Skip to content

Add HllUnion.update(HllUnion) #737

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

Open
leerho wants to merge 3 commits into
base: main
Choose a base branch
from
+55 −16
Open

Add HllUnion.update(HllUnion) #737

Show file tree
Hide file tree
Changes from all commits
Commits
Show all changes
3 commits
Select commit Hold shift + click to select a range
730f0ce
Adds union of unions.
leerho Jun 30, 2026
f98acf0
suppress printing
leerho Jun 30, 2026
16a8a31
Update PR#737
leerho Jun 30, 2026
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
10 changes: 9 additions & 1 deletion src/main/java/org/apache/datasketches/hll/HllUnion.java
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -176,7 +176,7 @@ public double getEstimate() {
checkRebuildCurMinNumKxQ(gadget);
return gadget.getEstimate();
}

/**
* Gets the effective <i>lgConfigK</i> for the HllUnion operator, which may be less than
* <i>lgMaxK</i>.
Expand Down Expand Up @@ -320,6 +320,14 @@ public void update(final HllSketch sketch) {
gadget.hllSketchImpl = unionImpl(sketch, gadget, lgMaxK);
}

/**
* Update this HllUnion operator with the given HllUnion.
* @param union the given HllUnion.
*/
public void update(final HllUnion union) {
gadget.hllSketchImpl = unionImpl(union.gadget, gadget, lgMaxK);
}

@Override
void couponUpdate(final int coupon) {
if (coupon == EMPTY) { return; }
Expand Down
61 changes: 46 additions & 15 deletions src/test/java/org/apache/datasketches/hll/UnionCaseTest.java
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -31,6 +31,8 @@
import static org.testng.Assert.assertTrue;

import java.lang.foreign.MemorySegment;
//import java.lang.invoke.MethodHandles;
//import java.lang.invoke.VarHandle;

import org.apache.datasketches.common.SketchesStateException;
import org.testng.annotations.Test;
Expand All @@ -42,8 +44,8 @@ public class UnionCaseTest {
private static final String LS = System.getProperty("line.separator");
long v = 0;
final static int maxLgK = 12;
HllSketch source;
//HllUnion union;
HllSketch skSource;
HllUnion uSource;
String hfmt = "%10s%10s%10s%10s%10s%10s%10s%10s%10s%10s%10s" + LS;
String hdr = String.format(hfmt, "caseNum","srcLgKStr","gdtLgKStr","srcType","gdtType",
"srcSeg","gdtSeg","srcMode","gdtMode","srcOoof","gdtOoof");
Expand All @@ -52,48 +54,68 @@ public class UnionCaseTest {
public void checkAllCases() {
print(hdr);
for (int i = 0; i < 24; i++) {
checkCase(i, HLL_4, false);
checkCase(i, HLL_4, false, false);
}
println("");

print(hdr);
for (int i = 0; i < 24; i++) {
checkCase(i, HLL_6, false);
checkCase(i, HLL_6, false, false);
}
println("");

print(hdr);
for (int i = 0; i < 24; i++) {
checkCase(i, HLL_8, false);
checkCase(i, HLL_8, false, false);
}
println("");

print(hdr);
for (int i = 0; i < 24; i++) {
checkCase(i, HLL_4, true);
checkCase(i, HLL_8, false, true); //srcUnion
}
println("");

print(hdr);
for (int i = 0; i < 24; i++) {
checkCase(i, HLL_4, true, false);
}
println("");

print(hdr);
for (int i = 0; i < 24; i++) {
checkCase(i, HLL_6, true);
checkCase(i, HLL_6, true, false);
}
println("");

print(hdr);
for (int i = 0; i < 24; i++) {
checkCase(i, HLL_8, true);
checkCase(i, HLL_8, true, false);
}
println("");

print(hdr);
for (int i = 0; i < 24; i++) {
checkCase(i, HLL_8, true, true); //srcUnion
}
println("");
}

private void checkCase(final int caseNum, final TgtHllType srcType, final boolean srcSeg) {
source = getSource(caseNum, srcType, srcSeg);
private void checkCase(final int caseNum, final TgtHllType srcType, final boolean srcSeg, final boolean srcUnion) {
if (srcUnion) {
uSource = getUnionSrc(caseNum);
} else {
skSource = getSkSource(caseNum, srcType, srcSeg);
}
final boolean gdtSeg = (caseNum & 1) > 0;
final HllUnion union = getUnion(caseNum, gdtSeg);
union.update(source);
if (srcUnion) {
union.update(uSource);
} else {
union.update(skSource);
}
final int totalU = getSrcCount(caseNum, maxLgK) + getUnionCount(caseNum);
output(caseNum, source, union, totalU);
output(caseNum, skSource, union, totalU);

@proost proost Jul 1, 2026

Copy link
Copy Markdown
Member

Choose a reason for hiding this comment

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

Looks like skSource -> srcUnion ? uSource : skSource correct.

}

private void output(final int caseNum, final HllSketch source, final HllUnion union, final int totalU) {
Expand Down Expand Up @@ -121,7 +143,7 @@ private void output(final int caseNum, final HllSketch source, final HllUnion un
assertTrue(err < rse, "Err: " + err + ", RSE: " + rse);
}

private HllSketch getSource(final int caseNum, final TgtHllType tgtHllType, final boolean useMemorySegment) {
private HllSketch getSkSource(final int caseNum, final TgtHllType tgtHllType, final boolean useMemorySegment) {
final int srcLgK = getSrcLgK(caseNum, maxLgK);
final int srcU = getSrcCount(caseNum, maxLgK);
if (useMemorySegment) {
Expand All @@ -131,9 +153,18 @@ private HllSketch getSource(final int caseNum, final TgtHllType tgtHllType, fina
}
}

private HllUnion getUnionSrc(final int caseNum) {
final int srcLgK = getSrcLgK(caseNum, maxLgK);
final int srcU = getSrcCount(caseNum, maxLgK);
final HllSketch sk = buildHeapSketch(srcLgK, HLL_8, srcU);
final HllUnion u = new HllUnion(maxLgK);
u.update(sk);
return u;
}

private HllUnion getUnion(final int caseNum, final boolean useMemorySegment) {
final int unionU = getUnionCount(caseNum);
return (useMemorySegment) ? buildMemorSegmentUnion(maxLgK, unionU) : buildHeapUnion(maxLgK, unionU);
return (useMemorySegment) ? buildMemorySegmentUnion(maxLgK, unionU) : buildHeapUnion(maxLgK, unionU);
}

private static int getUnionCount(final int caseNum) {
Expand Down Expand Up @@ -394,7 +425,7 @@ private HllUnion buildHeapUnion(final int lgMaxK, final int n) {
return u;
}

private HllUnion buildMemorSegmentUnion(final int lgMaxK, final int n) {
private HllUnion buildMemorySegmentUnion(final int lgMaxK, final int n) {
final int bytes = HllSketch.getMaxUpdatableSerializationBytes(lgMaxK, TgtHllType.HLL_8);
final MemorySegment wseg = MemorySegment.ofArray(new byte[bytes]);
final HllUnion u = new HllUnion(lgMaxK, wseg);
Expand Down
Loading