CODEWITHSUNDEEP
scalaapache-sparkhadoopsqoop
Big_Data
Big_Data

Reputation: 322

java.lang.NumberFormatException: For input string: "|"

I have imported a table into HDFS as

fields-terminated-by '|'

sqoop import \
--connect jdbc:mysql://connection \
--username  \
--password  \
--table products \
--as-textfile \
--target-dir /user/username/productsdemo \
--fields-terminated-by '|'

after that, I am trying to read it as RDD using spark-shell version 1.6.2

var productsRDD = sc.textFile("/user/username/productsdemo")

and converting it into a data frame 

var productsDF = productsRDD.map(product =>{ 
var o = product.split("|");
products(o(0).toInt,o(1).toInt,o(2),o(3),o(4).toFloat,o(5))
}).toDF("product_id", "product_category_id","product_name","product_description","product_price","product_image" )

But When I try to print the output it is throwing the below exception.

java.lang.NumberFormatException: For input string: "|"

Why I am getting this error can anyone help me out of this?

Upvotes: 0

Views: 532

Answers (1)

flyingfox
flyingfox

Reputation: 13506

split are use regex to do the split string, since | is a special character in regex means OR You need to use \\| instead of | when split

var o = product.split("\\|");

Upvotes: 2

Related Questions