feature #465 [Examples][Store] Implement indexing pipeline (OskarStark)

This PR was squashed before being merged into the main branch.

Discussion
----------

[Examples][Store] Implement indexing pipeline

| Q             | A
| ------------- | ---
| Bug fix?      | no
| New feature?  | yes
| Docs?         | no
| Issues        | Fixes #429
| License       | MIT

### Needs
* symfony/ai#468

### `./runner indexer`
<img width="1832" height="664" alt="CleanShot 2025-09-08 at 17 11 53@2x" src="https://github.com/user-attachments/assets/19f32f93-f4b5-4b90-86a4-74d1623f78f1" />

### `bin/console app:blog:embed` (removed)
<img width="1780" height="306" alt="CleanShot 2025-09-08 at 17 12 45@2x" src="https://github.com/user-attachments/assets/4c8905be-f7b3-455b-8d20-112fe9f77136" />

### `bin/console app:blog:query`
<img width="1752" height="1168" alt="CleanShot 2025-09-08 at 17 13 34@2x" src="https://github.com/user-attachments/assets/45211bfb-366e-413b-b80f-33a757c63e31" />

### `bin/console ai:store:index blog`
<img width="1930" height="388" alt="CleanShot 2025-09-08 at 22 41 00@2x" src="https://github.com/user-attachments/assets/1d27d5e4-de23-4c3a-8465-a87eba79f2fc" />

Commits
-------

29349c11 [Examples][Store] Implement indexing pipeline

Author: chr-hertel

CLAUDE.md

@@ -36,7 +36,7 @@ composer install
 echo "OPENAI_API_KEY='sk-...'" > .env.local
 
 # Initialize vector store
-symfony console app:blog:embed -vv
+symfony console ai:store:index blog -vv
 
 # Test vector store
 symfony console app:blog:query

README.md

@@ -74,7 +74,7 @@ The [Chroma DB](https://www.trychroma.com/) is a vector store that is used to st
 To initialize the Chroma DB, you need to run the following command:
 
 ```shell
-symfony console app:blog:embed -vv
+symfony console ai:store:index blog -vv
 ```
 
 Now you should be able to run the test command and get some results:

config/packages/ai.yaml

@@ -59,7 +59,11 @@ ai:
                 class: 'Symfony\AI\Platform\Bridge\OpenAi\Embeddings'
                 name: !php/const Symfony\AI\Platform\Bridge\OpenAi\Embeddings::TEXT_ADA_002
     indexer:
-        default:
+        blog:
+            loader: 'App\Blog\FeedLoader'
+            source: 'https://feeds.feedburner.com/symfony/blog'
+            transformers:
+                - 'Symfony\AI\Store\Document\Transformer\TextTrimTransformer'
             vectorizer: 'ai.vectorizer.openai_embeddings'
             store: 'ai.store.chroma_db.symfonycon'
 
@@ -75,3 +79,5 @@ services:
     Symfony\AI\Agent\Toolbox\Tool\Wikipedia: ~
     Symfony\AI\Agent\Toolbox\Tool\SimilaritySearch:
         $vectorizer: '@ai.vectorizer.openai_embeddings'
+    
+    Symfony\AI\Store\Document\Transformer\TextTrimTransformer: ~

src/Blog/Command/EmbedCommand.php

@@ -11,23 +11,33 @@
 
 namespace App\Blog;
 
+use Symfony\AI\Store\Document\LoaderInterface;
+use Symfony\AI\Store\Document\Metadata;
+use Symfony\AI\Store\Document\TextDocument;
+use Symfony\AI\Store\Exception\InvalidArgumentException;
 use Symfony\Component\DomCrawler\Crawler;
 use Symfony\Component\Uid\Uuid;
 use Symfony\Contracts\HttpClient\HttpClientInterface;
 
-class FeedLoader
+final class FeedLoader implements LoaderInterface
 {
     public function __construct(
         private HttpClientInterface $httpClient,
     ) {
     }
 
     /**
-     * @return Post[]
+     * @param ?string              $source  RSS feed URL
+     * @param array<string, mixed> $options
+     *
+     * @return iterable<TextDocument>
      */
-    public function load(): array
+    public function load(?string $source, array $options = []): iterable
     {
-        $result = $this->httpClient->request('GET', 'https://feeds.feedburner.com/symfony/blog');
+        if (null === $source) {
+            throw new InvalidArgumentException('FeedLoader requires a RSS feed URL as source, null given.');
+        }
+        $result = $this->httpClient->request('GET', $source);
 
         $posts = [];
         $crawler = new Crawler($result->getContent());
@@ -44,6 +54,8 @@ public function load(): array
             );
         });
 
-        return $posts;
+        foreach ($posts as $post) {
+            yield new TextDocument($post->id, $post->toString(), new Metadata($post->toArray()));
+        }
     }
 }