diff --git a/Cargo.lock b/Cargo.lock
index cf6085bdd8..b491895cf5 100644
--- a/Cargo.lock
+++ b/Cargo.lock
@@ -8940,8 +8940,9 @@ dependencies = [
 
 [[package]]
 name = "sqlness"
-version = "0.4.3"
-source = "git+https://github.com/CeresDB/sqlness.git?rev=a4663365795d2067eb53966c383e1bb0c89c7627#a4663365795d2067eb53966c383e1bb0c89c7627"
+version = "0.5.0"
+source = "registry+https://github.com/rust-lang/crates.io-index"
+checksum = "0860f149718809371602b42573693e1ed2b1d0aed35fe69e04e4e4e9918d81f7"
 dependencies = [
  "async-trait",
  "derive_builder 0.11.2",
diff --git a/tests/cases/distributed/alter/rename_table.result b/tests/cases/distributed/alter/rename_table.result
index dbc77584e6..6e703c1010 100644
--- a/tests/cases/distributed/alter/rename_table.result
+++ b/tests/cases/distributed/alter/rename_table.result
@@ -25,6 +25,7 @@ SELECT * from t;
 |   | 4 |
 +---+---+
 
+-- TODO(LFC): Port test cases from standalone env when distribute rename table is implemented (#723).
 ALTER TABLE t RENAME new_table;
 
 Affected Rows: 0
@@ -33,6 +34,7 @@ DROP TABLE t;
 
 Error: 4001(TableNotFound), Table not found: greptime.public.t
 
+-- TODO: this clause should success
 -- SQLNESS REPLACE details.*
 DROP TABLE new_table;
 
diff --git a/tests/cases/distributed/optimizer/filter_push_down.result b/tests/cases/distributed/optimizer/filter_push_down.result
index 6859a0b7ed..4c06353723 100644
--- a/tests/cases/distributed/optimizer/filter_push_down.result
+++ b/tests/cases/distributed/optimizer/filter_push_down.result
@@ -180,6 +180,16 @@ SELECT i FROM (SELECT * FROM integers i1 UNION SELECT * FROM integers i2) a WHER
 | 3 |
 +---+
 
+-- TODO(LFC): Somehow the following SQL does not order by column 1 under new DataFusion occasionally. Should further investigate it. Comment it out temporarily.
+-- expected:
+--  +---+---+--------------+
+--  | a | b | ROW_NUMBER() |
+--  +---+---+--------------+
+--  | 1 | 1 | 1            |
+--  | 2 | 2 | 5            |
+--  | 3 | 3 | 9            |
+--  +---+---+--------------+
+-- SELECT * FROM (SELECT i1.i AS a, i2.i AS b, row_number() OVER (ORDER BY i1.i, i2.i) FROM integers i1, integers i2 WHERE i1.i IS NOT NULL AND i2.i IS NOT NULL) a1 WHERE a=b ORDER BY 1;
 SELECT * FROM (SELECT 0=1 AS cond FROM integers i1, integers i2) a1 WHERE cond ORDER BY 1;
 
 ++
diff --git a/tests/cases/distributed/tql-explain-analyze/analyze.result b/tests/cases/distributed/tql-explain-analyze/analyze.result
index 4087943ce8..1cd0c5c988 100644
--- a/tests/cases/distributed/tql-explain-analyze/analyze.result
+++ b/tests/cases/distributed/tql-explain-analyze/analyze.result
@@ -2,10 +2,12 @@ CREATE TABLE test(i DOUBLE, j TIMESTAMP TIME INDEX, k STRING PRIMARY KEY);
 
 Affected Rows: 0
 
+-- insert two points at 1ms and one point at 2ms
 INSERT INTO test VALUES (1, 1, "a"), (1, 1, "b"), (2, 2, "a");
 
 Affected Rows: 3
 
+-- analyze at 0s, 5s and 10s. No point at 0s.
 -- SQLNESS REPLACE (metrics.*) REDACTED
 -- SQLNESS REPLACE (RoundRobinBatch.*) REDACTED
 -- SQLNESS REPLACE (-+) -
diff --git a/tests/cases/distributed/tql-explain-analyze/explain.result b/tests/cases/distributed/tql-explain-analyze/explain.result
index cac729473b..2be8b54bfa 100644
--- a/tests/cases/distributed/tql-explain-analyze/explain.result
+++ b/tests/cases/distributed/tql-explain-analyze/explain.result
@@ -2,10 +2,12 @@ CREATE TABLE test(i DOUBLE, j TIMESTAMP TIME INDEX, k STRING PRIMARY KEY);
 
 Affected Rows: 0
 
+-- insert two points at 1ms and one point at 2ms
 INSERT INTO test VALUES (1, 1, "a"), (1, 1, "b"), (2, 2, "a");
 
 Affected Rows: 3
 
+-- explain at 0s, 5s and 10s. No point at 0s.
 -- SQLNESS REPLACE (RoundRobinBatch.*) REDACTED
 -- SQLNESS REPLACE (peer-.*) REDACTED
 TQL EXPLAIN (0, 10, '5s') test;
diff --git a/tests/cases/standalone/common/aggregate/distinct_order_by.result b/tests/cases/standalone/common/aggregate/distinct_order_by.result
index bacfd3badb..81649b776a 100644
--- a/tests/cases/standalone/common/aggregate/distinct_order_by.result
+++ b/tests/cases/standalone/common/aggregate/distinct_order_by.result
@@ -15,6 +15,14 @@ SELECT DISTINCT i%2 FROM integers ORDER BY 1;
 | 1                     |
 +-----------------------+
 
+-- TODO(LFC): Failed to run under new DataFusion
+-- expected:
+--  +-----------------------+
+--  | integers.i % Int64(2) |
+--  +-----------------------+
+--  | 1                     |
+--  | 0                     |
+--  +-----------------------+
 SELECT DISTINCT i % 2 FROM integers WHERE i<3 ORDER BY i;
 
 Error: 3000(PlanQuery), Error during planning: For SELECT DISTINCT, ORDER BY expressions i must appear in select list
diff --git a/tests/cases/standalone/common/insert/insert.result b/tests/cases/standalone/common/insert/insert.result
index 0de3fcda45..45f6a9137b 100644
--- a/tests/cases/standalone/common/insert/insert.result
+++ b/tests/cases/standalone/common/insert/insert.result
@@ -21,6 +21,7 @@ SELECT * FROM integers;
 | 1970-01-01T00:00:00.005 |
 +-------------------------+
 
+-- Test insert with long string constant
 CREATE TABLE IF NOT EXISTS presentations (
     presentation_date TIMESTAMP,
     author VARCHAR NOT NULL,
diff --git a/tests/cases/standalone/common/order/nulls_first.result b/tests/cases/standalone/common/order/nulls_first.result
index 8e0d8a733f..1bab6062f9 100644
--- a/tests/cases/standalone/common/order/nulls_first.result
+++ b/tests/cases/standalone/common/order/nulls_first.result
@@ -36,6 +36,10 @@ SELECT * FROM test ORDER BY i NULLS LAST, j NULLS FIRST;
 |   | 1 | 2 |
 +---+---+---+
 
+-- TODO(ruihang): The following two SQL will fail under distributed mode with error
+-- Error: 1003(Internal), status: Internal, message: "Failed to collect recordbatch, source: Failed to poll stream, source: Arrow error: Invalid argument error: batches[0] schema is different with argument schema.\n            batches[0] schema: Schema { fields: [Field { name: \"i\", data_type: Int32, nullable: true, dict_id: 0, dict_is_ordered: false, metadata: {} }, Field { name: \"j\", data_type: Int32, nullable: true, dict_id: 0, dict_is_ordered: false, metadata: {} }, Field { name: \"t\", data_type: Int64, nullable: false, dict_id: 0, dict_is_ordered: false, metadata: {\"greptime:time_index\": \"true\"} }], metadata: {\"greptime:version\": \"0\"} },\n            argument schema: Schema { fields: [Field { name: \"i\", data_type: Int32, nullable: true, dict_id: 0, dict_is_ordered: false, metadata: {} }, Field { name: \"j\", data_type: Int32, nullable: true, dict_id: 0, dict_is_ordered: false, metadata: {} }, Field { name: \"t\", data_type: Int64, nullable: false, dict_id: 0, dict_is_ordered: false, metadata: {\"greptime:time_index\": \"true\"} }], metadata: {} }\n            ", details: [], metadata: MetadataMap { headers: {"inner_error_code": "Internal"} }
+-- SELECT i, j, row_number() OVER (PARTITION BY i ORDER BY j NULLS FIRST) FROM test ORDER BY i NULLS FIRST, j NULLS FIRST;
+-- SELECT i, j, row_number() OVER (PARTITION BY i ORDER BY j NULLS LAST) FROM test ORDER BY i NULLS FIRST, j NULLS FIRST;
 SELECT * FROM test ORDER BY i NULLS FIRST, j NULLS LAST LIMIT 2;
 
 +---+---+---+
diff --git a/tests/cases/standalone/common/order/order_by.result b/tests/cases/standalone/common/order/order_by.result
index 640c593ac5..18210bfc53 100644
--- a/tests/cases/standalone/common/order/order_by.result
+++ b/tests/cases/standalone/common/order/order_by.result
@@ -192,10 +192,13 @@ SELECT a-10 AS k FROM test UNION SELECT a-10 AS l FROM test ORDER BY k;
 | 3 |
 +---+
 
+-- ORDER BY on alias in right-most query
+-- CONTROVERSIAL: SQLite allows both "k" and "l" to be referenced here, Postgres and MonetDB give an error.
 SELECT a-10 AS k FROM test UNION SELECT a-10 AS l FROM test ORDER BY l;
 
 Error: 3000(PlanQuery), No field named l. Valid fields are k.
 
+-- Not compatible with duckdb, work in gretimedb
 SELECT a-10 AS k FROM test UNION SELECT a-10 AS l FROM test ORDER BY 1-k;
 
 +---+
@@ -206,10 +209,18 @@ SELECT a-10 AS k FROM test UNION SELECT a-10 AS l FROM test ORDER BY 1-k;
 | 1 |
 +---+
 
+-- Not compatible with duckdb, give an error in greptimedb
+-- TODO(LFC): Failed to meet the expected error:
+-- expected:
+--   Error: 3000(PlanQuery), Schema error: No field named 'a'. Valid fields are 'k'.
 SELECT a-10 AS k FROM test UNION SELECT a-10 AS l FROM test ORDER BY a-10;
 
 Error: 3000(PlanQuery), Error during planning: For SELECT DISTINCT, ORDER BY expressions a must appear in select list
 
+-- Not compatible with duckdb, give an error in greptimedb
+-- TODO(LFC): Failed to meet the expected error:
+-- expected:
+--   Error: 3000(PlanQuery), Schema error: No field named 'a'. Valid fields are 'k'.
 SELECT a-10 AS k FROM test UNION SELECT a-11 AS l FROM test ORDER BY a-11;
 
 Error: 3000(PlanQuery), Error during planning: For SELECT DISTINCT, ORDER BY expressions a must appear in select list
diff --git a/tests/cases/standalone/common/order/order_by_exceptions.result b/tests/cases/standalone/common/order/order_by_exceptions.result
index f14bb99c01..d75f2c0437 100644
--- a/tests/cases/standalone/common/order/order_by_exceptions.result
+++ b/tests/cases/standalone/common/order/order_by_exceptions.result
@@ -10,10 +10,12 @@ SELECT a FROM test ORDER BY 2;
 
 Error: 3000(PlanQuery), Error during planning: Order by column out of bounds, specified: 2, max: 1
 
+-- Not work in greptimedb
 SELECT a FROM test ORDER BY 'hello', a;
 
 Error: 1003(Internal), Error during planning: Sort operation is not applicable to scalar value hello
 
+-- Ambiguous reference in union alias, give and error in duckdb, but works in greptimedb
 SELECT a AS k, b FROM test UNION SELECT a, b AS k FROM test ORDER BY k;
 
 +----+----+
@@ -38,6 +40,10 @@ SELECT a % 2, b FROM test UNION SELECT b, a % 2 AS k ORDER BY a % 2;
 
 Error: 3000(PlanQuery), No field named b.
 
+-- Works duckdb, but not work in greptimedb
+-- TODO(LFC): Failed to meet the expected error:
+-- expected:
+--   Error: 3000(PlanQuery), Schema error: No field named 'a'. Valid fields are 'test.a % Int64(2)', 'b'.
 SELECT a % 2, b FROM test UNION SELECT a % 2 AS k, b FROM test ORDER BY a % 2;
 
 Error: 3000(PlanQuery), Error during planning: For SELECT DISTINCT, ORDER BY expressions a must appear in select list
diff --git a/tests/cases/standalone/common/tql/aggr_over_time.result b/tests/cases/standalone/common/tql/aggr_over_time.result
index 1608262bff..5b73fc5445 100644
--- a/tests/cases/standalone/common/tql/aggr_over_time.result
+++ b/tests/cases/standalone/common/tql/aggr_over_time.result
@@ -1,3 +1,7 @@
+-- Port from functions.test L607 - L630, commit 001ee2620e094970e5657ce39275b2fccdbd1359
+-- Include stddev/stdvar over time
+-- load 10s
+--   metric 0 8 8 2 3
 create table metric (ts timestamp(3) time index, val double);
 
 Affected Rows: 0
@@ -23,6 +27,8 @@ select * from metric;
 | 1970-01-01T00:00:40 | 3.0 |
 +---------------------+-----+
 
+-- eval instant at 1m stdvar_over_time(metric[1m])
+--   {} 10.56
 tql eval (60, 61, '10s') stdvar_over_time(metric[1m]);
 
 +---------------------+-------------------------------------+
@@ -31,6 +37,8 @@ tql eval (60, 61, '10s') stdvar_over_time(metric[1m]);
 | 1970-01-01T00:01:00 | 10.559999999999999                  |
 +---------------------+-------------------------------------+
 
+-- eval instant at 1m stddev_over_time(metric[1m])
+--   {} 3.249615
 tql eval (60, 60, '1s') stddev_over_time(metric[1m]);
 
 +---------------------+-------------------------------------+
@@ -39,6 +47,8 @@ tql eval (60, 60, '1s') stddev_over_time(metric[1m]);
 | 1970-01-01T00:01:00 | 3.249615361854384                   |
 +---------------------+-------------------------------------+
 
+-- eval instant at 1m stddev_over_time((metric[1m]))
+--   {} 3.249615
 tql eval (60, 60, '1s') stddev_over_time((metric[1m]));
 
 +---------------------+-------------------------------------+
@@ -51,6 +61,8 @@ drop table metric;
 
 Affected Rows: 1
 
+-- load 10s
+--   metric 1.5990505637277868 1.5990505637277868 1.5990505637277868
 create table metric (ts timestamp(3) time index, val double);
 
 Affected Rows: 0
@@ -63,6 +75,8 @@ insert into metric values
 
 Affected Rows: 4
 
+-- eval instant at 1m stdvar_over_time(metric[1m])
+--   {} 0
 tql eval (60, 60, '1s') stdvar_over_time(metric[1m]);
 
 +---------------------+-------------------------------------+
@@ -71,6 +85,8 @@ tql eval (60, 60, '1s') stdvar_over_time(metric[1m]);
 | 1970-01-01T00:01:00 | 0.47943050725465364                 |
 +---------------------+-------------------------------------+
 
+-- eval instant at 1m stddev_over_time(metric[1m])
+--   {} 0
 tql eval (60, 60, '1s') stddev_over_time(metric[1m]);
 
 +---------------------+-------------------------------------+
@@ -83,6 +99,12 @@ drop table metric;
 
 Affected Rows: 1
 
+-- Port from functions.test L632 - L680, commit 001ee2620e094970e5657ce39275b2fccdbd1359
+-- Include quantile over time
+-- load 10s
+-- 	data{test="two samples"} 0 1
+-- 	data{test="three samples"} 0 1 2
+-- 	data{test="uneven samples"} 0 1 4
 create table data (ts timestamp(3) time index, val double, test string primary key);
 
 Affected Rows: 0
@@ -99,10 +121,58 @@ insert into data values
 
 Affected Rows: 8
 
+-- eval instant at 1m quantile_over_time(0, data[1m])
+-- 	{test="two samples"} 0
+-- 	{test="three samples"} 0
+-- 	{test="uneven samples"} 0
+-- tql eval (60, 60, '1s') quantile_over_time(0, data[1m]);
+-- eval instant at 1m quantile_over_time(0.5, data[1m])
+-- 	{test="two samples"} 0.5
+-- 	{test="three samples"} 1
+-- 	{test="uneven samples"} 1
+-- tql eval (60, 60, '1s') quantile_over_time(0.5, data[1m]);
+-- eval instant at 1m quantile_over_time(0.75, data[1m])
+-- 	{test="two samples"} 0.75
+-- 	{test="three samples"} 1.5
+-- 	{test="uneven samples"} 2.5
+-- tql eval (60, 60, '1s') quantile_over_time(0.75, data[1m]);
+-- eval instant at 1m quantile_over_time(0.8, data[1m])
+-- 	{test="two samples"} 0.8
+-- 	{test="three samples"} 1.6
+-- 	{test="uneven samples"} 2.8
+-- tql eval (60, 60, '1s') quantile_over_time(0.8, data[1m]);
+-- eval instant at 1m quantile_over_time(1, data[1m])
+-- 	{test="two samples"} 1
+-- 	{test="three samples"} 2
+-- 	{test="uneven samples"} 4
+-- tql eval (60, 60, '1s') quantile_over_time(1, data[1m]);
+-- eval instant at 1m quantile_over_time(-1, data[1m])
+-- 	{test="two samples"} -Inf
+-- 	{test="three samples"} -Inf
+-- 	{test="uneven samples"} -Inf
+-- tql eval (60, 60, '1s') quantile_over_time(-1, data[1m]);
+-- eval instant at 1m quantile_over_time(2, data[1m])
+-- 	{test="two samples"} +Inf
+-- 	{test="three samples"} +Inf
+-- 	{test="uneven samples"} +Inf
+-- tql eval (60, 60, '1s') quantile_over_time(2, data[1m]);
+-- eval instant at 1m (quantile_over_time(2, (data[1m])))
+-- 	{test="two samples"} +Inf
+-- 	{test="three samples"} +Inf
+-- 	{test="uneven samples"} +Inf
+-- tql eval (60, 60, '1s') (quantile_over_time(2, (data[1m])));
 drop table data;
 
 Affected Rows: 1
 
+-- Port from functions.test L773 - L802, commit 001ee2620e094970e5657ce39275b2fccdbd1359
+-- Include max/min/last over time
+-- load 10s
+-- 	data{type="numbers"} 2 0 3
+-- 	data{type="some_nan"} 2 0 NaN
+-- 	data{type="some_nan2"} 2 NaN 1
+-- 	data{type="some_nan3"} NaN 0 1
+-- 	data{type="only_nan"} NaN NaN NaN
 create table data (ts timestamp(3) time index, val double, ty string primary key);
 
 Affected Rows: 0
@@ -126,6 +196,27 @@ insert into data values
 
 Affected Rows: 15
 
+-- eval instant at 1m min_over_time(data[1m])
+-- 	{type="numbers"} 0
+-- 	{type="some_nan"} 0
+-- 	{type="some_nan2"} 1
+-- 	{type="some_nan3"} 0
+-- 	{type="only_nan"} NaN
+-- tql eval (60, 60, '1s') min_over_time(data[1m]);
+-- eval instant at 1m max_over_time(data[1m])
+-- 	{type="numbers"} 3
+-- 	{type="some_nan"} 2
+-- 	{type="some_nan2"} 2
+-- 	{type="some_nan3"} 1
+-- 	{type="only_nan"} NaN
+-- tql eval (60, 60, '1s') max_over_time(data[1m]);
+-- eval instant at 1m last_over_time(data[1m])
+-- 	data{type="numbers"} 3
+-- 	data{type="some_nan"} NaN
+-- 	data{type="some_nan2"} 1
+-- 	data{type="some_nan3"} 1
+-- 	data{type="only_nan"} NaN
+-- tql eval (60, 60, '1s') last_over_time(data[1m]);
 drop table data;
 
 Affected Rows: 1
diff --git a/tests/cases/standalone/common/tql/basic.result b/tests/cases/standalone/common/tql/basic.result
index 09ce38cba5..46f600f9ce 100644
--- a/tests/cases/standalone/common/tql/basic.result
+++ b/tests/cases/standalone/common/tql/basic.result
@@ -2,11 +2,13 @@ CREATE TABLE test(i DOUBLE, j TIMESTAMP TIME INDEX, k STRING PRIMARY KEY);
 
 Affected Rows: 0
 
+-- insert two points at 1ms and one point at 2ms
 INSERT INTO test VALUES (1, 1, "a"), (1, 1, "b"), (2, 2, "a");
 
 Affected Rows: 3
 
 -- SQLNESS SORT_RESULT 2 1
+-- evaluate at 0s, 5s and 10s. No point at 0s.
 TQL EVAL (0, 10, '5s') test;
 
 +-----+---------------------+---+
@@ -18,6 +20,7 @@ TQL EVAL (0, 10, '5s') test;
 | 2.0 | 1970-01-01T00:00:10 | a |
 +-----+---------------------+---+
 
+-- the point at 1ms will be shadowed by the point at 2ms
 TQL EVAL (0, 10, '5s') test{k="a"};
 
 +-----+---------------------+---+
diff --git a/tests/cases/standalone/common/tql/operator.result b/tests/cases/standalone/common/tql/operator.result
index 360d2bc48a..1ad0461978 100644
--- a/tests/cases/standalone/common/tql/operator.result
+++ b/tests/cases/standalone/common/tql/operator.result
@@ -1,3 +1,9 @@
+-- Port from operators.test L607 - L630, commit 001ee2620e094970e5657ce39275b2fccdbd1359
+-- Include atan2
+-- load 5m
+--     trigy{} 10
+--     trigx{} 20
+--     trigNaN{} NaN
 create table trigy (ts timestamp(3) time index, val double);
 
 Affected Rows: 0
@@ -22,6 +28,8 @@ insert into trignan values (0, 'NaN'::double);
 
 Affected Rows: 1
 
+-- eval instant at 5m trigy atan2 trigx
+--     trigy{} 0.4636476090008061
 tql eval (300, 300, '1s') trigy atan2 trigx;
 
 +---------------------+----------------------------+
@@ -30,11 +38,16 @@ tql eval (300, 300, '1s') trigy atan2 trigx;
 | 1970-01-01T00:05:00 | 0.4636476090008061         |
 +---------------------+----------------------------+
 
+-- eval instant at 5m trigy atan2 trigNaN
+--     trigy{} NaN
+-- This query doesn't have result because `trignan` is NaN and will be filtered out.
 tql eval (300, 300, '1s') trigy atan2 trignan;
 
 ++
 ++
 
+-- eval instant at 5m 10 atan2 20
+--     0.4636476090008061
 tql eval (300, 300, '1s') 10 atan2 20;
 
 +---------------------+--------------------+
@@ -43,6 +56,8 @@ tql eval (300, 300, '1s') 10 atan2 20;
 | 1970-01-01T00:05:00 | 0.4636476090008061 |
 +---------------------+--------------------+
 
+-- eval instant at 5m 10 atan2 NaN
+--     NaN
 tql eval (300, 300, '1s') 10 atan2 NaN;
 
 +---------------------+-------+
diff --git a/tests/cases/standalone/common/types/blob.result b/tests/cases/standalone/common/types/blob.result
index a35e89bc11..1673d99fe2 100644
--- a/tests/cases/standalone/common/types/blob.result
+++ b/tests/cases/standalone/common/types/blob.result
@@ -2,6 +2,7 @@ CREATE TABLE blobs (b BYTEA, t timestamp time index);
 
 Affected Rows: 0
 
+--Insert valid hex strings--
 INSERT INTO blobs VALUES('\xaa\xff\xaa'::BYTEA, 1), ('\xAA\xFF\xAA\xAA\xFF\xAA'::BYTEA, 2), ('\xAA\xFF\xAA\xAA\xFF\xAA\xAA\xFF\xAA'::BYTEA, 3);
 
 Affected Rows: 3
@@ -16,6 +17,7 @@ SELECT * FROM blobs;
 | 5c7841415c7846465c7841415c7841415c7846465c7841415c7841415c7846465c784141 | 1970-01-01T00:00:00.003 |
 +--------------------------------------------------------------------------+-------------------------+
 
+--Insert valid hex strings, lower case--
 DELETE FROM blobs;
 
 Affected Rows: 3
@@ -34,6 +36,7 @@ SELECT * FROM blobs;
 | 5c7861615c7866665c7861615c7861615c7866665c7861615c7861615c7866665c786161 | 1970-01-01T00:00:00.003 |
 +--------------------------------------------------------------------------+-------------------------+
 
+--Insert valid hex strings with number and letters--
 DELETE FROM blobs;
 
 Affected Rows: 3
@@ -52,10 +55,12 @@ SELECT * FROM blobs;
 | 5c78616131313939616131313939616131313939 | 1970-01-01T00:00:00.003 |
 +------------------------------------------+-------------------------+
 
+--Insert invalid hex strings (invalid hex chars: G, H, I)--
 INSERT INTO blobs VALUES('\xGA\xFF\xAA'::BYTEA, 4);
 
 Affected Rows: 1
 
+--Insert invalid hex strings (odd # of chars)--
 INSERT INTO blobs VALUES('\xA'::BYTEA, 4);
 
 Affected Rows: 1
diff --git a/tests/cases/standalone/cte/cte.result b/tests/cases/standalone/cte/cte.result
index 5dedec787e..562883a9ef 100644
--- a/tests/cases/standalone/cte/cte.result
+++ b/tests/cases/standalone/cte/cte.result
@@ -59,6 +59,7 @@ with cte1 as (select 42), cte1 as (select 42) select * FROM cte1;
 
 Error: 3000(PlanQuery), sql parser error: WITH query name "cte1" specified more than once
 
+-- reference to CTE before its actually defined, it's not supported by datafusion
 with cte3 as (select ref2.j as i from cte1 as ref2), cte1 as (Select i as j from a), cte2 as (select ref.j+1 as k from cte1 as ref) select * from cte2 union all select * FROM cte3;
 
 Error: 3000(PlanQuery), Error during planning: Table not found: greptime.public.cte1
@@ -96,6 +97,7 @@ SELECT 1 UNION ALL (WITH cte AS (SELECT 42) SELECT * FROM cte) order by 1;
 | 42       |
 +----------+
 
+-- Recursive CTEs are not supported in datafusion
 WITH RECURSIVE cte(d) AS (
 		SELECT 1
 	UNION ALL
@@ -109,6 +111,7 @@ SELECT max(d) FROM cte;
 
 Error: 3000(PlanQuery), This feature is not implemented: Recursive CTEs are not supported
 
+-- Nested aliases is not supported in datafusion
 with cte (a) as (
     select 1
 )
diff --git a/tests/cases/standalone/cte/cte_in_cte.result b/tests/cases/standalone/cte/cte_in_cte.result
index d9b18ebe2e..b20ba4b40c 100644
--- a/tests/cases/standalone/cte/cte_in_cte.result
+++ b/tests/cases/standalone/cte/cte_in_cte.result
@@ -50,14 +50,19 @@ with cte1 as (Select i as j from a) select * from (with cte2 as (select max(j) a
 | 42 |
 +----+
 
+-- Refer to CTE in subquery expression,
+-- this feature is not implemented in datafusion
 with cte1 as (Select i as j from a) select * from cte1 where j = (with cte2 as (select max(j) as j from cte1) select j from cte2);
 
 Error: 3001(EngineExecuteQuery), This feature is not implemented: Physical plan does not support logical expression (<subquery>)
 
+-- Refer to same-named CTE in a subquery expression
+-- this feature is not implemented in datafusion
 with cte as (Select i as j from a) select * from cte where j = (with cte as (select max(j) as j from cte) select j from cte);
 
 Error: 3000(PlanQuery), sql parser error: WITH query name "cte" specified more than once
 
+-- self-refer to non-existent cte-
 with cte as (select * from cte) select * from cte;
 
 Error: 3000(PlanQuery), Error during planning: Table not found: greptime.public.cte
diff --git a/tests/cases/standalone/optimizer/filter_push_down.result b/tests/cases/standalone/optimizer/filter_push_down.result
index e48471107f..85ded032de 100644
--- a/tests/cases/standalone/optimizer/filter_push_down.result
+++ b/tests/cases/standalone/optimizer/filter_push_down.result
@@ -187,6 +187,16 @@ SELECT i FROM (SELECT * FROM integers i1 UNION SELECT * FROM integers i2) a WHER
 | 3 |
 +---+
 
+-- TODO(LFC): Somehow the following SQL does not order by column 1 under new DataFusion occasionally. Should further investigate it. Comment it out temporarily.
+-- expected:
+--  +---+---+--------------+
+--  | a | b | ROW_NUMBER() |
+--  +---+---+--------------+
+--  | 1 | 1 | 1            |
+--  | 2 | 2 | 5            |
+--  | 3 | 3 | 9            |
+--  +---+---+--------------+
+-- SELECT * FROM (SELECT i1.i AS a, i2.i AS b, row_number() OVER (ORDER BY i1.i, i2.i) FROM integers i1, integers i2 WHERE i1.i IS NOT NULL AND i2.i IS NOT NULL) a1 WHERE a=b ORDER BY 1;
 SELECT * FROM (SELECT 0=1 AS cond FROM integers i1, integers i2) a1 WHERE cond ORDER BY 1;
 
 ++
diff --git a/tests/cases/standalone/tql-explain-analyze/analyze.result b/tests/cases/standalone/tql-explain-analyze/analyze.result
index d8e767d84b..db3dcb265d 100644
--- a/tests/cases/standalone/tql-explain-analyze/analyze.result
+++ b/tests/cases/standalone/tql-explain-analyze/analyze.result
@@ -2,10 +2,12 @@ CREATE TABLE test(i DOUBLE, j TIMESTAMP TIME INDEX, k STRING PRIMARY KEY);
 
 Affected Rows: 0
 
+-- insert two points at 1ms and one point at 2ms
 INSERT INTO test VALUES (1, 1, "a"), (1, 1, "b"), (2, 2, "a");
 
 Affected Rows: 3
 
+-- analyze at 0s, 5s and 10s. No point at 0s.
 -- SQLNESS REPLACE (metrics.*) REDACTED
 -- SQLNESS REPLACE (RoundRobinBatch.*) REDACTED
 -- SQLNESS REPLACE (-+) -
diff --git a/tests/cases/standalone/tql-explain-analyze/explain.result b/tests/cases/standalone/tql-explain-analyze/explain.result
index 8a1c23626b..a9f501ff78 100644
--- a/tests/cases/standalone/tql-explain-analyze/explain.result
+++ b/tests/cases/standalone/tql-explain-analyze/explain.result
@@ -2,10 +2,12 @@ CREATE TABLE test(i DOUBLE, j TIMESTAMP TIME INDEX, k STRING PRIMARY KEY);
 
 Affected Rows: 0
 
+-- insert two points at 1ms and one point at 2ms
 INSERT INTO test VALUES (1, 1, "a"), (1, 1, "b"), (2, 2, "a");
 
 Affected Rows: 3
 
+-- explain at 0s, 5s and 10s. No point at 0s.
 -- SQLNESS REPLACE (RoundRobinBatch.*) REDACTED
 TQL EXPLAIN (0, 10, '5s') test;
 
diff --git a/tests/runner/Cargo.toml b/tests/runner/Cargo.toml
index f9e0bdd032..d14566f6f0 100644
--- a/tests/runner/Cargo.toml
+++ b/tests/runner/Cargo.toml
@@ -13,6 +13,6 @@ common-grpc = { path = "../../src/common/grpc" }
 common-query = { path = "../../src/common/query" }
 common-time = { path = "../../src/common/time" }
 serde.workspace = true
-sqlness = { git = "https://github.com/CeresDB/sqlness.git", rev = "a4663365795d2067eb53966c383e1bb0c89c7627" }
+sqlness = { version = "0.5" }
 tinytemplate = "1.2"
 tokio.workspace = true