NutchMap.java 文件源码

@Override
public void map(LongWritable key, Text value, Mapper.Context context) throws IOException, InterruptedException {
    TrecOLParser document = new TrecOLParser(value.toString());
    documentAnalyzed = new MapWritable();
    if (document.isParsed()) {
        this.tokenizer.tokenize(document.getDocContent());
        while (this.tokenizer.hasMoreTokens()) {
            IntWritable counter = CastingTypes.zero;
            String newTerm = this.tokenizer.nextToken();
            Text term = new Text(newTerm);
            if (documentAnalyzed.containsKey(term)) {
                counter = CastingTypes.strToIntWr(documentAnalyzed.get(term).toString());
            }
            documentAnalyzed.put(term, CastingTypes.intToIntWr(counter.get()+1));
        }
        if ( ! documentAnalyzed.isEmpty()) {
            context.write(CastingTypes.strToIntWr(document.getDocId()), documentAnalyzed);
        }
    }
}