/* (non-Javadoc)
* @see org.apache.hadoop.mapreduce.Reducer#setup(org.apache.hadoop.mapreduce.Reducer.Context)
*/
protected void setup(Context context) throws IOException, InterruptedException {
// load schema
Configuration conf = context.getConfiguration();
String filePath = conf.get("schema.file.path");
FileSystem dfs = FileSystem.get(conf);
Path src = new Path(filePath);
FSDataInputStream fs = dfs.open(src);
ObjectMapper mapper = new ObjectMapper();
schema = mapper.readValue(fs, MixedTypeSchema.class);
firstTypeSize = schema.getEntityByType(0).getFieldCount();
firstIdOrdinal = schema.getEntityByType(0).getIdField().getOrdinal();
secondIdOrdinal = schema.getEntityByType(1).getIdField().getOrdinal();
Field field = schema.getEntityByType(0).getClassAttributeField();
if (null != field) {
firstClassAttrOrdinal = field.getOrdinal();
secondClassAttrOrdinal = schema.getEntityByType(0).getClassAttributeField().getOrdinal();
}
fields = schema.getEntityByType(0).getFields();
targetFields = schema.getEntityByType(1).getFields();
scale = context.getConfiguration().getInt("distance.scale", 1000);
distStrategy = schema.createDistanceStrategy(scale);
fieldDelimRegex = context.getConfiguration().get("field.delim.regex", "\\[\\]");
fieldDelim = context.getConfiguration().get("field.delim", ",");
textSimStrategy = schema.createTextSimilarityStrategy();
outputVerbose = context.getConfiguration().getBoolean("sim.output.verbose", true);
identifyWithFilePrefix =
context.getConfiguration().getBoolean("identify.with.file.prefix", false);
System.out.println(
"firstTypeSize: "
+ firstTypeSize
+ " firstIdOrdinal:"
+ firstIdOrdinal
+ " secondIdOrdinal:"
+ secondIdOrdinal
+ " Source field count:"
+ fields.size()
+ " Target field count:"
+ targetFields.size());
}
