Hi guys,
I`m working on creation external Hive tables from Avro files using column-based
approach:
CREATE TABLE kst (
string1 string,
string2 string,
int1 int,
boolean1 boolean,
long1 bigint,
float1 float,
double1 double,
inner_record1
struct<int_in_inner_record1:int,string_in_inner_record1:string>,
enum1 string,
array1 array<string>,
map1 map<string,string>,
union1 uniontype<float,boolean,string>,
fixed1 binary,
null1 void,
unionnullint int,
bytes1 binary)
PARTITIONED BY (ds string)
STORED AS AVRO;
And faced with case sensitive issue:
If I have column in Avro as String1, it will be not found in dataset and I will
get default NULL, cause Hive will transform it to lowercase “string1”.
Similar issue was with Parquet https://issues.apache.org/jira/browse/HIVE-7554
and was resolved by extending properties.
Is there any work around except using avro.schema.url or avro.schema.literal ?
