Fix linting

Author: NoahStapp

libs/langchain-mongodb-retrievers/README.md

@@ -13,7 +13,7 @@ pip install -U langchain-mongodb-retriever
 ## Using MongoDBAtlasVectorSearch
 ```python
 import os
-from langchain_mongodb import MongoDBAtlasVectorSearch
+from langchain_mongodb_retrievers import MongoDBAtlasVectorSearch
 from langchain_openai import OpenAIEmbeddings
 
 # Pull MongoDB Atlas URI from environment variables

…godb/tests/integration_tests/test_mmr.py …vers/tests/integration_tests/test_mmr.pylibs/langchain-mongodb/tests/integration_tests/test_mmr.py renamed to libs/langchain-mongodb-retrievers/tests/integration_tests/test_mmr.py libs/langchain-mongodb/tests/integration_tests/test_mmr.py renamed to libs/langchain-mongodb-retrievers/tests/integration_tests/test_mmr.py

@@ -4,8 +4,8 @@
 from typing import Generator, Sequence, Union
 
 import pytest
-from langchain.chains.query_constructor.schema import AttributeInfo
-from langchain.retrievers.self_query.base import SelfQueryRetriever
+from langchain_classic.chains.query_constructor.schema import AttributeInfo
+from langchain_classic.retrievers.self_query.base import SelfQueryRetriever
 from langchain_core.documents import Document
 from langchain_mongodb import index
 from langchain_openai import AzureChatOpenAI, ChatOpenAI

libs/langchain-mongodb-retrievers/tests/integration_tests/test_retriever_selfquerying.py

@@ -9,30 +9,3 @@ pip install -U langchain-mongodb
 
 # Usage
 - [Integrate Atlas Vector Search with LangChain](https://www.mongodb.com/docs/atlas/atlas-vector-search/ai-integrations/langchain/#get-started-with-the-langchain-integration) for a walkthrough on using your first LangChain implementation with MongoDB Atlas.
-
-## Using MongoDBAtlasVectorSearch
-```python
-import os
-from langchain_mongodb import MongoDBAtlasVectorSearch
-from langchain_openai import OpenAIEmbeddings
-
-# Pull MongoDB Atlas URI from environment variables
-MONGODB_ATLAS_CONNECTION_STRING = os.environ["MONGODB_CONNECTION_STRING"]
-DB_NAME = "langchain_db"
-COLLECTION_NAME = "test"
-VECTOR_SEARCH_INDEX_NAME = "index_name"
-
-MODEL_NAME = "text-embedding-3-large"
-OPENAI_API_KEY =  os.environ["OPENAI_API_KEY"]
-
-
-vectorstore = MongoDBAtlasVectorSearch.from_connection_string(
-    connection_string=MONGODB_ATLAS_CONNECTION_STRING,
-    namespace=DB_NAME + "." + COLLECTION_NAME,
-    embedding=OpenAIEmbeddings(model=MODEL_NAME),
-    index_name=VECTOR_SEARCH_INDEX_NAME,
-)
-
-retrieved_docs = vectorstore.similarity_search(
-    "How do I deploy MongoDBAtlasVectorSearch in our production environment?")
-```

libs/langchain-mongodb/README.md

@@ -2,8 +2,7 @@
 
 import os
 from copy import deepcopy
-from time import monotonic, sleep
-from typing import Any, Dict, Generator, Iterable, List, Mapping, Optional, Union, cast
+from typing import Any, Dict, List, Mapping, Optional, cast
 
 from bson import ObjectId
 from langchain_core.callbacks.manager import (
@@ -27,9 +26,7 @@
 from pymongo.operations import SearchIndexModel
 from pymongo.results import BulkWriteResult, DeleteResult, InsertManyResult
 
-from langchain_mongodb import MongoDBAtlasVectorSearch
 from langchain_mongodb.agent_toolkit.database import MongoDBDatabase
-from langchain_mongodb.cache import MongoDBAtlasSemanticCache
 
 TIMEOUT = 120
 INTERVAL = 0.5
@@ -55,53 +52,6 @@ def create_llm() -> BaseChatModel:
     return ChatOllama(model="llama3:8b", cache=False, seed=12345)
 
 
-class PatchedMongoDBAtlasVectorSearch(MongoDBAtlasVectorSearch):
-    def bulk_embed_and_insert_texts(
-        self,
-        texts: Union[List[str], Iterable[str]],
-        metadatas: Union[List[dict], Generator[dict, Any, Any]],
-        ids: Optional[List[str]] = None,
-    ) -> List:
-        """Patched insert_texts that waits for data to be indexed before returning"""
-        ids_inserted = super().bulk_embed_and_insert_texts(texts, metadatas, ids)
-        n_docs = self.collection.count_documents({})
-        start = monotonic()
-        while monotonic() - start <= TIMEOUT:
-            if (
-                len(self.similarity_search("sandwich", k=n_docs, oversampling_factor=1))
-                == n_docs
-            ):
-                return ids_inserted
-            else:
-                sleep(INTERVAL)
-        raise TimeoutError(f"Failed to embed, insert, and index texts in {TIMEOUT}s.")
-
-    def _similarity_search_with_score(self, query_vector, **kwargs):
-        # Remove the _ids for testing purposes.
-        docs = super()._similarity_search_with_score(query_vector, **kwargs)
-        for doc, _ in docs:
-            del doc.metadata["_id"]
-        return docs
-
-    def delete(self, ids: Optional[List[str]] = None, **kwargs: Any) -> Optional[bool]:
-        ret = super().delete(ids, **kwargs)
-        n_docs = self.collection.count_documents({})
-        start = monotonic()
-        while monotonic() - start <= TIMEOUT:
-            if (
-                len(
-                    self.similarity_search(
-                        "sandwich", k=max(n_docs, 1), oversampling_factor=1
-                    )
-                )
-                == n_docs
-            ):
-                return ret
-            else:
-                sleep(INTERVAL)
-        raise TimeoutError(f"Failed to embed, insert, and index texts in {TIMEOUT}s.")
-
-
 class ConsistentFakeEmbeddings(Embeddings):
     """Fake embeddings functionality for testing."""
 
@@ -327,27 +277,6 @@ def update_one(  # type: ignore
         elif upsert:
             self._data.append({**find_query, **set_options})
 
-    def _execute_cache_aggregation_query(self, *args, **kwargs) -> List[Dict[str, Any]]:  # type: ignore
-        """Helper function only to be used for MongoDBAtlasSemanticCache Testing
-
-        Returns:
-            List[Dict[str, Any]]: Aggregation query result
-        """
-        pipeline: List[Dict[str, Any]] = args[0]
-        params = pipeline[0]["$vectorSearch"]
-        embedding = params["queryVector"]
-        # Assumes MongoDBAtlasSemanticCache.LLM == "llm_string"
-        llm_string = params["filter"][MongoDBAtlasSemanticCache.LLM]["$eq"]
-
-        acc = []
-        for document in self._data:
-            if (
-                document.get("embedding") == embedding
-                and document.get(MongoDBAtlasSemanticCache.LLM) == llm_string
-            ):
-                acc.append(document)
-        return acc
-
     def bulk_write(self, requests, **kwargs):
         upserted = []
         for ind, request in enumerate(requests):