built in - Hadoop - detect last record of the input split -

I have a MapRadus job, whose map work uses TextInputFormat. I want to know that in the work of the map, when reaching the end of division (i.e. the last record is settled in the map function).

I know that there are some built-in counter (example: the mapped input records counter that counts the input records consumed by all mapper but I do not need it).

Can I use any of these at the counter?

If not, do you know how can I get this information in my maps work?

You define your argument as Mapper.cleanup (context) method (or < Code> Mapper.close () for old mapped APIs), it is said later that the final record has been processed by your map method.