walterddr commented on code in PR #10423:
URL: https://github.com/apache/pinot/pull/10423#discussion_r1152504977
##########
pinot-spi/src/main/java/org/apache/pinot/spi/data/FieldSpec.java:
##########
@@ -45,6 +45,7 @@
@SuppressWarnings("unused")
public abstract class FieldSpec implements Comparable<FieldSpec>, Serializable
{
public static final int DEFAULT_MAX_LENGTH = 512;
+ public static final boolean DEFAULT_NULLABILITY = false;
Review Comment:
that's the idea.
and we will gradually deprecate the table level nullability and switch to
use column level. in which case we need to
1. migate the schema to have nullable column setting
2. turn off table level enable nullablity.
there's a whole lot of problem.
1. what happen to all the previously null setup columns in previous segments?
2. how do we handle column become nullable but some segments generated
previously doesn't have the vector
3. I can think of a lot of other corner cases.
This PR doesn't address any of these. It only give a more fine-grain
configuration telling the query engine some columns are nullable while others
are not --> in contrast to a table level, where query engine HAVE TO assume all
columns are nullable, and generate less efficient plan
--
This is an automated message from the Apache Git Service.
To respond to the message, please log on to GitHub and use the
URL above to go to the specific comment.
To unsubscribe, e-mail: [email protected]
For queries about this service, please contact Infrastructure at:
[email protected]
---------------------------------------------------------------------
To unsubscribe, e-mail: [email protected]
For additional commands, e-mail: [email protected]
