clintropolis commented on code in PR #14262:
URL: https://github.com/apache/druid/pull/14262#discussion_r1191777219
##########
processing/src/main/java/org/apache/druid/segment/AutoTypeColumnIndexer.java:
##########
@@ -282,6 +282,10 @@ public ColumnCapabilities getColumnCapabilities()
private ColumnType getLogicalType()
{
+ if (fieldIndexers.isEmpty()) {
+ // we didn't see anything, so we can be anything, so why not a string?
+ return ColumnType.STRING;
Review Comment:
I think it would be useful since null really is like its own thing, but i
think the change has some implications since we probably also want a
`ValueType.NULL` to go along with it and then a lot of places to handle the
case.
Ive been considering this for quite a while now because it would make a lot
of things a lot nicer, particularly with non-vectorized expressions where we
have to treat `null` as a `STRING` unless something externally told us to treat
it as some other type, so there are lots of strange places where we allow
string type to participate in things which might otherwise be invalid because
we don't have a null type.
--
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]
