From 82b08045fd3c2348978fd6ba79b881ac0b3d8a34 Mon Sep 17 00:00:00 2001
From: Jim Ferenczi <jimczi@apache.org>
Date: Thu, 3 Oct 2024 12:21:43 +0100
Subject: [PATCH] Prevent flattening of ordered and unordered interval sources
 (#13819)

This commit removes the flattening of ordered and unordered interval sources, as it alters the gap visibility for parent intervals. For example, ordered("a", ordered("b", "c")) should result in a different gap compared to ordered("a", "b", "c").

Phrase/Block operators will continue to flatten their sub-sources since this does not affect the inner gap (which is always 0 in the case of blocks).
---
 .../intervals/BlockIntervalsSource.java       |  1 +
 .../intervals/OrderedIntervalsSource.java     | 14 +----------
 .../intervals/UnorderedIntervalsSource.java   | 14 +----------
 .../queries/intervals/TestIntervalQuery.java  | 24 +++++++++++++++++++
 4 files changed, 27 insertions(+), 26 deletions(-)

diff --git a/lucene/queries/src/java/org/apache/lucene/queries/intervals/BlockIntervalsSource.java b/lucene/queries/src/java/org/apache/lucene/queries/intervals/BlockIntervalsSource.java
index 91a5e94d5f2b..efb03b154ef3 100644
--- a/lucene/queries/src/java/org/apache/lucene/queries/intervals/BlockIntervalsSource.java
+++ b/lucene/queries/src/java/org/apache/lucene/queries/intervals/BlockIntervalsSource.java
@@ -38,6 +38,7 @@ private static List<IntervalsSource> flatten(List<IntervalsSource> sources) {
     List<IntervalsSource> flattened = new ArrayList<>();
     for (IntervalsSource s : sources) {
       if (s instanceof BlockIntervalsSource) {
+        // Block sources can be flattened because they do not increase the gap (gap = 0)
         flattened.addAll(((BlockIntervalsSource) s).subSources);
       } else {
         flattened.add(s);
diff --git a/lucene/queries/src/java/org/apache/lucene/queries/intervals/OrderedIntervalsSource.java b/lucene/queries/src/java/org/apache/lucene/queries/intervals/OrderedIntervalsSource.java
index 65fa6d033957..a89df974b662 100644
--- a/lucene/queries/src/java/org/apache/lucene/queries/intervals/OrderedIntervalsSource.java
+++ b/lucene/queries/src/java/org/apache/lucene/queries/intervals/OrderedIntervalsSource.java
@@ -30,25 +30,13 @@ static IntervalsSource build(List<IntervalsSource> sources) {
     if (sources.size() == 1) {
       return sources.get(0);
     }
-    List<IntervalsSource> rewritten = deduplicate(flatten(sources));
+    List<IntervalsSource> rewritten = deduplicate(sources);
     if (rewritten.size() == 1) {
       return rewritten.get(0);
     }
     return new OrderedIntervalsSource(rewritten);
   }
 
-  private static List<IntervalsSource> flatten(List<IntervalsSource> sources) {
-    List<IntervalsSource> flattened = new ArrayList<>();
-    for (IntervalsSource s : sources) {
-      if (s instanceof OrderedIntervalsSource) {
-        flattened.addAll(((OrderedIntervalsSource) s).subSources);
-      } else {
-        flattened.add(s);
-      }
-    }
-    return flattened;
-  }
-
   private static List<IntervalsSource> deduplicate(List<IntervalsSource> sources) {
     List<IntervalsSource> deduplicated = new ArrayList<>();
     List<IntervalsSource> current = new ArrayList<>();
diff --git a/lucene/queries/src/java/org/apache/lucene/queries/intervals/UnorderedIntervalsSource.java b/lucene/queries/src/java/org/apache/lucene/queries/intervals/UnorderedIntervalsSource.java
index 132ab4b3976e..d2c708b4feed 100644
--- a/lucene/queries/src/java/org/apache/lucene/queries/intervals/UnorderedIntervalsSource.java
+++ b/lucene/queries/src/java/org/apache/lucene/queries/intervals/UnorderedIntervalsSource.java
@@ -33,7 +33,7 @@ static IntervalsSource build(List<IntervalsSource> sources) {
     if (sources.size() == 1) {
       return sources.get(0);
     }
-    List<IntervalsSource> rewritten = deduplicate(flatten(sources));
+    List<IntervalsSource> rewritten = deduplicate(sources);
     if (rewritten.size() == 1) {
       return rewritten.get(0);
     }
@@ -55,18 +55,6 @@ private static List<IntervalsSource> deduplicate(List<IntervalsSource> sources)
     return deduplicated;
   }
 
-  private static List<IntervalsSource> flatten(List<IntervalsSource> sources) {
-    List<IntervalsSource> flattened = new ArrayList<>();
-    for (IntervalsSource s : sources) {
-      if (s instanceof UnorderedIntervalsSource) {
-        flattened.addAll(((UnorderedIntervalsSource) s).subSources);
-      } else {
-        flattened.add(s);
-      }
-    }
-    return flattened;
-  }
-
   private UnorderedIntervalsSource(List<IntervalsSource> sources) {
     super(sources);
   }
diff --git a/lucene/queries/src/test/org/apache/lucene/queries/intervals/TestIntervalQuery.java b/lucene/queries/src/test/org/apache/lucene/queries/intervals/TestIntervalQuery.java
index af2ec0230c23..a0539a0f8520 100644
--- a/lucene/queries/src/test/org/apache/lucene/queries/intervals/TestIntervalQuery.java
+++ b/lucene/queries/src/test/org/apache/lucene/queries/intervals/TestIntervalQuery.java
@@ -338,6 +338,18 @@ public void testNestedOrInUnorderedMaxGaps() throws IOException {
     checkHits(q, new int[] {6, 7});
   }
 
+  public void testUnorderedWithNoGap() throws IOException {
+    Query q =
+        new IntervalQuery(
+            field,
+            Intervals.maxgaps(
+                0,
+                Intervals.unordered(
+                    Intervals.term("w3"),
+                    Intervals.unordered(Intervals.term("w1"), Intervals.term("w5")))));
+    checkHits(q, new int[] {0});
+  }
+
   public void testOrderedWithGaps() throws IOException {
     Query q =
         new IntervalQuery(
@@ -360,6 +372,18 @@ public void testOrderedWithGaps2() throws IOException {
     checkHits(q, new int[] {12});
   }
 
+  public void testOrderedWithNoGap() throws IOException {
+    Query q =
+        new IntervalQuery(
+            field,
+            Intervals.maxgaps(
+                0,
+                Intervals.ordered(
+                    Intervals.ordered(Intervals.term("w1"), Intervals.term("w4")),
+                    Intervals.term("w5"))));
+    checkHits(q, new int[] {0});
+  }
+
   public void testNestedOrInContainedBy() throws IOException {
     Query q =
         new IntervalQuery(