/**
* Creates a new data row reader for sparse format. The attributes indices must not be set. If
* they are, they are reassigned new values when this constructor is called!
*
* @param factory Factory used to create {@link DataRow} instances.
* @param format One Out of FORMAT_XY, FORMAT_YX, FORMAT_PREFIX, and FORMAT_SEPARATE_FILE.
* @param prefixMap Maps prefixes to special attribute names (e.g. "l" to
* "label").
* @param attributeSet Set of regular and special attributes.
* @param attributeReader Reader for the data
* @param labelReader Reader for the labels. Only necessary if format is FORMAT_SEPARATE_FILE.
* @param sampleSize sample size, may be -1 for no limit.
* @param useQuotesForNominalValues Determines whether nominal values are surrounded by quotes or
* not. If <code>useQuotesForNominalValues == true</code> the first and last character of the
* nominal values are ignored.
* @param quoteChar The char that is used to surround nominal values.
*/
public SparseFormatDataRowReader(
DataRowFactory factory,
int format,
Map<String, String> prefixMap,
AttributeSet attributeSet,
Reader attributeReader,
Reader labelReader,
int sampleSize,
boolean useQuotesForNominalValues,
char quoteChar) {
super(factory);
this.format = format;
this.prefixMap = prefixMap;
this.attributeSet = attributeSet;
if (attributeSet == null) {
throw new IllegalArgumentException("AttributeSet must not be null.");
}
this.dimension = attributeSet.getAllAttributes().size();
this.maxNumber = sampleSize;
this.inAttributes = new BufferedReader(attributeReader);
if (format == FORMAT_SEPARATE_FILE) {
if (labelReader == null)
throw new IllegalArgumentException(
"labelReader must not be null if format is 'separate_file'!");
this.inLabels = new BufferedReader(labelReader);
}
if (format != FORMAT_NO_LABEL) {
if (attributeSet.getSpecialAttribute("label") == null) {
throw new IllegalArgumentException(
"If format is not no_label, label attribute must be defined.");
}
}
this.useQuotesForNominalValues = useQuotesForNominalValues;
this.quoteChar = quoteChar;
}
