static DefaultSimilarity()
{
for (int i = 0; i < 256; i++)
{
NORM_TABLE[i] = SmallFloat.Byte315ToFloat((sbyte)i);
}
}
static BM25Similarity()
{
for (int i = 0; i < 256; i++)
{
float f = SmallFloat.Byte315ToFloat((sbyte)i);
NORM_TABLE[i] = 1.0f / (f * f);
}
}
/// <summary>
/// 得到文档某字段的Norm
/// </summary>
/// <param name="reader"></param>
/// <param name="field"></param>
/// <param name="docId"></param>
/// <returns></returns>
public static float GetNorm(IndexReader reader, string field, int docId)
{
return(SmallFloat.Byte315ToFloat(reader.Norms(field)[docId]));
}
/// <summary>
/// The default implementation encodes <code>boost / sqrt(length)</code>
/// with <seealso cref="SmallFloat#floatToByte315(float)"/>. this is compatible with
/// Lucene's default implementation. If you change this, then you should
/// change <seealso cref="#decodeNormValue(byte)"/> to match.
/// </summary>
protected internal virtual sbyte EncodeNormValue(float boost, int fieldLength)
{
return(SmallFloat.FloatToByte315(boost / (float)Math.Sqrt(fieldLength)));
}
/// <summary>Encodes a normalization factor for storage in an index.
///
/// <p/>The encoding uses a three-bit mantissa, a five-bit exponent, and
/// the zero-exponent point at 15, thus
/// representing values from around 7x10^9 to 2x10^-9 with about one
/// significant decimal digit of accuracy. Zero is also represented.
/// Negative numbers are rounded up to zero. Values too large to represent
/// are rounded down to the largest representable value. Positive values too
/// small to represent are rounded up to the smallest positive representable
/// value.
///
/// </summary>
/// <seealso cref="AbstractField.Boost" />
/// <seealso cref="Lucene.Net.Util.SmallFloat" />
public static byte EncodeNorm(float f)
{
return((byte)SmallFloat.FloatToByte315(f));
}
/// <summary>
/// Encodes a normalization factor for storage in an index.
/// <p>
/// The encoding uses a three-bit mantissa, a five-bit exponent, and the
/// zero-exponent point at 15, thus representing values from around 7x10^9 to
/// 2x10^-9 with about one significant decimal digit of accuracy. Zero is also
/// represented. Negative numbers are rounded up to zero. Values too large to
/// represent are rounded down to the largest representable value. Positive
/// values too small to represent are rounded up to the smallest positive
/// representable value.
/// </summary>
/// <seealso cref= Lucene.Net.Document.Field#setBoost(float) </seealso>
/// <seealso cref= Lucene.Net.Util.SmallFloat </seealso>
public override sealed long EncodeNormValue(float f)
{
return(SmallFloat.FloatToByte315(f));
}
