gecko/modules/libjar/nsZipArchive.cpp

/* -*- Mode: C++; tab-width: 4; indent-tabs-mode: nil; c-basic-offset: 2 -*- */
/* ***** BEGIN LICENSE BLOCK *****
 * Version: MPL 1.1/GPL 2.0/LGPL 2.1
 *
 * The contents of this file are subject to the Mozilla Public License Version
 * 1.1 (the "License"); you may not use this file except in compliance with
 * the License. You may obtain a copy of the License at
 * http://www.mozilla.org/MPL/
 *
 * Software distributed under the License is distributed on an "AS IS" basis,
 * WITHOUT WARRANTY OF ANY KIND, either express or implied. See the License
 * for the specific language governing rights and limitations under the
 * License.
 *
 * The Original Code is Mozilla Communicator client code, released
 * March 31, 1998.
 *
 * The Initial Developer of the Original Code is
 * Netscape Communications Corporation.
 * Portions created by the Initial Developer are Copyright (C) 1998
 * the Initial Developer. All Rights Reserved.
 *
 * Contributor(s):
 *   Daniel Veditz <dveditz@netscape.com>
 *   Samir Gehani <sgehani@netscape.com>
 *   Mitch Stoltz <mstoltz@netscape.com>
 *   Jeroen Dobbelaere <jeroen.dobbelaere@acunia.com>
 *   Jeff Walden <jwalden+code@mit.edu>
 *   Taras Glek <tglek@mozilla.com>
 *
 * Alternatively, the contents of this file may be used under the terms of
 * either the GNU General Public License Version 2 or later (the "GPL"), or
 * the GNU Lesser General Public License Version 2.1 or later (the "LGPL"),
 * in which case the provisions of the GPL or the LGPL are applicable instead
 * of those above. If you wish to allow use of your version of this file only
 * under the terms of either the GPL or the LGPL, and not to allow others to
 * use your version of this file under the terms of the MPL, indicate your
 * decision by deleting the provisions above and replace them with the notice
 * and other provisions required by the GPL or the LGPL. If you do not delete
 * the provisions above, a recipient may use your version of this file under
 * the terms of any one of the MPL, the GPL or the LGPL.
 *
 * ***** END LICENSE BLOCK ***** */

/*
 * This module implements a simple archive extractor for the PKZIP format.
 *
 * The underlying nsZipArchive is NOT thread-safe. Do not pass references
 * or pointers to it across thread boundaries.
 */


#define ZFILE_CREATE    PR_WRONLY | PR_CREATE_FILE
#define READTYPE  PRInt32
#include "zlib.h"
#include "nsISupportsUtils.h"
#include "nsRecyclingAllocator.h"
#include "prio.h"
#include "plstr.h"
#include "prlog.h"
#include "stdlib.h"
#include "nsWildCard.h"
#include "zipfile.h"
#include "zipstruct.h"
#include "nsZipArchive.h"

/**
 * Globals
 *
 * Global allocator used with zlib. Destroyed in module shutdown.
 */
#define NBUCKETS 6
#define BY4ALLOC_ITEMS 320
nsRecyclingAllocator *gZlibAllocator = NULL;

// For placement new used for arena allocations of zip file list
#include NEW_H
#define ZIP_ARENABLOCKSIZE (1*1024)

#ifdef XP_UNIX
    #include <sys/types.h>
    #include <sys/stat.h>
    #include <limits.h>
    #include <unistd.h>
#elif defined(XP_WIN) || defined(XP_OS2)
    #include <io.h>
#elif defined(XP_BEOS)
    #include <unistd.h>
#endif

#ifdef __SYMBIAN32__
    #include <sys/syslimits.h>
#endif /*__SYMBIAN32__*/


#ifndef XP_UNIX /* we need some constants defined in limits.h and unistd.h */
#  ifndef S_IFMT
#    define S_IFMT 0170000
#  endif
#  ifndef S_IFLNK
#    define S_IFLNK  0120000
#  endif
#  ifndef PATH_MAX
#    define PATH_MAX 1024
#  endif
#endif  /* XP_UNIX */

#include "zipfile.h"
#include "zipstruct.h"
#include "nsZipArchive.h"

static PRUint16 xtoint(unsigned char *ii);
static PRUint32 xtolong(unsigned char *ll);
static PRUint16 ExtractMode(unsigned char *ll);
static PRUint32 HashName(const char* aName);
#if defined(XP_UNIX) || defined(XP_BEOS)
static PRBool IsSymlink(unsigned char *ll);
static nsresult ResolveSymlink(const char *path);
#endif

//***********************************************************
// Allocators for use with zlib
//
// These are allocators that are performance tuned for
// use with zlib. Our use of zlib for every file we read from
// the jar file when running navigator, we do these allocation.
// alloc 24
// alloc 64
// alloc 11520
// alloc 32768
// alloc 1216 [304x4] max
// alloc 76   [19x4]
// free  76   [19x4]
// alloc 1152 [288x4]
// free  1152 [288x4]
// free  1216 [304x4]
// alloc 28
// free  28
// free  32768
// free  11520
// free  64
// free  24
//
// The pool will allocate these as:
//
//          32,768
//          11,520
//           1,280 [320x4] - shared by first x4 alloc, 28
//           1,280 [320x4] - shared by second and third x4 alloc
//              64
//              24
//          ------
//          46,936
//
// And almost all of the file reads happen serially. Hence this
// allocator tries to keep one set of memory needed for one file around
// and reused the same blocks for other file reads.
//
// The interesting question is when should be free this ?
// - memory pressure should be one.
// - after startup of navigator
// - after startup of mail
// In general, this allocator should be enabled before
// we startup and disabled after we startup if memory is a concern.
//***********************************************************

static void *
zlibAlloc(void *opaque, uInt items, uInt size)
{
  nsRecyclingAllocator *zallocator = (nsRecyclingAllocator *)opaque;
  if (zallocator) {
    // Bump up x4 allocations
    PRUint32 realitems = items;
    if (size == 4 && items < BY4ALLOC_ITEMS)
      realitems = BY4ALLOC_ITEMS;
     return zallocator->Calloc(realitems, size);
  }
  else
    return calloc(items, size);
}

static void
zlibFree(void *opaque, void *ptr)
{
  nsRecyclingAllocator *zallocator = (nsRecyclingAllocator *)opaque;
  if (zallocator)
    zallocator->Free(ptr);
  else
    free(ptr);
  return;
}

nsresult gZlibInit(z_stream *zs)
{
  memset(zs, 0, sizeof(z_stream));
  //-- ensure we have our zlib allocator for better performance
  if (!gZlibAllocator) {
    gZlibAllocator = new nsRecyclingAllocator(NBUCKETS, NS_DEFAULT_RECYCLE_TIMEOUT, "libjar");
  }
  if (gZlibAllocator) {
    zs->zalloc = zlibAlloc;
    zs->zfree = zlibFree;
    zs->opaque = gZlibAllocator;
  }
  int zerr = inflateInit2(zs, -MAX_WBITS);
  if (zerr != Z_OK) return ZIP_ERR_MEMORY;

  return ZIP_OK;
}

nsZipHandle::nsZipHandle()
  : mFd(nsnull)
  , mFileData(nsnull)
  , mLen(0)
  , mMap(nsnull)
  , mRefCnt(0)
{
  MOZ_COUNT_CTOR(nsZipHandle);
}

NS_IMPL_THREADSAFE_ADDREF(nsZipHandle)
NS_IMPL_THREADSAFE_RELEASE(nsZipHandle)

nsresult nsZipHandle::Init(PRFileDesc *fd, nsZipHandle **ret)
{
  PRInt64 size = PR_Available64(fd);
  if (size >= PR_INT32_MAX)
    return NS_ERROR_FILE_TOO_BIG;

  PRFileMap *map = PR_CreateFileMap(fd, size, PR_PROT_READONLY);

  if (!map)
    return NS_ERROR_FAILURE;

  nsZipHandle *handle = new nsZipHandle();
  if (!handle)
    return NS_ERROR_OUT_OF_MEMORY;

  handle->mFd = fd;
  handle->mMap = map;
  handle->mLen = (PRUint32) size;
  handle->mFileData = (PRUint8*) PR_MemMap(map, 0, handle->mLen);
  handle->AddRef();
  *ret = handle;
  return NS_OK;
}

PRInt32 nsZipHandle::Read(PRUint32 aPosition, void *aBuffer, PRUint32 aCount)
{
  if (mLen < 0 || aPosition+aCount > (PRUint32) mLen)
    return -1;
  PRInt32 count = PR_MIN(aCount, mLen - aPosition);
  memcpy(aBuffer, mFileData + aPosition, count);
  return count;
}

nsZipHandle::~nsZipHandle()
{
  if (mFd) {
    PR_MemUnmap(mFileData, mLen);
    PR_CloseFileMap(mMap);
    PR_Close(mFd);
    mFd = 0;
  }
  MOZ_COUNT_DTOR(nsZipHandle);
}

//***********************************************************
//      nsZipArchive  --  public methods
//***********************************************************


//---------------------------------------------
//  nsZipArchive::OpenArchive
//---------------------------------------------
nsresult nsZipArchive::OpenArchive(PRFileDesc * fd)
{
  nsresult rv = nsZipHandle::Init(fd, getter_AddRefs(mFd));
  if (NS_FAILED(rv))
    return rv;

  // Initialize our arena
  PL_INIT_ARENA_POOL(&mArena, "ZipArena", ZIP_ARENABLOCKSIZE);

  //-- get table of contents for archive
  return BuildFileList();
}

//---------------------------------------------
//  nsZipArchive::Test
//---------------------------------------------
nsresult nsZipArchive::Test(const char *aEntryName)
{
  nsZipItem* currItem;

  if (aEntryName) // only test specified item
  {
    currItem = GetItem(aEntryName);
    if (!currItem)
      return ZIP_ERR_FNF;
    //-- don't test (synthetic) directory items
    if (currItem->isDirectory)
      return ZIP_OK;
    return ExtractFile(currItem, 0, 0);
  }

  // test all items in archive
  for (int i = 0; i < ZIP_TABSIZE; i++) {
    for (currItem = mFiles[i]; currItem; currItem = currItem->next) {
      //-- don't test (synthetic) directory items
      if (currItem->isDirectory)
        continue;
      nsresult rv = ExtractFile(currItem, 0, 0);
      if (rv != ZIP_OK)
        return rv;
    }
  }

  return ZIP_OK;
}

//---------------------------------------------
//  nsZipArchive::CloseArchive
//---------------------------------------------
nsresult nsZipArchive::CloseArchive()
{
  if (mFd) {
    PL_FinishArenaPool(&mArena);
  }

  // CAUTION:
  // We don't need to delete each of the nsZipItem as the memory for
  // the zip item and the filename it holds are both allocated from the Arena.
  // Hence, destroying the Arena is like destroying all the memory
  // for all the nsZipItem in one shot. But if the ~nsZipItem is doing
  // anything more than cleaning up memory, we should start calling it.
  // Let us also cleanup the mFiles table for re-use on the next 'open' call
  for (int i = 0; i < ZIP_TABSIZE; i++) {
    mFiles[i] = 0;
  }
  mFd = NULL;
  mBuiltSynthetics = PR_FALSE;
  return ZIP_OK;
}

//---------------------------------------------
// nsZipArchive::GetItem
//---------------------------------------------
nsZipItem*  nsZipArchive::GetItem(const char * aEntryName)
{
  if (aEntryName) {
    //-- If the request is for a directory, make sure that synthetic entries 
    //-- are created for the directories without their own entry.
    if (!mBuiltSynthetics) {
        PRUint32 len = strlen(aEntryName);
        if ((len > 0) && (aEntryName[len-1] == '/')) {
            if (BuildSynthetics() != ZIP_OK)
                return 0;
        }
    }

    nsZipItem* item = mFiles[ HashName(aEntryName) ];
    while (item) {
      if (!strcmp(aEntryName, item->name))
        return item; //-- found it
      item = item->next;
    }
  }
  return 0;
}

//---------------------------------------------
// nsZipArchive::ExtractFile
// This extracts the item to the filehandle provided.
// If 'aFd' is null, it only tests the extraction.
// On extraction error(s) it removes the file.
// When needed, it also resolves the symlink.
//---------------------------------------------
nsresult nsZipArchive::ExtractFile(nsZipItem *item, const char *outname,
                                   PRFileDesc* aFd)
{
  if (!item)
    return ZIP_ERR_PARAM;
  if (!mFd)
    return ZIP_ERR_GENERAL;

  // Directory extraction is handled in nsJAR::Extract,
  // so the item to be extracted should never be a directory
  PR_ASSERT(!item->isDirectory);

  //-- move to the start of file's data
  if (!MaybeReadItem(item))
    return ZIP_ERR_CORRUPT;

  nsSeekableZipHandle fd;
  if (!fd.Open(mFd.get(), item->dataOffset, item->size))
    return ZIP_ERR_CORRUPT;

  nsresult rv;

  //-- extract the file using the appropriate method
  switch(item->compression)
  {
    case STORED:
      rv = CopyItemToDisk(item->size, item->crc32, fd, aFd);
      break;

    case DEFLATED:
      rv = InflateItem(item, fd, aFd);
      break;

    default:
      //-- unsupported compression type
      rv = ZIP_ERR_UNSUPPORTED;
  }

  //-- delete the file on errors, or resolve symlink if needed
  if (aFd) {
    PR_Close(aFd);
    if (rv != ZIP_OK)
      PR_Delete(outname);
#if defined(XP_UNIX) || defined(XP_BEOS)
    else if (item->isSymlink)
      rv = ResolveSymlink(outname);
#endif
  }

  return rv;
}

//---------------------------------------------
// nsZipArchive::FindInit
//---------------------------------------------
PRInt32
nsZipArchive::FindInit(const char * aPattern, nsZipFind **aFind)
{
  if (!aFind)
    return ZIP_ERR_PARAM;

  // null out param in case an error happens
  *aFind = NULL;

  PRBool  regExp = PR_FALSE;
  char*   pattern = 0;

  // Create synthetic directory entries on demand
  nsresult rv = BuildSynthetics();
  if (rv != ZIP_OK)
    return rv;

  // validate the pattern
  if (aPattern)
  {
    switch (NS_WildCardValid((char*)aPattern))
    {
      case INVALID_SXP:
        return ZIP_ERR_PARAM;

      case NON_SXP:
        regExp = PR_FALSE;
        break;

      case VALID_SXP:
        regExp = PR_TRUE;
        break;

      default:
        // undocumented return value from RegExpValid!
        PR_ASSERT(PR_FALSE);
        return ZIP_ERR_PARAM;
    }

    pattern = PL_strdup(aPattern);
    if (!pattern)
      return ZIP_ERR_MEMORY;
  }

  *aFind = new nsZipFind(this, pattern, regExp);
  if (!*aFind) {
    PL_strfree(pattern);
    return ZIP_ERR_MEMORY;
  }

  return ZIP_OK;
}



//---------------------------------------------
// nsZipFind::FindNext
//---------------------------------------------
nsresult nsZipFind::FindNext(const char ** aResult)
{
  if (!mArchive || !aResult)
    return ZIP_ERR_PARAM;

  *aResult = 0;

  // we start from last match, look for next
  while (mSlot < ZIP_TABSIZE)
  {
    // move to next in current chain, or move to new slot
    mItem = mItem ? mItem->next : mArchive->mFiles[mSlot];

    PRBool found = PR_FALSE;
    if (!mItem)
      ++mSlot;                          // no more in this chain, move to next slot
    else if (!mPattern)
      found = PR_TRUE;            // always match
    else if (mRegExp)
      found = (NS_WildCardMatch(mItem->name, mPattern, PR_FALSE) == MATCH);
    else
      found = (PL_strcmp(mItem->name, mPattern) == 0);

    if (found) {
      *aResult = mItem->name;
      return ZIP_OK;
    }
  }

  return ZIP_ERR_FNF;
}

#if defined(XP_UNIX) || defined(XP_BEOS)
//---------------------------------------------
// ResolveSymlink
//---------------------------------------------
static nsresult ResolveSymlink(const char *path)
{
  PRFileDesc * fIn = PR_Open(path, PR_RDONLY, 0000);
  if (!fIn)
    return ZIP_ERR_DISK;

  char buf[PATH_MAX+1];
  PRInt32 length = PR_Read(fIn, (void*)buf, PATH_MAX);
  PR_Close(fIn);

  if ( (length <= 0)
    || ((buf[length] = 0, PR_Delete(path)) != 0)
    || (symlink(buf, path) != 0))
  {
     return ZIP_ERR_DISK;
  }
  return ZIP_OK;
}
#endif

//***********************************************************
//      nsZipArchive  --  private implementation
//***********************************************************

#define BR_BUF_SIZE 1024 /* backward read buffer size */

//---------------------------------------------
//  nsZipArchive::CreateZipItem
//---------------------------------------------
nsZipItem* nsZipArchive::CreateZipItem(PRUint16 namelen)
{
  // sizeof(nsZipItem) includes space for name's null byte
  // Arena allocate the nsZipItem
  void *mem;
  PL_ARENA_ALLOCATE(mem, &mArena, sizeof(nsZipItem)+namelen);
  return (nsZipItem*)mem;
}

//---------------------------------------------
//  nsZipArchive::BuildFileList
//---------------------------------------------
nsresult nsZipArchive::BuildFileList()
{
  PRUint8   *buf;
  //-----------------------------------------------------------------------
  // locate the central directory via the End record
  //-----------------------------------------------------------------------

  //-- get archive size using end pos
  PRInt32  pos = (PRInt32) mFd->mLen;

  PRInt32 central = -1;
  while (central == -1)
  {
    //-- read backwards in 1K-sized chunks (unless file is less than 1K)
    PRInt32  bufsize = pos > BR_BUF_SIZE ? BR_BUF_SIZE : pos;
    pos -= bufsize;

    buf = mFd->mFileData + pos;

    //-- scan for ENDSIG
    PRUint8 *endp = buf + bufsize;
    for (endp -= ZIPEND_SIZE; endp >= buf; endp--)
    {
      if (xtolong(endp) == ENDSIG)
      { 
        //-- Seek to start of central directory
        central = xtolong(((ZipEnd *) endp)->offset_central_dir);
        break;
      }
    }

    if (central != -1)
      break;

    if (pos <= 0)
      //-- We're at the beginning of the file, and still no sign
      //-- of the end signature.  File must be corrupted!
      return ZIP_ERR_CORRUPT;

    //-- backward read must overlap ZipEnd length
    pos += ZIPEND_SIZE;

  } /* while looking for end signature */


  //-------------------------------------------------------
  // read the central directory headers
  //-------------------------------------------------------
  PRInt32 byteCount = mFd->mLen - central;
  buf = mFd->mFileData + central;
  pos = 0;
  PRUint32 sig = xtolong(buf);
  while (sig == CENTRALSIG) {
    //-- make sure we've read enough
    if (byteCount - pos < ZIPCENTRAL_SIZE)
      return ZIP_ERR_CORRUPT;

    //-------------------------------------------------------
    // read the fixed-size data
    //-------------------------------------------------------
    ZipCentral* central = (ZipCentral*)(buf+pos);

    PRUint16 namelen = xtoint(central->filename_len);
    PRUint16 extralen = xtoint(central->extrafield_len);
    PRUint16 commentlen = xtoint(central->commentfield_len);

    //-- sanity check variable sizes and refuse to deal with
    //-- anything too big: it's likely a corrupt archive
    if (namelen > BR_BUF_SIZE || extralen > BR_BUF_SIZE || commentlen > 2*BR_BUF_SIZE)
      return ZIP_ERR_CORRUPT;

    nsZipItem* item = CreateZipItem(namelen);
    if (!item)
      return ZIP_ERR_MEMORY;

    item->headerOffset  = xtolong(central->localhdr_offset);
    item->dataOffset    = 0;
    item->size          = xtolong(central->size);
    item->realsize      = xtolong(central->orglen);
    item->crc32         = xtolong(central->crc32);
    item->time          = xtoint(central->time);
    item->date          = xtoint(central->date);
    item->isSynthetic   = PR_FALSE;
    item->hasDataOffset = PR_FALSE;

    PRUint16 compression = xtoint(central->method);
    item->compression   = (compression < UNSUPPORTED) ? (PRUint8)compression
                                                      : UNSUPPORTED;

    item->mode = ExtractMode(central->external_attributes);
#if defined(XP_UNIX) || defined(XP_BEOS)
    // Check if item is a symlink
    item->isSymlink = IsSymlink(central->external_attributes);
#endif

    pos += ZIPCENTRAL_SIZE;
    //-------------------------------------------------------
    // get the item name
    //-------------------------------------------------------
    memcpy(item->name, buf+pos, namelen);
    item->name[namelen] = 0;
    //-- an item whose name ends with '/' is a directory
    item->isDirectory = ('/' == item->name[namelen - 1]);

    //-- add item to file table
    //-- note that an explicit entry for a directory will override
    //-- a fake entry created for that directory (as in the case
    //-- of processing foo/bar.txt and then foo/) -- this will
    //-- preserve an explicit directory's metadata at the cost of
    //-- an extra nsZipItem (and that only happens if we process a
    //-- file inside that directory before processing the directory
    //-- entry itself)
    PRUint32 hash = HashName(item->name);
    item->next = mFiles[hash];
    mFiles[hash] = item;

    //-------------------------------------------------------
    // set up to process the next item at the top of loop
    //-------------------------------------------------------
    pos += namelen + extralen + commentlen;
    sig = xtolong(buf+pos);
  } /* while reading central directory records */

  if (sig != ENDSIG)
    return ZIP_ERR_CORRUPT;

  return ZIP_OK;
}

//---------------------------------------------
//  nsZipArchive::BuildSynthetics
//---------------------------------------------
nsresult nsZipArchive::BuildSynthetics()
{
  if (mBuiltSynthetics)
    return ZIP_OK;
  mBuiltSynthetics = PR_TRUE;

  // Create synthetic entries for any missing directories.
  // Do this when all ziptable has scanned to prevent double entries.
  for (int i = 0; i < ZIP_TABSIZE; ++i)
  {
    for (nsZipItem* item = mFiles[i]; item != 0; item = item->next)
    {
      if (item->isSynthetic)
        continue;
    
      //-- add entries for directories in the current item's path
      //-- go from end to beginning, because then we can stop trying
      //-- to create diritems if we find that the diritem we want to
      //-- create already exists
      //-- start just before the last char so as to not add the item
      //-- twice if it's a directory
      PRUint16 namelen = strlen(item->name);
      for (char* p = item->name + namelen - 2; p >= item->name; p--)
      {
        if ('/' != *p)
          continue;

        // See whether we need to create any more implicit directories,
        // because if we don't we can avoid a lot of work.
        // We can even avoid (de)allocating space for a bogus dirname with
        // a little trickery -- save the char at item->name[dirnamelen],
        // set it to 0, compare the strings, and restore the saved
        // char when done
        const PRUint32 dirnamelen = p + 1 - item->name;
        const char savedChar = item->name[dirnamelen];
        item->name[dirnamelen] = 0;

        // Is the directory in the file table?
        PRUint32 hash = HashName(item->name);
        PRBool found = PR_FALSE;
        for (nsZipItem* zi = mFiles[hash]; zi != NULL; zi = zi->next)
        {
          if (0 == strcmp(item->name, zi->name))
          {
            // we've already added this dir and all its parents
            found = PR_TRUE;
            break;
          }
        }

        // restore the char immediately
        item->name[dirnamelen] = savedChar;

        // if the directory was found, break out of the directory
        // creation loop now that we know all implicit directories
        // are there -- otherwise, start creating the zip item
        if (found)
          break;

        nsZipItem* diritem = CreateZipItem(dirnamelen);
        if (!diritem)
          return ZIP_ERR_MEMORY;

        memcpy(diritem->name, item->name, dirnamelen);
        diritem->name[dirnamelen] = 0;

        diritem->isDirectory = PR_TRUE;
        diritem->isSynthetic = PR_TRUE;
        diritem->compression = STORED;
        diritem->size = diritem->realsize = 0;
        diritem->crc32 = 0;
        diritem->mode = 0755;

        // Set an obviously wrong last-modified date/time, because
        // finding something more accurate like the most recent
        // last-modified date/time of the dir's contents is a lot
        // of effort.  The date/time corresponds to 1980-01-01 00:00.
        diritem->time = 0;
        diritem->date = 1 + (1 << 5) + (0 << 9);

        // add diritem to the file table
        diritem->next = mFiles[hash];
        mFiles[hash] = diritem;
      } /* end processing of dirs in item's name */
    }
  }
  return ZIP_OK;
}

nsZipHandle* nsZipArchive::GetFD(nsZipItem* aItem)
{
  if (!mFd || !MaybeReadItem(aItem))
    return NULL;
  return mFd.get();
}

//---------------------------------------------
// nsZipArchive::MaybeReadItem
//---------------------------------------------
bool nsZipArchive::MaybeReadItem(nsZipItem* aItem)
{
  PR_ASSERT (aItem);

  //-- the first time an item is used we need to calculate its offset
  if (!aItem->hasDataOffset)
  {
    //-- read local header to get variable length values and calculate
    //-- the real data offset
    //--
    //-- NOTE: extralen is different in central header and local header
    //--       for archives created using the Unix "zip" utility. To set
    //--       the offset accurately we need the _local_ extralen.
    if (!mFd || !mFd->mLen > aItem->headerOffset + ZIPLOCAL_SIZE)
      return false;

    ZipLocal   *Local =     (ZipLocal*)(mFd->mFileData + aItem->headerOffset);
    //check limits here
    if ((xtolong(Local->signature) != LOCALSIG))
    {
      //-- read error or local header not found
      return false;
    }

    aItem->dataOffset = aItem->headerOffset +
                        ZIPLOCAL_SIZE +
                        xtoint(Local->filename_len) +
                        xtoint(Local->extrafield_len);
    aItem->hasDataOffset = PR_TRUE;
  }

  return true;
}

//---------------------------------------------
// nsZipArchive::CopyItemToDisk
//---------------------------------------------
nsresult
nsZipArchive::CopyItemToDisk(PRUint32 itemSize, PRUint32 itemCrc, 
                             nsSeekableZipHandle &fd, PRFileDesc* outFD)
/*
 * This function copies an archive item to disk, to the
 * file specified by outFD. If outFD is zero, the extracted data is
 * not written, only checked for CRC, so this is in effect same as 'Test'.
 */
{
  PRUint32    chunk, pos, crc;
  char buf[ZIP_BUFLEN];

  //-- initialize crc
  crc = crc32(0L, Z_NULL, 0);

  //-- copy chunks until file is done
  for (pos = 0; pos < itemSize; pos += chunk)
  {
    chunk = (itemSize - pos < ZIP_BUFLEN) ? (itemSize - pos) : ZIP_BUFLEN;
    
    if (fd.Read(buf, chunk) != (READTYPE)chunk)
    {
      //-- unexpected end of data in archive
      return ZIP_ERR_CORRUPT;
    }

    //-- incrementally update crc32
    crc = crc32(crc, (const unsigned char*)buf, chunk);

    if (outFD && PR_Write(outFD, buf, chunk) < (READTYPE)chunk)
    {
      //-- Couldn't write all the data (disk full?)
      return ZIP_ERR_DISK;
    }
  }

  //-- verify crc32
  if (crc != itemCrc)
      return ZIP_ERR_CORRUPT;

  return ZIP_OK;
}


//---------------------------------------------
// nsZipArchive::InflateItem
//---------------------------------------------
nsresult nsZipArchive::InflateItem(const nsZipItem* aItem, nsSeekableZipHandle &fd, PRFileDesc* outFD)
/*
 * This function inflates an archive item to disk, to the
 * file specified by outFD. If outFD is zero, the extracted data is
 * not written, only checked for CRC, so this is in effect same as 'Test'.
 */
{
  PR_ASSERT(aItem);

  //-- allocate deflation buffers
  Bytef inbuf[ZIP_BUFLEN];
  Bytef outbuf[ZIP_BUFLEN];

  //-- set up the inflate
  z_stream    zs;
  nsresult status = gZlibInit(&zs);
  if (status != ZIP_OK)
    return ZIP_ERR_GENERAL;

  //-- inflate loop
  zs.next_out = outbuf;
  zs.avail_out = ZIP_BUFLEN;

  PRUint32  size = aItem->size;
  PRUint32  outpos = 0;
  PRUint32  crc = crc32(0L, Z_NULL, 0);
  int       zerr = Z_OK;
  while (zerr == Z_OK)
  {
    PRBool      bRead = PR_FALSE;
    PRBool      bWrote= PR_FALSE;

    if (zs.avail_in == 0 && zs.total_in < size)
    {
      //-- no data to inflate yet still more in file:
      //-- read another chunk of compressed data
      PRUint32 chunk = (size-zs.total_in < ZIP_BUFLEN) ? size-zs.total_in : ZIP_BUFLEN;

      if (fd.Read(inbuf, chunk) != (READTYPE)chunk)
      {
        //-- unexpected end of data
        status = ZIP_ERR_CORRUPT;
        break;
      }

      zs.next_in  = inbuf;
      zs.avail_in = chunk;
      bRead       = PR_TRUE;
    }

    if (zs.avail_out == 0)
    {
      //-- write inflated buffer to disk and make space
      if (outFD && PR_Write(outFD, outbuf, ZIP_BUFLEN) < ZIP_BUFLEN)
      {
        //-- Couldn't write all the data (disk full?)
        status = ZIP_ERR_DISK;
        break;
      }

      outpos = zs.total_out;
      zs.next_out  = outbuf;
      zs.avail_out = ZIP_BUFLEN;
      bWrote       = PR_TRUE;
    }

    if(bRead || bWrote)
    {
      Bytef* old_next_out = zs.next_out;

      zerr = inflate(&zs, Z_PARTIAL_FLUSH);

      //-- incrementally update crc32
      crc = crc32(crc, (const unsigned char*)old_next_out, zs.next_out - old_next_out);
    }
    else
      zerr = Z_STREAM_END;

  } // while

  //-- verify crc32
  if ((status == ZIP_OK) && (crc != aItem->crc32))
  {
      status = ZIP_ERR_CORRUPT;
      goto cleanup;
  }

  //-- write last inflated bit to disk
  if (zerr == Z_STREAM_END && outpos < zs.total_out)
  {
    PRUint32 chunk = zs.total_out - outpos;
    if (outFD && PR_Write(outFD, outbuf, chunk) < (READTYPE)chunk)
      status = ZIP_ERR_DISK;
  }

  //-- convert zlib error to return value
  if (status == ZIP_OK && zerr != Z_OK && zerr != Z_STREAM_END)
  {
    status = (zerr == Z_MEM_ERROR) ? ZIP_ERR_MEMORY : ZIP_ERR_CORRUPT;
  }

  //-- if found no errors make sure we've converted the whole thing
  PR_ASSERT(status != ZIP_OK || zs.total_in == aItem->size);
  PR_ASSERT(status != ZIP_OK || zs.total_out == aItem->realsize);

cleanup:
  //-- free zlib internal state
  inflateEnd(&zs);

  return status;
}

//------------------------------------------
// nsZipArchive constructor and destructor
//------------------------------------------

nsZipArchive::nsZipArchive() :
  mBuiltSynthetics(PR_FALSE)
{
  MOZ_COUNT_CTOR(nsZipArchive);

  // initialize the table to NULL
  memset(mFiles, 0, sizeof mFiles);
}

nsZipArchive::~nsZipArchive()
{
  CloseArchive();

  MOZ_COUNT_DTOR(nsZipArchive);
}


//------------------------------------------
// nsZipFind constructor and destructor
//------------------------------------------

nsZipFind::nsZipFind(nsZipArchive* aZip, char* aPattern, PRBool aRegExp) : 
  mArchive(aZip),
  mPattern(aPattern),
  mItem(0),
  mSlot(0),
  mRegExp(aRegExp)
{
  MOZ_COUNT_CTOR(nsZipFind);
}

nsZipFind::~nsZipFind()
{
  PL_strfree(mPattern);

  MOZ_COUNT_DTOR(nsZipFind);
}

//------------------------------------------
// helper functions
//------------------------------------------

/* 
 * HashName 
 *
 * returns a hash key for the entry name 
 */
static PRUint32 HashName(const char* aName)
{
  PR_ASSERT(aName != 0);

  PRUint32 val = 0;
  for (PRUint8* c = (PRUint8*)aName; *c != 0; c++) {
    val = val*37 + *c;
  }

  return (val % ZIP_TABSIZE);
}

/*
 *  x t o i n t
 *
 *  Converts a two byte ugly endianed integer
 *  to our platform's integer.
 */
static PRUint16 xtoint (unsigned char *ii)
{
  return (PRUint16) ((ii [0]) | (ii [1] << 8));
}

/*
 *  x t o l o n g
 *
 *  Converts a four byte ugly endianed integer
 *  to our platform's integer.
 */
static PRUint32 xtolong (unsigned char *ll)
{
  return (PRUint32)( (ll [0] <<  0) |
                     (ll [1] <<  8) |
                     (ll [2] << 16) |
                     (ll [3] << 24) );
}

/*
 * ExtractMode
 *
 * Extracts bits 17-24 from a 32-bit unsigned long
 * representation of the external attributes field.
 * Subsequently it tacks on the implicit user-read
 * bit.
 */
static PRUint16 ExtractMode(unsigned char *ll)
{
    return ((PRUint16)(ll[2])) | 0x0100;
}

#if defined(XP_UNIX) || defined(XP_BEOS)
/*
 *
 *  Return true if the attributes are for a symbolic link
 *
 */

static PRBool IsSymlink(unsigned char *ll)
{
  return ((xtoint(ll+2) & S_IFMT) == S_IFLNK);
}
#endif