/// <summary>
/// 二分法模糊搜索
/// </summary>
/// <param name="dictIndexList"></param>
/// <param name="word"></param>
/// <returns></returns>
private List <int> BinaryQuery(List <DictIndex> dictIndexList, string word)
{
int low = 0;
int high = dictIndexList.Count - 1;
List <int> simVal = new List <int>();
while ((low <= high) && (low <= dictIndexList.Count - 1) && (high <= dictIndexList.Count - 1))
{
int middle = (high + low) >> 1;
DictIndex midVal = dictIndexList[middle];
if (midVal.word.ToLower().StartsWith(word.ToLower()))
{
simVal.Add(middle);
}
if (CompareTo(midVal.word, word) < 0)
{
low = middle + 1;
}
else if (CompareTo(midVal.word, word) > 0)
{
high = middle - 1;
}
else
{
return(simVal);
}
}
return(simVal);
}
/// <summary>
/// 二分查找算法
/// </summary>
/// <returns></returns>
private int BinarySearch(List <DictIndex> dictIndexList, string word)
{
int low = 0;
int high = dictIndexList.Count - 1;
while ((low <= high) && (low <= dictIndexList.Count - 1) && (high <= dictIndexList.Count - 1))
{
int middle = (high + low) >> 1;
DictIndex midVal = dictIndexList[middle];
if (CompareTo(midVal.word, word) < 0)
{
low = middle + 1;
}
else if (CompareTo(midVal.word, word) > 0)
{
high = middle - 1;
}
else
{
return(middle);
}
}
return(-1);
}
private List <DictIndex> GetIndex()
{
//List<DictIndex> dictIndexList = new List<DictIndex>();
// the maximun length of a word must less 256
// 256 bytes(word) + 1 byte('\0') + 4 bytes(offset) + 4 bytes(def size)
byte[] bytes = new byte[256 + 1 + 4 + 4];
int currentPos = 0;
while (IdxStream.Read(bytes, 0, bytes.Length) > 0)
{
int j = 0;
bool isWordPart = true;
bool isOffsetPart = false;
bool isSizePart = false;
string word = null;
long offset = 0; // offset of a word in data file
long size = 0; // size of word's defition
int wordLength = 0; // the byte(s) length of a word
for (int i = 0; i < bytes.Length; i++)
{
if (isWordPart)
{
if (bytes[i] == 0)
{
wordLength = i;
word = System.Text.Encoding.UTF8.GetString(bytes, j, i - j);
j = i;
isWordPart = false;
isOffsetPart = true;
}
continue;
}
if (isOffsetPart)
{
i += 3;// skip the split token: '\0'
j++;
if (i >= bytes.Length)
{
i = bytes.Length - 1;
}
offset = ByteConverter.Unsigned4BytesToInt(bytes, j);
j = i + 1;
isOffsetPart = false;
isSizePart = true;
continue;
}
if (isSizePart)
{
i += 3;
if (i >= bytes.Length)
{
i = bytes.Length - 1;
}
size = ByteConverter.Unsigned4BytesToInt(bytes, j);
j = i + 1;
isSizePart = false;
isWordPart = true;
}
DictIndex dictIndex = new DictIndex();
dictIndex.word = word;
dictIndex.offset = offset;
dictIndex.size = size;
dictIndexList.Add(dictIndex);
// skip current index entry
int indexSize = wordLength + 1 + 4 + 4;
IdxStream.Seek(indexSize + currentPos, SeekOrigin.Begin);
currentPos += indexSize;
break;
}
}
return(dictIndexList);
}
