Skip to content
New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

planner/core: support partition pruning for partition expression floor(unix_timestamp()) (#16402) #16521

Merged
merged 4 commits into from
Apr 20, 2020
Merged
Show file tree
Hide file tree
Changes from all commits
Commits
File filter

Filter by extension

Filter by extension

Conversations
Failed to load comments.
Loading
Jump to
Jump to file
Failed to load files.
Loading
Diff view
Diff view
174 changes: 174 additions & 0 deletions cmd/explaintest/r/partition_pruning.result
Original file line number Diff line number Diff line change
Expand Up @@ -4218,3 +4218,177 @@ id count task operator info
TableReader_8 10.00 root data:Selection_7
└─Selection_7 10.00 cop eq(test.t.name, "1")
└─TableScan_6 10000.00 cop table:t, partition:p1, range:[-inf,+inf], keep order:false, stats:pseudo
drop table if exists t;
create table t (ts timestamp(3) not null default current_timestamp(3))
partition by range (floor(unix_timestamp(ts))) (
partition p0 values less than (unix_timestamp('2020-04-05 00:00:00')),
partition p1 values less than (unix_timestamp('2020-04-15 00:00:00')),
partition p2 values less than (unix_timestamp('2020-04-25 00:00:00'))
);
explain select * from t where ts = '2020-04-06 00:00:00' -- p1;
id count task operator info
TableReader_8 10.00 root data:Selection_7
└─Selection_7 10.00 cop eq(test.t.ts, 2020-04-06 00:00:00.000000)
└─TableScan_6 10000.00 cop table:t, partition:p1, range:[-inf,+inf], keep order:false, stats:pseudo
explain select * from t where ts = '2020-04-05 00:00:00.001' -- p1;
id count task operator info
TableReader_8 10.00 root data:Selection_7
└─Selection_7 10.00 cop eq(test.t.ts, 2020-04-05 00:00:00.001000)
└─TableScan_6 10000.00 cop table:t, partition:p1, range:[-inf,+inf], keep order:false, stats:pseudo
explain select * from t where ts > '2020-04-15 00:00:00' -- p2;
id count task operator info
TableReader_8 3333.33 root data:Selection_7
└─Selection_7 3333.33 cop gt(test.t.ts, 2020-04-15 00:00:00.000000)
└─TableScan_6 10000.00 cop table:t, partition:p2, range:[-inf,+inf], keep order:false, stats:pseudo
explain select * from t where ts > '2020-04-14 23:59:59.999' -- p1,p2;
id count task operator info
Union_8 6666.67 root
├─TableReader_11 3333.33 root data:Selection_10
│ └─Selection_10 3333.33 cop gt(test.t.ts, 2020-04-14 23:59:59.999000)
│ └─TableScan_9 10000.00 cop table:t, partition:p1, range:[-inf,+inf], keep order:false, stats:pseudo
└─TableReader_14 3333.33 root data:Selection_13
└─Selection_13 3333.33 cop gt(test.t.ts, 2020-04-14 23:59:59.999000)
└─TableScan_12 10000.00 cop table:t, partition:p2, range:[-inf,+inf], keep order:false, stats:pseudo
explain select * from t where ts > '2020-04-15 00:00:00.001' -- p2;
id count task operator info
TableReader_8 3333.33 root data:Selection_7
└─Selection_7 3333.33 cop gt(test.t.ts, 2020-04-15 00:00:00.001000)
└─TableScan_6 10000.00 cop table:t, partition:p2, range:[-inf,+inf], keep order:false, stats:pseudo
explain select * from t where ts > '2020-04-26 00:00:00.001' -- dual;
id count task operator info
TableDual_6 0.00 root rows:0
explain select * from t where ts >= '2020-04-04 12:22:32' -- p0,p1,p2;
id count task operator info
Union_9 10000.00 root
├─TableReader_12 3333.33 root data:Selection_11
│ └─Selection_11 3333.33 cop ge(test.t.ts, 2020-04-04 12:22:32.000000)
│ └─TableScan_10 10000.00 cop table:t, partition:p0, range:[-inf,+inf], keep order:false, stats:pseudo
├─TableReader_15 3333.33 root data:Selection_14
│ └─Selection_14 3333.33 cop ge(test.t.ts, 2020-04-04 12:22:32.000000)
│ └─TableScan_13 10000.00 cop table:t, partition:p1, range:[-inf,+inf], keep order:false, stats:pseudo
└─TableReader_18 3333.33 root data:Selection_17
└─Selection_17 3333.33 cop ge(test.t.ts, 2020-04-04 12:22:32.000000)
└─TableScan_16 10000.00 cop table:t, partition:p2, range:[-inf,+inf], keep order:false, stats:pseudo
explain select * from t where ts >= '2020-04-05 00:00:00' -- p1,p2;
id count task operator info
Union_8 6666.67 root
├─TableReader_11 3333.33 root data:Selection_10
│ └─Selection_10 3333.33 cop ge(test.t.ts, 2020-04-05 00:00:00.000000)
│ └─TableScan_9 10000.00 cop table:t, partition:p1, range:[-inf,+inf], keep order:false, stats:pseudo
└─TableReader_14 3333.33 root data:Selection_13
└─Selection_13 3333.33 cop ge(test.t.ts, 2020-04-05 00:00:00.000000)
└─TableScan_12 10000.00 cop table:t, partition:p2, range:[-inf,+inf], keep order:false, stats:pseudo
explain select * from t where ts >= '2020-04-25 00:00:00' -- dual;
id count task operator info
TableDual_6 0.00 root rows:0
explain select * from t where ts < '2020-04-25 00:00:00' -- p0,p1,p2;
id count task operator info
Union_9 9970.00 root
├─TableReader_12 3323.33 root data:Selection_11
│ └─Selection_11 3323.33 cop lt(test.t.ts, 2020-04-25 00:00:00.000000)
│ └─TableScan_10 10000.00 cop table:t, partition:p0, range:[-inf,+inf], keep order:false, stats:pseudo
├─TableReader_15 3323.33 root data:Selection_14
│ └─Selection_14 3323.33 cop lt(test.t.ts, 2020-04-25 00:00:00.000000)
│ └─TableScan_13 10000.00 cop table:t, partition:p1, range:[-inf,+inf], keep order:false, stats:pseudo
└─TableReader_18 3323.33 root data:Selection_17
└─Selection_17 3323.33 cop lt(test.t.ts, 2020-04-25 00:00:00.000000)
└─TableScan_16 10000.00 cop table:t, partition:p2, range:[-inf,+inf], keep order:false, stats:pseudo
explain select * from t where ts < '2020-04-15 00:00:00.001' -- p0,p1,p2;
id count task operator info
Union_9 9970.00 root
├─TableReader_12 3323.33 root data:Selection_11
│ └─Selection_11 3323.33 cop lt(test.t.ts, 2020-04-15 00:00:00.001000)
│ └─TableScan_10 10000.00 cop table:t, partition:p0, range:[-inf,+inf], keep order:false, stats:pseudo
├─TableReader_15 3323.33 root data:Selection_14
│ └─Selection_14 3323.33 cop lt(test.t.ts, 2020-04-15 00:00:00.001000)
│ └─TableScan_13 10000.00 cop table:t, partition:p1, range:[-inf,+inf], keep order:false, stats:pseudo
└─TableReader_18 3323.33 root data:Selection_17
└─Selection_17 3323.33 cop lt(test.t.ts, 2020-04-15 00:00:00.001000)
└─TableScan_16 10000.00 cop table:t, partition:p2, range:[-inf,+inf], keep order:false, stats:pseudo
explain select * from t where ts < '2020-04-15 00:00:00' -- expect perfect : p0,p1, obtain: p0,p1,p2;
id count task operator info
Union_9 9970.00 root
├─TableReader_12 3323.33 root data:Selection_11
│ └─Selection_11 3323.33 cop lt(test.t.ts, 2020-04-15 00:00:00.000000)
│ └─TableScan_10 10000.00 cop table:t, partition:p0, range:[-inf,+inf], keep order:false, stats:pseudo
├─TableReader_15 3323.33 root data:Selection_14
│ └─Selection_14 3323.33 cop lt(test.t.ts, 2020-04-15 00:00:00.000000)
│ └─TableScan_13 10000.00 cop table:t, partition:p1, range:[-inf,+inf], keep order:false, stats:pseudo
└─TableReader_18 3323.33 root data:Selection_17
└─Selection_17 3323.33 cop lt(test.t.ts, 2020-04-15 00:00:00.000000)
└─TableScan_16 10000.00 cop table:t, partition:p2, range:[-inf,+inf], keep order:false, stats:pseudo
explain select * from t where ts < '2020-04-14 23:59:59.999' -- p0,p1;
id count task operator info
Union_8 6646.67 root
├─TableReader_11 3323.33 root data:Selection_10
│ └─Selection_10 3323.33 cop lt(test.t.ts, 2020-04-14 23:59:59.999000)
│ └─TableScan_9 10000.00 cop table:t, partition:p0, range:[-inf,+inf], keep order:false, stats:pseudo
└─TableReader_14 3323.33 root data:Selection_13
└─Selection_13 3323.33 cop lt(test.t.ts, 2020-04-14 23:59:59.999000)
└─TableScan_12 10000.00 cop table:t, partition:p1, range:[-inf,+inf], keep order:false, stats:pseudo
explain select * from t where ts < '2020-04-03 00:00:00' -- p0;
id count task operator info
TableReader_8 3323.33 root data:Selection_7
└─Selection_7 3323.33 cop lt(test.t.ts, 2020-04-03 00:00:00.000000)
└─TableScan_6 10000.00 cop table:t, partition:p0, range:[-inf,+inf], keep order:false, stats:pseudo
explain select * from t where ts < '2021-05-03 00:00:00' -- p0,p1,p2;
id count task operator info
Union_9 9970.00 root
├─TableReader_12 3323.33 root data:Selection_11
│ └─Selection_11 3323.33 cop lt(test.t.ts, 2021-05-03 00:00:00.000000)
│ └─TableScan_10 10000.00 cop table:t, partition:p0, range:[-inf,+inf], keep order:false, stats:pseudo
├─TableReader_15 3323.33 root data:Selection_14
│ └─Selection_14 3323.33 cop lt(test.t.ts, 2021-05-03 00:00:00.000000)
│ └─TableScan_13 10000.00 cop table:t, partition:p1, range:[-inf,+inf], keep order:false, stats:pseudo
└─TableReader_18 3323.33 root data:Selection_17
└─Selection_17 3323.33 cop lt(test.t.ts, 2021-05-03 00:00:00.000000)
└─TableScan_16 10000.00 cop table:t, partition:p2, range:[-inf,+inf], keep order:false, stats:pseudo
explain select * from t where ts <= '2020-04-05 00:00:00' -- p0,p1;
id count task operator info
Union_8 6646.67 root
├─TableReader_11 3323.33 root data:Selection_10
│ └─Selection_10 3323.33 cop le(test.t.ts, 2020-04-05 00:00:00.000000)
│ └─TableScan_9 10000.00 cop table:t, partition:p0, range:[-inf,+inf], keep order:false, stats:pseudo
└─TableReader_14 3323.33 root data:Selection_13
└─Selection_13 3323.33 cop le(test.t.ts, 2020-04-05 00:00:00.000000)
└─TableScan_12 10000.00 cop table:t, partition:p1, range:[-inf,+inf], keep order:false, stats:pseudo
explain select * from t where ts <= '2020-04-03 00:00:00' -- p0;
id count task operator info
TableReader_8 3323.33 root data:Selection_7
└─Selection_7 3323.33 cop le(test.t.ts, 2020-04-03 00:00:00.000000)
└─TableScan_6 10000.00 cop table:t, partition:p0, range:[-inf,+inf], keep order:false, stats:pseudo
explain select * from t where ts <= '2020-04-14 23:59:59.123' -- p0,p1;
id count task operator info
Union_8 6646.67 root
├─TableReader_11 3323.33 root data:Selection_10
│ └─Selection_10 3323.33 cop le(test.t.ts, 2020-04-14 23:59:59.123000)
│ └─TableScan_9 10000.00 cop table:t, partition:p0, range:[-inf,+inf], keep order:false, stats:pseudo
└─TableReader_14 3323.33 root data:Selection_13
└─Selection_13 3323.33 cop le(test.t.ts, 2020-04-14 23:59:59.123000)
└─TableScan_12 10000.00 cop table:t, partition:p1, range:[-inf,+inf], keep order:false, stats:pseudo
explain select * from t where ts <= '2020-04-25 00:00:00' -- p0,p1,p2;
id count task operator info
Union_9 9970.00 root
├─TableReader_12 3323.33 root data:Selection_11
│ └─Selection_11 3323.33 cop le(test.t.ts, 2020-04-25 00:00:00.000000)
│ └─TableScan_10 10000.00 cop table:t, partition:p0, range:[-inf,+inf], keep order:false, stats:pseudo
├─TableReader_15 3323.33 root data:Selection_14
│ └─Selection_14 3323.33 cop le(test.t.ts, 2020-04-25 00:00:00.000000)
│ └─TableScan_13 10000.00 cop table:t, partition:p1, range:[-inf,+inf], keep order:false, stats:pseudo
└─TableReader_18 3323.33 root data:Selection_17
└─Selection_17 3323.33 cop le(test.t.ts, 2020-04-25 00:00:00.000000)
└─TableScan_16 10000.00 cop table:t, partition:p2, range:[-inf,+inf], keep order:false, stats:pseudo
explain select * from t where ts > '2020-04-25 00:00:00' or ts < '2020-01-02 00:00:00' -- p0;
id count task operator info
TableReader_8 6656.67 root data:Selection_7
└─Selection_7 6656.67 cop or(gt(test.t.ts, 2020-04-25 00:00:00.000000), lt(test.t.ts, 2020-01-02 00:00:00.000000))
└─TableScan_6 10000.00 cop table:t, partition:p0, range:[-inf,+inf], keep order:false, stats:pseudo
explain select * from t where ts > '2020-04-02 00:00:00' and ts < '2020-04-07 00:00:00' -- p0,p1;
id count task operator info
Union_8 500.00 root
├─TableReader_11 250.00 root data:Selection_10
│ └─Selection_10 250.00 cop gt(test.t.ts, 2020-04-02 00:00:00.000000), lt(test.t.ts, 2020-04-07 00:00:00.000000)
│ └─TableScan_9 10000.00 cop table:t, partition:p0, range:[-inf,+inf], keep order:false, stats:pseudo
└─TableReader_14 250.00 root data:Selection_13
└─Selection_13 250.00 cop gt(test.t.ts, 2020-04-02 00:00:00.000000), lt(test.t.ts, 2020-04-07 00:00:00.000000)
└─TableScan_12 10000.00 cop table:t, partition:p1, range:[-inf,+inf], keep order:false, stats:pseudo
36 changes: 36 additions & 0 deletions cmd/explaintest/t/partition_pruning.test
Original file line number Diff line number Diff line change
Expand Up @@ -977,6 +977,42 @@ PARTITION BY RANGE COLUMNS(a) (

desc select * from t where a = 11 and b = 1 or a = 12 and b = 1;


drop table if exists t;
create table t (id int, name varchar(20)) partition by hash(id) partitions 128;
explain SELECT * FROM t partition (p1) where name = '1';

#
# MySQL doesn't support partition pruning for 'floor(unix_timestamp(ts))' but it works on TiDB
# https://github.com/pingcap/tidb/issues/16354
#
drop table if exists t;
create table t (ts timestamp(3) not null default current_timestamp(3))
partition by range (floor(unix_timestamp(ts))) (
partition p0 values less than (unix_timestamp('2020-04-05 00:00:00')),
partition p1 values less than (unix_timestamp('2020-04-15 00:00:00')),
partition p2 values less than (unix_timestamp('2020-04-25 00:00:00'))
);

explain select * from t where ts = '2020-04-06 00:00:00' -- p1;
explain select * from t where ts = '2020-04-05 00:00:00.001' -- p1;
explain select * from t where ts > '2020-04-15 00:00:00' -- p2;
explain select * from t where ts > '2020-04-14 23:59:59.999' -- p1,p2;
explain select * from t where ts > '2020-04-15 00:00:00.001' -- p2;
explain select * from t where ts > '2020-04-26 00:00:00.001' -- dual;
explain select * from t where ts >= '2020-04-04 12:22:32' -- p0,p1,p2;
explain select * from t where ts >= '2020-04-05 00:00:00' -- p1,p2;
explain select * from t where ts >= '2020-04-25 00:00:00' -- dual;
explain select * from t where ts < '2020-04-25 00:00:00' -- p0,p1,p2;
explain select * from t where ts < '2020-04-15 00:00:00.001' -- p0,p1,p2;
explain select * from t where ts < '2020-04-15 00:00:00' -- expect perfect : p0,p1, obtain: p0,p1,p2;
explain select * from t where ts < '2020-04-14 23:59:59.999' -- p0,p1;
explain select * from t where ts < '2020-04-03 00:00:00' -- p0;
explain select * from t where ts < '2021-05-03 00:00:00' -- p0,p1,p2;
explain select * from t where ts <= '2020-04-05 00:00:00' -- p0,p1;
explain select * from t where ts <= '2020-04-03 00:00:00' -- p0;
explain select * from t where ts <= '2020-04-14 23:59:59.123' -- p0,p1;
explain select * from t where ts <= '2020-04-25 00:00:00' -- p0,p1,p2;
explain select * from t where ts > '2020-04-25 00:00:00' or ts < '2020-01-02 00:00:00' -- p0;
explain select * from t where ts > '2020-04-02 00:00:00' and ts < '2020-04-07 00:00:00' -- p0,p1;

44 changes: 44 additions & 0 deletions planner/core/integration_test.go
Original file line number Diff line number Diff line change
Expand Up @@ -325,9 +325,53 @@ func (s *testIntegrationSuite) TestIssue16440(c *C) {
tk := testkit.NewTestKit(c, s.store)

tk.MustExec("use test")

tk.MustExec("drop table if exists t1")
tk.MustExec("create table t1(c0 int)")
tk.MustExec("INSERT INTO t1(c0) VALUES (NULL);")
tk.MustQuery("SELECT t1.c0 FROM t1 WHERE NOT t1.c0;").Check(testkit.Rows())
tk.MustExec("drop table t1")

tk.MustExec("drop table if exists t0, t1")
tk.MustExec("CREATE TABLE t0(t0 INT UNIQUE);")
tk.MustExec("CREATE TABLE t1(c0 FLOAT);")
tk.MustExec("INSERT INTO t1(c0) VALUES (0);")
tk.MustExec("INSERT INTO t0(t0) VALUES (NULL), (NULL);")
tk.MustQuery("SELECT t1.c0 FROM t1 LEFT JOIN t0 ON 1;").Check(testkit.Rows("0", "0"))

tk.MustExec("drop table if exists t0, t1")
tk.MustExec("CREATE TABLE t0(t0 INT);")
tk.MustExec("CREATE TABLE t1(c0 FLOAT);")
tk.MustExec("INSERT INTO t1(c0) VALUES (0);")
tk.MustExec("INSERT INTO t0(t0) VALUES (NULL), (NULL);")
tk.MustQuery("SELECT t1.c0 FROM t1 LEFT JOIN t0 ON 1;").Check(testkit.Rows("0", "0"))

tk.MustExec("drop table if exists t0, t1")
tk.MustExec("CREATE TABLE t0(t0 INT);")
tk.MustExec("CREATE TABLE t1(c0 FLOAT);")
tk.MustExec("create unique index idx on t0(t0);")
tk.MustExec("INSERT INTO t1(c0) VALUES (0);")
tk.MustExec("INSERT INTO t0(t0) VALUES (NULL), (NULL);")
tk.MustQuery("SELECT t1.c0 FROM t1 LEFT JOIN t0 ON 1;").Check(testkit.Rows("0", "0"))
}

func (s *testIntegrationSuite) TestFloorUnixTimestampPruning(c *C) {
tk := testkit.NewTestKit(c, s.store)
tk.MustExec("use test")
tk.MustExec("drop table if exists floor_unix_timestamp")
tk.MustExec(`create table floor_unix_timestamp (ts timestamp(3))
partition by range (floor(unix_timestamp(ts))) (
partition p0 values less than (unix_timestamp('2020-04-05 00:00:00')),
partition p1 values less than (unix_timestamp('2020-04-12 00:00:00')),
partition p2 values less than (unix_timestamp('2020-04-15 00:00:00')))`)
tk.MustExec("insert into floor_unix_timestamp values ('2020-04-04 00:00:00')")
tk.MustExec("insert into floor_unix_timestamp values ('2020-04-04 23:59:59.999')")
tk.MustExec("insert into floor_unix_timestamp values ('2020-04-05 00:00:00')")
tk.MustExec("insert into floor_unix_timestamp values ('2020-04-05 00:00:00.001')")
tk.MustExec("insert into floor_unix_timestamp values ('2020-04-12 01:02:03.456')")
tk.MustExec("insert into floor_unix_timestamp values ('2020-04-14 00:00:42')")
tk.MustQuery("select count(*) from floor_unix_timestamp where '2020-04-05 00:00:00.001' = ts").Check(testkit.Rows("1"))
tk.MustQuery("select * from floor_unix_timestamp where ts > '2020-04-05 00:00:00' order by ts").Check(testkit.Rows("2020-04-05 00:00:00.001", "2020-04-12 01:02:03.456", "2020-04-14 00:00:42.000"))
tk.MustQuery("select count(*) from floor_unix_timestamp where ts <= '2020-04-05 23:00:00'").Check(testkit.Rows("4"))
tk.MustQuery("select * from floor_unix_timestamp partition(p1, p2) where ts > '2020-04-14 00:00:00'").Check(testkit.Rows("2020-04-14 00:00:42.000"))
}
38 changes: 34 additions & 4 deletions planner/core/rule_partition_processor.go
Original file line number Diff line number Diff line change
Expand Up @@ -517,6 +517,19 @@ func makePartitionByFnCol(ds *DataSource, partitionExpr string) (*expression.Col
var fn *expression.ScalarFunction
switch raw := partExpr.(type) {
case *expression.ScalarFunction:
// Special handle for floor(unix_timestamp(ts)) as partition expression.
// This pattern is so common for timestamp(3) column as partition expression that it deserve an optimization.
if raw.FuncName.L == ast.Floor {
if ut, ok := raw.GetArgs()[0].(*expression.ScalarFunction); ok && ut.FuncName.L == ast.UnixTimestamp {
args := ut.GetArgs()
if len(args) == 1 {
if c, ok1 := args[0].(*expression.Column); ok1 {
return c, raw, nil
}
}
}
}

if _, ok := monotoneIncFuncs[raw.FuncName.L]; ok {
fn = raw
col = fn.GetArgs()[0].(*expression.Column)
Expand Down Expand Up @@ -622,10 +635,8 @@ func extractDataForPrune(sctx sessionctx.Context, expr expression.Expression, pa
var constExpr expression.Expression
if partFn != nil {
// If the partition expression is fn(col), change constExpr to fn(constExpr).
// No 'copy on write' for the expression here, this is a dangerous operation.
args := partFn.GetArgs()
args[0] = con
constExpr = partFn
constExpr = replaceColumnWithConst(partFn, con)

// Sometimes we need to relax the condition, < to <=, > to >=.
// For example, the following case doesn't hold:
// col < '2020-02-11 17:34:11' => to_days(col) < to_days(2020-02-11 17:34:11)
Expand All @@ -644,6 +655,25 @@ func extractDataForPrune(sctx sessionctx.Context, expr expression.Expression, pa
return ret, false
}

// replaceColumnWithConst change fn(col) to fn(const)
func replaceColumnWithConst(partFn *expression.ScalarFunction, con *expression.Constant) *expression.ScalarFunction {
args := partFn.GetArgs()
// The partition function may be floor(unix_timestamp(ts)) instead of a simple fn(col).
if partFn.FuncName.L == ast.Floor {
ut := args[0].(*expression.ScalarFunction)
if ut.FuncName.L == ast.UnixTimestamp {
args = ut.GetArgs()
args[0] = con
return partFn
}
}

// No 'copy on write' for the expression here, this is a dangerous operation.
args[0] = con
return partFn

}

// opposite turns > to <, >= to <= and so on.
func opposite(op string) string {
switch op {
Expand Down