feat: add pinecone store (#55)

Author: chr-hertel

README.md

@@ -46,7 +46,7 @@ Supported Stores
 * [x] [ChromaDB](https://trychroma.com)
 * [x] [Azure AI Search](https://azure.microsoft.com/en-us/products/ai-services/ai-search)
 * [x] [MongoDB Atlas Search](https://mongodb.com/products/platform/atlas-vector-search)
-* [ ] [Pinecone](https://pinecone.io)
+* [x] [Pinecone](https://pinecone.io)
 
 Provided Tools
 --------------

composer.json

@@ -38,8 +38,9 @@
         "symfony/var-dumper": "^6.4 || ^7.1"
     },
     "suggest": {
-        "mongodb/mongodb": "For using MongoDB Atlas as retrieval vector store.",
         "codewithkyrian/chromadb-php": "For using the ChromaDB as retrieval vector store.",
+        "mongodb/mongodb": "For using MongoDB Atlas as retrieval vector store.",
+        "probots-io/pinecone-php": "For using the Pinecone as retrieval vector store.",
         "symfony/clock": "For using the clock tool.",
         "symfony/css-selector": "For using the YouTube transcription tool.",
         "symfony/dom-crawler": "For using the YouTube transcription tool."

src/Document/Vector.php

@@ -9,7 +9,7 @@ final class Vector
     /**
      * @param list<float> $data
      */
-    private function __construct(
+    public function __construct(
         private readonly array $data,
         private ?int $dimensions = null,
     ) {

src/Store/MongoDB/Store.php

@@ -126,7 +126,7 @@ public function query(Vector $vector, array $options = []): array
 
         foreach ($results as $result) {
             $documents[] = Document::fromVector(
-                Vector::create1536($result[$this->vectorFieldName]),
+                new Vector($result[$this->vectorFieldName]),
                 $this->toUuid($result['_id']),
                 new Metadata($result['metadata'] ?? []),
             );

src/Store/Pinecone/Store.php

@@ -0,0 +1,80 @@
+<?php
+
+declare(strict_types=1);
+
+namespace PhpLlm\LlmChain\Store\Pinecone;
+
+use PhpLlm\LlmChain\Document\Document;
+use PhpLlm\LlmChain\Document\Metadata;
+use PhpLlm\LlmChain\Document\Vector;
+use PhpLlm\LlmChain\Store\VectorStoreInterface;
+use Probots\Pinecone\Client;
+use Probots\Pinecone\Resources\Data\VectorResource;
+use Psr\Log\LoggerInterface;
+use Symfony\Component\Uid\Uuid;
+
+final readonly class Store implements VectorStoreInterface
+{
+    /**
+     * @param array<string, mixed> $filter
+     */
+    public function __construct(
+        private Client $pinecone,
+        private LoggerInterface $logger,
+        private ?string $namespace = null,
+        private array $filter = [],
+        private int $topK = 3,
+    ) {
+    }
+
+    public function addDocument(Document $document): void
+    {
+        $this->addDocuments([$document]);
+    }
+
+    public function addDocuments(array $documents): void
+    {
+        $vectors = [];
+        foreach ($documents as $document) {
+            if (!$document->hasVector()) {
+                $this->logger->warning('Document {id} does not have a vector', ['id' => $document->id]);
+                continue;
+            }
+
+            $vectors[] = [
+                'id' => (string) $document->id,
+                'values' => $document->vector->getData(),
+                'metadata' => $document->metadata->getArrayCopy(),
+            ];
+        }
+
+        $this->getVectors()->upsert($vectors);
+    }
+
+    public function query(Vector $vector, array $options = []): array
+    {
+        $response = $this->getVectors()->query(
+            vector: $vector->getData(),
+            namespace: $options['namespace'] ?? $this->namespace,
+            filter: $options['filter'] ?? $this->filter,
+            topK: $options['topK'] ?? $this->topK,
+            includeValues: true,
+        );
+
+        $documents = [];
+        foreach ($response->json()['matches'] as $match) {
+            $documents[] = Document::fromVector(
+                new Vector($match['values']),
+                Uuid::fromString($match['id']),
+                new Metadata($match['metadata']),
+            );
+        }
+
+        return $documents;
+    }
+
+    private function getVectors(): VectorResource
+    {
+        return $this->pinecone->data()->vectors();
+    }
+}