Examples of com.thinkaurelius.titan.core.attribute.FullFloat

Package com.thinkaurelius.titan.core.attribute

Examples of com.thinkaurelius.titan.core.attribute.FullFloat

com.thinkaurelius.titan.core.attribute.FullFloat
Helper class to be used when wanting to store the full value spectrum of a float value.
By default, Titan will convert float values into a decimal format so that it can be stored in byte-order and be used for vertex-centric indices. However, that restricts the value spectrum of floats when registered using {@link com.thinkaurelius.titan.core.KeyMaker#dataType(Class)} with Float.class as argument.
When order preservation is not important, use FullFloat as the data type. @author Matthias Broecheler (me@matthiasb.com)

    @Override
    public FullFloat convert(Object value) {
        if (value instanceof Number) {
            double d = ((Number)value).doubleValue();
            if (d<Float.MIN_VALUE || d>Float.MAX_VALUE) throw new IllegalArgumentException("Value too large for float: " + value);
            return new FullFloat((float)d);
        } else if (value instanceof String) {
            return new FullFloat(Float.parseFloat((String) value));
        } else return null;
    }

View Full Code Here

TOP

Related Classes of com.thinkaurelius.titan.core.attribute.FullFloat

com.thinkaurelius.titan.graphdb.database.serialize.attribute.FullFloatHandler

All source code are property of their respective owners. Java is a trademark of Sun Microsystems, Inc and owned by ORACLE Inc. Contact coftware#gmail.com.