pydoop_mapreduce_wordcount_with_mapper_reducer_classes.py

1import pydoop.mapreduce.api as api
2import pydoop.mapreduce.pipes as pipes
3
4class Mapper(api.Mapper):
5    def map(self, context):
6        words = context.value.split()
7        for w in words:
8            context.emit(w, 1)
9
10class Reducer(api.Reducer):
11    def reduce(self, context):
12        s = sum(context.values)
13        context.emit(context.key, s)
14
15def __main__():
16    pipes.run_task(pipes.Factory(mapper_class=Mapper, reducer_class=Reducer))