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

fix: div and rem by negative zero #960

Merged
merged 6 commits into from
Sep 24, 2024
Merged
Show file tree
Hide file tree
Changes from 5 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
6 changes: 4 additions & 2 deletions docs/source/contributor-guide/development.md
Original file line number Diff line number Diff line change
Expand Up @@ -91,13 +91,15 @@ Comet has a plan stability testing framework that can be used to test the stabil
The plan stability testing framework is located in the `spark` module and can be run using the following command:

```sh
./mvnw -pl spark -Dsuites="org.apache.spark.sql.comet.CometTPCDSV1_4_PlanStabilitySuite" test
./mvnw -pl spark -Dsuites="org.apache.spark.sql.comet.CometTPCDSV1_4_PlanStabilitySuite" -nsu test
Copy link
Member

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

Why add -nsu?

Copy link
Contributor Author

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

It helps for faster turnaround

./mvnw -pl spark -Dsuites="org.apache.spark.sql.comet.CometTPCDSV1_4_PlanStabilitySuite" -Pspark-3.5 -nsu test
Copy link
Contributor

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

Since you added these steps to the testing commands, does it makes sense to duplicate them below for the generation commands?

Copy link
Contributor Author

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

Thanks, updated.

./mvnw -pl spark -Dsuites="org.apache.spark.sql.comet.CometTPCDSV1_4_PlanStabilitySuite" -Pspark-4.0 -nsu test
```

and
```sh
./mvnw -pl spark -Dsuites="org.apache.spark.sql.comet.CometTPCDSV2_7_PlanStabilitySuite" test
./mvnw -pl spark -Dsuites="org.apache.spark.sql.comet.CometTPCDSV2_7_PlanStabilitySuite" -nsu test
./mvnw -pl spark -Dsuites="org.apache.spark.sql.comet.CometTPCDSV2_7_PlanStabilitySuite" -Pspark-3.5 -nsu test
./mvnw -pl spark -Dsuites="org.apache.spark.sql.comet.CometTPCDSV2_7_PlanStabilitySuite" -Pspark-4.0 -nsu test
```

Expand Down
Original file line number Diff line number Diff line change
Expand Up @@ -25,7 +25,7 @@ import org.apache.spark.SparkConf
import org.apache.spark.internal.Logging
import org.apache.spark.network.util.ByteUnit
import org.apache.spark.sql.{SparkSession, SparkSessionExtensions}
import org.apache.spark.sql.catalyst.expressions.{EqualNullSafe, EqualTo, Expression, GreaterThan, GreaterThanOrEqual, KnownFloatingPointNormalized, LessThan, LessThanOrEqual, NamedExpression, PlanExpression}
import org.apache.spark.sql.catalyst.expressions.{Divide, DoubleLiteral, EqualNullSafe, EqualTo, Expression, FloatLiteral, GreaterThan, GreaterThanOrEqual, KnownFloatingPointNormalized, LessThan, LessThanOrEqual, NamedExpression, PlanExpression, Remainder}
import org.apache.spark.sql.catalyst.expressions.aggregate.{Final, Partial}
import org.apache.spark.sql.catalyst.optimizer.NormalizeNaNAndZero
import org.apache.spark.sql.catalyst.rules.Rule
Expand Down Expand Up @@ -873,12 +873,18 @@ class CometSparkSessionExtensions
LessThan(normalizeNaNAndZero(left), normalizeNaNAndZero(right))
case LessThanOrEqual(left, right) =>
LessThanOrEqual(normalizeNaNAndZero(left), normalizeNaNAndZero(right))
case Divide(left, right, evalMode) =>
Divide(left, normalizeNaNAndZero(right), evalMode)
case Remainder(left, right, evalMode) =>
Remainder(left, normalizeNaNAndZero(right), evalMode)
}
}

def normalizeNaNAndZero(expr: Expression): Expression = {
expr match {
case _: KnownFloatingPointNormalized => expr
case FloatLiteral(f) if !f.equals(-0.0f) => expr
Copy link
Contributor

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

Just clarifying: this is saying that as long as the literal isn't -0.0, we don't have to normalize it? And it doesn't look like there's any way to define a literal NaN, so we don't handle that case?

Copy link
Contributor Author

@kazuyukitanimura kazuyukitanimura Sep 24, 2024

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

Just clarifying: this is saying that as long as the literal isn't -0.0, we don't have to normalize it?

Exactly

And it doesn't look like there's any way to define a literal NaN, so we don't handle that case?

We can create a literal NaN, but Spark NormalizeNaNAndZero is not doing much for literal NaN. I think NaN normalization matters only when some expression results are NaN https://github.com/apache/spark/blob/master/sql/catalyst/src/main/scala/org/apache/spark/sql/catalyst/optimizer/NormalizeFloatingNumbers.scala#L194

case DoubleLiteral(d) if !d.equals(-0.0d) => expr
case _ =>
expr.dataType match {
case _: FloatType | _: DoubleType =>
Expand Down
Original file line number Diff line number Diff line change
Expand Up @@ -2636,7 +2636,11 @@ object QueryPlanSerde extends Logging with ShimQueryPlanSerde with CometExprShim

def nullIfWhenPrimitive(expression: Expression): Expression = if (isPrimitive(expression)) {
val zero = Literal.default(expression.dataType)
If(EqualTo(expression, zero), Literal.create(null, expression.dataType), expression)
expression match {
case _: Literal if expression != zero => expression
Copy link
Member

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

Is this related?

Copy link
Contributor Author

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

Indirectly related, this will avoid to put EqualTo for non zero literals for divisions

case _ =>
If(EqualTo(expression, zero), Literal.create(null, expression.dataType), expression)
}
} else {
expression
}
Expand Down
Original file line number Diff line number Diff line change
Expand Up @@ -129,7 +129,7 @@ Functions [2]: [sum(CASE WHEN (d_date#12 < 2000-03-11) THEN inv_quantity_on_hand

(22) CometFilter
Input [4]: [w_warehouse_name#7, i_item_id#9, inv_before#15, inv_after#16]
Condition : (CASE WHEN (inv_before#15 > 0) THEN (knownfloatingpointnormalized(normalizenanandzero((cast(inv_after#16 as double) / cast(inv_before#15 as double)))) >= knownfloatingpointnormalized(normalizenanandzero(0.666667))) END AND CASE WHEN (inv_before#15 > 0) THEN (knownfloatingpointnormalized(normalizenanandzero((cast(inv_after#16 as double) / cast(inv_before#15 as double)))) <= knownfloatingpointnormalized(normalizenanandzero(1.5))) END)
Condition : (CASE WHEN (inv_before#15 > 0) THEN (knownfloatingpointnormalized(normalizenanandzero((cast(inv_after#16 as double) / knownfloatingpointnormalized(normalizenanandzero(cast(inv_before#15 as double)))))) >= 0.666667) END AND CASE WHEN (inv_before#15 > 0) THEN (knownfloatingpointnormalized(normalizenanandzero((cast(inv_after#16 as double) / knownfloatingpointnormalized(normalizenanandzero(cast(inv_before#15 as double)))))) <= 1.5) END)

(23) CometTakeOrderedAndProject
Input [4]: [w_warehouse_name#7, i_item_id#9, inv_before#15, inv_after#16]
Expand Down
Original file line number Diff line number Diff line change
Expand Up @@ -110,7 +110,7 @@ ReadSchema: struct<hd_demo_sk:int,hd_buy_potential:string,hd_dep_count:int,hd_ve

(16) CometFilter
Input [4]: [hd_demo_sk#12, hd_buy_potential#13, hd_dep_count#14, hd_vehicle_count#15]
Condition : ((((isnotnull(hd_vehicle_count#15) AND ((hd_buy_potential#13 = >10000 ) OR (hd_buy_potential#13 = unknown ))) AND (hd_vehicle_count#15 > 0)) AND CASE WHEN (hd_vehicle_count#15 > 0) THEN (knownfloatingpointnormalized(normalizenanandzero((cast(hd_dep_count#14 as double) / cast(hd_vehicle_count#15 as double)))) > knownfloatingpointnormalized(normalizenanandzero(1.2))) END) AND isnotnull(hd_demo_sk#12))
Condition : ((((isnotnull(hd_vehicle_count#15) AND ((hd_buy_potential#13 = >10000 ) OR (hd_buy_potential#13 = unknown ))) AND (hd_vehicle_count#15 > 0)) AND CASE WHEN (hd_vehicle_count#15 > 0) THEN (knownfloatingpointnormalized(normalizenanandzero((cast(hd_dep_count#14 as double) / knownfloatingpointnormalized(normalizenanandzero(cast(hd_vehicle_count#15 as double)))))) > 1.2) END) AND isnotnull(hd_demo_sk#12))

(17) CometProject
Input [4]: [hd_demo_sk#12, hd_buy_potential#13, hd_dep_count#14, hd_vehicle_count#15]
Expand Down
Original file line number Diff line number Diff line change
Expand Up @@ -152,11 +152,11 @@ Functions [2]: [stddev_samp(cast(inv_quantity_on_hand#3 as double)), avg(inv_qua

(22) CometFilter
Input [5]: [w_warehouse_sk#7, i_item_sk#6, d_moy#11, stdev#17, mean#18]
Condition : CASE WHEN (knownfloatingpointnormalized(normalizenanandzero(mean#18)) = knownfloatingpointnormalized(normalizenanandzero(0.0))) THEN false ELSE (knownfloatingpointnormalized(normalizenanandzero((stdev#17 / mean#18))) > knownfloatingpointnormalized(normalizenanandzero(1.0))) END
Condition : CASE WHEN (knownfloatingpointnormalized(normalizenanandzero(mean#18)) = 0.0) THEN false ELSE (knownfloatingpointnormalized(normalizenanandzero((stdev#17 / knownfloatingpointnormalized(normalizenanandzero(mean#18))))) > 1.0) END

(23) CometProject
Input [5]: [w_warehouse_sk#7, i_item_sk#6, d_moy#11, stdev#17, mean#18]
Arguments: [w_warehouse_sk#7, i_item_sk#6, d_moy#11, mean#18, cov#19], [w_warehouse_sk#7, i_item_sk#6, d_moy#11, mean#18, CASE WHEN (knownfloatingpointnormalized(normalizenanandzero(mean#18)) = knownfloatingpointnormalized(normalizenanandzero(0.0))) THEN null ELSE (stdev#17 / mean#18) END AS cov#19]
Arguments: [w_warehouse_sk#7, i_item_sk#6, d_moy#11, mean#18, cov#19], [w_warehouse_sk#7, i_item_sk#6, d_moy#11, mean#18, CASE WHEN (knownfloatingpointnormalized(normalizenanandzero(mean#18)) = 0.0) THEN null ELSE (stdev#17 / knownfloatingpointnormalized(normalizenanandzero(mean#18))) END AS cov#19]

(24) Scan parquet spark_catalog.default.inventory
Output [4]: [inv_item_sk#20, inv_warehouse_sk#21, inv_quantity_on_hand#22, inv_date_sk#23]
Expand Down Expand Up @@ -238,11 +238,11 @@ Functions [2]: [stddev_samp(cast(inv_quantity_on_hand#22 as double)), avg(inv_qu

(41) CometFilter
Input [5]: [w_warehouse_sk#26, i_item_sk#25, d_moy#30, stdev#17, mean#18]
Condition : CASE WHEN (knownfloatingpointnormalized(normalizenanandzero(mean#18)) = knownfloatingpointnormalized(normalizenanandzero(0.0))) THEN false ELSE (knownfloatingpointnormalized(normalizenanandzero((stdev#17 / mean#18))) > knownfloatingpointnormalized(normalizenanandzero(1.0))) END
Condition : CASE WHEN (knownfloatingpointnormalized(normalizenanandzero(mean#18)) = 0.0) THEN false ELSE (knownfloatingpointnormalized(normalizenanandzero((stdev#17 / knownfloatingpointnormalized(normalizenanandzero(mean#18))))) > 1.0) END

(42) CometProject
Input [5]: [w_warehouse_sk#26, i_item_sk#25, d_moy#30, stdev#17, mean#18]
Arguments: [w_warehouse_sk#26, i_item_sk#25, d_moy#30, mean#36, cov#37], [w_warehouse_sk#26, i_item_sk#25, d_moy#30, mean#18 AS mean#36, CASE WHEN (knownfloatingpointnormalized(normalizenanandzero(mean#18)) = knownfloatingpointnormalized(normalizenanandzero(0.0))) THEN null ELSE (stdev#17 / mean#18) END AS cov#37]
Arguments: [w_warehouse_sk#26, i_item_sk#25, d_moy#30, mean#36, cov#37], [w_warehouse_sk#26, i_item_sk#25, d_moy#30, mean#18 AS mean#36, CASE WHEN (knownfloatingpointnormalized(normalizenanandzero(mean#18)) = 0.0) THEN null ELSE (stdev#17 / knownfloatingpointnormalized(normalizenanandzero(mean#18))) END AS cov#37]

(43) CometBroadcastExchange
Input [5]: [w_warehouse_sk#26, i_item_sk#25, d_moy#30, mean#36, cov#37]
Expand Down
Original file line number Diff line number Diff line change
Expand Up @@ -152,11 +152,11 @@ Functions [2]: [stddev_samp(cast(inv_quantity_on_hand#3 as double)), avg(inv_qua

(22) CometFilter
Input [5]: [w_warehouse_sk#7, i_item_sk#6, d_moy#11, stdev#17, mean#18]
Condition : (CASE WHEN (knownfloatingpointnormalized(normalizenanandzero(mean#18)) = knownfloatingpointnormalized(normalizenanandzero(0.0))) THEN false ELSE (knownfloatingpointnormalized(normalizenanandzero((stdev#17 / mean#18))) > knownfloatingpointnormalized(normalizenanandzero(1.0))) END AND CASE WHEN (knownfloatingpointnormalized(normalizenanandzero(mean#18)) = knownfloatingpointnormalized(normalizenanandzero(0.0))) THEN false ELSE (knownfloatingpointnormalized(normalizenanandzero((stdev#17 / mean#18))) > knownfloatingpointnormalized(normalizenanandzero(1.5))) END)
Condition : (CASE WHEN (knownfloatingpointnormalized(normalizenanandzero(mean#18)) = 0.0) THEN false ELSE (knownfloatingpointnormalized(normalizenanandzero((stdev#17 / knownfloatingpointnormalized(normalizenanandzero(mean#18))))) > 1.0) END AND CASE WHEN (knownfloatingpointnormalized(normalizenanandzero(mean#18)) = 0.0) THEN false ELSE (knownfloatingpointnormalized(normalizenanandzero((stdev#17 / knownfloatingpointnormalized(normalizenanandzero(mean#18))))) > 1.5) END)

(23) CometProject
Input [5]: [w_warehouse_sk#7, i_item_sk#6, d_moy#11, stdev#17, mean#18]
Arguments: [w_warehouse_sk#7, i_item_sk#6, d_moy#11, mean#18, cov#19], [w_warehouse_sk#7, i_item_sk#6, d_moy#11, mean#18, CASE WHEN (knownfloatingpointnormalized(normalizenanandzero(mean#18)) = knownfloatingpointnormalized(normalizenanandzero(0.0))) THEN null ELSE (stdev#17 / mean#18) END AS cov#19]
Arguments: [w_warehouse_sk#7, i_item_sk#6, d_moy#11, mean#18, cov#19], [w_warehouse_sk#7, i_item_sk#6, d_moy#11, mean#18, CASE WHEN (knownfloatingpointnormalized(normalizenanandzero(mean#18)) = 0.0) THEN null ELSE (stdev#17 / knownfloatingpointnormalized(normalizenanandzero(mean#18))) END AS cov#19]

(24) Scan parquet spark_catalog.default.inventory
Output [4]: [inv_item_sk#20, inv_warehouse_sk#21, inv_quantity_on_hand#22, inv_date_sk#23]
Expand Down Expand Up @@ -238,11 +238,11 @@ Functions [2]: [stddev_samp(cast(inv_quantity_on_hand#22 as double)), avg(inv_qu

(41) CometFilter
Input [5]: [w_warehouse_sk#26, i_item_sk#25, d_moy#30, stdev#17, mean#18]
Condition : CASE WHEN (knownfloatingpointnormalized(normalizenanandzero(mean#18)) = knownfloatingpointnormalized(normalizenanandzero(0.0))) THEN false ELSE (knownfloatingpointnormalized(normalizenanandzero((stdev#17 / mean#18))) > knownfloatingpointnormalized(normalizenanandzero(1.0))) END
Condition : CASE WHEN (knownfloatingpointnormalized(normalizenanandzero(mean#18)) = 0.0) THEN false ELSE (knownfloatingpointnormalized(normalizenanandzero((stdev#17 / knownfloatingpointnormalized(normalizenanandzero(mean#18))))) > 1.0) END

(42) CometProject
Input [5]: [w_warehouse_sk#26, i_item_sk#25, d_moy#30, stdev#17, mean#18]
Arguments: [w_warehouse_sk#26, i_item_sk#25, d_moy#30, mean#36, cov#37], [w_warehouse_sk#26, i_item_sk#25, d_moy#30, mean#18 AS mean#36, CASE WHEN (knownfloatingpointnormalized(normalizenanandzero(mean#18)) = knownfloatingpointnormalized(normalizenanandzero(0.0))) THEN null ELSE (stdev#17 / mean#18) END AS cov#37]
Arguments: [w_warehouse_sk#26, i_item_sk#25, d_moy#30, mean#36, cov#37], [w_warehouse_sk#26, i_item_sk#25, d_moy#30, mean#18 AS mean#36, CASE WHEN (knownfloatingpointnormalized(normalizenanandzero(mean#18)) = 0.0) THEN null ELSE (stdev#17 / knownfloatingpointnormalized(normalizenanandzero(mean#18))) END AS cov#37]

(43) CometBroadcastExchange
Input [5]: [w_warehouse_sk#26, i_item_sk#25, d_moy#30, mean#36, cov#37]
Expand Down
Original file line number Diff line number Diff line change
Expand Up @@ -110,7 +110,7 @@ ReadSchema: struct<hd_demo_sk:int,hd_buy_potential:string,hd_dep_count:int,hd_ve

(16) CometFilter
Input [4]: [hd_demo_sk#12, hd_buy_potential#13, hd_dep_count#14, hd_vehicle_count#15]
Condition : ((((isnotnull(hd_vehicle_count#15) AND ((hd_buy_potential#13 = >10000 ) OR (hd_buy_potential#13 = unknown ))) AND (hd_vehicle_count#15 > 0)) AND CASE WHEN (hd_vehicle_count#15 > 0) THEN (knownfloatingpointnormalized(normalizenanandzero((cast(hd_dep_count#14 as double) / cast(hd_vehicle_count#15 as double)))) > knownfloatingpointnormalized(normalizenanandzero(1.0))) END) AND isnotnull(hd_demo_sk#12))
Condition : ((((isnotnull(hd_vehicle_count#15) AND ((hd_buy_potential#13 = >10000 ) OR (hd_buy_potential#13 = unknown ))) AND (hd_vehicle_count#15 > 0)) AND CASE WHEN (hd_vehicle_count#15 > 0) THEN (knownfloatingpointnormalized(normalizenanandzero((cast(hd_dep_count#14 as double) / knownfloatingpointnormalized(normalizenanandzero(cast(hd_vehicle_count#15 as double)))))) > 1.0) END) AND isnotnull(hd_demo_sk#12))

(17) CometProject
Input [4]: [hd_demo_sk#12, hd_buy_potential#13, hd_dep_count#14, hd_vehicle_count#15]
Expand Down
Original file line number Diff line number Diff line change
Expand Up @@ -364,7 +364,7 @@ Arguments: [ss_sold_year#17, ss_item_sk#1, ss_customer_sk#2], [cs_sold_year#58,
Input [15]: [ss_sold_year#17, ss_item_sk#1, ss_customer_sk#2, ss_qty#18, ss_wc#19, ss_sp#20, ws_qty#39, ws_wc#40, ws_sp#41, cs_sold_year#58, cs_item_sk#43, cs_customer_sk#59, cs_qty#60, cs_wc#61, cs_sp#62]

(66) Project [codegen id : 1]
Output [12]: [round((cast(ss_qty#18 as double) / cast(coalesce((ws_qty#39 + cs_qty#60), 1) as double)), 2) AS ratio#63, ss_qty#18 AS store_qty#64, ss_wc#19 AS store_wholesale_cost#65, ss_sp#20 AS store_sales_price#66, (coalesce(ws_qty#39, 0) + coalesce(cs_qty#60, 0)) AS other_chan_qty#67, (coalesce(ws_wc#40, 0.00) + coalesce(cs_wc#61, 0.00)) AS other_chan_wholesale_cost#68, (coalesce(ws_sp#41, 0.00) + coalesce(cs_sp#62, 0.00)) AS other_chan_sales_price#69, ss_qty#18, ss_wc#19, ss_sp#20, ws_qty#39, cs_qty#60]
Output [12]: [round((cast(ss_qty#18 as double) / knownfloatingpointnormalized(normalizenanandzero(cast(coalesce((ws_qty#39 + cs_qty#60), 1) as double)))), 2) AS ratio#63, ss_qty#18 AS store_qty#64, ss_wc#19 AS store_wholesale_cost#65, ss_sp#20 AS store_sales_price#66, (coalesce(ws_qty#39, 0) + coalesce(cs_qty#60, 0)) AS other_chan_qty#67, (coalesce(ws_wc#40, 0.00) + coalesce(cs_wc#61, 0.00)) AS other_chan_wholesale_cost#68, (coalesce(ws_sp#41, 0.00) + coalesce(cs_sp#62, 0.00)) AS other_chan_sales_price#69, ss_qty#18, ss_wc#19, ss_sp#20, ws_qty#39, cs_qty#60]
Input [15]: [ss_sold_year#17, ss_item_sk#1, ss_customer_sk#2, ss_qty#18, ss_wc#19, ss_sp#20, ws_qty#39, ws_wc#40, ws_sp#41, cs_sold_year#58, cs_item_sk#43, cs_customer_sk#59, cs_qty#60, cs_wc#61, cs_sp#62]

(67) TakeOrderedAndProject
Expand Down
Original file line number Diff line number Diff line change
Expand Up @@ -259,7 +259,7 @@ Arguments: [item_id#21], [item_id#31], Inner, BuildRight

(46) CometProject
Input [5]: [item_id#21, sr_item_qty#22, cr_item_qty#20, item_id#31, wr_item_qty#32]
Arguments: [item_id#21, sr_item_qty#22, sr_dev#33, cr_item_qty#20, cr_dev#34, wr_item_qty#32, wr_dev#35, average#36], [item_id#21, sr_item_qty#22, (((cast(sr_item_qty#22 as double) / cast(((sr_item_qty#22 + cr_item_qty#20) + wr_item_qty#32) as double)) / 3.0) * 100.0) AS sr_dev#33, cr_item_qty#20, (((cast(cr_item_qty#20 as double) / cast(((sr_item_qty#22 + cr_item_qty#20) + wr_item_qty#32) as double)) / 3.0) * 100.0) AS cr_dev#34, wr_item_qty#32, (((cast(wr_item_qty#32 as double) / cast(((sr_item_qty#22 + cr_item_qty#20) + wr_item_qty#32) as double)) / 3.0) * 100.0) AS wr_dev#35, (cast(((sr_item_qty#22 + cr_item_qty#20) + wr_item_qty#32) as decimal(20,0)) / 3.0) AS average#36]
Arguments: [item_id#21, sr_item_qty#22, sr_dev#33, cr_item_qty#20, cr_dev#34, wr_item_qty#32, wr_dev#35, average#36], [item_id#21, sr_item_qty#22, (((cast(sr_item_qty#22 as double) / knownfloatingpointnormalized(normalizenanandzero(cast(((sr_item_qty#22 + cr_item_qty#20) + wr_item_qty#32) as double)))) / 3.0) * 100.0) AS sr_dev#33, cr_item_qty#20, (((cast(cr_item_qty#20 as double) / knownfloatingpointnormalized(normalizenanandzero(cast(((sr_item_qty#22 + cr_item_qty#20) + wr_item_qty#32) as double)))) / 3.0) * 100.0) AS cr_dev#34, wr_item_qty#32, (((cast(wr_item_qty#32 as double) / knownfloatingpointnormalized(normalizenanandzero(cast(((sr_item_qty#22 + cr_item_qty#20) + wr_item_qty#32) as double)))) / 3.0) * 100.0) AS wr_dev#35, (cast(((sr_item_qty#22 + cr_item_qty#20) + wr_item_qty#32) as decimal(20,0)) / 3.0) AS average#36]

(47) CometTakeOrderedAndProject
Input [8]: [item_id#21, sr_item_qty#22, sr_dev#33, cr_item_qty#20, cr_dev#34, wr_item_qty#32, wr_dev#35, average#36]
Expand Down
Original file line number Diff line number Diff line change
Expand Up @@ -136,7 +136,7 @@ Results [4]: [w_warehouse_name#7, i_item_id#9, sum(CASE WHEN (d_date#12 < 2000-0

(23) Filter [codegen id : 2]
Input [4]: [w_warehouse_name#7, i_item_id#9, inv_before#19, inv_after#20]
Condition : (CASE WHEN (inv_before#19 > 0) THEN (knownfloatingpointnormalized(normalizenanandzero((cast(inv_after#20 as double) / cast(inv_before#19 as double)))) >= knownfloatingpointnormalized(normalizenanandzero(0.666667))) END AND CASE WHEN (inv_before#19 > 0) THEN (knownfloatingpointnormalized(normalizenanandzero((cast(inv_after#20 as double) / cast(inv_before#19 as double)))) <= knownfloatingpointnormalized(normalizenanandzero(1.5))) END)
Condition : (CASE WHEN (inv_before#19 > 0) THEN (knownfloatingpointnormalized(normalizenanandzero((cast(inv_after#20 as double) / knownfloatingpointnormalized(normalizenanandzero(cast(inv_before#19 as double)))))) >= 0.666667) END AND CASE WHEN (inv_before#19 > 0) THEN (knownfloatingpointnormalized(normalizenanandzero((cast(inv_after#20 as double) / knownfloatingpointnormalized(normalizenanandzero(cast(inv_before#19 as double)))))) <= 1.5) END)

(24) TakeOrderedAndProject
Input [4]: [w_warehouse_name#7, i_item_id#9, inv_before#19, inv_after#20]
Expand Down
Original file line number Diff line number Diff line change
Expand Up @@ -110,7 +110,7 @@ ReadSchema: struct<hd_demo_sk:int,hd_buy_potential:string,hd_dep_count:int,hd_ve

(16) CometFilter
Input [4]: [hd_demo_sk#12, hd_buy_potential#13, hd_dep_count#14, hd_vehicle_count#15]
Condition : ((((isnotnull(hd_vehicle_count#15) AND ((hd_buy_potential#13 = >10000 ) OR (hd_buy_potential#13 = unknown ))) AND (hd_vehicle_count#15 > 0)) AND CASE WHEN (hd_vehicle_count#15 > 0) THEN (knownfloatingpointnormalized(normalizenanandzero((cast(hd_dep_count#14 as double) / cast(hd_vehicle_count#15 as double)))) > knownfloatingpointnormalized(normalizenanandzero(1.2))) END) AND isnotnull(hd_demo_sk#12))
Condition : ((((isnotnull(hd_vehicle_count#15) AND ((hd_buy_potential#13 = >10000 ) OR (hd_buy_potential#13 = unknown ))) AND (hd_vehicle_count#15 > 0)) AND CASE WHEN (hd_vehicle_count#15 > 0) THEN (knownfloatingpointnormalized(normalizenanandzero((cast(hd_dep_count#14 as double) / knownfloatingpointnormalized(normalizenanandzero(cast(hd_vehicle_count#15 as double)))))) > 1.2) END) AND isnotnull(hd_demo_sk#12))

(17) CometProject
Input [4]: [hd_demo_sk#12, hd_buy_potential#13, hd_dep_count#14, hd_vehicle_count#15]
Expand Down
Loading
Loading