pig.txt

stock = LOAD 'stocksdir/stockprices.csv' USING  PigStorage(',');

stock1 = FOREACH stock GENERATE $0,$2;
-------------------------------------------------------------------------------------------------------------------

stock = LOAD 'stocksdir/stockprices.csv' USING  PigStorage(',') AS (symbol:chararray,date:chararray,price:float);

Describe stock

stock1 = FOREACH stock GENERATE symbol,price;

Dump stock1

-------------------------------------------------------------------------------------------------------------------

stock1 = FOREACH stock GENERATE symbol,SQRT(price),price*0.1 as price1;

stock2 = filter stock1 by symbol == 'tcs';

stock3 = order stock2 by price desc;
dump stock3;
-------------------------------------------------------------------------------------------------------------------

stock_1 = group stock by symbol;

stock_2 = foreach stock_1 generate group,COUNT(stock);

stock_3 = foreach stock_1 generate UPPER(group),AVG(stock.price);

===================================================================================================================

stocks = LOAD 'stocksdir/stockprices.csv' USING  PigStorage(',') AS (symbol:chararray,date:chararray,price:float);

symmaster = LOAD 'stocksdir/symbolmaster.csv' USING  PigStorage(',') AS (symbol:chararray,cname:chararray,phone:chararray,hq:chararray);

joined = JOIN stocks by symbol , symmaster by symbol; //inner join

dump joined

describe joined

jproj = FOREACH joined GENERATE date,price, stocks::symbol;//no need for relation name in case of unique columns

dump jproj

jproj = FOREACH joined GENERATE date,price, stocks::symbol parallel 20;//set no of reducer(here 20)

set default_parallel 10; //set default reducer no 

STORE jproj into 'joinedout';

jproj = FOREACH joined GENERATE stocks::symbol,cname,price parallel 3;

STORE jproj into 'joinedout1' using PigStorage(',');

rjoined = JOIN stocks by symbol right , symmaster by symbol;//(right/left/full outer)

fjoined = JOIN stocks by symbol full outer , symmaster by symbol;

history