public virtual void Get(int docNumber, TermVectorMapper mapper) { // Check if no term vectors are available for this segment at all if (tvx != null) { //We need to offset by SeekTvx(docNumber); long tvdPosition = tvx.ReadLong(); tvd.Seek(tvdPosition); int fieldCount = tvd.ReadVInt(); // No fields are vectorized for this document if (fieldCount != 0) { System.String[] fields = ReadFields(fieldCount); long[] tvfPointers = ReadTvfPointers(fieldCount); mapper.SetDocumentNumber(docNumber); ReadTermVectors(fields, tvfPointers, mapper); } } else { //System.out.println("No tvx file"); } }
public virtual void Get(int docNum, System.String field, TermVectorMapper mapper) { if (tvx != null) { int fieldNumber = fieldInfos.FieldNumber(field); //We need to account for the FORMAT_SIZE at when seeking in the tvx //We don't need to do this in other seeks because we already have the // file pointer //that was written in another file SeekTvx(docNum); //System.out.println("TVX Pointer: " + tvx.getFilePointer()); long tvdPosition = tvx.ReadLong(); tvd.Seek(tvdPosition); int fieldCount = tvd.ReadVInt(); //System.out.println("Num Fields: " + fieldCount); // There are only a few fields per document. We opt for a full scan // rather then requiring that they be ordered. We need to read through // all of the fields anyway to get to the tvf pointers. int number = 0; int found = -1; for (int i = 0; i < fieldCount; i++) { if (format >= FORMAT_VERSION) { number = tvd.ReadVInt(); } else { number += tvd.ReadVInt(); } if (number == fieldNumber) { found = i; } } // This field, although valid in the segment, was not found in this // document if (found != -1) { // Compute position in the tvf file long position; if (format >= FORMAT_VERSION2) { position = tvx.ReadLong(); } else { position = tvd.ReadVLong(); } for (int i = 1; i <= found; i++) { position += tvd.ReadVLong(); } mapper.SetDocumentNumber(docNum); ReadTermVector(field, position, mapper); } else { //System.out.println("Fieldable not found"); } } else { //System.out.println("No tvx file"); } }