expression: fix compatibility of extract day_time unit functions (pin…

…gcap#21601)

Signed-off-by: lzmhhh123 <[email protected]>

cmd/explaintest/r/tpch.result

@@ -446,7 +446,7 @@ id	estRows	task	access object	operator info
 Sort_22	769.96	root		tpch.nation.n_name, tpch.nation.n_name, Column#50
 └─Projection_24	769.96	root		tpch.nation.n_name, tpch.nation.n_name, Column#50, Column#52
   └─HashAgg_27	769.96	root		group by:Column#50, tpch.nation.n_name, tpch.nation.n_name, funcs:sum(Column#51)->Column#52, funcs:firstrow(tpch.nation.n_name)->tpch.nation.n_name, funcs:firstrow(tpch.nation.n_name)->tpch.nation.n_name, funcs:firstrow(Column#50)->Column#50
-    └─Projection_28	1957240.42	root		tpch.nation.n_name, tpch.nation.n_name, extract(YEAR, tpch.lineitem.l_shipdate)->Column#50, mul(tpch.lineitem.l_extendedprice, minus(1, tpch.lineitem.l_discount))->Column#51
+    └─Projection_28	1957240.42	root		tpch.nation.n_name, tpch.nation.n_name, extract(YEAR, cast(tpch.lineitem.l_shipdate, var_string(10)))->Column#50, mul(tpch.lineitem.l_extendedprice, minus(1, tpch.lineitem.l_discount))->Column#51
       └─HashJoin_40	1957240.42	root		inner join, equal:[eq(tpch.customer.c_nationkey, tpch.nation.n_nationkey)], other cond:or(and(eq(tpch.nation.n_name, "JAPAN"), eq(tpch.nation.n_name, "INDIA")), and(eq(tpch.nation.n_name, "INDIA"), eq(tpch.nation.n_name, "JAPAN")))
         ├─TableReader_94(Build)	2.00	root		data:Selection_93
         │ └─Selection_93	2.00	cop[tikv]		or(eq(tpch.nation.n_name, "INDIA"), eq(tpch.nation.n_name, "JAPAN"))
@@ -519,7 +519,7 @@ Sort_29	719.02	root		Column#62
 └─Projection_31	719.02	root		Column#62, div(Column#64, Column#65)->Column#66
   └─HashAgg_34	719.02	root		group by:Column#78, funcs:sum(Column#75)->Column#64, funcs:sum(Column#76)->Column#65, funcs:firstrow(Column#77)->Column#62
     └─Projection_123	563136.02	root		case(eq(tpch.nation.n_name, INDIA), Column#63, 0)->Column#75, Column#63, Column#62, Column#62
-      └─Projection_35	563136.02	root		extract(YEAR, tpch.orders.o_orderdate)->Column#62, mul(tpch.lineitem.l_extendedprice, minus(1, tpch.lineitem.l_discount))->Column#63, tpch.nation.n_name
+      └─Projection_35	563136.02	root		extract(YEAR, cast(tpch.orders.o_orderdate, var_string(10)))->Column#62, mul(tpch.lineitem.l_extendedprice, minus(1, tpch.lineitem.l_discount))->Column#63, tpch.nation.n_name
         └─HashJoin_45	563136.02	root		inner join, equal:[eq(tpch.supplier.s_nationkey, tpch.nation.n_nationkey)]
           ├─TableReader_121(Build)	25.00	root		data:TableFullScan_120
           │ └─TableFullScan_120	25.00	cop[tikv]	table:n2	keep order:false
@@ -594,7 +594,7 @@ id	estRows	task	access object	operator info
 Sort_25	2406.00	root		tpch.nation.n_name, Column#53:desc
 └─Projection_27	2406.00	root		tpch.nation.n_name, Column#53, Column#55
   └─HashAgg_30	2406.00	root		group by:Column#53, tpch.nation.n_name, funcs:sum(Column#54)->Column#55, funcs:firstrow(tpch.nation.n_name)->tpch.nation.n_name, funcs:firstrow(Column#53)->Column#53
-    └─Projection_31	241379546.70	root		tpch.nation.n_name, extract(YEAR, tpch.orders.o_orderdate)->Column#53, minus(mul(tpch.lineitem.l_extendedprice, minus(1, tpch.lineitem.l_discount)), mul(tpch.partsupp.ps_supplycost, tpch.lineitem.l_quantity))->Column#54
+    └─Projection_31	241379546.70	root		tpch.nation.n_name, extract(YEAR, cast(tpch.orders.o_orderdate, var_string(10)))->Column#53, minus(mul(tpch.lineitem.l_extendedprice, minus(1, tpch.lineitem.l_discount)), mul(tpch.partsupp.ps_supplycost, tpch.lineitem.l_quantity))->Column#54
       └─HashJoin_42	241379546.70	root		inner join, equal:[eq(tpch.lineitem.l_orderkey, tpch.orders.o_orderkey)]
         ├─TableReader_117(Build)	75000000.00	root		data:TableFullScan_116
         │ └─TableFullScan_116	75000000.00	cop[tikv]	table:orders	keep order:false

expression/builtin_time.go

@@ -2631,10 +2631,15 @@ func (c *extractFunctionClass) getFunction(ctx sessionctx.Context, args []Expres
 	}
 	var bf baseBuiltinFunc
 	if isDatetimeUnit {
-		bf, err = newBaseBuiltinFuncWithTp(ctx, c.funcName, args, types.ETInt, types.ETString, types.ETDatetime)
+		decimalArg1 := int(types.MaxFsp)
+		if args[1].GetType().EvalType() != types.ETString {
+			decimalArg1 = 0
+		}
+		bf, err = newBaseBuiltinFuncWithTp(ctx, c.funcName, args, types.ETInt, types.ETString, types.ETString)
 		if err != nil {
 			return nil, err
 		}
+		bf.args[1].GetType().Decimal = decimalArg1
 		sig = &builtinExtractDatetimeSig{bf}
 		sig.setPbCode(tipb.ScalarFuncSig_ExtractDatetime)
 	} else {
@@ -2665,11 +2670,38 @@ func (b *builtinExtractDatetimeSig) evalInt(row chunk.Row) (int64, bool, error)
 	if isNull || err != nil {
 		return 0, isNull, err
 	}
-	dt, isNull, err := b.args[1].EvalTime(b.ctx, row)
+	dtStr, isNull, err := b.args[1].EvalString(b.ctx, row)
 	if isNull || err != nil {
 		return 0, isNull, err
 	}
+	sc := b.ctx.GetSessionVars().StmtCtx
+	switch strings.ToUpper(unit) {
+	case "DAY_MICROSECOND", "DAY_SECOND", "DAY_MINUTE", "DAY_HOUR":
+		dur, err := types.ParseDuration(sc, dtStr, types.GetFsp(dtStr))
+		if err != nil {
+			return 0, true, err
+		}
+		res, err := types.ExtractDurationNum(&dur, unit)
+		if err != nil {
+			return 0, true, err
+		}
+		dt, err := types.ParseDatetime(sc, dtStr)
+		if err != nil {
+			return res, false, nil
+		}
+		if dt.Hour() == dur.Hour() && dt.Minute() == dur.Minute() && dt.Second() == dur.Second() && dt.Year() > 0 {
+			res, err = types.ExtractDatetimeNum(&dt, unit)
+		}
+		return res, err != nil, err
+	}
+	dt, err := types.ParseDatetime(sc, dtStr)
+	if err != nil {
+		if !terror.ErrorEqual(err, types.ErrWrongValue) {
+			return 0, true, err
+		}
+	}
 	if dt.IsZero() {
+		dt.SetFsp(int8(b.args[1].GetType().Decimal))
 		if b.ctx.GetSessionVars().SQLMode.HasNoZeroDateMode() {
 			isNull, err := handleInvalidZeroTime(b.ctx, dt)
 			return 0, isNull, err

expression/builtin_time_vec.go

@@ -206,7 +206,8 @@ func (b *builtinSysDateWithoutFspSig) vecEvalTime(input *chunk.Chunk, result *ch
 }
 
 func (b *builtinExtractDatetimeSig) vectorized() bool {
-	return true
+	// TODO: to fix https://github.com/pingcap/tidb/issues/9716 in vectorized evaluation.
+	return false
 }
 
 func (b *builtinExtractDatetimeSig) vecEvalInt(input *chunk.Chunk, result *chunk.Column) error {

expression/builtin_time_vec_test.go

@@ -548,6 +548,7 @@ var vecBuiltinTimeCases = map[string][]vecExprBenchCase{
 	ast.LastDay: {
 		{retEvalType: types.ETDatetime, childrenTypes: []types.EvalType{types.ETDatetime}},
 	},
+	/* TODO: to fix https://github.com/pingcap/tidb/issues/9716 in vectorized evaluation.
 	ast.Extract: {
 		{retEvalType: types.ETInt, childrenTypes: []types.EvalType{types.ETString, types.ETDatetime}, geners: []dataGenerator{newDateTimeUnitStrGener(), nil}},
 		{retEvalType: types.ETInt, childrenTypes: []types.EvalType{types.ETString, types.ETDuration},
@@ -581,6 +582,7 @@ var vecBuiltinTimeCases = map[string][]vecExprBenchCase{
 			constants: []*Constant{{Value: types.NewStringDatum("HOUR_MINUTE"), RetType: types.NewFieldType(mysql.TypeString)}},
 		},
 	},
+	*/
 	ast.ConvertTz: {
 		{retEvalType: types.ETDatetime, childrenTypes: []types.EvalType{types.ETDatetime, types.ETString, types.ETString},
 			geners: []dataGenerator{nil, newNullWrappedGener(0.2, &tzStrGener{}), newNullWrappedGener(0.2, &tzStrGener{})}},

expression/integration_test.go

@@ -2065,6 +2065,14 @@ func (s *testIntegrationSuite2) TestTimeBuiltin(c *C) {
 	// for extract
 	result = tk.MustQuery(`select extract(day from '800:12:12'), extract(hour from '800:12:12'), extract(month from 20170101), extract(day_second from '2017-01-01 12:12:12')`)
 	result.Check(testkit.Rows("12 800 1 1121212"))
+	result = tk.MustQuery("select extract(day_microsecond from '2017-01-01 12:12:12'), extract(day_microsecond from '01 12:12:12'), extract(day_microsecond from '12:12:12'), extract(day_microsecond from '01 00:00:00.89')")
+	result.Check(testkit.Rows("1121212000000 361212000000 121212000000 240000890000"))
+	result = tk.MustQuery("select extract(day_second from '2017-01-01 12:12:12'), extract(day_second from '01 12:12:12'), extract(day_second from '12:12:12'), extract(day_second from '01 00:00:00.89')")
+	result.Check(testkit.Rows("1121212 361212 121212 240000"))
+	result = tk.MustQuery("select extract(day_minute from '2017-01-01 12:12:12'), extract(day_minute from '01 12:12:12'), extract(day_minute from '12:12:12'), extract(day_minute from '01 00:00:00.89')")
+	result.Check(testkit.Rows("11212 3612 1212 2400"))
+	result = tk.MustQuery("select extract(day_hour from '2017-01-01 12:12:12'), extract(day_hour from '01 12:12:12'), extract(day_hour from '12:12:12'), extract(day_hour from '01 00:00:00.89')")
+	result.Check(testkit.Rows("112 36 12 24"))
 
 	// for adddate, subdate
 	dateArithmeticalTests := []struct {

types/time.go

@@ -2175,6 +2175,14 @@ func ExtractDurationNum(d *Duration, unit string) (int64, error) {
 		return int64(d.Hour())*10000 + int64(d.Minute())*100 + int64(d.Second()), nil
 	case "HOUR_MINUTE":
 		return int64(d.Hour())*100 + int64(d.Minute()), nil
+	case "DAY_MICROSECOND":
+		return int64(d.Hour()*10000+d.Minute()*100+d.Second())*1000000 + int64(d.MicroSecond()), nil
+	case "DAY_SECOND":
+		return int64(d.Hour())*10000 + int64(d.Minute())*100 + int64(d.Second()), nil
+	case "DAY_MINUTE":
+		return int64(d.Hour())*100 + int64(d.Minute()), nil
+	case "DAY_HOUR":
+		return int64(d.Hour()), nil
 	default:
 		return 0, errors.Errorf("invalid unit %s", unit)
 	}

types/time_test.go

@@ -1562,6 +1562,10 @@ func (s *testTimeSuite) TestExtractDurationNum(c *C) {
 		{"HOUR_MICROSECOND", 31536},
 		{"HOUR_SECOND", 0},
 		{"HOUR_MINUTE", 0},
+		{"DAY_MICROSECOND", 31536},
+		{"DAY_SECOND", 0},
+		{"DAY_MINUTE", 0},
+		{"DAY_HOUR", 0},
 	}
 
 	for _, col := range tbl {