Fix OR filters returning duplicate documents when using multiple indexes

[Copilot]

OR filters with multiple indexed fields return the same document multiple times when it matches multiple clauses through different indexes. For example:

NitriteCollection items = db.getCollection("items");
items.createIndex(IndexOptions.indexOptions(IndexType.NON_UNIQUE), "field_a");
items.createIndex(IndexOptions.indexOptions(IndexType.NON_UNIQUE), "field_b");

Document doc = Document.createDocument();
doc.put("field_a", "A");
doc.put("field_b", "B");
items.insert(doc);

Filter orFilter = Filter.or(
    FluentFilter.where("field_a").eq("A"),
    FluentFilter.where("field_b").eq("B")
);

DocumentCursor cursor = items.find(orFilter);
// Returns 2 results instead of 1 - same document appears twice

Changes

• ReadOperations.java: Apply DistinctStream unconditionally for OR filters with subplans to deduplicate by NitriteId
• IssueTest.java: Add test case validating single document matching multiple OR clauses returns once
• CollectionFindByCompoundIndexTest.java: Update tests expecting duplicate behavior to reflect correct OR semantics

The fix ensures OR filters implement proper set union semantics where each document appears at most once regardless of how many conditions it satisfies.

Original prompt

---

This section details on the original issue you should resolve

<issue_title>Multiple indexes make records appear multiple times in result</issue_title>
<issue_description>Using 'or' with filters for fields in diffent indexes will cause the matching document to be produced once for each index.

NitriteCollection items = db.getCollection("items");
items.createIndex(IndexOptions.indexOptions(IndexType.NON_UNIQUE), "field_a");
items.createIndex(IndexOptions.indexOptions(IndexType.NON_UNIQUE), "field_b");
  	
Document doc = Document.createDocument();
doc.put("field_a", "A");
doc.put("field_b", "B");
items.insert(doc);

Filter aFilter = FluentFilter.where("field_a").eq("A");
Filter bFilter = FluentFilter.where("field_b").eq("B");
  	
Filter orFilter = Filter.or(aFilter, bFilter);

DocumentCursor cursor = items.find(orFilter);
Iterator<Document> docIter = cursor.iterator();
  	
List<Long> matches = new ArrayList<>();
while (docIter.hasNext()) {
    Document match = docIter.next();
    long id = match.getId().getIdValue();
    matches.add(id);
}
assertEquals("Single document must yiels single match", 1, matches.size());</issue_description>

Comments on the Issue (you are @copilot in this section)

• Fixes Multiple indexes make records appear multiple times in result #1168

---

✨ Let Copilot coding agent set things up for you — coding agent works faster and does higher quality work when set up for your repo.

[coderabbitai]

Important

Review skipped

Bot user detected.

To trigger a single review, invoke the @coderabbitai review command.

You can disable this status message by setting the reviews.review_status to false in the CodeRabbit configuration file.

---

_{Comment @coderabbitai help to get the list of available commands and usage tips.}

[Copilot]

Pull Request Overview

This PR fixes a bug where OR filters with multiple indexed fields incorrectly return duplicate documents when a single document matches multiple OR clauses through different indexes. The fix ensures proper set union semantics by always deduplicating results for OR filters.

• Applied DistinctStream unconditionally for OR filters with subplans in ReadOperations.java
• Added regression test in IssueTest.java to validate the fix
• Updated existing tests that incorrectly expected duplicate behavior

Reviewed Changes

Copilot reviewed 5 out of 5 changed files in this pull request and generated 1 comment.

Show a summary per file

File	Description
nitrite/src/main/java/org/dizitart/no2/collection/operation/ReadOperations.java	Removed conditional check and always apply DistinctStream for OR filters to prevent duplicates
nitrite/src/test/java/org/dizitart/no2/integration/collection/IssueTest.java	Added test case reproducing the reported issue with OR filters across multiple indexes
nitrite/src/test/java/org/dizitart/no2/integration/collection/CollectionFindByCompoundIndexTest.java	Updated expected result counts from 5 to 3 and 3 to 2 to reflect correct deduplication behavior
nitrite-rocksdb-adapter/src/test/java/org/dizitart/no2/integration/collection/CollectionFindByCompoundIndexTest.java	Updated expected result count from 5 to 3 for OR filter test
nitrite-mvstore-adapter/src/test/java/org/dizitart/no2/integration/collection/CollectionFindByCompoundIndexTest.java	Updated expected result count from 5 to 3 for OR filter test

[Copilot]

Calling cursor.toList() twice (lines 106 and 110) iterates through the cursor multiple times and creates multiple list copies. Since cursors are typically forward-only, cache the list in a variable and reuse it for both filter operations to improve performance.