Issue with st_dump?

[JimShady]

Two (almost identical) queries below. One in PostGIS returns 78 features, the same one in apache-sedona returns 1 feature. I don't understand why - can anyone help/advise please?

PostGIS query - returns 78 features

WITH a_table AS (
SELECT st_boundary(geom) geom
FROM    emea_eu_seu.albania_prefectures_2021
UNION ALL
SELECT st_boundary(geom) geom
FROM    emea_eu_seu.albania_municipalities_2021
UNION ALL
SELECT st_boundary(geom) geom
FROM    emea_eu_seu.albania_2_digit_postcode_areas_2021
),
b_table AS (
SELECT   geom
FROM     a_table
GROUP BY geom
),
c_table AS (
SELECT   ST_Union(geom) geom
FROM     b_table
),
d_table AS (
SELECT ST_Polygonize(geom) AS geom 
FROM c_table
),
e_table AS (
SELECT st_dump(geom) geom
FROM   d_table
)
SELECT  COUNT(*) FROM e_table

Sedona query - returns 1 feature

%sql

WITH a_table AS (
SELECT st_boundary(st_geomfromwkb(geom_wkb)) geom
FROM   hive_metastore.gfk_bronze.albania_prefectures_2021
UNION ALL
SELECT st_boundary(st_geomfromwkb(geom_wkb)) geom
FROM   hive_metastore.gfk_bronze.albania_municipalities_2021
UNION ALL
SELECT st_boundary(st_geomfromwkb(geom_wkb)) geom
FROM   hive_metastore.gfk_bronze.albania_2_digit_postcode_areas_2021
),
b_table AS (
SELECT   geom
FROM     a_table
GROUP BY geom
),
c_table AS (
SELECT   ST_Union_Aggr(geom) geom
FROM     b_table
),
d_table AS (
SELECT ST_Polygonize(geom) AS geom 
FROM c_table
),
e_table AS (
SELECT st_dump(geom) geom
FROM   d_table
)
SELECT  COUNT(*) FROM e_table

[JimShady]

After some more digging it appears that the issue is with st_dump. In apache-sedona it is not creating 78 rows (I think?) it might be returning an array of 78 items in one record? So perhaps I need to explode it?

[JimShady]

Figured it out - I need to run the explode() function after st_dump.