TimeWeightedVectorStoreRetriever keeps documents in memory after vector store insertion fails

Submission checklist

• This is a bug, not a usage question.
• I added a clear and descriptive title that summarizes this issue.
• I used the GitHub search to find a similar question and didn't find it.
• I am sure that this is a bug in LangChain rather than my code.
• The bug is not resolved by updating to the latest stable version of LangChain (or the specific integration package).
• This is not related to the langchain-community package.
• I posted a self-contained, minimal, reproducible example. A maintainer can copy it and run it AS IS.

Package (Required)

• langchain
• langchain-openai
• langchain-anthropic
• langchain-classic
• langchain-core
• langchain-model-profiles
• langchain-tests
• langchain-text-splitters
• langchain-chroma
• langchain-deepseek
• langchain-exa
• langchain-fireworks
• langchain-groq
• langchain-huggingface
• langchain-mistralai
• langchain-nomic
• langchain-ollama
• langchain-openrouter
• langchain-perplexity
• langchain-qdrant
• langchain-xai
• Other / not sure / general

Related Issues / PRs

No response

Reproduction Steps / Example Code (Python)

from typing import Any

from langchain_classic.retrievers.time_weighted_retriever import (
    TimeWeightedVectorStoreRetriever,
)
from langchain_core.documents import Document
from langchain_core.embeddings import Embeddings
from langchain_core.vectorstores import VectorStore


class FakeEmbeddings(Embeddings):
    def embed_documents(self, texts: list[str]) -> list[list[float]]:
        return [[0.0] for _ in texts]

    def embed_query(self, text: str) -> list[float]:
        return [0.0]


class FailingVectorStore(VectorStore):
    @property
    def embeddings(self) -> Embeddings:
        return FakeEmbeddings()

    def add_texts(
        self,
        texts: list[str],
        metadatas: list[dict] | None = None,
        **kwargs: Any,
    ) -> list[str]:
        raise RuntimeError("synthetic vector store failure")

    def similarity_search(
        self,
        query: str,
        k: int = 4,
        **kwargs: Any,
    ) -> list[Document]:
        return []


retriever = TimeWeightedVectorStoreRetriever(
    vectorstore=FailingVectorStore(),
)

document = Document(page_content="ghost memory")

try:
    retriever.add_documents([document])
except RuntimeError as error:
    print(error)

print(
    "memory_stream after failure:",
    [doc.page_content for doc in retriever.memory_stream],
)

Error Message and Stack Trace (if applicable)

synthetic vector store failure
memory_stream after failure: ['ghost memory']

Description

TimeWeightedVectorStoreRetriever.add_documents() adds documents to memory_stream before writing them to the vector store.

If vectorstore.add_documents() raises an exception, the method exits while the documents remain in memory_stream. The retriever therefore contains a document that was never successfully stored in the vector store.

The relevant order of operations is:

for i, doc in enumerate(dup_docs):
    if "last_accessed_at" not in doc.metadata:
        doc.metadata["last_accessed_at"] = current_time
    if "created_at" not in doc.metadata:
        doc.metadata["created_at"] = current_time
    doc.metadata["buffer_idx"] = len(self.memory_stream) + i

self.memory_stream.extend(dup_docs)
return self.vectorstore.add_documents(dup_docs, **kwargs)

### System Info

System Information
------------------
> OS:  Windows
> OS Version:  10.0.26200
> Python Version:  3.13.3 (main, May 30 2025, 05:37:00) [MSC v.1943 64 bit (AMD64)]

Package Information
-------------------
> langchain_core: 1.6.3
> langsmith: 0.12.5
> langchain_classic: 1.0.8
> langchain_huggingface: 1.2.2
> langchain_protocol: 0.0.19
> langchain_text_splitters: 1.1.2

Optional packages not installed
-------------------------------
> deepagents
> deepagents-cli

Other Dependencies
------------------
> anyio: 4.15.1
> distro: 1.9.0
> httpx: 0.28.1
> httpx2: 2.13.0
> huggingface-hub: 1.31.0
> jsonpatch: 1.33
> orjson: 3.12.0
> packaging: 26.3
> pydantic: 2.13.5
> pyyaml: 6.0.3
> requests: 2.34.2
> requests-toolbelt: 1.0.0
> sniffio: 1.3.1
> sqlalchemy: 2.0.54
> tenacity: 9.1.4
> tokenizers: 0.23.2
> typing-extensions: 4.16.0
> uuid-utils: 0.17.1
> websockets: 17.1
> xxhash: 4.0.1
> zstandard: 0.25.0

### Social handles (optional)

_No response_