Most occurred first character in the word of a file -PIG

Most occurred first character in the word of a  file

lines  = LOAD '/user/cloudera/my-friends' AS (line: chararray); 

tokens = FOREACH lines GENERATE flatten(TOKENIZE(line)) As token:chararray;

letters = FOREACH tokens  GENERATE SUBSTRING(token,0,1) As letter:chararray;

lettergrp = GROUP letters by letter; 

countletter  = FOREACH  lettergrp  GENERATE group,COUNT(letters);

OrderCnt = ORDER countletter BY $1 DESC; 

result = LIMIT OrderCnt 1;

STORE result into '/user/cloudera/dummy5556777777';