CAMEL-21540: Add PGVector component for PostgreSQL vector database

bom/camel-bom/pom.xml

@@ -1902,6 +1902,11 @@
         <artifactId>camel-pgevent</artifactId>
         <version>4.19.0-SNAPSHOT</version>
       </dependency>
+      <dependency>
+        <groupId>org.apache.camel</groupId>
+        <artifactId>camel-pgvector</artifactId>
+        <version>4.19.0-SNAPSHOT</version>
+      </dependency>
       <dependency>
         <groupId>org.apache.camel</groupId>
         <artifactId>camel-pinecone</artifactId>

catalog/camel-allcomponents/pom.xml

@@ -1692,6 +1692,11 @@
             <artifactId>camel-pgevent</artifactId>
             <version>${project.version}</version>
         </dependency>
+        <dependency>
+            <groupId>org.apache.camel</groupId>
+            <artifactId>camel-pgvector</artifactId>
+            <version>${project.version}</version>
+        </dependency>
         <dependency>
             <groupId>org.apache.camel</groupId>
             <artifactId>camel-pinecone</artifactId>

catalog/camel-catalog/src/generated/resources/org/apache/camel/catalog/components.properties

@@ -295,6 +295,7 @@ paho-mqtt5
 pdf
 pg-replication-slot
 pgevent
+pgvector
 pinecone
 platform-http
 plc4x

catalog/camel-catalog/src/generated/resources/org/apache/camel/catalog/components/pgvector.json

@@ -0,0 +1,50 @@
+{
+  "component": {
+    "kind": "component",
+    "name": "pgvector",
+    "title": "PGVector",
+    "description": "Perform operations on the PostgreSQL pgvector Vector Database.",
+    "deprecated": false,
+    "firstVersion": "4.19.0",
+    "label": "database,ai",
+    "javaType": "org.apache.camel.component.pgvector.PgVectorComponent",
+    "supportLevel": "Preview",
+    "groupId": "org.apache.camel",
+    "artifactId": "camel-pgvector",
+    "version": "4.19.0-SNAPSHOT",
+    "scheme": "pgvector",
+    "extendsScheme": "",
+    "syntax": "pgvector:collection",
+    "async": false,
+    "api": false,
+    "consumerOnly": false,
+    "producerOnly": true,
+    "lenientProperties": false,
+    "browsable": false,
+    "remote": true
+  },
+  "componentProperties": {
+    "configuration": { "index": 0, "kind": "property", "displayName": "Configuration", "group": "producer", "label": "", "required": false, "type": "object", "javaType": "org.apache.camel.component.pgvector.PgVectorConfiguration", "deprecated": false, "autowired": false, "secret": false, "description": "The configuration;" },
+    "dataSource": { "index": 1, "kind": "property", "displayName": "Data Source", "group": "producer", "label": "", "required": false, "type": "object", "javaType": "javax.sql.DataSource", "deprecated": false, "deprecationNote": "", "autowired": true, "secret": false, "configurationClass": "org.apache.camel.component.pgvector.PgVectorConfiguration", "configurationField": "configuration", "description": "The DataSource to use for connecting to the PostgreSQL database with pgvector extension." },
+    "dimension": { "index": 2, "kind": "property", "displayName": "Dimension", "group": "producer", "label": "producer", "required": false, "type": "integer", "javaType": "int", "deprecated": false, "deprecationNote": "", "autowired": false, "secret": false, "defaultValue": 384, "configurationClass": "org.apache.camel.component.pgvector.PgVectorConfiguration", "configurationField": "configuration", "description": "The dimension of the vectors to store." },
+    "distanceType": { "index": 3, "kind": "property", "displayName": "Distance Type", "group": "producer", "label": "producer", "required": false, "type": "enum", "javaType": "org.apache.camel.component.pgvector.PgVectorDistanceType", "enum": [ "COSINE", "EUCLIDEAN", "INNER_PRODUCT" ], "deprecated": false, "deprecationNote": "", "autowired": false, "secret": false, "defaultValue": "COSINE", "configurationClass": "org.apache.camel.component.pgvector.PgVectorConfiguration", "configurationField": "configuration", "description": "The distance type to use for similarity search." },
+    "lazyStartProducer": { "index": 4, "kind": "property", "displayName": "Lazy Start Producer", "group": "producer", "label": "producer", "required": false, "type": "boolean", "javaType": "boolean", "deprecated": false, "autowired": false, "secret": false, "defaultValue": false, "description": "Whether the producer should be started lazy (on the first message). By starting lazy you can use this to allow CamelContext and routes to startup in situations where a producer may otherwise fail during starting and cause the route to fail being started. By deferring this startup to be lazy then the startup failure can be handled during routing messages via Camel's routing error handlers. Beware that when the first message is processed then creating and starting the producer may take a little time and prolong the total processing time of the processing." },
+    "autowiredEnabled": { "index": 5, "kind": "property", "displayName": "Autowired Enabled", "group": "advanced", "label": "advanced", "required": false, "type": "boolean", "javaType": "boolean", "deprecated": false, "autowired": false, "secret": false, "defaultValue": true, "description": "Whether autowiring is enabled. This is used for automatic autowiring options (the option must be marked as autowired) by looking up in the registry to find if there is a single instance of matching type, which then gets configured on the component. This can be used for automatic configuring JDBC data sources, JMS connection factories, AWS Clients, etc." }
+  },
+  "headers": {
+    "CamelPgVectorAction": { "index": 0, "kind": "header", "displayName": "", "group": "producer", "label": "", "required": false, "javaType": "String", "enum": [ "CREATE_TABLE", "CREATE_INDEX", "DROP_TABLE", "UPSERT", "DELETE", "SIMILARITY_SEARCH" ], "deprecated": false, "deprecationNote": "", "autowired": false, "secret": false, "description": "The action to be performed.", "constantName": "org.apache.camel.component.pgvector.PgVectorHeaders#ACTION" },
+    "CamelPgVectorRecordId": { "index": 1, "kind": "header", "displayName": "", "group": "producer", "label": "", "required": false, "javaType": "String", "deprecated": false, "deprecationNote": "", "autowired": false, "secret": false, "description": "The id of the vector record.", "constantName": "org.apache.camel.component.pgvector.PgVectorHeaders#RECORD_ID" },
+    "CamelPgVectorQueryTopK": { "index": 2, "kind": "header", "displayName": "", "group": "producer", "label": "", "required": false, "javaType": "Integer", "deprecated": false, "deprecationNote": "", "autowired": false, "secret": false, "defaultValue": "3", "description": "The maximum number of results to return for similarity search.", "constantName": "org.apache.camel.component.pgvector.PgVectorHeaders#QUERY_TOP_K" },
+    "CamelPgVectorTextContent": { "index": 3, "kind": "header", "displayName": "", "group": "producer", "label": "", "required": false, "javaType": "String", "deprecated": false, "deprecationNote": "", "autowired": false, "secret": false, "description": "The text content to store alongside the vector embedding.", "constantName": "org.apache.camel.component.pgvector.PgVectorHeaders#TEXT_CONTENT" },
+    "CamelPgVectorMetadata": { "index": 4, "kind": "header", "displayName": "", "group": "producer", "label": "", "required": false, "javaType": "String", "deprecated": false, "deprecationNote": "", "autowired": false, "secret": false, "description": "The metadata associated with the vector record, stored as JSON.", "constantName": "org.apache.camel.component.pgvector.PgVectorHeaders#METADATA" },
+    "CamelPgVectorFilter": { "index": 5, "kind": "header", "displayName": "", "group": "producer", "label": "", "required": false, "javaType": "String", "deprecated": false, "deprecationNote": "", "autowired": false, "secret": false, "description": "Filter condition for similarity search. Applied as a SQL WHERE clause on the text_content and metadata columns. Supports parameterized queries using placeholders with values provided via the CamelPgVectorFilterParams header. WARNING: When not using parameterized queries, the filter value is appended directly as SQL. Never use untrusted input as the filter value without parameterization, as this could lead to SQL injection.", "constantName": "org.apache.camel.component.pgvector.PgVectorHeaders#FILTER" },
+    "CamelPgVectorFilterParams": { "index": 6, "kind": "header", "displayName": "", "group": "producer", "label": "", "required": false, "javaType": "java.util.List", "deprecated": false, "deprecationNote": "", "autowired": false, "secret": false, "description": "Parameter values for parameterized filter queries. Use with placeholders in the CamelPgVectorFilter header. Example: filter = 'text_content LIKE AND metadata::jsonb-'category' = ' with filterParams = List.of(%hello%, science).", "constantName": "org.apache.camel.component.pgvector.PgVectorHeaders#FILTER_PARAMS" }
+  },
+  "properties": {
+    "collection": { "index": 0, "kind": "path", "displayName": "Collection", "group": "producer", "label": "", "required": true, "type": "string", "javaType": "java.lang.String", "deprecated": false, "deprecationNote": "", "autowired": false, "secret": false, "description": "The collection (table) name" },
+    "dataSource": { "index": 1, "kind": "parameter", "displayName": "Data Source", "group": "producer", "label": "", "required": false, "type": "object", "javaType": "javax.sql.DataSource", "deprecated": false, "deprecationNote": "", "autowired": true, "secret": false, "configurationClass": "org.apache.camel.component.pgvector.PgVectorConfiguration", "configurationField": "configuration", "description": "The DataSource to use for connecting to the PostgreSQL database with pgvector extension." },
+    "dimension": { "index": 2, "kind": "parameter", "displayName": "Dimension", "group": "producer", "label": "producer", "required": false, "type": "integer", "javaType": "int", "deprecated": false, "deprecationNote": "", "autowired": false, "secret": false, "defaultValue": 384, "configurationClass": "org.apache.camel.component.pgvector.PgVectorConfiguration", "configurationField": "configuration", "description": "The dimension of the vectors to store." },
+    "distanceType": { "index": 3, "kind": "parameter", "displayName": "Distance Type", "group": "producer", "label": "producer", "required": false, "type": "enum", "javaType": "org.apache.camel.component.pgvector.PgVectorDistanceType", "enum": [ "COSINE", "EUCLIDEAN", "INNER_PRODUCT" ], "deprecated": false, "deprecationNote": "", "autowired": false, "secret": false, "defaultValue": "COSINE", "configurationClass": "org.apache.camel.component.pgvector.PgVectorConfiguration", "configurationField": "configuration", "description": "The distance type to use for similarity search." },
+    "lazyStartProducer": { "index": 4, "kind": "parameter", "displayName": "Lazy Start Producer", "group": "producer (advanced)", "label": "producer,advanced", "required": false, "type": "boolean", "javaType": "boolean", "deprecated": false, "autowired": false, "secret": false, "defaultValue": false, "description": "Whether the producer should be started lazy (on the first message). By starting lazy you can use this to allow CamelContext and routes to startup in situations where a producer may otherwise fail during starting and cause the route to fail being started. By deferring this startup to be lazy then the startup failure can be handled during routing messages via Camel's routing error handlers. Beware that when the first message is processed then creating and starting the producer may take a little time and prolong the total processing time of the processing." }
+  }
+}

catalog/camel-catalog/src/generated/resources/org/apache/camel/catalog/transformers.properties

@@ -32,6 +32,8 @@ infinispan-embeddings
 milvus-embeddings
 neo4j-embeddings
 neo4j-rag
+pgvector-embeddings
+pgvector-rag
 pinecone-embeddings
 pinecone-rag
 protobuf-binary

catalog/camel-catalog/src/generated/resources/org/apache/camel/catalog/transformers/pgvector-embeddings.json

@@ -0,0 +1,14 @@
+{
+  "transformer": {
+    "kind": "transformer",
+    "name": "pgvector:embeddings",
+    "title": "Pgvector (Embeddings)",
+    "description": "Prepares the message to become an object writable by PgVector component",
+    "deprecated": false,
+    "javaType": "org.apache.camel.component.pgvector.transform.PgVectorEmbeddingsDataTypeTransformer",
+    "groupId": "org.apache.camel",
+    "artifactId": "camel-pgvector",
+    "version": "4.19.0-SNAPSHOT"
+  }
+}
+

catalog/camel-catalog/src/generated/resources/org/apache/camel/catalog/transformers/pgvector-rag.json

@@ -0,0 +1,14 @@
+{
+  "transformer": {
+    "kind": "transformer",
+    "name": "pgvector:rag",
+    "title": "Pgvector (Rag)",
+    "description": "Prepares the PgVector similarity search results to become a List of String for LangChain4j RAG",
+    "deprecated": false,
+    "javaType": "org.apache.camel.component.pgvector.transform.PgVectorReverseEmbeddingsDataTypeTransformer",
+    "groupId": "org.apache.camel",
+    "artifactId": "camel-pgvector",
+    "version": "4.19.0-SNAPSHOT"
+  }
+}
+

components/camel-ai/camel-langchain4j-embeddings/pom.xml

@@ -87,6 +87,11 @@
             <artifactId>camel-milvus</artifactId>
             <scope>test</scope>
         </dependency>
+        <dependency>
+            <groupId>org.apache.camel</groupId>
+            <artifactId>camel-pgvector</artifactId>
+            <scope>test</scope>
+        </dependency>
         <dependency>
             <groupId>org.apache.camel</groupId>
             <artifactId>camel-pinecone</artifactId>
@@ -156,6 +161,12 @@
             <version>${project.version}</version>
             <scope>test</scope>
         </dependency>
+        <dependency>
+            <groupId>org.apache.camel</groupId>
+            <artifactId>camel-test-infra-postgres</artifactId>
+            <version>${project.version}</version>
+            <scope>test</scope>
+        </dependency>
         <dependency>
             <groupId>org.apache.camel</groupId>
             <artifactId>camel-test-infra-weaviate</artifactId>

components/camel-ai/camel-langchain4j-embeddings/src/main/docs/langchain4j-embeddings-component.adoc

@@ -268,6 +268,77 @@ YAML::
 ----
 ====
 
+==== Using with PGVector
+
+This example shows how to embed text and store it in PostgreSQL with pgvector:
+
+[tabs]
+====
+Java::
++
+[source,java]
+----
+from("direct:store")
+    .to("langchain4j-embeddings:embed")
+    .setHeader(PgVectorHeaders.ACTION).constant(PgVectorAction.UPSERT)
+    .transformDataType(new DataType("pgvector:embeddings"))
+    .to("pgvector:myCollection");
+----
+
+YAML::
++
+[source,yaml]
+----
+- route:
+    from:
+      uri: "direct:store"
+    steps:
+      - to: "langchain4j-embeddings:embed"
+      - setHeader:
+          name: CamelPgVectorAction
+          constant: UPSERT
+      - transform:
+          dataType: "pgvector:embeddings"
+      - to: "pgvector:myCollection"
+----
+====
+
+Similarity search with PGVector and RAG:
+
+[tabs]
+====
+Java::
++
+[source,java]
+----
+from("direct:search")
+    .to("langchain4j-embeddings:embed")
+    .transformDataType(new DataType("pgvector:embeddings"))
+    .setHeader(PgVectorHeaders.ACTION, constant(PgVectorAction.SIMILARITY_SEARCH))
+    .to("pgvector:myCollection")
+    .transformDataType(new DataType("pgvector:rag"));
+----
+
+YAML::
++
+[source,yaml]
+----
+- route:
+    from:
+      uri: "direct:search"
+    steps:
+      - to: "langchain4j-embeddings:embed"
+      - transform:
+          dataType: "pgvector:embeddings"
+      - setHeader:
+          name: CamelPgVectorAction
+          constant: SIMILARITY_SEARCH
+      - to: "pgvector:myCollection"
+      - transform:
+          dataType: "pgvector:rag"
+----
+====
+
 ==== Using with LangChain4j Embedding Store Component
 
 For a simpler integration, use the `langchain4j-embeddingstore` component which provides a unified interface: