[
https://issues.apache.org/jira/browse/CALCITE-4215?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=17189908#comment-17189908
]
Vladimir Sitnikov commented on CALCITE-4215:
--------------------------------------------
I'm afraid -1 would be much harder to notice with IDE :-(
We could borrow an idea {{get...OrNull}} from other libraries. We can have two
metadata APIs: one which is null returning, and another that returns a
good-looking default.
For instance: {{getRowCount}} (throws on null) vs {{getRowCountOrNull}} (return
nulls).
Alternative: {{getRowCount}} (returns 100 in case row count is unknown) vs
{{getRowCountOrNull}} (return null).
In other words, the default method could return something that makes sense in
most of the cases, and there might be a method that returns raw {{null}} for
that 0.1% when it matters.
In practice, we already have
{{org.apache.calcite.rel.metadata.RelMetadataQuery#getAverageColumnSizes}} vs
{{org.apache.calcite.rel.metadata.RelMetadataQuery#getAverageColumnSizesNotNull}},
however it would read easier if {{OrNull}} was explicitly present in the
method name.
> Ensure org.apache.calcite.schema.Statistic uses null vs emptyList
> appropriately
> -------------------------------------------------------------------------------
>
> Key: CALCITE-4215
> URL: https://issues.apache.org/jira/browse/CALCITE-4215
> Project: Calcite
> Issue Type: Sub-task
> Components: core
> Affects Versions: 1.25.0
> Reporter: Vladimir Sitnikov
> Priority: Major
> Labels: pull-request-available
> Time Spent: 10m
> Remaining Estimate: 0h
>
> null: statistic is *not* *known*
> emptyList: statistic is *known*, and the value is *empty* (e.g. no unique
> keys in the table)
--
This message was sent by Atlassian Jira
(v8.3.4#803005)
