178   nsresult Test(const char *aEntryName);
 179
 180   /**
 181    * Closes an open archive.
 182    */
 183   nsresult CloseArchive();
 184
 185   /**
 186    * GetItem
 187    * @param   aEntryName Name of file in the archive
 188    * @return  pointer to nsZipItem
 189    */
 190   nsZipItem* GetItem(const char * aEntryName);
 191
 192   /**
 193    * ExtractFile
 194    *
 195    * @param   zipEntry   Name of file in archive to extract
 196    * @param   outFD      Filedescriptor to write contents to
 197    * @param   outname    Name of file to write to
 198    * @return  status code
 199    */
 200   nsresult ExtractFile(nsZipItem * zipEntry, const char *outname, PRFileDesc * outFD);
 201
 202   /**
 203    * FindInit
 204    *
 205    * Initializes a search for files in the archive. FindNext() returns
 206    * the actual matches. The nsZipFind must be deleted when you're done
 207    *
 208    * @param   aPattern    a string or RegExp pattern to search for
 209    *                      (may be NULL to find all files in archive)
 210    * @param   aFind       a pointer to a pointer to a structure used
 211    *                      in FindNext.  In the case of an error this
 212    *                      will be set to NULL.
 213    * @return  status code
 214    */
 215   PRInt32 FindInit(const char * aPattern, nsZipFind** aFind);
 216
 217   /**
 218    * Moves the filepointer aFd to the start of data of the aItem.
 219    * @param   aItem       Pointer to nsZipItem
 220    * @param   aFd         The filepointer to move
 221    */
 222   nsresult  SeekToItem(nsZipItem* aItem, PRFileDesc* aFd);
 223
 224 private:
 225   //--- private members ---
 226
 227   nsZipItem*    mFiles[ZIP_TABSIZE];
 228 #ifndef STANDALONE
 229   PLArenaPool   mArena;
 230 #endif
 231
 232   // Used for central directory reading, and for Test and Extract
 233   PRFileDesc    *mFd;
 234
 235   // Whether we synthesized the directory entries
 236   PRPackedBool  mBuiltSynthetics;
 237
 238   //--- private methods ---
 239
 240   nsZipArchive& operator=(const nsZipArchive& rhs); // prevent assignments
 241   nsZipArchive(const nsZipArchive& rhs);            // prevent copies
 242
 243   nsZipItem*        CreateZipItem(PRUint16 namelen);
 244   nsresult          BuildFileList();
 245   nsresult          BuildSynthetics();
 246
 247   nsresult  CopyItemToDisk(PRUint32 size, PRUint32 crc, PRFileDesc* outFD);
 248   nsresult  InflateItem(const nsZipItem* aItem, PRFileDesc* outFD);
 249 };
 250
 251
 252 /**
 253  * nsZipFind
 254  *
 255  * a helper class for nsZipArchive, representing a search
 256  */
 257 class nsZipFind
 258 {
 259 public:
 260 #ifdef STANDALONE
 261   const PRInt32       kMagic;
 262 #endif
 263
 264   nsZipFind(nsZipArchive* aZip, char* aPattern, PRBool regExp);
 265   ~nsZipFind();
 266
 267   nsresult      FindNext(const char ** aResult);
 268
 269 private:
 270   nsZipArchive* mArchive;
 271   char*         mPattern;
 272   nsZipItem*    mItem;
 273   PRUint16      mSlot;
 274   PRPackedBool  mRegExp;
 275
 276   //-- prevent copies and assignments
 277   nsZipFind& operator=(const nsZipFind& rhs);
 278   nsZipFind(const nsZipFind& rhs);
 279 };
 280
 281 nsresult gZlibInit(z_stream *zs);
 282
 283 #endif /* nsZipArchive_h_ */