C-Loftus commented on code in PR #2125:
URL: https://github.com/apache/iceberg-go/pull/2125#discussion_r4214103599
##########
table/internal/parquet_files_test.go:
##########
@@ -3238,7 +3238,10 @@ func TestWriteDataFileGeoBounds(t *testing.T) {
assert.True(t, geomType.Equals(lit.Type()), "want geo type, got
%s", lit.Type())
assert.Equal(t, geoBoundBytes(5, 10),
lit.(iceberg.GeoLiteral).Value())
- // Null counts are still recorded for geo columns.
+ // Null counts are still recorded for geo columns. The Parquet
writer
+ // omits statistics for GEOMETRY/GEOGRAPHY columns, so these
come from
+ // the Arrow data rather than the footer.
+ assert.Equal(t, int64(0), df.NullValueCounts()[2])
Review Comment:
Done: both the full-mode and counts-mode cases now require.Contains the key
first, and the test data has a null geometry row, so the expected count is 1
rather than a defaulted 0.
--
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]
