Closed fllife closed 3 years ago
https://www.google.com/search?q=parquet+int96+julian
EDIT shorter URL
Also this is duplicated to #408
type.INT64ToTime
is compatible with Spark output. For details you can reference here.
You should check the encoding type in your INT96.
type.INT64ToTime
is compatible with Spark output. For details you can reference here.You should check the encoding type in your INT96.
Maybe encoding type is wrong? how should i check the encoding type in my INT96?
Maybe encoding type is wrong? how should i check the encoding type in my INT96?
If you don't want to write code, you can use a parquet-tools to dump schema of the parquet file, like the one in this repo, or the python one, or the Java one, or mine :).
Question, my parquet file schema has a
INT96
type, and I need to convert INT96 to timestamp, but when I read INT96 filed from file , the field string is messy code like�̌r) T�%
,so timestamp accuracy is lost after converted like1925-01-01 00:12:21.610
CODE:My main code is as follows:
so someone can tell me how can I read data to avoid messy code or how can I convert INT96 from parquet? thanks