Clean up compression settings when deleting compressed cagg

When deleting a cagg with compression on the materialization hypertable
the compression settings for that hypertable would not get removed when
dropping the cagg.

.unreleased/pr_6867

@@ -0,0 +1 @@
+Fixes: #6867 Clean up compression settings when deleting compressed cagg

src/ts_catalog/continuous_agg.c

@@ -38,6 +38,7 @@
 #include "time_bucket.h"
 #include "time_utils.h"
 #include "ts_catalog/catalog.h"
+#include "ts_catalog/compression_settings.h"
 #include "ts_catalog/continuous_agg.h"
 #include "ts_catalog/continuous_aggs_watermark.h"
 #include "utils.h"
@@ -1017,6 +1018,7 @@ drop_continuous_agg(FormData_continuous_agg *cadata, bool drop_user_view)
 	if (OidIsValid(mat_hypertable.objectId))
 	{
 		performDeletion(&mat_hypertable, DROP_CASCADE, 0);
+		ts_compression_settings_delete(mat_hypertable.objectId);
 		ts_hypertable_delete_by_id(cadata->mat_hypertable_id);
 	}
 

tsl/test/expected/cagg_ddl-13.out

@@ -2074,3 +2074,31 @@ SELECT * FROM conditions_daily ORDER BY bucket, avg;
  NYC      | Thu Nov 01 17:00:00 2018 PDT |  15
 (6 rows)
 
+-- check compression settings are cleaned up when deleting a cagg with compression
+CREATE TABLE cagg_cleanup(time timestamptz not null);
+SELECT table_name FROM create_hypertable('cagg_cleanup','time');
+  table_name  
+--------------
+ cagg_cleanup
+(1 row)
+
+INSERT INTO cagg_cleanup SELECT '2020-01-01';
+CREATE MATERIALIZED VIEW cagg1 WITH (timescaledb.continuous) AS SELECT time_bucket('1h',time) FROM cagg_cleanup GROUP BY 1;
+NOTICE:  refreshing continuous aggregate "cagg1"
+ALTER MATERIALIZED VIEW cagg1 SET (timescaledb.compress);
+NOTICE:  defaulting compress_orderby to time_bucket
+WARNING:  there was some uncertainty picking the default segment by for the hypertable: You do not have any indexes on columns that can be used for segment_by and thus we are not using segment_by for compression. Please make sure you are not missing any indexes
+NOTICE:  default segment by for hypertable "_materialized_hypertable_52" is set to ""
+SELECT count(compress_chunk(ch)) FROM show_chunks('cagg1') ch;
+ count 
+-------
+     1
+(1 row)
+
+DROP MATERIALIZED VIEW cagg1;
+NOTICE:  drop cascades to table _timescaledb_internal._hyper_52_68_chunk
+SELECT * FROM _timescaledb_catalog.compression_settings;
+ relid | segmentby | orderby | orderby_desc | orderby_nullsfirst 
+-------+-----------+---------+--------------+--------------------
+(0 rows)
+

tsl/test/expected/cagg_ddl-14.out

@@ -2074,3 +2074,31 @@ SELECT * FROM conditions_daily ORDER BY bucket, avg;
  NYC      | Thu Nov 01 17:00:00 2018 PDT |  15
 (6 rows)
 
+-- check compression settings are cleaned up when deleting a cagg with compression
+CREATE TABLE cagg_cleanup(time timestamptz not null);
+SELECT table_name FROM create_hypertable('cagg_cleanup','time');
+  table_name  
+--------------
+ cagg_cleanup
+(1 row)
+
+INSERT INTO cagg_cleanup SELECT '2020-01-01';
+CREATE MATERIALIZED VIEW cagg1 WITH (timescaledb.continuous) AS SELECT time_bucket('1h',time) FROM cagg_cleanup GROUP BY 1;
+NOTICE:  refreshing continuous aggregate "cagg1"
+ALTER MATERIALIZED VIEW cagg1 SET (timescaledb.compress);
+NOTICE:  defaulting compress_orderby to time_bucket
+WARNING:  there was some uncertainty picking the default segment by for the hypertable: You do not have any indexes on columns that can be used for segment_by and thus we are not using segment_by for compression. Please make sure you are not missing any indexes
+NOTICE:  default segment by for hypertable "_materialized_hypertable_52" is set to ""
+SELECT count(compress_chunk(ch)) FROM show_chunks('cagg1') ch;
+ count 
+-------
+     1
+(1 row)
+
+DROP MATERIALIZED VIEW cagg1;
+NOTICE:  drop cascades to table _timescaledb_internal._hyper_52_68_chunk
+SELECT * FROM _timescaledb_catalog.compression_settings;
+ relid | segmentby | orderby | orderby_desc | orderby_nullsfirst 
+-------+-----------+---------+--------------+--------------------
+(0 rows)
+

tsl/test/expected/cagg_ddl-15.out

@@ -2074,3 +2074,31 @@ SELECT * FROM conditions_daily ORDER BY bucket, avg;
  NYC      | Thu Nov 01 17:00:00 2018 PDT |  15
 (6 rows)
 
+-- check compression settings are cleaned up when deleting a cagg with compression
+CREATE TABLE cagg_cleanup(time timestamptz not null);
+SELECT table_name FROM create_hypertable('cagg_cleanup','time');
+  table_name  
+--------------
+ cagg_cleanup
+(1 row)
+
+INSERT INTO cagg_cleanup SELECT '2020-01-01';
+CREATE MATERIALIZED VIEW cagg1 WITH (timescaledb.continuous) AS SELECT time_bucket('1h',time) FROM cagg_cleanup GROUP BY 1;
+NOTICE:  refreshing continuous aggregate "cagg1"
+ALTER MATERIALIZED VIEW cagg1 SET (timescaledb.compress);
+NOTICE:  defaulting compress_orderby to time_bucket
+WARNING:  there was some uncertainty picking the default segment by for the hypertable: You do not have any indexes on columns that can be used for segment_by and thus we are not using segment_by for compression. Please make sure you are not missing any indexes
+NOTICE:  default segment by for hypertable "_materialized_hypertable_52" is set to ""
+SELECT count(compress_chunk(ch)) FROM show_chunks('cagg1') ch;
+ count 
+-------
+     1
+(1 row)
+
+DROP MATERIALIZED VIEW cagg1;
+NOTICE:  drop cascades to table _timescaledb_internal._hyper_52_68_chunk
+SELECT * FROM _timescaledb_catalog.compression_settings;
+ relid | segmentby | orderby | orderby_desc | orderby_nullsfirst 
+-------+-----------+---------+--------------+--------------------
+(0 rows)
+

tsl/test/expected/cagg_ddl-16.out

@@ -2074,3 +2074,31 @@ SELECT * FROM conditions_daily ORDER BY bucket, avg;
  NYC      | Thu Nov 01 17:00:00 2018 PDT |  15
 (6 rows)
 
+-- check compression settings are cleaned up when deleting a cagg with compression
+CREATE TABLE cagg_cleanup(time timestamptz not null);
+SELECT table_name FROM create_hypertable('cagg_cleanup','time');
+  table_name  
+--------------
+ cagg_cleanup
+(1 row)
+
+INSERT INTO cagg_cleanup SELECT '2020-01-01';
+CREATE MATERIALIZED VIEW cagg1 WITH (timescaledb.continuous) AS SELECT time_bucket('1h',time) FROM cagg_cleanup GROUP BY 1;
+NOTICE:  refreshing continuous aggregate "cagg1"
+ALTER MATERIALIZED VIEW cagg1 SET (timescaledb.compress);
+NOTICE:  defaulting compress_orderby to time_bucket
+WARNING:  there was some uncertainty picking the default segment by for the hypertable: You do not have any indexes on columns that can be used for segment_by and thus we are not using segment_by for compression. Please make sure you are not missing any indexes
+NOTICE:  default segment by for hypertable "_materialized_hypertable_52" is set to ""
+SELECT count(compress_chunk(ch)) FROM show_chunks('cagg1') ch;
+ count 
+-------
+     1
+(1 row)
+
+DROP MATERIALIZED VIEW cagg1;
+NOTICE:  drop cascades to table _timescaledb_internal._hyper_52_68_chunk
+SELECT * FROM _timescaledb_catalog.compression_settings;
+ relid | segmentby | orderby | orderby_desc | orderby_nullsfirst 
+-------+-----------+---------+--------------+--------------------
+(0 rows)
+

tsl/test/sql/cagg_ddl.sql.in

@@ -1320,3 +1320,14 @@ SELECT _timescaledb_functions.to_timestamp(_timescaledb_functions.cagg_watermark
 
 -- Should return ROWS because the watermark was reseted by the TRUNCATE
 SELECT * FROM conditions_daily ORDER BY bucket, avg;
+
+-- check compression settings are cleaned up when deleting a cagg with compression
+CREATE TABLE cagg_cleanup(time timestamptz not null);
+SELECT table_name FROM create_hypertable('cagg_cleanup','time');
+INSERT INTO cagg_cleanup SELECT '2020-01-01';
+CREATE MATERIALIZED VIEW cagg1 WITH (timescaledb.continuous) AS SELECT time_bucket('1h',time) FROM cagg_cleanup GROUP BY 1;
+ALTER MATERIALIZED VIEW cagg1 SET (timescaledb.compress);
+SELECT count(compress_chunk(ch)) FROM show_chunks('cagg1') ch;
+DROP MATERIALIZED VIEW cagg1;
+SELECT * FROM _timescaledb_catalog.compression_settings;
+