cmem-plugin-pgvector 0.6.2__tar.gz → 0.7.0__tar.gz

{cmem_plugin_pgvector-0.6.2 → cmem_plugin_pgvector-0.7.0}/PKG-INFO

@@ -1,6 +1,6 @@
 Metadata-Version: 2.3
 Name: cmem-plugin-pgvector
-Version: 0.6.2
+Version: 0.7.0
 Summary: Store and search for embedding vectors in a Postgres vector store.
 License: Apache-2.0
 Keywords: eccenca Corporate Memory,plugin
@@ -16,10 +16,10 @@ Classifier: Programming Language :: Python :: 3
 Classifier: Programming Language :: Python :: 3.11
 Classifier: Programming Language :: Python :: 3.12
 Classifier: Programming Language :: Python :: 3.13
-Requires-Dist: cmem-plugin-base (>=4.8.0,<5.0.0)
-Requires-Dist: langchain-community (>=0.3.2,<0.4.0)
-Requires-Dist: langchain-postgres (>=0.0.13,<0.0.14)
-Requires-Dist: psycopg[binary] (>=3.2.3,<4.0.0)
+Requires-Dist: cmem-plugin-base (>=4.12.1,<5.0.0)
+Requires-Dist: langchain-core (>=0.3.71,<0.4.0)
+Requires-Dist: langchain-postgres (>=0.0.15,<0.0.16)
+Requires-Dist: psycopg[binary] (>=3.2.9,<4.0.0)
 Description-Content-Type: text/markdown
 
 # cmem-plugin-pgvector
@@ -28,12 +28,10 @@ Store and search for embedding vectors in a Postgres vector store.
 
 [![eccenca Corporate Memory][cmem-shield]][cmem-link]
 
-This is a plugin for [eccenca](https://eccenca.com) [Corporate Memory](https://documentation.eccenca.com).
-
-You can install it with the [cmemc](https://eccenca.com/go/cmemc) command line client like this:
+This is a plugin for [eccenca](https://eccenca.com) [Corporate Memory](https://documentation.eccenca.com). You can install it with the [cmemc](https://eccenca.com/go/cmemc) command line client like this:
 
 ``` bash
-cmemc admin workspace python install cmem-plugin-llm
+cmemc admin workspace python install cmem-plugin-pgvector
 ```
 
 [![pypi version](https://img.shields.io/pypi/v/cmem-plugin-pgvector)](https://pypi.org/project/cmem-plugin-pgvector) [![license](https://img.shields.io/pypi/l/cmem-plugin-pgvector)](https://pypi.org/project/cmem-plugin-pgvector)

{cmem_plugin_pgvector-0.6.2 → cmem_plugin_pgvector-0.7.0}/README-public.md

@@ -4,12 +4,10 @@ Store and search for embedding vectors in a Postgres vector store.
 
 [![eccenca Corporate Memory][cmem-shield]][cmem-link]
 
-This is a plugin for [eccenca](https://eccenca.com) [Corporate Memory](https://documentation.eccenca.com).
-
-You can install it with the [cmemc](https://eccenca.com/go/cmemc) command line client like this:
+This is a plugin for [eccenca](https://eccenca.com) [Corporate Memory](https://documentation.eccenca.com). You can install it with the [cmemc](https://eccenca.com/go/cmemc) command line client like this:
 
 ``` bash
-cmemc admin workspace python install cmem-plugin-llm
+cmemc admin workspace python install cmem-plugin-pgvector
 ```
 
 [![pypi version](https://img.shields.io/pypi/v/cmem-plugin-pgvector)](https://pypi.org/project/cmem-plugin-pgvector) [![license](https://img.shields.io/pypi/l/cmem-plugin-pgvector)](https://pypi.org/project/cmem-plugin-pgvector)

{cmem_plugin_pgvector-0.6.2 → cmem_plugin_pgvector-0.7.0}/cmem_plugin_pgvector/search_task.py

@@ -13,13 +13,27 @@ from cmem_plugin_base.dataintegration.ports import (
     FixedNumberOfInputs,
     FixedSchemaPort,
 )
-from cmem_plugin_base.dataintegration.types import IntParameterType
+from cmem_plugin_base.dataintegration.types import EnumParameterType, IntParameterType
 from langchain_core.documents import Document
+from langchain_core.embeddings import Embeddings
 from langchain_postgres import PGVector
+from langchain_postgres.vectorstores import DistanceStrategy
 
 from cmem_plugin_pgvector.commons import DatabaseParams
 
 
+class DummyEmbeddings(Embeddings):
+    """Dummy embedding model"""
+
+    def embed_query(self, text: str) -> list[float]:
+        """Embed a query"""
+        raise NotImplementedError
+
+    def embed_documents(self, texts: list[str]) -> list[list[float]]:
+        """Embed a list of documents"""
+        raise NotImplementedError
+
+
 @Plugin(
     label="Search Vector Embeddings",
     description="Search for top-k metadata stored in Postgres Vector Store (PGVector).",
@@ -73,6 +87,14 @@ The results in this output are structured like this:
             default_value=10,
             param_type=IntParameterType(),
         ),
+        PluginParameter(
+            name="distance_strategy",
+            label="Distance Strategy",
+            description="The distance strategy to use. (default: COSINE)",
+            param_type=EnumParameterType(enum_type=DistanceStrategy),
+            default_value=DistanceStrategy.COSINE,
+            advanced=True,
+        ),
     ],
 )
 class PGVectorSearchPlugin(WorkflowPlugin):
@@ -92,6 +114,7 @@ class PGVectorSearchPlugin(WorkflowPlugin):
     report: ExecutionReport
     search_result_path: str
     top_k: int
+    distance_strategy: DistanceStrategy
 
     def __init__(  # noqa: PLR0913
         self,
@@ -104,6 +127,7 @@ class PGVectorSearchPlugin(WorkflowPlugin):
         search_result_path: str = "_search_result",
         embedding_query_path: str = "_embedding",
         top_k: int = 10,
+        distance_strategy: DistanceStrategy = DistanceStrategy.COSINE,
     ) -> None:
         self.collection_name = collection_name
         self.user = user
@@ -113,6 +137,7 @@ class PGVectorSearchPlugin(WorkflowPlugin):
         self.embedding_query_path = embedding_query_path
         self.search_result_path = search_result_path
         self.top_k = top_k
+        self.distance_strategy = distance_strategy
 
         str_password = self.password = password if isinstance(password, str) else password.decrypt()
         self.connection_string = (
@@ -126,9 +151,10 @@ class PGVectorSearchPlugin(WorkflowPlugin):
         self.db = PGVector(
             collection_name=self.collection_name,
             connection=self.connection_string,
-            embeddings=None,  # type: ignore  # noqa: PGH003
+            embeddings=DummyEmbeddings(),
             use_jsonb=True,
             pre_delete_collection=False,
+            distance_strategy=distance_strategy,
         )
         self._setup_ports()
 
@@ -170,15 +196,17 @@ class PGVectorSearchPlugin(WorkflowPlugin):
             pass
         return False
 
-    def _docs_to_json(self, docs: list[tuple[Document, float]]) -> list:
+    @staticmethod
+    def _docs_to_json(docs: list[tuple[Document, float]]) -> list:
         """Convert a list of Documents to a list of metadata"""
         doc_list: list = []
         for doc_tuple in docs:
-            json_entity = {}
-            json_entity["id"] = doc_tuple[0].id
-            json_entity["metadata"] = str(doc_tuple[0].metadata)
-            json_entity["_embedding_source"] = doc_tuple[0].page_content
-            json_entity["distance"] = str(doc_tuple[1])
+            json_entity = {
+                "id": doc_tuple[0].id,
+                "metadata": str(doc_tuple[0].metadata),
+                "_embedding_source": doc_tuple[0].page_content,
+                "distance": str(doc_tuple[1]),
+            }
             doc_list.append(json_entity)
         return doc_list
 

{cmem_plugin_pgvector-0.6.2 → cmem_plugin_pgvector-0.7.0}/pyproject.toml

@@ -1,6 +1,6 @@
 [tool.poetry]
 name = "cmem-plugin-pgvector"
-version = "0.6.2"
+version = "0.7.0"
 license = "Apache-2.0"
 description = "Store and search for embedding vectors in a Postgres vector store."
 authors = ["eccenca GmbH <cmempy-developer@eccenca.com>"]
@@ -16,27 +16,28 @@ keywords = [
 
 [tool.poetry.dependencies]# if you need to change python version here, change it also in .python-version
 python = "^3.11"
-langchain-community = "^0.3.2"
-langchain-postgres = "^0.0.13"
-psycopg = {extras = ["binary"], version = "^3.2.3"}
+langchain-postgres = "^0.0.15"
+psycopg = {extras = ["binary"], version = "^3.2.9"}
+langchain-core = "^0.3.71"
 
 [tool.poetry.dependencies.cmem-plugin-base]
-version = "^4.8.0"
+version = "^4.12.1"
 allow-prereleases = false
 
 [tool.poetry.group.dev.dependencies.cmem-cmemc]
-version = "^24.3.0"
+version = ">=24.2.0"
 
 [tool.poetry.group.dev.dependencies]
-genbadge = {extras = ["coverage"], version = "^1.1.1"}
-mypy = "^1.14.1"
-pip = "^25.0"
-pytest = "^8.3.4"
-pytest-cov = "^6.0.0"
+deptry = "^0.23.0"
+genbadge = {extras = ["coverage"], version = "^1.1.2"}
+mypy = "^1.16.1"
+pip = "^25.1.1"
+pytest = "^8.4.1"
+pytest-cov = "^6.2.1"
 pytest-dotenv = "^0.5.2"
 pytest-html = "^4.1.1"
 pytest-memray = { version = "^1.7.0",  markers = "platform_system != 'Windows'" }
-ruff = "^0.9.4"
+ruff = "^0.12.0"
 safety = "^1.10.3"
 aiohttp = "^3.10.11"
 
@@ -96,4 +97,3 @@ ignore = [
     "S101",    # use of assert detected
     "TRY003",  # Avoid specifying long messages outside the exception class
 ]
-