Open jonvex opened 1 day ago
HUDI-8036 changed the output schema and treated a symptom of the real issue but not the issue itself.
The root of the issue is that we are not parsing the partition path correctly and into the correct data types. This will be fixed in subsequent prs
I also had to modify the complex keygen sql tests a bit because now the correct partition column value and data type occurs
Spark reads timestamp partition cols in their correct datatype
low
N/A
hudi-bot
commented
1 day ago
Change Logs
HUDI-8036 changed the output schema and treated a symptom of the real issue but not the issue itself.
The root of the issue is that we are not parsing the partition path correctly and into the correct data types. This will be fixed in subsequent prs
I also had to modify the complex keygen sql tests a bit because now the correct partition column value and data type occurs
Impact
Spark reads timestamp partition cols in their correct datatype
Risk level (write none, low medium or high below)
low
Documentation Update
N/A
Contributor's checklist