Move partialize functions to _timescaledb_functions schema

To increase schema security we do not want to mix our own internal objects with user objects. Since chunks are created in the _timescaledb_internal schema our internal functions should live in a different dedicated schema. This patch make the necessary adjustments for the following functions: - finalize_agg_ffunc(internal,text,name,name,name[],bytea,anyelement) - finalize_agg_sfunc(internal,text,name,name,name[],bytea,anyelement) - partialize_agg(anyelement) - finalize_agg(text,name,name,name[][],bytea,anyelement)
lkshminarayanan · Aug 29, 2023 · 3b6dc7d · 3b6dc7d
1 parent 623381c
commit 3b6dc7d
Show file tree

Hide file tree

Showing 25 changed files with 708 additions and 2,683 deletions.
diff --git a/sql/partialize_finalize.sql b/sql/partialize_finalize.sql
@@ -2,24 +2,24 @@
 -- Please see the included NOTICE for copyright information and
 -- LICENSE-APACHE for a copy of the license.
 
-CREATE OR REPLACE FUNCTION _timescaledb_internal.partialize_agg(arg ANYELEMENT)
+CREATE OR REPLACE FUNCTION _timescaledb_functions.partialize_agg(arg ANYELEMENT)
 RETURNS BYTEA AS '@MODULE_PATHNAME@', 'ts_partialize_agg' LANGUAGE C STABLE PARALLEL SAFE;
 
-CREATE OR REPLACE FUNCTION _timescaledb_internal.finalize_agg_sfunc(
+CREATE OR REPLACE FUNCTION _timescaledb_functions.finalize_agg_sfunc(
 tstate internal, aggfn TEXT, inner_agg_collation_schema NAME, inner_agg_collation_name NAME, inner_agg_input_types NAME[][], inner_agg_serialized_state BYTEA, return_type_dummy_val ANYELEMENT)
 RETURNS internal
 AS '@MODULE_PATHNAME@', 'ts_finalize_agg_sfunc'
 LANGUAGE C IMMUTABLE;
 
-CREATE OR REPLACE FUNCTION _timescaledb_internal.finalize_agg_ffunc(
+CREATE OR REPLACE FUNCTION _timescaledb_functions.finalize_agg_ffunc(
 tstate internal, aggfn TEXT, inner_agg_collation_schema NAME, inner_agg_collation_name NAME, inner_agg_input_types NAME[][], inner_agg_serialized_state BYTEA, return_type_dummy_val ANYELEMENT)
 RETURNS anyelement
 AS '@MODULE_PATHNAME@', 'ts_finalize_agg_ffunc'
 LANGUAGE C IMMUTABLE;
 
-CREATE OR REPLACE AGGREGATE _timescaledb_internal.finalize_agg(agg_name TEXT,  inner_agg_collation_schema NAME,  inner_agg_collation_name NAME, inner_agg_input_types NAME[][], inner_agg_serialized_state BYTEA, return_type_dummy_val anyelement) (
-    SFUNC = _timescaledb_internal.finalize_agg_sfunc,
+CREATE OR REPLACE AGGREGATE _timescaledb_functions.finalize_agg(agg_name TEXT,  inner_agg_collation_schema NAME,  inner_agg_collation_name NAME, inner_agg_input_types NAME[][], inner_agg_serialized_state BYTEA, return_type_dummy_val anyelement) (
+    SFUNC = _timescaledb_functions.finalize_agg_sfunc,
     STYPE = internal,
-    FINALFUNC = _timescaledb_internal.finalize_agg_ffunc,
+    FINALFUNC = _timescaledb_functions.finalize_agg_ffunc,
     FINALFUNC_EXTRA
 );
diff --git a/sql/updates/latest-dev.sql b/sql/updates/latest-dev.sql
@@ -122,6 +122,8 @@ DECLARE
     cagg_migrate_execute_refresh_new_cagg, cagg_migrate_execute_copy_data, cagg_migrate_execute_override_cagg,
     cagg_migrate_execute_drop_old_cagg, cagg_migrate_execute_plan,
 
+    finalize_agg,
+
     hypertable_invalidation_log_delete, invalidation_cagg_log_add_entry, invalidation_hyper_log_add_entry,
     invalidation_process_cagg_log, invalidation_process_hypertable_log, materialization_invalidation_log_delete,
 
@@ -134,7 +136,13 @@ DECLARE
   }';
 BEGIN
   FOR foid, kind IN
-    SELECT oid, CASE WHEN prokind = 'f' THEN 'FUNCTION' ELSE 'PROCEDURE' END FROM pg_proc WHERE proname = ANY(funcs) AND pronamespace = '_timescaledb_internal'::regnamespace
+    SELECT oid,
+    CASE
+      WHEN prokind = 'f' THEN 'FUNCTION'
+      WHEN prokind = 'a' THEN 'AGGREGATE'
+      ELSE 'PROCEDURE'
+    END
+    FROM pg_proc WHERE proname = ANY(funcs) AND pronamespace = '_timescaledb_internal'::regnamespace
   LOOP
     EXECUTE format('ALTER %s %s SET SCHEMA _timescaledb_functions', kind, foid);
   END LOOP;
@@ -155,3 +163,7 @@ ALTER FUNCTION _timescaledb_internal.get_partition_hash(val anyelement) SET SCHE
 
 UPDATE _timescaledb_catalog.dimension SET partitioning_func_schema = '_timescaledb_functions' WHERE partitioning_func_schema = '_timescaledb_internal' AND partitioning_func IN ('get_partition_for_key','get_partition_hash');
 
+ALTER FUNCTION _timescaledb_internal.finalize_agg_ffunc(internal,text,name,name,name[],bytea,anyelement) SET SCHEMA _timescaledb_functions;
+ALTER FUNCTION _timescaledb_internal.finalize_agg_sfunc(internal,text,name,name,name[],bytea,anyelement) SET SCHEMA _timescaledb_functions;
+ALTER FUNCTION _timescaledb_internal.partialize_agg(anyelement) SET SCHEMA _timescaledb_functions;
+
diff --git a/sql/updates/reverse-dev.sql b/sql/updates/reverse-dev.sql
@@ -157,3 +157,8 @@ ALTER FUNCTION _timescaledb_functions.get_partition_hash(val anyelement) SET SCH
 
 UPDATE _timescaledb_catalog.dimension SET partitioning_func_schema = '_timescaledb_internal' WHERE partitioning_func_schema = '_timescaledb_functions' AND partitioning_func IN ('get_partition_for_key','get_partition_hash');
 
+ALTER FUNCTION _timescaledb_functions.finalize_agg_ffunc(internal,text,name,name,name[],bytea,anyelement) SET SCHEMA _timescaledb_internal;
+ALTER FUNCTION _timescaledb_functions.finalize_agg_sfunc(internal,text,name,name,name[],bytea,anyelement) SET SCHEMA _timescaledb_internal;
+ALTER FUNCTION _timescaledb_functions.partialize_agg(anyelement) SET SCHEMA _timescaledb_internal;
+ALTER AGGREGATE _timescaledb_functions.finalize_agg(text,name,name,name[][],bytea,anyelement) SET SCHEMA _timescaledb_internal;
+
diff --git a/src/planner/partialize.c b/src/planner/partialize.c
@@ -37,7 +37,7 @@ typedef struct PartializeWalkerState
  *
  * The partialize function is an expression of the form:
  *
- * _timescaledb_internal.partialize_agg(avg(temp))
+ * _timescaledb_functions.partialize_agg(avg(temp))
  *
  * where avg(temp) can be replaced by any aggregate that can be partialized.
  *
@@ -115,7 +115,7 @@ has_partialize_function(Node *node, PartializeAggFixAggref fix_aggref)
 									.looking_for_agg = false,
 									.fix_aggref = fix_aggref,
 									.fnoid = InvalidOid };
-	List *name = list_make2(makeString(INTERNAL_SCHEMA_NAME), makeString(TS_PARTIALFN));
+	List *name = list_make2(makeString(FUNCTIONS_SCHEMA_NAME), makeString(TS_PARTIALFN));
 
 	partialfnoid = LookupFuncName(name, lengthof(argtyp), argtyp, false);
 	Assert(OidIsValid(partialfnoid));
@@ -193,7 +193,7 @@ partialize_agg_paths(RelOptInfo *rel)
  * instance:
  *
  *  SELECT time_bucket('1 day', time), device,
- *  _timescaledb_internal.partialize_agg(avg(temp))
+ *  _timescaledb_functions.partialize_agg(avg(temp))
  *  GROUP BY 1, 2;
  *
  * Would compute the partial aggregate of avg(temp).

diff --git a/tsl/src/continuous_aggs/common.h b/tsl/src/continuous_aggs/common.h
@@ -37,7 +37,6 @@
 #include "ts_catalog/catalog.h"
 #include "ts_catalog/continuous_agg.h"
 
-#define TS_PARTIALFN "partialize_agg"
 #define CONTINUOUS_AGG_MAX_JOIN_RELATIONS 2
 #define DEFAULT_MATPARTCOLUMN_NAME "time_partition_col"
 

diff --git a/tsl/src/continuous_aggs/finalize.c b/tsl/src/continuous_aggs/finalize.c
@@ -11,6 +11,7 @@
 #include "finalize.h"
 #include "create.h"
 #include "common.h"
+#include <partialize_finalize.h>
 
 typedef struct CAggHavingCxt
 {
@@ -53,11 +54,11 @@ get_partialize_funcexpr(Aggref *agg)
 	FuncExpr *partialize_fnexpr;
 	Oid partfnoid, partargtype;
 	partargtype = ANYELEMENTOID;
-	partfnoid =
-		LookupFuncName(list_make2(makeString(INTERNAL_SCHEMA_NAME), makeString(TS_PARTIALFN)),
-					   1,
-					   &partargtype,
-					   false);
+	partfnoid = LookupFuncName(list_make2(makeString(FUNCTIONS_SCHEMA_NAME),
+										  makeString(PARTIALIZE_FUNC_NAME)),
+							   1,
+							   &partargtype,
+							   false);
 	partialize_fnexpr = makeFuncExpr(partfnoid,
 									 BYTEAOID,
 									 list_make1(agg), /*args*/
@@ -194,7 +195,7 @@ get_finalize_function_oid(void)
 	Oid finalfnoid;
 	Oid finalfnargtypes[] = { TEXTOID,	NAMEOID,	  NAMEOID, get_array_type(NAMEOID),
 							  BYTEAOID, ANYELEMENTOID };
-	List *funcname = list_make2(makeString(INTERNAL_SCHEMA_NAME), makeString(FINALFN));
+	List *funcname = list_make2(makeString(FUNCTIONS_SCHEMA_NAME), makeString(FINALFN));
 	int nargs = sizeof(finalfnargtypes) / sizeof(finalfnargtypes[0]);
 	finalfnoid = LookupFuncName(funcname, nargs, finalfnargtypes, false);
 	return finalfnoid;

diff --git a/tsl/src/fdw/deparse.c b/tsl/src/fdw/deparse.c
@@ -2973,7 +2973,7 @@ deparseAggref(Aggref *node, deparse_expr_cxt *context)
 
 	/* Find aggregate name from aggfnoid which is a pg_proc entry */
 	if (partial_agg)
-		appendStringInfoString(buf, INTERNAL_SCHEMA_NAME "." PARTIALIZE_FUNC_NAME "(");
+		appendStringInfoString(buf, FUNCTIONS_SCHEMA_NAME "." PARTIALIZE_FUNC_NAME "(");
 
 	appendFunctionName(node->aggfnoid, context);
 	appendStringInfoChar(buf, '(');

diff --git a/tsl/src/partialize_finalize.c b/tsl/src/partialize_finalize.c
@@ -161,7 +161,7 @@ collation_oid_from_name(char *schema_name, char *collation_name)
  * partial state of numeric aggregates also changing format.
  *
  * If a user that has stored partials (by using Continuous Aggregates or calling
- * _timescaledb_internal.finalize_agg()) upgrades to PG14 then the partial state deserialization
+ * _timescaledb_functions.finalize_agg()) upgrades to PG14 then the partial state deserialization
  * will lead to errors due to the mismatch with the PG14 code.
  *
  * For F_NUMERIC_AVG_DESERIALIZE and F_NUMERIC_DESERIALIZE the length of the serialized aggregate