//====================================================================
// Func Name : AddItem
// Description: Add a word item to the dictionary
// Parameters : sWord: the word
// nHandle:the handle number
// nFrequency: the frequency
// Returns : success or fail
//====================================================================
public bool AddItem(string sWord, int nPOS, int nFrequency)
{
int nPos, nFoundPos;
WordChain pRet, pTemp, pNext;
string sWordAdd;
//预处理,去掉词的前后的空格
if (!PreProcessing(ref sWord, out nPos, out sWordAdd))
{
return(false);
}
if (FindInOriginalTable(nPos, sWordAdd, nPOS, out nFoundPos))
{
//The word exists in the original table, so add the frequency
//Operation in the index table and its items
if (indexTable[nPos].WordItems[nFoundPos].nFrequency == -1)
{
//The word item has been removed
indexTable[nPos].WordItems[nFoundPos].nFrequency = nFrequency;
if (modifyTable == null)
{
modifyTable = new ModifyTableItem[Predefine.CC_NUM];
}
modifyTable[nPos].nDelete -= 1;
}
else
{
indexTable[nPos].WordItems[nFoundPos].nFrequency += nFrequency;
}
return(true);
}
//The items not exists in the index table.
//As following, we have to find the item whether exists in the modify data region
//If exists, change the frequency .or else add a item
if (modifyTable == null)
{
modifyTable = new ModifyTableItem[Predefine.CC_NUM];
for (int i = 0; i < Predefine.CC_NUM; i++)
{
modifyTable[i] = new ModifyTableItem();
}
}
if (FindInModifyTable(nPos, sWordAdd, nPOS, out pRet))
{
if (pRet != null)
{
pRet = pRet.next;
}
else
{
pRet = modifyTable[nPos].pWordItemHead;
}
pRet.data.nFrequency += nFrequency;
return(true);
}
//find the proper position to add the word to the modify data table and link
pTemp = new WordChain(); //Allocate the word chain node
pTemp.data = new WordItem();
pTemp.data.nPOS = nPOS; //store the handle
pTemp.data.nWordLen = Utility.GetWordLength(sWordAdd);
pTemp.data.sWord = sWordAdd;
pTemp.data.nFrequency = nFrequency;
pTemp.next = null;
if (pRet != null)
{
pNext = pRet.next; //Get the next item before the current item
pRet.next = pTemp; //link the node to the chain
}
else
{
pNext = modifyTable[nPos].pWordItemHead;
modifyTable[nPos].pWordItemHead = pTemp; //Set the pAdd as the head node
}
pTemp.next = pNext; //Very important!!!! or else it will lose some node
modifyTable[nPos].nCount++; //the number increase by one
return(true);
}
private void MergeAndSaveIndexTableItem(BinaryWriter writer, IndexTableItem item, ModifyTableItem modifyItem)
{
int j, nCount; //频率、词长、读取词性
WordChain pCur;
//计算修改后有效词块的数目
nCount = item.nCount + modifyItem.nCount - modifyItem.nDelete;
writer.Write(nCount);
pCur = modifyItem.pWordItemHead;
j = 0;
//对原表中的词块和修改表中的词块进行遍历,并把修改后的添加到原表中
while (pCur != null && j < item.nCount)
{
//如果修改表中的词小于原表中对应位置的词或者长度相等但nHandle值比原表中的小,则把修改表中的写入到词典文件当中.
if (Utility.CCStringCompare(pCur.data.sWord, item.WordItems[j].sWord) < 0 ||
((pCur.data.sWord == item.WordItems[j].sWord) &&
(pCur.data.nPOS < item.WordItems[j].nPOS)))
{
//Output the modified data to the file
SaveWordItem(writer, pCur.data);
pCur = pCur.next;
}
//频度nFrequecy等于-1说明该词已被删除,跳过它
else if (item.WordItems[j].nFrequency == -1)
{
j++;
}
//如果修改表中的词长度比原表中的长度大或 长度相等但句柄值要多,就把原表的词写入的词典文件中
else if (Utility.CCStringCompare(pCur.data.sWord, item.WordItems[j].sWord) > 0 ||
((pCur.data.sWord == item.WordItems[j].sWord) &&
(pCur.data.nPOS > item.WordItems[j].nPOS)))
{
//Output the index table data to the file
SaveWordItem(writer, item.WordItems[j]);
j++;
}
}
//如果归并结束后indexTable有剩余，则继续写完indexTable中的数据
if (j < item.nCount)
{
for (int i = j; i < item.nCount; i++)
{
if (item.WordItems[j].nFrequency != -1)
{
SaveWordItem(writer, item.WordItems[i]);
}
}
}
//否则继续写完modifyTable中的数据
else
{
while (pCur != null)
{
//Output the modified data to the file
SaveWordItem(writer, pCur.data);
pCur = pCur.next;
}
}
}
private void MergeAndSaveIndexTableItem(BinaryWriter writer, IndexTableItem item, ModifyTableItem modifyItem)
{
int j, nCount; //频率、词长、读取词性
WordChain pCur;
//计算修改后有效词块的数目
nCount = item.nCount + modifyItem.nCount - modifyItem.nDelete;
writer.Write(nCount);
pCur = modifyItem.pWordItemHead;
j = 0;
//对原表中的词块和修改表中的词块进行遍历,并把修改后的添加到原表中
while (pCur != null && j < item.nCount)
{
//如果修改表中的词小于原表中对应位置的词或者长度相等但nHandle值比原表中的小,则把修改表中的写入到词典文件当中.
if (Utility.CCStringCompare(pCur.data.sWord, item.WordItems[j].sWord) < 0 ||
((pCur.data.sWord == item.WordItems[j].sWord) &&
(pCur.data.nPOS < item.WordItems[j].nPOS)))
{
//Output the modified data to the file
SaveWordItem(writer, pCur.data);
pCur = pCur.next;
}
//频度nFrequecy等于-1说明该词已被删除,跳过它
else if (item.WordItems[j].nFrequency == -1)
j++;
//如果修改表中的词长度比原表中的长度大或 长度相等但句柄值要多,就把原表的词写入的词典文件中
else if (Utility.CCStringCompare(pCur.data.sWord, item.WordItems[j].sWord) > 0 ||
((pCur.data.sWord == item.WordItems[j].sWord) &&
(pCur.data.nPOS > item.WordItems[j].nPOS)))
{
//Output the index table data to the file
SaveWordItem(writer, item.WordItems[j]);
j++;
}
}
//如果归并结束后indexTable有剩余，则继续写完indexTable中的数据
if (j < item.nCount)
{
for (int i = j; i < item.nCount; i++)
if (item.WordItems[j].nFrequency != -1)
SaveWordItem(writer, item.WordItems[i]);
}
//否则继续写完modifyTable中的数据
else
while (pCur != null)
{
//Output the modified data to the file
SaveWordItem(writer, pCur.data);
pCur = pCur.next;
}
}
//====================================================================
// Func Name : AddItem
// Description: Add a word item to the dictionary
// Parameters : sWord: the word
// nHandle:the handle number
// nFrequency: the frequency
// Returns : success or fail
//====================================================================
public bool AddItem(string sWord, int nPOS, int nFrequency)
{
int nPos, nFoundPos;
WordChain pRet, pTemp, pNext;
string sWordAdd;
//预处理,去掉词的前后的空格
if (!PreProcessing(ref sWord, out nPos, out sWordAdd))
return false;
if (FindInOriginalTable(nPos, sWordAdd, nPOS, out nFoundPos))
{
//The word exists in the original table, so add the frequency
//Operation in the index table and its items
if (indexTable[nPos].WordItems[nFoundPos].nFrequency == -1)
{
//The word item has been removed
indexTable[nPos].WordItems[nFoundPos].nFrequency = nFrequency;
if (modifyTable == null)
modifyTable = new ModifyTableItem[Predefine.CC_NUM];
modifyTable[nPos].nDelete -= 1;
}
else
indexTable[nPos].WordItems[nFoundPos].nFrequency += nFrequency;
return true;
}
//The items not exists in the index table.
//As following, we have to find the item whether exists in the modify data region
//If exists, change the frequency .or else add a item
if (modifyTable == null)
{
modifyTable = new ModifyTableItem[Predefine.CC_NUM];
for (int i = 0; i < Predefine.CC_NUM; i++)
modifyTable[i] = new ModifyTableItem();
}
if (FindInModifyTable(nPos, sWordAdd, nPOS, out pRet))
{
if (pRet != null)
pRet = pRet.next;
else
pRet = modifyTable[nPos].pWordItemHead;
pRet.data.nFrequency += nFrequency;
return true;
}
//find the proper position to add the word to the modify data table and link
pTemp = new WordChain(); //Allocate the word chain node
pTemp.data = new WordItem();
pTemp.data.nPOS = nPOS; //store the handle
pTemp.data.nWordLen = Utility.GetWordLength(sWordAdd);
pTemp.data.sWord = sWordAdd;
pTemp.data.nFrequency = nFrequency;
pTemp.next = null;
if (pRet != null)
{
pNext = pRet.next; //Get the next item before the current item
pRet.next = pTemp; //link the node to the chain
}
else
{
pNext = modifyTable[nPos].pWordItemHead;
modifyTable[nPos].pWordItemHead = pTemp; //Set the pAdd as the head node
}
pTemp.next = pNext; //Very important!!!! or else it will lose some node
modifyTable[nPos].nCount++; //the number increase by one
return true;
}
private void MergeAndSaveIndexTableItem(BinaryWriter writer, IndexTableItem item, ModifyTableItem modifyItem)
{
int j, nCount; //Ƶ�ʡ��ʳ�����ȡ����
WordChain pCur;
//�����ĺ���Ч�ʿ����Ŀ
nCount = item.nCount + modifyItem.nCount - modifyItem.nDelete;
writer.Write(nCount);
pCur = modifyItem.pWordItemHead;
j = 0;
//��ԭ���еĴʿ���ı��еĴʿ���б���,�����ĺ����ӵ�ԭ����
while (pCur != null && j < item.nCount)
{
//����ı��еĴ�С��ԭ���ж�Ӧλ�õĴʻ��߳�����ȵ�nHandleֵ��ԭ���е�С,����ı��е�д�뵽�ʵ��ļ�����.
if (Utility.CCStringCompare(pCur.data.sWord, item.WordItems[j].sWord) < 0 ||
((pCur.data.sWord == item.WordItems[j].sWord) &&
(pCur.data.nPOS < item.WordItems[j].nPOS)))
{
//Output the modified data to the file
SaveWordItem(writer, pCur.data);
pCur = pCur.next;
}
//Ƶ��nFrequecy����-1˵���ô��ѱ�ɾ��,������
else if (item.WordItems[j].nFrequency == -1)
j++;
//����ı��еĴʳ��ȱ�ԭ���еij��ȴ�� ������ȵ����ֵҪ��,�Ͱ�ԭ��Ĵ�д��Ĵʵ��ļ���
else if (Utility.CCStringCompare(pCur.data.sWord, item.WordItems[j].sWord) > 0 ||
((pCur.data.sWord == item.WordItems[j].sWord) &&
(pCur.data.nPOS > item.WordItems[j].nPOS)))
{
//Output the index table data to the file
SaveWordItem(writer, item.WordItems[j]);
j++;
}
}
//����鲢������indexTable��ʣ�࣬�����д��indexTable�е�����
if (j < item.nCount)
{
for (int i = j; i < item.nCount; i++)
if (item.WordItems[j].nFrequency != -1)
SaveWordItem(writer, item.WordItems[i]);
}
//��������modifyTable�����
else
while (pCur != null)
{
//Output the modified data to the file
SaveWordItem(writer, pCur.data);
pCur = pCur.next;
}
}
