Flink CSV file reader fails to cast a LongType into a PojoType -
part of code trying execute in flink:
val pages = env.readcsvfile[(long)]("/home/ppi.csv", fielddelimiter = "\t", includedfields = array(1))
i want use pages
other purpose when compile, flink throws me error message
exception in thread "main" java.lang.classcastexception:
org.apache.flink.api.common.typeinfo.integertypeinfo cannot cast to
org.apache.flink.api.java.typeutils.pojotypeinfo
by way using 0.9 snapshot version of flink. in right direction highly appreciated.
if read csv file, return type scala tuple read fields. in example, reading single field give tuple1. try specify parentheses surrounding "long":
readcsvfile[(long)]
in scala can specify tuples 2 or more fields using parentheses. need write instead
readcsvfile[tuple1[long]]
the exception thrown because, flink's csvinputformat tries interpret non-tuple types pojo types.
Comments
Post a Comment