You are viewing a plain text version of this content. The canonical link for it is here.
Posted to commits@datasketches.apache.org by jm...@apache.org on 2024/02/06 08:29:25 UTC

(datasketches-java) branch ebpps updated: use partial item flag in serialization

This is an automated email from the ASF dual-hosted git repository.

jmalkin pushed a commit to branch ebpps
in repository https://gitbox.apache.org/repos/asf/datasketches-java.git


The following commit(s) were added to refs/heads/ebpps by this push:
     new 2d5769a7 use partial item flag in serialization
2d5769a7 is described below

commit 2d5769a7285cfe7ff6f38da49b22dce5fe479059
Author: jmalkin <78...@users.noreply.github.com>
AuthorDate: Tue Feb 6 00:29:15 2024 -0800

    use partial item flag in serialization
---
 .../org/apache/datasketches/sampling/EbppsItemsSketch.java    | 11 ++++++++---
 .../java/org/apache/datasketches/sampling/PreambleUtil.java   |  1 +
 2 files changed, 9 insertions(+), 3 deletions(-)

diff --git a/src/main/java/org/apache/datasketches/sampling/EbppsItemsSketch.java b/src/main/java/org/apache/datasketches/sampling/EbppsItemsSketch.java
index 8b44aa08..7a1a8757 100644
--- a/src/main/java/org/apache/datasketches/sampling/EbppsItemsSketch.java
+++ b/src/main/java/org/apache/datasketches/sampling/EbppsItemsSketch.java
@@ -21,6 +21,7 @@ package org.apache.datasketches.sampling;
 
 import static org.apache.datasketches.sampling.PreambleUtil.EBPPS_SER_VER;
 import static org.apache.datasketches.sampling.PreambleUtil.EMPTY_FLAG_MASK;
+import static org.apache.datasketches.sampling.PreambleUtil.HAS_PARTIAL_ITEM_MASK;
 
 import java.util.ArrayList;
 import java.util.Arrays;
@@ -121,6 +122,7 @@ public class EbppsItemsSketch<T> {
     final int familyId = PreambleUtil.extractFamilyID(srcMem);
     final int flags = PreambleUtil.extractFlags(srcMem);
     final boolean isEmpty = (flags & EMPTY_FLAG_MASK) != 0;
+    final boolean hasPartialItem = (flags & HAS_PARTIAL_ITEM_MASK) != 0;
 
     // Check values
     if (isEmpty) {
@@ -189,8 +191,11 @@ public class EbppsItemsSketch<T> {
     final List<T> itemsList = Arrays.asList(rawItems);
     final ArrayList<T> data;
     final T partialItem;
-    if (numFullItems < numTotalItems) {
-      data = new ArrayList<>(itemsList.subList(0, numFullItems));
+    if (hasPartialItem) {
+      if (numFullItems >= numTotalItems)
+        throw new SketchesArgumentException("Possible Corruption: Expected partial item but none found");
+
+        data = new ArrayList<>(itemsList.subList(0, numFullItems));
       partialItem = itemsList.get(numFullItems); // 0-based, so last item
     } else {
       data = new ArrayList<>(itemsList);
@@ -487,7 +492,7 @@ public class EbppsItemsSketch<T> {
     if (empty) {
       PreambleUtil.insertFlags(mem, EMPTY_FLAG_MASK);        // Byte 3
     } else {
-      PreambleUtil.insertFlags(mem, 0);
+      PreambleUtil.insertFlags(mem, sample_.hasPartialItem() ? HAS_PARTIAL_ITEM_MASK : 0);
     }
     PreambleUtil.insertK(mem, k_);                           // Bytes 4-7
     
diff --git a/src/main/java/org/apache/datasketches/sampling/PreambleUtil.java b/src/main/java/org/apache/datasketches/sampling/PreambleUtil.java
index 454514fe..19faa8d3 100644
--- a/src/main/java/org/apache/datasketches/sampling/PreambleUtil.java
+++ b/src/main/java/org/apache/datasketches/sampling/PreambleUtil.java
@@ -209,6 +209,7 @@ final class PreambleUtil {
   //static final int BIG_ENDIAN_FLAG_MASK = 1;
   //static final int READ_ONLY_FLAG_MASK  = 2;
   static final int EMPTY_FLAG_MASK      = 4;
+  static final int HAS_PARTIAL_ITEM_MASK = 8; // EBPPS only
   static final int GADGET_FLAG_MASK     = 128;
 
   //Other constants


---------------------------------------------------------------------
To unsubscribe, e-mail: commits-unsubscribe@datasketches.apache.org
For additional commands, e-mail: commits-help@datasketches.apache.org