Conversation
Signed-off-by: Adam Gutglick <adam@spiraldb.com>
Merging this PR will degrade performance by 12.34%
Performance Changes
Comparing 
Footnotes
|
Signed-off-by: Adam Gutglick <adam@spiraldb.com>
Signed-off-by: Adam Gutglick <adam@spiraldb.com>
Signed-off-by: Adam Gutglick <adam@spiraldb.com>
| } | ||
|
|
||
| /// Creates a new VariantArray with the given nullability. | ||
| pub fn new_nullable(child: ArrayRef, nullability: Nullability) -> Self { |
There was a problem hiding this comment.
Do we know that VariantArray is going to have a child ArrayRef or is that just so you can implement some of the required methods?
Also, we probably just want new to take a Nullablity directly instead of having a new_nullable to be consistent with the other nested arrays.
There was a problem hiding this comment.
The general idea is that the variant array has a single child which should be a more concrete encoding of the "variant" family.
There was a problem hiding this comment.
also fixed the new thing
Signed-off-by: Adam Gutglick <adam@spiraldb.com>
|
|
||
| fn stats(array: &Self::Array) -> StatsSetRef<'_> { | ||
| array.stats_set.to_ref(array.as_ref()) | ||
| array.child.statistics() |
There was a problem hiding this comment.
I think this is the right thing here, might change as I go deeper into the specific encoding
This PR introduces a minimal new canonical array for the new variant DType, its not very useable at the moment but I figured its a piece of mostly boilerplate code that is worth splitting out of the bigger branch. --------- Signed-off-by: Adam Gutglick <adam@spiraldb.com>
2 participants
This PR introduces a minimal new canonical array for the new variant DType, its not very useable at the moment but I figured its a piece of mostly boilerplate code that is worth splitting out of the bigger branch.