Posted by
VenuReddy on
Dec 26, 2020; 8:39am
URL: http://apache-carbondata-dev-mailing-list-archive.168.s1.nabble.com/DISCUSSION-Geo-spatial-index-algorithm-improvement-and-UDFs-enhancement-tp104717p105126.html
+1
Good to see Geo enhancements.
Just few points/queries :
1. Util UDFs seem to take origin latitude and grid size as argments as well.
Shall we inherit them from table specified in the query during query
processing? Probably can avoid invalid/inconsistent origin latitude and grid
size values given as UDF arguments(i.e., not same values as in
tableproperties).
2. Regarding the point - *"Allowing flexibility to user to specify the
spatial index value when 'LOAD' and 'INSERT INTO' without generating it
implicitly based on the configured table properties(i.e., grid size, origin
latitude etc)"*
Wouldn't the query results vary when user configured spatial indexes are
different than that of generated ones ? Also when insert into target_table
select * from source_table, both source_table and target_table may not
necessarily have the same geo specific tableproperties. May need to generate
the spatial index value based on the target_table properties.
3. After this algorithm improvements, polygon query result match with the
current version ? If no, suggest to capture the difference in the doc.
Regards,
Venu
--
Sent from:
http://apache-carbondata-dev-mailing-list-archive.1130556.n5.nabble.com/