sqoop-import --connect jdbc:mysql://ms.itversity.com:3306/retail_db \
--username retail_user \
--password itversity \
--table products \
--fields-terminated-by "|" \
--lines-terminated-by "\n" \
--warehouse-dir "/user/krishnatejathatavarthi/teja_aruns/solutions02/" \
--as-textfile

hadoop fs -chmod -R 765 /user/krishnatejathatavarthi/teja_aruns/solutions02

from pyspark import SparkContext,SparkConf,Row,HiveContext
import avro.schema
conf=SparkConf().setAppName("importnexport").setMaster("yarn-client")
sc=SparkContext(conf=conf)
sqlContext=HiveContext(sc)

productsRDD=sc.textFile("/user/krishnatejathatavarthi/teja_aruns/solutions02/products")
productsDF = productsRDD.map(lambda rec: Row(product_id=int(rec.split("|")[0]),product_category_id=int(rec.split("|")[1]),product_name=rec.split("|")[2],product_desc=rec.split("|")[3],product_price=float(rec.split("|")[4]),product_image=rec.split("|")[5])).toDF()
productsDF.registerTempTable("products")
sqlContext.setConf("spark.sql.shuffle.partitions","2")
sqlResult=sqlContext.sql("select product_category_id,max(product_price) max_price, count(product_id) total_orders, avg(product_price) avg_price, min(product_price) min_price from products where product_price < 100 group by product_category_id order by product_category_id")
sqlContext.setConf("spark.sql.avro.compression.codec","snappy")
sqlResult.save("/user/krishnatejathatavarthi/teja_aruns/solutions02/solution","com.databricks.spark.avro")