xref: /aoo41x/main/tools/workben/hashtbl.cxx (revision 89b56da7)

/**************************************************************
 *
 * Licensed to the Apache Software Foundation (ASF) under one
 * or more contributor license agreements.  See the NOTICE file
 * distributed with this work for additional information
 * regarding copyright ownership.  The ASF licenses this file
 * to you under the Apache License, Version 2.0 (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.apache.org/licenses/LICENSE-2.0
 *
 * Unless required by applicable law or agreed to in writing,
 * software distributed under the License is distributed on an
 * "AS IS" BASIS, WITHOUT WARRANTIES OR CONDITIONS OF ANY
 * KIND, either express or implied.  See the License for the
 * specific language governing permissions and limitations
 * under the License.
 *
 *************************************************************/



// MARKER(update_precomp.py): autogen include statement, do not remove
#include "precompiled_tools.hxx"

#include <tlgen.hxx>
#include "hashtbl.hxx"

#include <algorithm>

// -------------------------------------------------------------
// class HashItem
//
class HashItem
{
    enum ETag { TAG_EMPTY, TAG_USED, TAG_DELETED };

    void*   m_pObject;
    ETag    m_Tag;
    String  m_Key;

public:
    HashItem() { m_Tag = TAG_EMPTY; m_pObject = NULL; }

    BOOL IsDeleted() const
    {   return m_Tag == TAG_DELETED; }

    BOOL IsEmpty() const
    {   return m_Tag == TAG_DELETED || m_Tag == TAG_EMPTY; }

    BOOL IsFree() const
    {   return m_Tag == TAG_EMPTY; }

    BOOL IsUsed() const
    {   return m_Tag == TAG_USED; }

    void Delete()
    { m_Tag = TAG_DELETED; m_Key = ""; m_pObject = NULL; }

    String const& GetKey() const
    { return m_Key; }

    void* GetObject() const
    { return m_pObject; }

    void SetObject(String const Key, void *pObject)
    { m_Tag = TAG_USED; m_Key = Key; m_pObject = pObject; }
};

// #define MIN(a,b) (a)<(b)?(a):(b)
// #define MAX(a,b) (a)>(b)?(a):(b)

// -------------------------------------------------------------
// class HashTable
//

/*static*/ double HashTable::m_defMaxLoadFactor = 0.8;
/*static*/ double HashTable::m_defDefGrowFactor = 2.0;

HashTable::HashTable(ULONG lSize, BOOL bOwner, double dMaxLoadFactor, double dGrowFactor)
{
    m_lSize          = lSize;
    m_bOwner         = bOwner;
    m_lElem          = 0;
    m_dMaxLoadFactor = std::max(0.5,std::min(1.0,dMaxLoadFactor));  // 0.5 ... 1.0
    m_dGrowFactor    = std::max(1.3,(5.0,dGrowFactor));     // 1.3 ... 5.0
    m_pData          = new HashItem [lSize];

// Statistik
#ifdef DBG_UTIL
	m_aStatistic.m_lSingleHash = 0;
	m_aStatistic.m_lDoubleHash = 0;
	m_aStatistic.m_lProbe	   = 0;
#endif
}

HashTable::~HashTable()
{
    // Wenn die HashTable der Owner der Objecte ist,
    // m�ssen die Destruktoren separat gerufen werden.
    // Dies geschieht �ber die virtuelle Methode OnDeleteObject()
    //
    // Problem: Virtuelle Funktionen sind im Destructor nicht virtuell!!
    //          Der Code mu� deshalb ins Macro

    /*
    if (m_bOwner)
    {
        for (ULONG i=0; i<GetSize(); i++)
        {
            void *pObject = GetObjectAt(i);

            if (pObject != NULL)
                OnDeleteObject(pObject());
        }
    }
    */

    // Speicher f�r HashItems freigeben
    delete [] m_pData;
}

void* HashTable::GetObjectAt(ULONG lPos) const
// Gibt Objekt zur�ck, wenn es eines gibt, sonst NULL;
{
    DBG_ASSERT(lPos<m_lSize, "HashTable::GetObjectAt()");

    HashItem *pItem = &m_pData[lPos];

    return pItem->IsUsed() ? pItem->GetObject() : NULL;
}

void HashTable::OnDeleteObject(void*)
{
    DBG_ERROR("HashTable::OnDeleteObject(void*) nicht �berladen");
}

ULONG HashTable::Hash(String const& Key) const
{
	/*
    ULONG lHash = 0;
    ULONG i,n;

    for (i=0,n=Key.Len(); i<n; i++)
    {
        lHash *= 256L;
        lHash += (ULONG)(USHORT)Key.GetStr()[i];
        lHash %= m_lSize;
    }
    return lHash;
	*/

	// Hashfunktion von P.J. Weinberger
	// aus dem "Drachenbuch" von Aho/Sethi/Ullman
    ULONG i,n;
	ULONG h = 0;
	ULONG g = 0;

    for (i=0,n=Key.Len(); i<n; i++)
	{
		h = (h<<4) + (ULONG)(USHORT)Key.GetStr()[i];
		g = h & 0xf0000000;

		if (g != 0)
		{
			h = h ^ (g >> 24);
			h = h ^ g;
		}
	}

	return h % m_lSize;
}

ULONG HashTable::DHash(String const& Key, ULONG lOldHash) const
{
    ULONG lHash = lOldHash;
    ULONG i,n;

    for (i=0,n=Key.Len(); i<n; i++)
    {
        lHash *= 256L;
        lHash += (ULONG)(USHORT)Key.GetStr()[i];
        lHash %= m_lSize;
    }
    return lHash;

/*    return
		(
			lHash
		+	(char)Key.GetStr()[0] * 256
		+	(char)Key.GetStr()[Key.Len()-1]
		+	1
		)
		% m_lSize;
*/
}

ULONG HashTable::Probe(ULONG lPos) const
// gibt den Folgewert von lPos zur�ck
{
    lPos++; if (lPos==m_lSize) lPos=0;
    return lPos;
}

BOOL HashTable::IsFull() const
{
    return m_lElem>=m_lSize;
}

BOOL HashTable::Insert(String const& Key, void* pObject)
// pre:  Key ist nicht im Dictionary enthalten, sonst return FALSE
//       Dictionary ist nicht voll, sonst return FALSE
// post: pObject ist unter Key im Dictionary; m_nElem wurde erh�ht
{
    SmartGrow();

    if (IsFull())
    {
        DBG_ERROR("HashTable::Insert() is full");
        return FALSE;
    }

    if (FindPos(Key) != NULL )
        return FALSE;

    ULONG     lPos  = Hash(Key);
    HashItem *pItem = &m_pData[lPos];

    // first hashing
    //
    if (pItem->IsEmpty())
    {
        pItem->SetObject(Key, pObject);
        m_lElem++;

		#ifdef DBG_UTIL
		m_aStatistic.m_lSingleHash++;
		#endif

        return TRUE;
    }

    // double hashing
    //
    lPos  = DHash(Key,lPos);
    pItem = &m_pData[lPos];

    if (pItem->IsEmpty())
    {
        pItem->SetObject(Key, pObject);
        m_lElem++;

		#ifdef DBG_UTIL
		m_aStatistic.m_lDoubleHash++;
		#endif

        return TRUE;
    }

    // linear probing
    //
    do
    {
		#ifdef DBG_UTIL
		m_aStatistic.m_lProbe++;
		#endif

        lPos  = Probe(lPos);
        pItem = &m_pData[lPos];
    }
    while(!pItem->IsEmpty());

    pItem->SetObject(Key, pObject);
    m_lElem++;
    return TRUE;
}

HashItem* HashTable::FindPos(String const& Key) const
// sucht den Key; gibt Refrenz auf den Eintrag (gefunden)
// oder NULL (nicht gefunden) zur�ck
//
// pre:  -
// post: -
{
    // first hashing
    //
    ULONG     lPos  = Hash(Key);
    HashItem *pItem = &m_pData[lPos];

    if (pItem->IsUsed()
    &&  pItem->GetKey() == Key)
    {
        return pItem;
    }

    // double hashing
    //
    if (pItem->IsDeleted() || pItem->IsUsed())
    {
        lPos  = DHash(Key,lPos);
        pItem = &m_pData[lPos];

        if (pItem->IsUsed()
        &&  pItem->GetKey() == Key)
        {
            return pItem;
        }

        // linear probing
        //
        if (pItem->IsDeleted() || pItem->IsUsed())
        {
            ULONG n      = 0;
            BOOL  bFound = FALSE;
            BOOL  bEnd   = FALSE;

            do
            {
                n++;
                lPos   = Probe(lPos);
                pItem  = &m_pData[lPos];

                bFound =  pItem->IsUsed()
                       && pItem->GetKey() == Key;

                bEnd = !(n<m_lSize || pItem->IsFree());
            }
            while(!bFound && !bEnd);

            return bFound ? pItem : NULL;
        }
    }

    // nicht gefunden
    //
    return NULL;
}

void* HashTable::Find(String const& Key) const
// Gibt Verweis des Objektes zur�ck, das unter Key abgespeichert ist,
// oder NULL wenn nicht vorhanden.
//
// pre:  -
// post: -
{
    HashItem *pItem = FindPos(Key);

    if (pItem != NULL
    &&  pItem->GetKey() == Key)
        return pItem->GetObject();
    else
        return NULL;
}

void* HashTable::Delete(String const& Key)
// L�scht Objekt, das unter Key abgespeichert ist und gibt Verweis
// darauf zur�ck.
// Gibt NULL zur�ck, wenn Key nicht vorhanden ist.
//
// pre:  -
// post: Objekt ist nicht mehr enthalten; m_lElem dekrementiert
//       Wenn die HashTable der Owner ist, wurde das Object gel�scht
{
    HashItem *pItem = FindPos(Key);

    if (pItem != NULL
    &&  pItem->GetKey() == Key)
    {
        void* pObject = pItem->GetObject();

        if (m_bOwner)
            OnDeleteObject(pObject);

        pItem->Delete();
        m_lElem--;
        return pObject;
    }
    else
    {
        return NULL;
    }
}

double HashTable::CalcLoadFactor() const
// prozentuale Belegung der Hashtabelle berechnen
{
    return double(m_lElem) / double(m_lSize);
}

void HashTable::SmartGrow()
// Achtung: da die Objekte umkopiert werden, darf die OnDeleteObject-Methode
//          nicht gerufen werden
{
    double dLoadFactor = CalcLoadFactor();

    if (dLoadFactor <= m_dMaxLoadFactor)
        return; // nothing to grow

    ULONG     lOldSize = m_lSize;              // alte Daten sichern
    HashItem* pOldData = m_pData;

    m_lSize = ULONG (m_dGrowFactor * m_lSize); // neue Gr��e
    m_pData = new HashItem[m_lSize];           // neue Daten holen

    // kein Speicher:
    // Zustand "Tabelle voll" wird in Insert abgefangen
    //
    if (m_pData == NULL)
    {
        m_lSize = lOldSize;
        m_pData = pOldData;
        return;
    }

    m_lElem = 0;                               // noch keine neuen Daten

    // Umkopieren der Daten
    //
    for (ULONG i=0; i<lOldSize; i++)
    {
        HashItem *pItem = &pOldData[i];

        if (pItem->IsUsed())
            Insert(pItem->GetKey(),pItem->GetObject());
    }

    delete [] pOldData;
}

// Iterator ---------------------------------------------------------
//

HashTableIterator::HashTableIterator(HashTable const& aTable)
: m_aTable(aTable)
{
	m_lAt = 0;
}

void* HashTableIterator::GetFirst()
{
	m_lAt = 0;
	return FindValidObject(TRUE /* forward */);
}

void* HashTableIterator::GetLast()
{
	m_lAt = m_aTable.GetSize() -1;
	return FindValidObject(FALSE /* backward */);
}

void* HashTableIterator::GetNext()
{
	if (m_lAt+1 >= m_aTable.GetSize())
		return NULL;

	m_lAt++;
	return FindValidObject(TRUE /* forward */);
}

void* HashTableIterator::GetPrev()
{
	if (m_lAt <= 0)
		return NULL;

	m_lAt--;
	return FindValidObject(FALSE /* backward */);
}

void* HashTableIterator::FindValidObject(BOOL bForward)
// Sucht nach einem vorhandenen Objekt ab der aktuellen
// Position.
//
// pre:  ab inkl. m_lAt soll die Suche beginnen
// post: if not found then
//			if bForward == TRUE then
//				m_lAt == m_aTable.GetSize() -1
//			else
//				m_lAt == 0
//		 else
//			m_lAt ist die gefundene Position
{
	void *pObject = m_aTable.GetObjectAt(m_lAt);

	if (pObject != NULL)
		return pObject;

	while (pObject == NULL
	   && (bForward ? ((m_lAt+1) < m_aTable.GetSize())
					:   m_lAt    > 0))
	{
		if (bForward)
			m_lAt++;
		else
			m_lAt--;

		pObject = m_aTable.GetObjectAt(m_lAt);
	}

#ifdef DBG_UTIL

	if (pObject == NULL)
	{
		DBG_ASSERT(bForward ? m_lAt == m_aTable.GetSize() -1 : m_lAt == 0,
			"HashTableIterator::FindValidObject()");
	}

#endif

	return pObject;
}