jt2594838 commented on code in PR #18606:
URL: https://github.com/apache/iotdb/pull/18606#discussion_r4023237185


##########
iotdb-core/calc-commons/src/main/java/org/apache/iotdb/calc/execution/operator/source/relational/aggregation/grouped/GroupedVarianceAccumulator.java:
##########
@@ -52,6 +55,10 @@ public class GroupedVarianceAccumulator implements 
GroupedAccumulator {
   public GroupedVarianceAccumulator(
       TSDataType seriesDataType, VarianceAccumulator.VarianceType 
varianceType) {
     this.seriesDataType = seriesDataType;
+    this.doubleValueConverter =
+        TypeServices.NUMERIC_BATCH_READER_SERVICE

Review Comment:
   Use the native numeric batch reader for grouped variance, retaining the 
existing recurrence callback and logical row order rather than changing the 
numerical algorithm. NumericMomentBatchTest covers moments/variance across 
tree, table and grouped implementations, including masked/null inputs. This 
candidate was retained after mixed-type and fixed-DOUBLE measurements; related 
candidates with measured regressions were restored.



##########
iotdb-core/calc-commons/src/main/java/org/apache/iotdb/calc/execution/operator/source/relational/aggregation/TableCentralMomentAccumulator.java:
##########
@@ -51,47 +54,21 @@ public class TableCentralMomentAccumulator implements 
TableAccumulator {
   public TableCentralMomentAccumulator(
       TSDataType seriesDataType, CentralMomentAccumulator.MomentType 
momentType) {
     this.seriesDataType = seriesDataType;
+    this.doubleValueConverter =
+        TypeServices.NUMERIC_BATCH_READER_SERVICE

Review Comment:
   Use the native numeric batch reader for table central moment, retaining the 
existing recurrence callback and logical row order rather than changing the 
numerical algorithm. NumericMomentBatchTest covers moments/variance across 
tree, table and grouped implementations, including masked/null inputs. This 
candidate was retained after mixed-type and fixed-DOUBLE measurements; related 
candidates with measured regressions were restored.



##########
iotdb-core/calc-commons/src/main/java/org/apache/iotdb/calc/execution/operator/source/relational/aggregation/grouped/rate/GroupedNaiveDeltaAccumulator.java:
##########
@@ -59,31 +61,28 @@ public void setGroupCount(long groupCount) {
   @Override
   public void addInput(int[] groupIds, Column[] arguments, AggregationMask 
mask) {
     RateFunctionValidation.validateArgumentCount(arguments, 
RateFunctionType.DELTA);
-    int selectedCount = mask.getSelectedPositionCount();
-    int[] selectedPositions = mask.isSelectAll() ? null : 
mask.getSelectedPositions();
-    for (int index = 0; index < selectedCount; index++) {
-      int position = mask.isSelectAll() ? index : selectedPositions[index];
-      if (arguments[0].isNull(position)) {
-        continue;
-      }
-      int groupId = groupIds[position];
-      double value =
-          RateFunctionValidation.readValue(
-              arguments[0], position, valueDataType, RateFunctionType.DELTA);
-      long time =
-          RateFunctionValidation.readRequiredTime(
-              arguments[1], position, RateFunctionType.DELTA, 2);
-      long currentWindowStart =
-          RateFunctionValidation.readRequiredTime(
-              arguments[2], position, RateFunctionType.DELTA, 3);
-      long currentWindowEnd =
-          RateFunctionValidation.readRequiredTime(
-              arguments[3], position, RateFunctionType.DELTA, 4);
-      RateFunctionValidation.validateWindow(
-          RateFunctionType.DELTA, time, currentWindowStart, currentWindowEnd);
-      initializeOrValidateWindow(groupId, currentWindowStart, 
currentWindowEnd);
-      samples.add(groupId, time, value);
-    }
+    TypeServices.RATE_INPUT_SERVICE

Review Comment:
   GroupedNaiveDeltaAccumulator now dispatches to a native batch reader through 
RATE_INPUT_SERVICE. It preserves value validation before time/window validation 
and keeps the existing state update callback, including logical group positions 
where applicable. RateBatchInputTest exercises all 16 implementations with 
native and wrapped columns, masks/nulls and validation cases. This is one of 
the 12 retained rate candidates; the other four were restored after 
fixed-DOUBLE benchmarks showed regressions.



##########
iotdb-core/calc-commons/src/main/java/org/apache/iotdb/calc/execution/operator/source/relational/aggregation/grouped/rate/GroupedNaiveIncreaseAccumulator.java:
##########
@@ -59,31 +61,28 @@ public void setGroupCount(long groupCount) {
   @Override
   public void addInput(int[] groupIds, Column[] arguments, AggregationMask 
mask) {
     RateFunctionValidation.validateArgumentCount(arguments, 
RateFunctionType.INCREASE);
-    int selectedCount = mask.getSelectedPositionCount();
-    int[] selectedPositions = mask.isSelectAll() ? null : 
mask.getSelectedPositions();
-    for (int index = 0; index < selectedCount; index++) {
-      int position = mask.isSelectAll() ? index : selectedPositions[index];
-      if (arguments[0].isNull(position)) {
-        continue;
-      }
-      int groupId = groupIds[position];
-      double value =
-          RateFunctionValidation.readValue(
-              arguments[0], position, valueDataType, 
RateFunctionType.INCREASE);
-      long time =
-          RateFunctionValidation.readRequiredTime(
-              arguments[1], position, RateFunctionType.INCREASE, 2);
-      long currentWindowStart =
-          RateFunctionValidation.readRequiredTime(
-              arguments[2], position, RateFunctionType.INCREASE, 3);
-      long currentWindowEnd =
-          RateFunctionValidation.readRequiredTime(
-              arguments[3], position, RateFunctionType.INCREASE, 4);
-      RateFunctionValidation.validateWindow(
-          RateFunctionType.INCREASE, time, currentWindowStart, 
currentWindowEnd);
-      initializeOrValidateWindow(groupId, currentWindowStart, 
currentWindowEnd);
-      samples.add(groupId, time, value);
-    }
+    TypeServices.RATE_INPUT_SERVICE

Review Comment:
   GroupedNaiveIncreaseAccumulator now dispatches to a native batch reader 
through RATE_INPUT_SERVICE. It preserves value validation before time/window 
validation and keeps the existing state update callback, including logical 
group positions where applicable. RateBatchInputTest exercises all 16 
implementations with native and wrapped columns, masks/nulls and validation 
cases. This is one of the 12 retained rate candidates; the other four were 
restored after fixed-DOUBLE benchmarks showed regressions.



##########
iotdb-core/calc-commons/src/main/java/org/apache/iotdb/calc/execution/operator/source/relational/aggregation/grouped/rate/GroupedNaiveIrateAccumulator.java:
##########
@@ -58,22 +60,19 @@ public void setGroupCount(long groupCount) {
   @Override
   public void addInput(int[] groupIds, Column[] arguments, AggregationMask 
mask) {
     RateFunctionValidation.validateArgumentCount(arguments, 
RateFunctionType.IRATE);
-    int selectedCount = mask.getSelectedPositionCount();
-    int[] selectedPositions = mask.isSelectAll() ? null : 
mask.getSelectedPositions();
-    for (int index = 0; index < selectedCount; index++) {
-      int position = mask.isSelectAll() ? index : selectedPositions[index];
-      if (arguments[0].isNull(position)) {
-        continue;
-      }
-      int groupId = groupIds[position];
-      double value =
-          RateFunctionValidation.readValue(
-              arguments[0], position, valueDataType, RateFunctionType.IRATE);
-      long time =
-          RateFunctionValidation.readRequiredTime(
-              arguments[1], position, RateFunctionType.IRATE, 2);
-      samples.add(groupId, time, value);
-    }
+    TypeServices.RATE_INPUT_SERVICE

Review Comment:
   GroupedNaiveIrateAccumulator now dispatches to a native batch reader through 
RATE_INPUT_SERVICE. It preserves value validation before time/window validation 
and keeps the existing state update callback, including logical group positions 
where applicable. RateBatchInputTest exercises all 16 implementations with 
native and wrapped columns, masks/nulls and validation cases. This is one of 
the 12 retained rate candidates; the other four were restored after 
fixed-DOUBLE benchmarks showed regressions.



##########
iotdb-core/calc-commons/src/main/java/org/apache/iotdb/calc/execution/operator/source/relational/aggregation/grouped/rate/GroupedNaiveRateAccumulator.java:
##########
@@ -59,28 +61,28 @@ public void setGroupCount(long groupCount) {
   @Override
   public void addInput(int[] groupIds, Column[] arguments, AggregationMask 
mask) {
     RateFunctionValidation.validateArgumentCount(arguments, 
RateFunctionType.RATE);
-    int selectedCount = mask.getSelectedPositionCount();
-    int[] selectedPositions = mask.isSelectAll() ? null : 
mask.getSelectedPositions();
-    for (int index = 0; index < selectedCount; index++) {
-      int position = mask.isSelectAll() ? index : selectedPositions[index];
-      if (arguments[0].isNull(position)) {
-        continue;
-      }
-      int groupId = groupIds[position];
-      double value =
-          RateFunctionValidation.readValue(
-              arguments[0], position, valueDataType, RateFunctionType.RATE);
-      long time =
-          RateFunctionValidation.readRequiredTime(arguments[1], position, 
RateFunctionType.RATE, 2);
-      long currentWindowStart =
-          RateFunctionValidation.readRequiredTime(arguments[2], position, 
RateFunctionType.RATE, 3);
-      long currentWindowEnd =
-          RateFunctionValidation.readRequiredTime(arguments[3], position, 
RateFunctionType.RATE, 4);
-      RateFunctionValidation.validateWindow(
-          RateFunctionType.RATE, time, currentWindowStart, currentWindowEnd);
-      initializeOrValidateWindow(groupId, currentWindowStart, 
currentWindowEnd);
-      samples.add(groupId, time, value);
-    }
+    TypeServices.RATE_INPUT_SERVICE

Review Comment:
   GroupedNaiveRateAccumulator now dispatches to a native batch reader through 
RATE_INPUT_SERVICE. It preserves value validation before time/window validation 
and keeps the existing state update callback, including logical group positions 
where applicable. RateBatchInputTest exercises all 16 implementations with 
native and wrapped columns, masks/nulls and validation cases. This is one of 
the 12 retained rate candidates; the other four were restored after 
fixed-DOUBLE benchmarks showed regressions.



##########
iotdb-core/calc-commons/src/main/java/org/apache/iotdb/calc/execution/operator/source/relational/aggregation/grouped/rate/GroupedOrderedIrateAccumulator.java:
##########
@@ -68,22 +70,19 @@ public void setGroupCount(long groupCount) {
   @Override
   public void addInput(int[] groupIds, Column[] arguments, AggregationMask 
mask) {
     RateFunctionValidation.validateArgumentCount(arguments, 
RateFunctionType.IRATE);
-    int selectedCount = mask.getSelectedPositionCount();
-    int[] selectedPositions = mask.isSelectAll() ? null : 
mask.getSelectedPositions();
-    for (int index = 0; index < selectedCount; index++) {
-      int position = mask.isSelectAll() ? index : selectedPositions[index];
-      if (arguments[0].isNull(position)) {
-        continue;
-      }
-      int groupId = groupIds[position];
-      double value =
-          RateFunctionValidation.readValue(
-              arguments[0], position, valueDataType, RateFunctionType.IRATE);
-      long time =
-          RateFunctionValidation.readRequiredTime(
-              arguments[1], position, RateFunctionType.IRATE, 2);
-      update(groupId, time, value);
-    }
+    TypeServices.RATE_INPUT_SERVICE

Review Comment:
   GroupedOrderedIrateAccumulator now dispatches to a native batch reader 
through RATE_INPUT_SERVICE. It preserves value validation before time/window 
validation and keeps the existing state update callback, including logical 
group positions where applicable. RateBatchInputTest exercises all 16 
implementations with native and wrapped columns, masks/nulls and validation 
cases. This is one of the 12 retained rate candidates; the other four were 
restored after fixed-DOUBLE benchmarks showed regressions.



##########
iotdb-core/calc-commons/src/main/java/org/apache/iotdb/calc/execution/operator/source/relational/aggregation/rate/NaiveDeltaAccumulator.java:
##########
@@ -58,30 +60,27 @@ public TableAccumulator copy() {
   @Override
   public void addInput(Column[] arguments, AggregationMask mask) {
     RateFunctionValidation.validateArgumentCount(arguments, 
RateFunctionType.DELTA);
-    int selectedCount = mask.getSelectedPositionCount();
-    int[] selectedPositions = mask.isSelectAll() ? null : 
mask.getSelectedPositions();
-    for (int index = 0; index < selectedCount; index++) {
-      int position = mask.isSelectAll() ? index : selectedPositions[index];
-      if (arguments[0].isNull(position)) {
-        continue;
-      }
-      double value =
-          RateFunctionValidation.readValue(
-              arguments[0], position, valueDataType, RateFunctionType.DELTA);
-      long time =
-          RateFunctionValidation.readRequiredTime(
-              arguments[1], position, RateFunctionType.DELTA, 2);
-      long currentWindowStart =
-          RateFunctionValidation.readRequiredTime(
-              arguments[2], position, RateFunctionType.DELTA, 3);
-      long currentWindowEnd =
-          RateFunctionValidation.readRequiredTime(
-              arguments[3], position, RateFunctionType.DELTA, 4);
-      RateFunctionValidation.validateWindow(
-          RateFunctionType.DELTA, time, currentWindowStart, currentWindowEnd);
-      initializeOrValidateWindow(currentWindowStart, currentWindowEnd);
-      samples.add(time, value);
-    }
+    TypeServices.RATE_INPUT_SERVICE

Review Comment:
   NaiveDeltaAccumulator now dispatches to a native batch reader through 
RATE_INPUT_SERVICE. It preserves value validation before time/window validation 
and keeps the existing state update callback, including logical group positions 
where applicable. RateBatchInputTest exercises all 16 implementations with 
native and wrapped columns, masks/nulls and validation cases. This is one of 
the 12 retained rate candidates; the other four were restored after 
fixed-DOUBLE benchmarks showed regressions.



-- 
This is an automated message from the Apache Git Service.
To respond to the message, please log on to GitHub and use the
URL above to go to the specific comment.

To unsubscribe, e-mail: [email protected]

For queries about this service, please contact Infrastructure at:
[email protected]

Reply via email to