README.md

 

Java library for the Qdrant vector search engine.

Qdrant Java Client

Java client library with handy utility methods and overloads for interfacing with Qdrant.

📥 Installation

Not yet published.

Important

Requires Java 8 or above.

To install the library, add the following lines to your build config file.

Maven

<dependency>
  <groupId>io.qdrant</groupId>
  <artifactId>client</artifactId>
  <version>1.0</version>
</dependency>

Scala SBT

libraryDependencies += "io.qdrant" % "client" % "1.0"

Gradle

implementation 'io.qdrant:client:1.0'

📖 Documentation

• QdrantClient Reference
• Utility Methods Reference

🔌 Connecting to Qdrant

Note

The library uses Qdrant's GRPC interface. The default port being 6334.

Uses TLS if the URL protocol is https, plaintext otherwise.

Connecting to a local Qdrant instance

import io.qdrant.client.QdrantClient;

QdrantClient client = new QdrantClient("http://localhost:6334");

Connecting to Qdrant cloud

import io.qdrant.client.QdrantClient;

QdrantClient client = new QdrantClient("https://xyz-eg.eu-central.aws.cloud.qdrant.io:6334", "<your-api-key>");

🧪 Example Usage

Click to expand example

You can connect to Qdrant by instantiating a QdrantClient instance.

import io.qdrant.client.QdrantClient;

QdrantClient client = new QdrantClient("http://localhost:6334");

System.out.println(client.listCollections());

Output:

collections {
name: "Documents"
}
collections {
name: "some_collection"
}
time: 7.04541E-4

We can now perform operations on the DB. Like creating a collection, adding a point.

The library offers handy utility methods for constructing GRPC structures.

import java.util.Arrays;
import java.util.HashMap;
import java.util.List;
import java.util.Map;

import io.qdrant.client.utils.*;

String collectionName = "Documents";

client.recreateCollection(collectionName, 6, Distance.Cosine);

Map<String, Object> map = new HashMap<>();
map.put("name", "John Doe");
map.put("age", 42);
map.put("married", true);

PointStruct point =
  PointUtil.point(
    0,
    VectorUtil.toVector(0.0f, 0.1f, 0.2f, 0.3f, 0.4f, 0.5f),
    PayloadUtil.toPayload(map));
List<PointStruct> points = Arrays.asList(point);
client.upsertPoints(collectionName, points, null);

Performing a search on the vectors with filtering

import io.qdrant.client.grpc.Points.Filter;
import io.qdrant.client.grpc.Points.SearchPoints;
import io.qdrant.client.grpc.Points.SearchResponse;

import io.qdrant.client.utils.*;

Filter filter = FilterUtil.must(FilterUtil.fieldCondition("age", FilterUtil.match(42)));
      
SearchPoints request = SearchPoints.newBuilder()
        .setCollectionName(collectionName)
        .addAllVector(Arrays.asList(0.0f, 0.1f, 0.2f, 0.3f, 0.4f, 0.5f))
        .setFilter(filter)
        .setWithPayload(SelectorUtil.withPayload())
        .setLimit(10)
        .build();
SearchResponse result = client.searchPoints(request);

ScoredPoint result = results.getResult(0);
            
System.out.println("Similarity: " + result.getScore());
System.out.println("Payload: " + PayloadUtil.toMap(result.getPayload()));

Output:

Similarity: 0.9999999
Payload: {name=John Doe, married=true, age=42}

⚖️ LICENSE

Apache 2.0 © 2023