Skip to content

JAVA-6099 Add new fields for Auto embedding #1936

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
strogiyotec wants to merge 5 commits into
base: main
Choose a base branch
from
+296 −98
Open

JAVA-6099 Add new fields for Auto embedding #1936

Show file tree
Hide file tree
Changes from all commits
Commits
Show all changes
5 commits
Select commit Hold shift + click to select a range
9065008
JAVA-6099 Add new fields for Auto embedding
Apr 9, 2026
c140c58
disable test case
Apr 9, 2026
b674eb9
upgrade embedding test cases
Apr 9, 2026
a01a336
JAVA-6099 address PR comments
Apr 16, 2026
615a4d0
JAVA-6099 remove unused methods from tests
Apr 16, 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
217 changes: 217 additions & 0 deletions ...c/src/test/functional/com/mongodb/client/AbstractAtlasSearchIndexManagementProseTest.java
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -31,8 +31,11 @@
import org.junit.jupiter.api.Assertions;
import org.junit.jupiter.api.Assumptions;
import org.junit.jupiter.api.BeforeEach;
import org.junit.jupiter.api.Disabled;
import org.junit.jupiter.api.DisplayName;
import org.junit.jupiter.api.Test;
import org.junit.jupiter.params.ParameterizedTest;
import org.junit.jupiter.params.provider.ValueSource;

import java.util.List;
import java.util.Map;
Expand Down Expand Up @@ -79,6 +82,8 @@ public abstract class AbstractAtlasSearchIndexManagementProseTest {
"{"
+ " mappings: { dynamic: true }"
+ "}");
private static final String AUTO_EMBED_FIELD_PATH = "plot";
private static final String AUTO_EMBED_INDEX_NAME = "voyage_4";
private static final Document VECTOR_SEARCH_DEFINITION = Document.parse(
"{"
+ " fields: ["
Expand Down Expand Up @@ -281,6 +286,218 @@ public void shouldRequireExplicitTypeToCreateVectorSearchIndex() {
VECTOR_SEARCH_DEFINITION));
}

@Test
@DisplayName("should fail when invalid model name was used for auto embedding index")
void shouldFailWhenInvalidModelNameWasUsed() {
//TODO-JAVA-6059 remove this assumption when auto embedding is generally available
Assumptions.assumeTrue(false);

assertThrows(
MongoCommandException.class,
() -> createAutoEmbeddingIndex("test"),
"Valid voyage model name was not used"
);
}

@Test
@DisplayName("should fail to create auto embedding index without model")
void shouldFailToCreateAutoEmbeddingIndexWithoutModel() {
//TODO-JAVA-6059 remove this assumption when auto embedding is generally available
Assumptions.assumeTrue(false);

SearchIndexModel indexModel = new SearchIndexModel(
AUTO_EMBED_INDEX_NAME,
new Document(
"fields",
singletonList(
new Document("type", "autoEmbed")
.append("modality", "text")
.append("path", AUTO_EMBED_FIELD_PATH)
)),
SearchIndexType.vectorSearch()
);
assertThrows(
MongoCommandException.class,
() -> collection.createSearchIndexes(singletonList(indexModel)),
"Expected index creation to fail because model is not specified"
);
}

@ParameterizedTest(name = "should create auto embedding index with {0} quantization")
@ValueSource(strings = {"float", "scalar", "binary", "binaryNoRescore"})
void shouldCreateAutoEmbeddingIndexWithQuantization(final String quantization) {
//TODO-JAVA-6059 remove this assumption when auto embedding is generally available
Assumptions.assumeTrue(false);

final String indexName = AUTO_EMBED_INDEX_NAME + "_" + quantization;
SearchIndexModel indexModel = new SearchIndexModel(
indexName,
new Document(
"fields",
singletonList(
new Document("type", "autoEmbed")
.append("modality", "text")
.append("path", AUTO_EMBED_FIELD_PATH)
.append("model", "voyage-4-large")
.append("quantization", quantization)
)),
SearchIndexType.vectorSearch()
);
List<String> result = collection.createSearchIndexes(singletonList(indexModel));
Assertions.assertFalse(result.isEmpty());
}

@Test
@DisplayName("should create auto embedding index with custom numDimensions")
@Disabled("Currently numDimensions can't be used, it fails with server error:"
+ " 'Invalid numDimensions value for autoEmbed field. Expected an integer.'")
void shouldCreateAutoEmbeddingIndexWithCustomNumDimensions() {
//TODO-JAVA-6059 remove this assumption when auto embedding is generally available
Assumptions.assumeTrue(false);

SearchIndexModel indexModel = new SearchIndexModel(
AUTO_EMBED_INDEX_NAME,
new Document(
"fields",
singletonList(
new Document("type", "autoEmbed")
.append("modality", "text")
.append("path", AUTO_EMBED_FIELD_PATH)
.append("model", "voyage-4-large")
.append("numDimensions", 512)
)),
SearchIndexType.vectorSearch()
);
List<String> result = collection.createSearchIndexes(singletonList(indexModel));
Assertions.assertFalse(result.isEmpty());
}

@Test
@DisplayName("should create auto embedding index with filter field")
void shouldCreateAutoEmbeddingIndexWithFilterField() {
//TODO-JAVA-6059 remove this assumption when auto embedding is generally available
Assumptions.assumeTrue(false);

SearchIndexModel indexModel = new SearchIndexModel(
AUTO_EMBED_INDEX_NAME,
new Document(
"fields",
asList(
new Document("type", "autoEmbed")
.append("modality", "text")
.append("path", AUTO_EMBED_FIELD_PATH)
.append("model", "voyage-4-large"),
new Document("type", "filter")
.append("path", "director")
)),
SearchIndexType.vectorSearch()
);
List<String> result = collection.createSearchIndexes(singletonList(indexModel));
Assertions.assertFalse(result.isEmpty());
}

@Test
@DisplayName("should fail when mixing vector and autoEmbed types in the same index")
void shouldFailWhenMixingVectorAndAutoEmbedTypes() {
//TODO-JAVA-6059 remove this assumption when auto embedding is generally available
Assumptions.assumeTrue(false);

SearchIndexModel indexModel = new SearchIndexModel(
AUTO_EMBED_INDEX_NAME,
new Document(
"fields",
asList(
new Document("type", "autoEmbed")
.append("modality", "text")
.append("path", AUTO_EMBED_FIELD_PATH)
.append("model", "voyage-4-large"),
new Document("type", "vector")
.append("path", "plot_embedding")
.append("numDimensions", 1024)
.append("similarity", "cosine")
)),
SearchIndexType.vectorSearch()
);
assertThrows(
MongoCommandException.class,
() -> collection.createSearchIndexes(singletonList(indexModel)),
"Expected index creation to fail because vector and autoEmbed types cannot be mixed"
);
}

@Test
@DisplayName("should fail when duplicate paths are used in auto embedding index")
void shouldFailWhenDuplicatePathsAreUsed() {
//TODO-JAVA-6059 remove this assumption when auto embedding is generally available
Assumptions.assumeTrue(false);

SearchIndexModel indexModel = new SearchIndexModel(
AUTO_EMBED_INDEX_NAME,
new Document(
"fields",
asList(
new Document("type", "autoEmbed")
.append("modality", "text")
.append("path", AUTO_EMBED_FIELD_PATH)
.append("model", "voyage-4-large"),
new Document("type", "autoEmbed")
.append("modality", "text")
.append("path", AUTO_EMBED_FIELD_PATH)
.append("model", "voyage-4-large")
)),
SearchIndexType.vectorSearch()
);
assertThrows(
MongoCommandException.class,
() -> collection.createSearchIndexes(singletonList(indexModel)),
"Expected index creation to fail because of duplicate paths"
);
}

@Test
@DisplayName("should fail when autoEmbed field is used as filter field")
void shouldFailWhenAutoEmbedFieldUsedAsFilterField() {
//TODO-JAVA-6059 remove this assumption when auto embedding is generally available
Assumptions.assumeTrue(false);

SearchIndexModel indexModel = new SearchIndexModel(
AUTO_EMBED_INDEX_NAME,
new Document(
"fields",
asList(
new Document("type", "autoEmbed")
.append("modality", "text")
.append("path", AUTO_EMBED_FIELD_PATH)
.append("model", "voyage-4-large"),
new Document("type", "filter")
.append("path", AUTO_EMBED_FIELD_PATH)
)),
SearchIndexType.vectorSearch()
);
assertThrows(
MongoCommandException.class,
() -> collection.createSearchIndexes(singletonList(indexModel)),
"Expected index creation to fail because autoEmbed field cannot be used as a filter field"
);
}

private void createAutoEmbeddingIndex(final String modelName) {
SearchIndexModel indexModel = new SearchIndexModel(
AUTO_EMBED_INDEX_NAME,
new Document(
"fields",
singletonList(
new Document("type", "autoEmbed")
.append("modality", "text")
.append("model", modelName)
.append("path", AUTO_EMBED_FIELD_PATH)
)),
SearchIndexType.vectorSearch()
);
List<String> result = collection.createSearchIndexes(singletonList(indexModel));
Assertions.assertFalse(result.isEmpty());
}

private void assertIndexDeleted() throws InterruptedException {
int attempts = MAX_WAIT_ATTEMPTS;
while (collection.listSearchIndexes().first() != null && checkAttempt(attempts--)) {
Expand Down
Loading