HIVE_BAD_DATA with Parquet BINARY

0

Hi,

I saved some Parquet files (with Spark), and some of the fields were converted to Timestamp types. That appears to be saved as INT96, or BINARY, inside Parquet.

When I declare the columns as TIMESTAMP in Amazon Athena, and then try to invoke them, I receive this error message:

HIVE_BAD_DATA: Field scheduled_content_start_time's type BINARY in parquet is incompatible with type bigint defined in table schema

However, if I try to declare the column as type BINARY, it tells be that BINARY is not supported for Parquet file format.

How can I resolve this?

Thanks,
Michael

msjg
gefragt vor 7 Jahren1876 Aufrufe
2 Antworten
0

you could use string

beantwortet vor 7 Jahren
0

Hi,

Good day.

But can you query a string as you would a timestamp tough?

Regards
JJ

beantwortet vor 6 Jahren

Du bist nicht angemeldet. Anmelden um eine Antwort zu veröffentlichen.

Eine gute Antwort beantwortet die Frage klar, gibt konstruktives Feedback und fördert die berufliche Weiterentwicklung des Fragenstellers.

Richtlinien für die Beantwortung von Fragen