Hi.
While investigating some issue, I noticed that direct SQL in Hive would
usually fail if default partition is involved, and fall back to JDO; JDO
filter pruning will exclude the default partition.
I didn’t realize the latter was the case… Is it the expected behavior with
the default partition? I
{code}
- even more so when {{part}} is an integer column.
Would it be better to allow 'bad' values to map onto directory-friendly strings
like {{__HIVE_NULL_PARTITION_KEY__}} etc?
> let user specify "bad file" rather than HIVE_DEFAULT_PAR
