Skip to content

prep for 5.0.2, fix KllItemsSketch level 0 sorting issue #529

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.

Sign up for GitHub

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

Merged
jmalkin merged 1 commit into from
Mar 14, 2024
Merged

prep for 5.0.2, fix KllItemsSketch level 0 sorting issue #529

Show file tree
Hide file tree
Changes from all commits
Commits
Show all changes
1 commit
Select commit
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
4 changes: 2 additions & 2 deletions pom.xml
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -33,7 +33,7 @@ under the License.

<groupId>org.apache.datasketches</groupId>
<artifactId>datasketches-java</artifactId>
<version>5.0.1</version>
<version>5.0.2</version>
<packaging>jar</packaging>

<name>${project.artifactId}</name>
Expand Down Expand Up @@ -734,7 +734,7 @@ under the License.
</pluginManagement>
</build>
</profile>

<profile>
<id>check-cpp-historical-files</id>
<build>
Expand Down
3 changes: 1 addition & 2 deletions src/main/java/org/apache/datasketches/kll/KllItemsHelper.java
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -409,7 +409,7 @@ private static <T> int[] generalItemsCompress(

// level zero might not be sorted, so we must sort it if we wish to compact it
if ((curLevel == 0) && !isLevelZeroSorted) {
Arrays.sort(inBuf, adjBeg, adjBeg + adjPop);
Arrays.sort((T[])inBuf, adjBeg, adjBeg + adjPop, comp);
}

if (popAbove == 0) { // Level above is empty, so halve up
Expand Down Expand Up @@ -486,4 +486,3 @@ private static <T> void populateItemWorkArrays(
// }

}

31 changes: 31 additions & 0 deletions src/test/java/org/apache/datasketches/kll/KllItemsSketchTest.java
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -22,6 +22,7 @@
import static java.lang.Math.ceil;
import static org.apache.datasketches.kll.KllSketch.SketchStructure.*;
import static org.apache.datasketches.kll.KllSketch.SketchType.*;
import static org.apache.datasketches.quantilescommon.LongsAsOrderableStrings.getString;
import static org.apache.datasketches.quantilescommon.QuantileSearchCriteria.EXCLUSIVE;
import static org.apache.datasketches.quantilescommon.QuantileSearchCriteria.INCLUSIVE;
import static org.testng.Assert.assertEquals;
Expand All @@ -31,6 +32,7 @@
import static org.testng.Assert.fail;

import java.util.Comparator;
import java.util.Random;

import org.apache.datasketches.common.ArrayOfStringsSerDe;
import org.apache.datasketches.common.SketchesArgumentException;
Expand All @@ -42,6 +44,7 @@
import org.apache.datasketches.quantilescommon.DoublesSortedView;
import org.apache.datasketches.quantilescommon.GenericSortedView;
import org.apache.datasketches.quantilescommon.GenericSortedViewIterator;
import org.apache.datasketches.quantilescommon.QuantilesGenericSketchIterator;
import org.testng.annotations.Test;

@SuppressWarnings("unused")
Expand Down Expand Up @@ -751,6 +754,34 @@ public void checkSortedViewAfterReset() {
try { sk.getSortedView(); fail(); } catch (SketchesArgumentException e) { }
}

@Test
//There is no guarantee that L0 is sorted after a merge.
//The issue is, during a merge, L0 must be sorted prior to a compaction to a higher level.
//Otherwise the higher levels would not be sorted properly.
public void checkL0SortDuringMerge() throws NumberFormatException {
final Random rand = new Random();
final KllItemsSketch<String> sk1 = KllItemsSketch.newHeapInstance(8, Comparator.reverseOrder(), serDe);
final KllItemsSketch<String> sk2 = KllItemsSketch.newHeapInstance(8, Comparator.reverseOrder(), serDe);
final int n = 26; //don't change this
for (int i = 1; i <= n; i++ ) {
final int j = rand.nextInt(n) + 1;
sk1.update(getString(j, 3));
sk2.update(getString(j +100, 3));
}
sk1.merge(sk2);
println(sk1.toString(true, true)); //L1 and above should be sorted in reverse. Ignore L0.
final int lvl1size = sk1.levelsArr[2] - sk1.levelsArr[1];
final QuantilesGenericSketchIterator<String> itr = sk1.iterator();
itr.next();
int prev = Integer.parseInt(itr.getQuantile().trim());
for (int i = 1; i < lvl1size; i++) {
if (itr.next()) {
int v = Integer.parseInt(itr.getQuantile().trim());
assertTrue(v <= prev);
prev = v;
}
}
}

private final static boolean enablePrinting = false;

Expand Down