windows-xp/Source/XPSP1/NT/base/ntsetup/win95upg/common/migutil/hash.c

/*++

Copyright (c) 1998 Microsoft Corporation

Module Name:

    hash.c

Abstract:

    Replacement routines for the string table functions in setupapi.dll.
    This routines are much more easy to work with.

Author:

    Jim Schmidt (jimschm)   22-Dec-1998

Revision History:

    jimschm     23-Nov-1999  Removed setup api compatibility, fixed enum
                             to be insertion order
    ovidiut     14-Oct-1999  New coding conventions + Win64 compliance.
    marcw        2-Sep-1999  Moved over from Win9xUpg project.

--*/

#include "pch.h"

//
// Includes
//

// None

#define DBG_HASH    "HashTable"

//
// Strings
//

#define S_HASHTABLE "HashTable"

//
// Constants
//

#define BUCKETS                 1009
#define HASHTABLE_SIGNATURE     0x122398ff

//
// Macros
//

#ifdef DEBUG
    #define ASSERT_TABLE_IS_VALID(table)        MYASSERT(pTableIsValid(table))
#else
    #define ASSERT_TABLE_IS_VALID(table)
#endif

//
// Types
//

typedef struct _tagBUCKETITEM {
    struct _tagBUCKETITEM *Next;
    struct _tagBUCKETITEM *NextLink, *PrevLink;
    INT Locked;
    WORD StringSize;
    // string follows StringSize
    // optional data follows string
} BUCKETITEM, *PBUCKETITEM;

typedef struct {
    struct _tagBUCKETITEM *Next;
    struct _tagBUCKETITEM *NextLink, *PrevLink;
    INT Locked;
    PVOID String;
    // optional data follows struct
} BUCKETITEM_EXTERN_STR, *PBUCKETITEM_EXTERN_STR;

typedef struct {
    DWORD Signature;
    BOOL Unicode;
    BOOL ExternalStrings;
    BOOL CaseSensitive;
    POOLHANDLE Pool;
    PBUCKETITEM *Bucket;
    PBUCKETITEM FirstLink;
    PBUCKETITEM LastLink;
    PBUCKETITEM DelayedDelete;
    UINT ExtraDataSize;
    UINT MinimumStringBytes;
    UINT MaximumStringBytes;
    UINT Buckets;
} HASHTABLESTRUCT, *PHASHTABLESTRUCT;


//
// Globals
//

// None

//
// Macro expansion list
//

// None

//
// Private function prototypes
//

// None

//
// Macro expansion definition
//

// None

//
// Code
//



#ifdef DEBUG

BOOL
pTableIsValid (
    IN      HASHTABLE Table
    )
{
    BOOL b = TRUE;

    if (!Table) {
        return FALSE;
    }

    __try {
        if (((PHASHTABLESTRUCT) Table)->Signature != HASHTABLE_SIGNATURE) {
            b = FALSE;
        }
    }
    __except (TRUE) {
        b = FALSE;
    }

    return b;
}

#endif


/*++

Routine Description:

  pComputeHashValue adds all the character values of the string, shifting to
  maintain order.

Arguments:

  String - Specifies the string to compute the hash value for

Return Value:

  The hash value, within the range of 0 to BUCKETS - 1.

--*/

UINT
pComputeHashValueA (
    IN      PCSTR String,
    IN      UINT Buckets
    )
{
    UINT hashValue = 0;

    while (*String) {
        hashValue = _rotl (hashValue, 2);
        hashValue += (UINT) *String;
        String++;
    }

    hashValue %= Buckets;

    return hashValue;
}


UINT
pComputeHashValueW (
    IN      PCWSTR String,
    IN      UINT Buckets
    )
{
    UINT hashValue = 0;

    while (*String) {
        hashValue = _rotl (hashValue, 2);
        hashValue += (UINT) *String;
        String++;
    }

    hashValue %= Buckets;

    return hashValue;
}


UINT
pComputePrefixHashValueA (
    IN      PCSTR String,
    IN      UINT Size,
    IN      UINT Buckets
    )
{
    UINT hashValue = 0;
    PCSTR end;

    end = (PCSTR) ((PBYTE) String + Size);

    while (String < end) {
        hashValue = _rotl (hashValue, 2);
        hashValue += (UINT) *String;
        String++;
    }

    hashValue %= Buckets;

    return hashValue;
}


UINT
pComputePrefixHashValueW (
    IN      PCWSTR String,
    IN      UINT Size,
    IN      UINT Buckets
    )
{
    UINT hashValue = 0;
    PCWSTR end;

    end = (PCWSTR) ((PBYTE) String + Size);

    while (String < end) {
        hashValue = _rotl (hashValue, 2);
        hashValue += (UINT) *String;
        String++;
    }

    hashValue %= Buckets;

    return hashValue;
}


HASHTABLE
HtAllocExAW (
    IN      BOOL CaseSensitive,
    IN      BOOL Unicode,
    IN      BOOL ExternalStrings,
    IN      UINT ExtraDataSize,
    IN      UINT Buckets
    )

/*++

Routine Description:

  AllocateHashTableEx creates a hash table.  If ExtraDataSize is non-zero,
  each hash table entry gets an allocation of ExtraDataSize added to it.

Arguments:

  CaseSensitive - Specifies TRUE if the hash table is case-sensitive, FALSE
                  if all strings should be stored and compared in lower-case
                  only

  Unicode - Specifies TRUE to allocate a UNICODE hash table, or FALSE to
            allocate an ANSI table.  None of the routines in this file do any
            sort of UNICODE/ANSI converstion.

  ExternalStrings - Specifies TRUE if the strings belong to memory maintained
                    by the caller

  ExtraDataSize - Specifies the size of binary data associated with the
                  table item, or 0 for none.

Return Value:

  A handle to the string table.

--*/

{
    PHASHTABLESTRUCT hashTable;
    POOLHANDLE pool;

    if (!Buckets) {
        Buckets = BUCKETS;
    }

    pool = PoolMemInitNamedPool (S_HASHTABLE);
    MYASSERT (pool);

    PoolMemDisableTracking (pool);

    hashTable = (PHASHTABLESTRUCT) PoolMemGetAlignedMemory (
                                        pool,
                                        sizeof (HASHTABLESTRUCT) + (sizeof (PBUCKETITEM) * Buckets)
                                        );
    MYASSERT (hashTable);

    hashTable->Signature = HASHTABLE_SIGNATURE;
    hashTable->CaseSensitive = CaseSensitive;
    hashTable->Unicode = Unicode;
    hashTable->ExternalStrings = ExternalStrings;
    hashTable->Pool = pool;
    hashTable->Bucket = (PBUCKETITEM *) ((PBYTE) hashTable + sizeof (HASHTABLESTRUCT));
    hashTable->FirstLink = NULL;
    hashTable->LastLink = NULL;
    hashTable->ExtraDataSize = ExtraDataSize;
    hashTable->MinimumStringBytes = (UINT) -1;
    hashTable->MaximumStringBytes = 0;
    hashTable->Buckets = Buckets;

    //
    // Zero out all of the bucket structures.
    //
    ZeroMemory (hashTable->Bucket, sizeof (PBUCKETITEM) * Buckets);

    return (HASHTABLE) hashTable;
}


VOID
HtFree (
    IN      HASHTABLE HashTable
    )

/*++

Routine Description:

  HtFree releases all resources associated with a string table.

Arguments:

  None.

Return Value:

  None.

--*/

{
    PHASHTABLESTRUCT table = (PHASHTABLESTRUCT) HashTable;

    if (table) {
        ASSERT_TABLE_IS_VALID (HashTable);
        PoolMemEmptyPool (table->Pool);
        PoolMemDestroyPool (table->Pool);
    }
}


PBUCKETITEM
pHtFindStringA (
    IN      HASHTABLE HashTable,
    IN      PCSTR String,
    OUT     PVOID ExtraDataBuffer,              OPTIONAL
    IN      BOOL AlreadyLowercase,
    OUT     PUINT OutHashValue,
    OUT     PBUCKETITEM *PrevBucketItem
    )

/*++

Routine Description:

  pHtFindString implements the hash table lookup routine.  It returns the
  pointer to the bucket item or NULL if the item was not found.

Arguments:

  HashTable        - Specifies the handle to the hash table
  String           - Specifies the string to find.  If this string is
                     case-insensitive but has already been lowercased, then make
                     sure to pass TRUE in the AlreadyLowercase argument.
  ExtraDataBuffer  - A buffer that receives the bytes stored as extra data with
                     the found item; caller must size this according to the
                     extra data size specified to HtAllocExAW
  AlreadyLowercase - Specifies TRUE if String is already lower case
  OutHashValue     - Receives the hash value.  This is non optional for
                     efficiency.
  PrevBucketItem    - Receives the previous bucket item

Return Value:

  The pointer to the bucket item or NULL if no item was found.

--*/

{
    PHASHTABLESTRUCT table = (PHASHTABLESTRUCT) HashTable;
    PSTR dupStr = NULL;
    UINT hashValue;
    PBUCKETITEM item;
    PCSTR p1, p2;
    PCVOID storedDataPtr;

    ASSERT_TABLE_IS_VALID (HashTable);

    *PrevBucketItem = NULL;

    if (!AlreadyLowercase && !table->CaseSensitive) {
        dupStr = DuplicateTextA (String);
        if (!dupStr) {
            MYASSERT (FALSE);
            return NULL;
        }

        (void) CharLowerA (dupStr);
        String = dupStr;
    }

    hashValue = pComputeHashValueA (String, table->Buckets);

    item = table->Bucket[hashValue];

    while (item) {

        if (table->ExternalStrings) {
            p1 = (PCSTR) ((PBUCKETITEM_EXTERN_STR) item)->String;
        } else {
            p1 = (PCSTR) ((PBYTE) item + sizeof (BUCKETITEM));
        }

        p2 = String;

        while (*p1) {
            if (*p1 != *p2) {
                break;
            }

            p1++;
            p2++;
        }

        if (*p1 == 0 && *p2 == 0) {
            break;
        }

        *PrevBucketItem = item;
        item = item->Next;
    }

    if (item && ExtraDataBuffer) {
        (void) HtGetExtraData (HashTable, (HASHITEM)item, &storedDataPtr);

        CopyMemory (
            (PBYTE) ExtraDataBuffer,
            (PBYTE) storedDataPtr,
            table->ExtraDataSize
            );
    }

    FreeTextA (dupStr);

    *OutHashValue = hashValue;

    return item;
}


PBUCKETITEM
pHtFindStringW (
    IN      HASHTABLE HashTable,
    IN      PCWSTR String,
    OUT     PVOID ExtraDataBuffer,              OPTIONAL
    IN      BOOL AlreadyLowercase,
    OUT     PUINT OutHashValue,
    OUT     PBUCKETITEM *PrevBucketItem
    )
{
    PHASHTABLESTRUCT table = (PHASHTABLESTRUCT) HashTable;
    PWSTR dupStr = NULL;
    UINT hashValue;
    PBUCKETITEM item;
    PCWSTR p1, p2;
    PCVOID storedDataPtr;

    ASSERT_TABLE_IS_VALID (HashTable);

    *PrevBucketItem = NULL;

    if (!AlreadyLowercase && !table->CaseSensitive) {
        dupStr = DuplicateTextW (String);
        if (!dupStr) {
            MYASSERT (FALSE);
            return NULL;
        }

        (void) _wcslwr (dupStr);
        String = dupStr;
    }

    hashValue = pComputeHashValueW (String, table->Buckets);

    item = table->Bucket[hashValue];
    while (item) {

        if (table->ExternalStrings) {
            p1 = (PCWSTR) ((PBUCKETITEM_EXTERN_STR) item)->String;
        } else {
            p1 = (PCWSTR) ((PBYTE) item + sizeof (BUCKETITEM));
        }

        p2 = String;

        while (*p1) {
            if (*p1 != *p2) {
                break;
            }

            p1++;
            p2++;
        }

        if (*p1 == 0 && *p2 == 0) {
            break;
        }

        *PrevBucketItem = item;
        item = item->Next;
    }

    if (item && ExtraDataBuffer) {
        (void) HtGetExtraData (HashTable, (HASHITEM)item, &storedDataPtr);

        CopyMemory (
            (PBYTE) ExtraDataBuffer,
            (PBYTE) storedDataPtr,
            table->ExtraDataSize
            );
    }

    FreeTextW (dupStr);

    *OutHashValue = hashValue;

    return item;
}


PBUCKETITEM
pHtFindPrefixA (
    IN      HASHTABLE HashTable,
    IN      PCSTR String,
    IN      PCSTR BufferEnd,
    OUT     PVOID ExtraDataBuffer,              OPTIONAL
    IN      BOOL AlreadyLowercase,
    OUT     PUINT OutHashValue
    )

/*++

Routine Description:

  pHtFindPrefix implements a hash table lookup routine that tests each hash
  table entry, character-by-character, until a match is found, or until the
  hash table maximum is reached.  It returns the pointer to the bucket item or
  NULL if the item was not found.

Arguments:

  HashTable        - Specifies the handle to the hash table
  String           - Specifies the string to find.  If this string is
                     case-insensitive but has already been lowercased, then make
                     sure to pass TRUE in the AlreadyLowercase argument.
  BufferEnd        - Specifies the end of the string buffer, which may be longer
                     than all entries in the hash table, or it may be shorter.
  ExtraDataBuffer  - A buffer that receives the bytes stored as extra data with
                     the found item; caller must size this according to the
                     extra data size specified to HtAllocExAW
  AlreadyLowercase - Specifies TRUE if String is in lower-case, FALSE otherwise.
  OutHashValue     - Receives the hash value.  This is non optional for
                     efficiency.  If pHtFindPrefix does not find a match,
                     this value will be set to zero.

Return Value:

  The pointer to the bucket item or NULL if no item was found.

--*/

{
    PHASHTABLESTRUCT table = (PHASHTABLESTRUCT) HashTable;
    PSTR dupStr = NULL;
    UINT hashValue = 0;
    PBUCKETITEM item = NULL;
    PCSTR p1, p2;
    PCSTR p1End;
    PCVOID storedDataPtr;
    UINT maxBytes;
    UINT currentBytes;
    PCSTR shortestEnd;

    ASSERT_TABLE_IS_VALID (HashTable);

    maxBytes = (UINT)((PBYTE) BufferEnd - (PBYTE) String);
    maxBytes = min (maxBytes, table->MaximumStringBytes);

    if (!maxBytes || table->MinimumStringBytes == (UINT) -1) {
        return NULL;
    }

    if (!AlreadyLowercase && !table->CaseSensitive) {
        dupStr = AllocTextA (maxBytes / sizeof (CHAR));
        if (!dupStr) {
            MYASSERT (FALSE);
            return NULL;
        }

        StringCopyByteCountA (dupStr, String, maxBytes + sizeof (CHAR));
        CharLowerA (dupStr);
        String = dupStr;
    }

    BufferEnd = (PCSTR) ((PBYTE) String + maxBytes);
    shortestEnd = (PCSTR) ((PBYTE) String + table->MinimumStringBytes);
    if (shortestEnd == String) {
        shortestEnd = _mbsinc (shortestEnd);
    }

    while (BufferEnd >= shortestEnd) {

        currentBytes = (UINT)((PBYTE) BufferEnd - (PBYTE) String);

        hashValue = pComputePrefixHashValueA (String, currentBytes, table->Buckets);

        item = table->Bucket[hashValue];

        while (item) {

            if ((item->StringSize - sizeof (CHAR)) == currentBytes) {

                if (table->ExternalStrings) {
                    p1 = (PCSTR) ((PBUCKETITEM_EXTERN_STR) item)->String;
                } else {
                    p1 = (PCSTR) ((PBYTE) item + sizeof (BUCKETITEM));
                }

                p1End = (PCSTR) ((PBYTE) p1 + currentBytes);
                p2 = String;

                while (p1 < p1End) {
                    if (*p1 != *p2) {
                        break;
                    }

                    p1++;
                    p2++;
                }

                if (p1 == p1End) {
                    break;
                }
            }

            item = item->Next;
        }

        if (item) {
            break;
        }

        BufferEnd = _mbsdec2 (String, BufferEnd);
    }

    if (item && ExtraDataBuffer) {
        (void) HtGetExtraData (HashTable, (HASHITEM)item, &storedDataPtr);

        CopyMemory (
            (PBYTE) ExtraDataBuffer,
            (PBYTE) storedDataPtr,
            table->ExtraDataSize
            );
    }

    FreeTextA (dupStr);

    *OutHashValue = hashValue;

    return item;
}


PBUCKETITEM
pHtFindPrefixW (
    IN      HASHTABLE HashTable,
    IN      PCWSTR String,
    IN      PCWSTR BufferEnd,
    OUT     PVOID ExtraDataBuffer,              OPTIONAL
    IN      BOOL AlreadyLowercase,
    OUT     PUINT OutHashValue
    )
{
    PHASHTABLESTRUCT table = (PHASHTABLESTRUCT) HashTable;
    PWSTR dupStr = NULL;
    UINT hashValue = 0;
    PBUCKETITEM item = NULL;
    PCWSTR p1, p2;
    PCWSTR p1End;
    PCVOID storedDataPtr;
    UINT maxBytes;
    PCWSTR shortestEnd;
    UINT currentBytes;

    ASSERT_TABLE_IS_VALID (HashTable);

    maxBytes = (UINT)((PBYTE) BufferEnd - (PBYTE) String);
    maxBytes = min (maxBytes, table->MaximumStringBytes);

    if (!maxBytes || table->MinimumStringBytes == (UINT) -1) {
        return NULL;
    }

    if (!AlreadyLowercase && !table->CaseSensitive) {
        dupStr = AllocTextW (maxBytes / sizeof (WCHAR));
        if (!dupStr) {
            MYASSERT (FALSE);
            return NULL;
        }

        StringCopyByteCountW (dupStr, String, maxBytes + sizeof (WCHAR));
        _wcslwr (dupStr);
        String = dupStr;
    }

    BufferEnd = (PCWSTR) ((PBYTE) String + maxBytes);
    shortestEnd = (PCWSTR) ((PBYTE) String + table->MinimumStringBytes);
    if (shortestEnd == String) {
        shortestEnd++;
    }

    while (BufferEnd >= shortestEnd) {

        currentBytes = (UINT)((PBYTE) BufferEnd - (PBYTE) String);

        hashValue = pComputePrefixHashValueW (String, currentBytes, table->Buckets);

        item = table->Bucket[hashValue];

        while (item) {

            if ((item->StringSize - sizeof (WCHAR)) == currentBytes) {

                if (table->ExternalStrings) {
                    p1 = (PCWSTR) ((PBUCKETITEM_EXTERN_STR) item)->String;
                } else {
                    p1 = (PCWSTR) ((PBYTE) item + sizeof (BUCKETITEM));
                }

                p1End = (PCWSTR) ((PBYTE) p1 + currentBytes);
                p2 = String;

                while (p1 < p1End) {
                    if (*p1 != *p2) {
                        break;
                    }

                    p1++;
                    p2++;
                }

                if (p1 == p1End) {
                    break;
                }
            }

            item = item->Next;
        }

        if (item) {
            break;
        }

        BufferEnd--;
    }

    if (item && ExtraDataBuffer) {
        (void) HtGetExtraData (HashTable, (HASHITEM)item, &storedDataPtr);

        CopyMemory (
            (PBYTE) ExtraDataBuffer,
            (PBYTE) storedDataPtr,
            table->ExtraDataSize
            );
    }

    FreeTextW (dupStr);

    *OutHashValue = hashValue;

    return item;
}


HASHITEM
HtAddStringExA (
    IN      HASHTABLE HashTable,
    IN      PCSTR String,
    IN      PCVOID ExtraData,           OPTIONAL
    IN      BOOL AlreadyLowercase
    )

/*++

Routine Description:

  HtAddStringEx adds a string to the hash table, and copies ExtraData to the
  new hash table entry.  If String is already in the hash table, the ExtraData
  is updated.

Arguments:

  HashTable        - Specifies the handle to the hash table, as returned from
                     AllocateHashTable.
  String           - Specifies the string to add to the table
  ExtraData        - Specifies the source binary data to be copied to the hash
                     table entry
  AlreadyLowercase - Specifies TRUE String is in all lowercase

Return Value:

  Returns the pointer to the bucket item allocated or update.

--*/

{
    PHASHTABLESTRUCT table = (PHASHTABLESTRUCT) HashTable;
    PBUCKETITEM item;
    PBUCKETITEM_EXTERN_STR externItem;
    PBUCKETITEM existingItem;
    PSTR dupStr = NULL;
    HASHITEM rc = NULL;
    UINT size;
    UINT hashValue;
    UINT strSize;
    PCVOID storedDataPtr;
    PBUCKETITEM dontCare;

    ASSERT_TABLE_IS_VALID (HashTable);

    if (table->Unicode) {
        DEBUGMSG ((DBG_WHOOPS, "Cannot add ANSI string to UNICODE table"));
        return 0;
    }

    if (!AlreadyLowercase && !table->CaseSensitive) {
        dupStr = DuplicateTextA (String);
        if (!dupStr) {
            MYASSERT (FALSE);
            return 0;
        }

        CharLowerA (dupStr);
        String = dupStr;
    }

    existingItem = pHtFindStringA (HashTable, String, NULL, TRUE, &hashValue, &dontCare);

    if (existingItem) {

        rc = (HASHITEM) existingItem;

    } else {

        //
        // item does not exist, add it now
        //

        strSize = SizeOfStringA (String);

        if (table->ExternalStrings) {

            size = sizeof (BUCKETITEM_EXTERN_STR) + table->ExtraDataSize;

            externItem = (PBUCKETITEM_EXTERN_STR) PoolMemGetAlignedMemory (table->Pool, size);
            MYASSERT (externItem);

            externItem->Locked = 0;

            externItem->Next = table->Bucket[hashValue];
            table->Bucket[hashValue] = (PBUCKETITEM) externItem;

            if (table->LastLink) {
                table->LastLink->NextLink = (PBUCKETITEM) externItem;
            }
            externItem->PrevLink = table->LastLink;
            table->LastLink = (PBUCKETITEM) externItem;
            externItem->NextLink = NULL;

            if (!table->FirstLink) {
                table->FirstLink = (PBUCKETITEM) externItem;
            }

            rc = (HASHITEM) externItem;

        } else {

            size = sizeof (BUCKETITEM) + strSize + table->ExtraDataSize;

            item = (PBUCKETITEM) PoolMemGetAlignedMemory (table->Pool, size);
            MYASSERT (item);

            item->Locked = 0;

            item->Next = table->Bucket[hashValue];
            table->Bucket[hashValue] = item;

            item->StringSize = (WORD) strSize;
            CopyMemory ((PBYTE) item + sizeof (BUCKETITEM), String, strSize);

            if (table->LastLink) {
                table->LastLink->NextLink = item;
            }
            item->PrevLink = table->LastLink;
            table->LastLink = item;
            item->NextLink = NULL;

            if (!table->FirstLink) {
                table->FirstLink = item;
            }

            rc = (HASHITEM) item;
        }

        strSize -= sizeof (CHAR);
        table->MaximumStringBytes = max (table->MaximumStringBytes, strSize);
        table->MinimumStringBytes = min (table->MinimumStringBytes, strSize);
    }

    MYASSERT (rc);
    (void) HtGetExtraData (HashTable, rc, &storedDataPtr);

    if (ExtraData) {

        CopyMemory (
            (PBYTE) storedDataPtr,
            (PBYTE) ExtraData,
            table->ExtraDataSize
            );

    } else if (!existingItem) {

        ZeroMemory (
            (PBYTE) storedDataPtr,
            table->ExtraDataSize
            );
    }

    FreeTextA (dupStr);

    return rc;
}


HASHITEM
HtAddStringExW (
    IN      HASHTABLE HashTable,
    IN      PCWSTR String,
    IN      PCVOID ExtraData,           OPTIONAL
    IN      BOOL AlreadyLowercase
    )
{
    PHASHTABLESTRUCT table = (PHASHTABLESTRUCT) HashTable;
    PBUCKETITEM item;
    PBUCKETITEM_EXTERN_STR externItem;
    PBUCKETITEM existingItem;
    PWSTR dupStr = NULL;
    HASHITEM rc = NULL;
    UINT size;
    UINT hashValue;
    UINT strSize;
    PCVOID storedDataPtr;
    PBUCKETITEM dontCare;

    ASSERT_TABLE_IS_VALID (HashTable);

    if (!table->Unicode) {
        DEBUGMSG ((DBG_WHOOPS, "Cannot add UNICODE string to ANSI table"));
        return 0;
    }

    if (!AlreadyLowercase && !table->CaseSensitive) {
        dupStr = DuplicateTextW (String);
        if (!dupStr) {
            MYASSERT (FALSE);
            return 0;
        }

        _wcslwr (dupStr);
        String = dupStr;
    }

    existingItem = pHtFindStringW (HashTable, String, NULL, TRUE, &hashValue, &dontCare);

    if (existingItem) {

        rc = (HASHITEM) existingItem;

    } else {

        //
        // item does not exist, add it now
        //

        strSize = SizeOfStringW (String);

        if (table->ExternalStrings) {

            size = sizeof (BUCKETITEM_EXTERN_STR) + table->ExtraDataSize;

            externItem = (PBUCKETITEM_EXTERN_STR) PoolMemGetAlignedMemory (table->Pool, size);
            MYASSERT (externItem);

            externItem->Locked = 0;

            externItem->Next = table->Bucket[hashValue];
            table->Bucket[hashValue] = (PBUCKETITEM) externItem;

            if (table->LastLink) {
                table->LastLink->NextLink = (PBUCKETITEM) externItem;
            }
            externItem->PrevLink = table->LastLink;
            table->LastLink = (PBUCKETITEM) externItem;
            externItem->NextLink = NULL;

            if (!table->FirstLink) {
                table->FirstLink = (PBUCKETITEM) externItem;
            }

            rc = (HASHITEM) externItem;

        } else {

            size = sizeof (BUCKETITEM) + strSize + table->ExtraDataSize;

            item = (PBUCKETITEM) PoolMemGetAlignedMemory (table->Pool, size);
            MYASSERT (item);

            item->Locked = 0;

            item->Next = table->Bucket[hashValue];
            table->Bucket[hashValue] = item;

            item->StringSize = (WORD) strSize;
            CopyMemory ((PBYTE) item + sizeof (BUCKETITEM), String, strSize);

            if (table->LastLink) {
                table->LastLink->NextLink = item;
            }
            item->PrevLink = table->LastLink;
            table->LastLink = item;
            item->NextLink = NULL;

            if (!table->FirstLink) {
                table->FirstLink = item;
            }

            rc = (HASHITEM) item;
        }

        strSize -= sizeof (WCHAR);
        table->MaximumStringBytes = max (table->MaximumStringBytes, strSize);
        table->MinimumStringBytes = min (table->MinimumStringBytes, strSize);
    }

    MYASSERT (rc);
    (void) HtGetExtraData (HashTable, rc, &storedDataPtr);

    if (ExtraData) {

        CopyMemory (
            (PBYTE) storedDataPtr,
            (PBYTE) ExtraData,
            table->ExtraDataSize
            );

    } else if (!existingItem) {

        ZeroMemory (
            (PBYTE) storedDataPtr,
            table->ExtraDataSize
            );
    }

    FreeTextW (dupStr);

    return rc;
}


VOID
pRemoveHashItem (
    IN      PHASHTABLESTRUCT Table,
    IN      UINT BucketNum,
    IN      PBUCKETITEM PrevItem,
    IN      PBUCKETITEM ItemToDelete
    )
{
    if (!PrevItem) {
        MYASSERT (Table->Bucket[BucketNum] == ItemToDelete);
        Table->Bucket[BucketNum] = ItemToDelete->Next;
    } else {
        PrevItem->Next = ItemToDelete->Next;
    }

    if (ItemToDelete->PrevLink) {
        ItemToDelete->PrevLink->NextLink = ItemToDelete->NextLink;
    } else {
        Table->FirstLink = ItemToDelete->Next;
    }

    if (ItemToDelete->NextLink) {
        ItemToDelete->NextLink->PrevLink = ItemToDelete->PrevLink;
    } else {
        Table->LastLink = ItemToDelete->PrevLink;
    }

    if (ItemToDelete->Locked) {
        ItemToDelete->Next = Table->DelayedDelete;
        Table->DelayedDelete = ItemToDelete;
    } else {
        PoolMemReleaseMemory (Table->Pool, ItemToDelete);
    }
}


BOOL
HtRemoveItem (
    IN      HASHTABLE HashTable,
    IN      HASHITEM Item
    )
{
    PHASHTABLESTRUCT table = (PHASHTABLESTRUCT) HashTable;
    UINT bucketNumber;
    PBUCKETITEM prevItem;
    PBUCKETITEM thisItem;
    PCSTR ansiStr;
    PCWSTR unicodeStr;

    if (!Item) {
        return FALSE;
    }

    //
    // Find prev bucket item
    //

    if (table->Unicode) {
        unicodeStr = HtGetStringFromItemW (Item);
        MYASSERT (unicodeStr);

        thisItem = pHtFindStringW (
                        HashTable,
                        unicodeStr,
                        NULL,
                        TRUE,
                        &bucketNumber,
                        &prevItem
                        );

    } else {
        ansiStr = HtGetStringFromItemA (Item);
        MYASSERT (ansiStr);

        thisItem = pHtFindStringA (
                        HashTable,
                        ansiStr,
                        NULL,
                        TRUE,
                        &bucketNumber,
                        &prevItem
                        );

    }

    MYASSERT (Item == thisItem);

    if (Item != thisItem) {
        return FALSE;
    }

    pRemoveHashItem (table, bucketNumber, prevItem, thisItem);

    return TRUE;
}


BOOL
HtRemoveStringA (
    IN      HASHTABLE HashTable,
    IN      PCSTR AnsiString
    )
{
    PHASHTABLESTRUCT table = (PHASHTABLESTRUCT) HashTable;
    UINT bucketNumber;
    PBUCKETITEM prevItem;
    PBUCKETITEM thisItem;

    if (table->Unicode) {
        DEBUGMSG ((DBG_WHOOPS, "Cannot delete UNICODE table with ANSI api"));
        return FALSE;
    }

    thisItem = pHtFindStringA (
                    HashTable,
                    AnsiString,
                    NULL,
                    FALSE,
                    &bucketNumber,
                    &prevItem
                    );

    if (!thisItem) {
        return FALSE;
    }

    pRemoveHashItem (table, bucketNumber, prevItem, thisItem);

    return TRUE;
}



BOOL
HtRemoveStringW (
    IN      HASHTABLE HashTable,
    IN      PCWSTR UnicodeString
    )
{
    PHASHTABLESTRUCT table = (PHASHTABLESTRUCT) HashTable;
    UINT bucketNumber;
    PBUCKETITEM prevItem;
    PBUCKETITEM thisItem;

    if (!table->Unicode) {
        DEBUGMSG ((DBG_WHOOPS, "Cannot delete ANSI table with UNICODE api"));
        return FALSE;
    }

    thisItem = pHtFindStringW (
                    HashTable,
                    UnicodeString,
                    NULL,
                    FALSE,
                    &bucketNumber,
                    &prevItem
                    );

    if (!thisItem) {
        return FALSE;
    }

    pRemoveHashItem (table, bucketNumber, prevItem, thisItem);

    return TRUE;
}



/*++

Routine Description:

  HtFindStringEx is the external entry point for pHtFindString.

Arguments:

  HashTable        - Specifies the hash table handle, as returned by
                     AllocateHashTable.
  String           - Specifies the string to find
  ExtraDataBuffer  - Receives the extra data associated with the found item
  AlreadyLowercase - Specifies TRUE if the String is in lowercase

Return Value:

  A pointer to the bucket item or NULL if the string was not found.

--*/

HASHITEM
HtFindStringExA (
    IN      HASHTABLE HashTable,
    IN      PCSTR String,
    OUT     PVOID ExtraDataBuffer,              OPTIONAL
    IN      BOOL AlreadyLowercase
    )
{
    UINT dontCare;
    PBUCKETITEM dontCare2;

    return (HASHITEM) pHtFindStringA (
                            HashTable,
                            String,
                            ExtraDataBuffer,
                            AlreadyLowercase,
                            &dontCare,
                            &dontCare2
                            );
}


HASHITEM
HtFindStringExW (
    IN      HASHTABLE HashTable,
    IN      PCWSTR String,
    OUT     PVOID ExtraDataBuffer,              OPTIONAL
    IN      BOOL AlreadyLowercase
    )
{
    UINT dontCare;
    PBUCKETITEM dontCare2;

    return (HASHITEM) pHtFindStringW (
                            HashTable,
                            String,
                            ExtraDataBuffer,
                            AlreadyLowercase,
                            &dontCare,
                            &dontCare2
                            );
}


/*++

Routine Description:

  HtFindStringEx is the external entry point for pHtFindString.

Arguments:

  HashTable        - Specifies the hash table handle, as returned by
                     AllocateHashTable.
  String           - Specifies the string to find
  BufferEnd        - Specifies the end of the buffer for String
  ExtraDataBuffer  - Receives the extra data associated with the found item
  AlreadyLowercase - Specifies TRUE if String is in all lowercase

Return Value:

  A pointer to the bucket item or NULL if the string was not found.

--*/

HASHITEM
HtFindPrefixExA (
    IN      HASHTABLE HashTable,
    IN      PCSTR String,
    IN      PCSTR BufferEnd,
    OUT     PVOID ExtraDataBuffer,              OPTIONAL
    IN      BOOL AlreadyLowercase
    )
{
    UINT dontCare;

    return (HASHITEM) pHtFindPrefixA (
                            HashTable,
                            String,
                            BufferEnd,
                            ExtraDataBuffer,
                            AlreadyLowercase,
                            &dontCare
                            );
}


HASHITEM
HtFindPrefixExW (
    IN      HASHTABLE HashTable,
    IN      PCWSTR String,
    IN      PCWSTR BufferEnd,
    OUT     PVOID ExtraDataBuffer,              OPTIONAL
    IN      BOOL AlreadyLowercase
    )
{
    UINT dontCare;

    return (HASHITEM) pHtFindPrefixW (
                            HashTable,
                            String,
                            BufferEnd,
                            ExtraDataBuffer,
                            AlreadyLowercase,
                            &dontCare
                            );
}


BOOL
HtGetExtraData (
    IN      HASHTABLE HashTable,
    IN      HASHITEM Index,
    OUT     PCVOID *ExtraData
    )

/*++

Routine Description:

  HtGetExtraData gets the extra data associated with a bucket item.
  The caller must supply the ID as returned from HtFindStringEx or
  HtAddStringEx.  This routine is useful when ExtraData is large, and
  the normal find routine would be slow because of the CopyMemory code path.

Arguments:

  HashTable - Specifies the handle to the hash table
  Index     - Specifies the offset as returned from HtFindStringEx or
              HtAddStringEx
  ExtraData - Receives the extra data pointer (it does NOT copy the data to
              the buffer).

Return Value:

  TRUE if ExtraData was set, FALSE otherwise.

--*/

{
    PHASHTABLESTRUCT table = (PHASHTABLESTRUCT) HashTable;
    PBUCKETITEM item;
    PBUCKETITEM_EXTERN_STR externStrItem;

    ASSERT_TABLE_IS_VALID (HashTable);

    if (!Index) {
        return FALSE;
    }

    if (table->ExternalStrings) {

        externStrItem = (PBUCKETITEM_EXTERN_STR) Index;
        *ExtraData = (PCVOID) ((PBYTE) externStrItem + sizeof (PBUCKETITEM_EXTERN_STR));

    } else {

        item = (PBUCKETITEM) Index;
        *ExtraData = (PCVOID) ((PBYTE) item + sizeof (BUCKETITEM) + item->StringSize);

    }

    return TRUE;
}


BOOL
HtCopyStringData (
    IN      HASHTABLE HashTable,
    IN      HASHITEM Index,
    OUT     PVOID ExtraDataBuffer
    )

/*++

Routine Description:

  HtCopyStringData gets the extra data associated with a bucket item
  and copies it to the caller's buffer.

Arguments:

  HashTable       - Specifies the handle to the hash table
  Index           - Specifies the offset as returned from HtFindStringEx or
                    HtAddStringEx
  ExtraDataBuffer - Receives the extra data

Return Value:

  TRUE if ExtraDataBuffer was copied, FALSE otherwise.

--*/

{
    PHASHTABLESTRUCT table = (PHASHTABLESTRUCT) HashTable;
    PCVOID storedDataPtr;

    ASSERT_TABLE_IS_VALID (HashTable);

    if (!HtGetExtraData (HashTable, Index, &storedDataPtr)) {
        return FALSE;
    }

    CopyMemory (
        (PBYTE) ExtraDataBuffer,
        (PBYTE) storedDataPtr,
        table->ExtraDataSize
        );

    return TRUE;
}


BOOL
HtSetStringData (
    IN      HASHTABLE HashTable,
    IN      HASHITEM Index,
    IN      PCVOID ExtraData
    )

/*++

Routine Description:

  HtSetStringData copies new extra data to the specified hash table entry.

Arguments:

  HashTable - Specifies the handle to the hash table
  Index     - Specifies the offset as returned from HtFindStringEx or
              HtAddStringEx
  ExtraData - Specifies the extra data

Return Value:

  TRUE if the item was updated, FALSE otherwise.

--*/

{
    PHASHTABLESTRUCT table = (PHASHTABLESTRUCT) HashTable;
    PCVOID storedDataPtr;

    ASSERT_TABLE_IS_VALID (HashTable);

    if (!HtGetExtraData (HashTable, Index, &storedDataPtr)) {
        return FALSE;
    }

    CopyMemory (
        (PBYTE) storedDataPtr,
        (PBYTE) ExtraData,
        table->ExtraDataSize
        );

    return TRUE;
}


BOOL
EnumFirstHashTableStringA (
    OUT     PHASHTABLE_ENUMA EnumPtr,
    IN      HASHTABLE HashTable
    )

/*++

Routine Description:

  EnumFirstHashTableString begins an enumeration of the hash table structure.
  The return order is random.  Also, do not modify the hash table while an
  enumeration is active.

Arguments:

  EnumPtr   - Receives the string, extra data and offset for the first item
              in the hash table.
  HashTable - Specifies the handle of the hash table to enumerate.

Return Value:

  TRUE if an item was enumerated, FALSE otherwise.

--*/

{
    PHASHTABLESTRUCT table = (PHASHTABLESTRUCT) HashTable;

    ASSERT_TABLE_IS_VALID (HashTable);

    if (table->Unicode) {
        DEBUGMSG ((DBG_WHOOPS, "Cannot enum UNICODE table with ANSI wrapper"));
        return FALSE;
    }

    ZeroMemory (EnumPtr, sizeof (HASHTABLE_ENUMA));

    EnumPtr->Internal = (HASHTABLE) table;

    return EnumNextHashTableStringA (EnumPtr);
}


BOOL
EnumFirstHashTableStringW (
    OUT     PHASHTABLE_ENUMW EnumPtr,
    IN      HASHTABLE HashTable
    )
{
    PHASHTABLESTRUCT table = (PHASHTABLESTRUCT) HashTable;

    ASSERT_TABLE_IS_VALID (HashTable);

    if (!table->Unicode) {
        DEBUGMSG ((DBG_WHOOPS, "Cannot enum ANSI table with UNICODE wrapper"));
        return FALSE;
    }

    ZeroMemory (EnumPtr, sizeof (HASHTABLE_ENUMW));

    EnumPtr->Internal = (HASHTABLE) table;

    return EnumNextHashTableStringW (EnumPtr);
}


BOOL
EnumNextHashTableStringA (
    IN OUT  PHASHTABLE_ENUMA EnumPtr
    )

/*++

Routine Description:

  EnumNextHashTableString continues an enumeration started by
  EnumFirstHashTableString.  Call the routine until it returns FALSE.

Arguments:

  EnumPtr - Specifies the structure of an active enumeration.  Receives
            updated string, extra data and offset members.

Return Value:

  TRUE if another item was enumerated, FALSE if no items remain.

--*/

{
    PHASHTABLESTRUCT table = (PHASHTABLESTRUCT) EnumPtr->Internal;
    PBUCKETITEM item;
    PBUCKETITEM_EXTERN_STR externItem;

    if (!EnumPtr->Internal) {
        return FALSE;
    }

    ASSERT_TABLE_IS_VALID (EnumPtr->Internal);

    if (!EnumPtr->Index) {
        item = table->FirstLink;
    } else {
        item = (PBUCKETITEM) EnumPtr->Index;
        item->Locked -= 1;
        MYASSERT (item->Locked >= 0);
        item = item->NextLink;
    }

    if (item) {
        //
        // Return a valid item
        //

        item->Locked += 1;
        EnumPtr->Index = (HASHITEM) item;

        if (table->ExternalStrings) {
            externItem = (PBUCKETITEM_EXTERN_STR) item;
            EnumPtr->String = (PCSTR) (externItem->String);
        } else {
            EnumPtr->String = (PCSTR) ((PBYTE) item + sizeof (BUCKETITEM));
        }

        if (table->ExtraDataSize) {
            MYASSERT (EnumPtr->Index);
            (void) HtGetExtraData (EnumPtr->Internal, EnumPtr->Index, &EnumPtr->ExtraData);
        }

        return TRUE;
    }

    EnumPtr->Internal = 0;

    return FALSE;
}


BOOL
EnumNextHashTableStringW (
    IN OUT  PHASHTABLE_ENUMW EnumPtr
    )
{
    PHASHTABLESTRUCT table = (PHASHTABLESTRUCT) EnumPtr->Internal;
    PBUCKETITEM item;
    PBUCKETITEM_EXTERN_STR externItem;

    if (!EnumPtr->Internal) {
        return FALSE;
    }

    ASSERT_TABLE_IS_VALID (EnumPtr->Internal);

    if (!EnumPtr->Index) {
        item = table->FirstLink;
    } else {
        item = (PBUCKETITEM) EnumPtr->Index;
        item->Locked -= 1;
        MYASSERT (item->Locked >= 0);
        item = item->NextLink;
    }

    if (item) {
        //
        // Return a valid item
        //

        item->Locked += 1;
        EnumPtr->Index = (HASHITEM) item;

        if (table->ExternalStrings) {
            externItem = (PBUCKETITEM_EXTERN_STR) item;
            EnumPtr->String = (PCWSTR) (externItem->String);
        } else {
            EnumPtr->String = (PCWSTR) ((PBYTE) item + sizeof (BUCKETITEM));
        }

        if (table->ExtraDataSize) {
            MYASSERT (EnumPtr->Index);
            (void) HtGetExtraData (EnumPtr->Internal, EnumPtr->Index, &EnumPtr->ExtraData);
        }

        return TRUE;
    }

    EnumPtr->Internal = 0;

    return FALSE;
}


VOID
AbortHashTableEnumA (
    IN      PHASHTABLE_ENUMA EnumPtr
    )
{
    PHASHTABLESTRUCT table = (PHASHTABLESTRUCT) EnumPtr->Internal;
    PBUCKETITEM item;
    PBUCKETITEM nextItem;
    PBUCKETITEM prevItem;

    if (!EnumPtr->Internal) {
        return;
    }

    ASSERT_TABLE_IS_VALID (EnumPtr->Internal);

    if (EnumPtr->Index) {
        item = (PBUCKETITEM) EnumPtr->Index;
        item->Locked -= 1;
        MYASSERT (item->Locked >= 0);
    }

    //
    // Evaluate delayed delete items; remove those that are no longer locked
    //

    if (table->DelayedDelete) {
        item = table->DelayedDelete;
        prevItem = NULL;

        while (item) {

            nextItem = item->Next;

            if (!item->Locked) {
                PoolMemReleaseMemory (table->Pool, item);
            } else {
                if (prevItem) {
                    prevItem->Next = item;
                } else {
                    table->DelayedDelete = item;
                }

                prevItem = item;
            }

            item = nextItem;
        }

        if (prevItem) {
            prevItem->Next = NULL;
        } else {
            table->DelayedDelete = NULL;
        }
    }

    ZeroMemory (EnumPtr, sizeof (HASHTABLE_ENUMA));
}


VOID
AbortHashTableEnumW (
    IN      PHASHTABLE_ENUMW EnumPtr
    )
{
    PHASHTABLESTRUCT table = (PHASHTABLESTRUCT) EnumPtr->Internal;
    PBUCKETITEM item;
    PBUCKETITEM prevItem;
    PBUCKETITEM nextItem;

    if (!EnumPtr->Internal) {
        return;
    }

    ASSERT_TABLE_IS_VALID (EnumPtr->Internal);

    if (EnumPtr->Index) {
        item = (PBUCKETITEM) EnumPtr->Index;
        item->Locked -= 1;
        MYASSERT (item->Locked >= 0);
    }

    //
    // Evaluate delayed delete items; remove those that are no longer locked
    //

    if (table->DelayedDelete) {
        item = table->DelayedDelete;
        prevItem = NULL;

        while (item) {

            nextItem = item->Next;

            if (!item->Locked) {
                PoolMemReleaseMemory (table->Pool, item);
            } else {
                if (prevItem) {
                    prevItem->Next = item;
                } else {
                    table->DelayedDelete = item;
                }

                prevItem = item;
            }

            item = nextItem;
        }

        if (prevItem) {
            prevItem->Next = NULL;
        } else {
            table->DelayedDelete = NULL;
        }
    }

    ZeroMemory (EnumPtr, sizeof (HASHTABLE_ENUMW));
}


BOOL
EnumHashTableWithCallbackA (
    IN      HASHTABLE HashTable,
    IN      PHASHTABLE_CALLBACK_ROUTINEA Proc,
    IN      LPARAM lParam
    )

/*++

Routine Description:

  EnumHashTableWithCallback implements a setupapi-style enumerator.  The
  callback routine is called for each item in the string table, and if the
  callback routine returns FALSE, the enumeration ends.

Arguments:

  HashTable - Specifies the handle to the table to enumerate
  Proc      - Specifies the callback procedure address
  lParam    - Specifies a value to pass to the callback, and is intended only
              for use by the caller.

Return Value:

  Always TRUE.

--*/

{
    PHASHTABLESTRUCT table = (PHASHTABLESTRUCT) HashTable;
    HASHTABLE_ENUMA e;

    ASSERT_TABLE_IS_VALID (HashTable);

    if (EnumFirstHashTableStringA (&e, HashTable)) {
        do {
            if (!Proc (HashTable, e.Index, e.String, (PVOID) e.ExtraData, table->ExtraDataSize, lParam)) {
                AbortHashTableEnumA (&e);
                break;
            }
        } while (EnumNextHashTableStringA (&e));
    }

    return TRUE;
}


BOOL
EnumHashTableWithCallbackW (
    IN      HASHTABLE HashTable,
    IN      PHASHTABLE_CALLBACK_ROUTINEW Proc,
    IN      LPARAM lParam
    )
{
    PHASHTABLESTRUCT table = (PHASHTABLESTRUCT) HashTable;
    HASHTABLE_ENUMW e;

    ASSERT_TABLE_IS_VALID (HashTable);

    if (EnumFirstHashTableStringW (&e, HashTable)) {
        do {
            if (!Proc (HashTable, e.Index, e.String, (PVOID) e.ExtraData, table->ExtraDataSize, lParam)) {
                AbortHashTableEnumW (&e);
                break;
            }
        } while (EnumNextHashTableStringW (&e));
    }

    return TRUE;
}


PCSTR
HtGetStringFromItemA (
    IN      HASHITEM Item
    )
{
    if (!Item) {
        return NULL;
    }

    return (PCSTR) ((PBYTE) Item + sizeof (BUCKETITEM));
}


PCWSTR
HtGetStringFromItemW (
    IN      HASHITEM Item
    )
{
    if (!Item) {
        return NULL;
    }

    return (PCWSTR) ((PBYTE) Item + sizeof (BUCKETITEM));
}


BOOL
HtIsEmpty (
    IN      HASHTABLE HashTable
    )
{
    PHASHTABLESTRUCT table = (PHASHTABLESTRUCT) HashTable;

    return (table->FirstLink == NULL);
}


BOOL
HtWriteToFile (
    IN      HASHTABLE HashTable,
    IN      HANDLE OutputFile,
    IN      HASHTABLEOUTPUTFLAGS Flags
    )
{
    PHASHTABLESTRUCT table = (PHASHTABLESTRUCT) HashTable;
    BOOL result = FALSE;
    DWORD bytesWritten;
    PBUCKETITEM item;
    PCWSTR unicodeStr;
    PCSTR ansiStr;
    DWORD dontCare;

    ASSERT_TABLE_IS_VALID (HashTable);

    __try {
        if (Flags & WRITE_UNICODE_HEADER) {
            if (table->Unicode) {
                if ((!WriteFile (OutputFile, "\xff\xfe", 2, &bytesWritten, NULL)) ||
                    (bytesWritten != 2)
                    ) {
                    DEBUGMSG ((DBG_ERROR, "Error writing hash table to output file"));
                    __leave;
                }
            }
        }

        if (Flags & REVERSE_ORDER) {
            item = table->LastLink;
        } else {
            item = table->FirstLink;
        }

        while (item) {

            if (table->Unicode) {

                unicodeStr = HtGetStringFromItemW (item);

                if (!WriteFile (OutputFile, unicodeStr, ByteCountW (unicodeStr), &dontCare, NULL)) {
                    __leave;
                }

                if (!WriteFile (OutputFile, L"\r\n", 4, &dontCare, NULL)) {
                    __leave;
                }

            } else {

                ansiStr = HtGetStringFromItemA (item);

                if (!WriteFile (OutputFile, ansiStr, ByteCountA (ansiStr), &dontCare, NULL)) {
                    __leave;
                }

                if (!WriteFile (OutputFile, "\r\n", 2, &dontCare, NULL)) {
                    __leave;
                }

            }

            if (Flags & REVERSE_ORDER) {
                item = item->PrevLink;
            } else {
                item = item->NextLink;
            }
        }

        result = TRUE;
    }
    __finally {
    }

    return result;



}