style: more refs fixes (#33730)

Author: mdrxy

libs/core/langchain_core/caches.py

@@ -2,8 +2,8 @@
 
 Distinct from provider-based [prompt caching](https://docs.langchain.com/oss/python/langchain/models#prompt-caching).
 
-!!! warning
-    This is a beta feature! Please be wary of deploying experimental code to production
+!!! warning "Beta feature"
+    This is a beta feature. Please be wary of deploying experimental code to production
     unless you've taken appropriate precautions.
 
 A cache is useful for two reasons:
@@ -49,17 +49,18 @@ def lookup(self, prompt: str, llm_string: str) -> RETURN_VAL_TYPE | None:
         """Look up based on `prompt` and `llm_string`.
 
         A cache implementation is expected to generate a key from the 2-tuple
-        of prompt and llm_string (e.g., by concatenating them with a delimiter).
+        of `prompt` and `llm_string` (e.g., by concatenating them with a delimiter).
 
         Args:
             prompt: A string representation of the prompt.
                 In the case of a chat model, the prompt is a non-trivial
                 serialization of the prompt into the language model.
             llm_string: A string representation of the LLM configuration.
+
                 This is used to capture the invocation parameters of the LLM
                 (e.g., model name, temperature, stop tokens, max tokens, etc.).
-                These invocation parameters are serialized into a string
-                representation.
+
+                These invocation parameters are serialized into a string representation.
 
         Returns:
             On a cache miss, return `None`. On a cache hit, return the cached value.
@@ -78,8 +79,10 @@ def update(self, prompt: str, llm_string: str, return_val: RETURN_VAL_TYPE) -> N
                 In the case of a chat model, the prompt is a non-trivial
                 serialization of the prompt into the language model.
             llm_string: A string representation of the LLM configuration.
+
                 This is used to capture the invocation parameters of the LLM
                 (e.g., model name, temperature, stop tokens, max tokens, etc.).
+
                 These invocation parameters are serialized into a string
                 representation.
             return_val: The value to be cached. The value is a list of `Generation`
@@ -94,15 +97,17 @@ async def alookup(self, prompt: str, llm_string: str) -> RETURN_VAL_TYPE | None:
         """Async look up based on `prompt` and `llm_string`.
 
         A cache implementation is expected to generate a key from the 2-tuple
-        of prompt and llm_string (e.g., by concatenating them with a delimiter).
+        of `prompt` and `llm_string` (e.g., by concatenating them with a delimiter).
 
         Args:
             prompt: A string representation of the prompt.
                 In the case of a chat model, the prompt is a non-trivial
                 serialization of the prompt into the language model.
             llm_string: A string representation of the LLM configuration.
+
                 This is used to capture the invocation parameters of the LLM
                 (e.g., model name, temperature, stop tokens, max tokens, etc.).
+
                 These invocation parameters are serialized into a string
                 representation.
 
@@ -125,8 +130,10 @@ async def aupdate(
                 In the case of a chat model, the prompt is a non-trivial
                 serialization of the prompt into the language model.
             llm_string: A string representation of the LLM configuration.
+
                 This is used to capture the invocation parameters of the LLM
                 (e.g., model name, temperature, stop tokens, max tokens, etc.).
+
                 These invocation parameters are serialized into a string
                 representation.
             return_val: The value to be cached. The value is a list of `Generation`

libs/core/langchain_core/documents/__init__.py

@@ -1,7 +1,28 @@
-"""Documents module.
+"""Documents module for data retrieval and processing workflows.
 
-**Document** module is a collection of classes that handle documents
-and their transformations.
+This module provides core abstractions for handling data in retrieval-augmented
+generation (RAG) pipelines, vector stores, and document processing workflows.
+
+!!! warning "Documents vs. message content"
+    This module is distinct from `langchain_core.messages.content`, which provides
+    multimodal content blocks for **LLM chat I/O** (text, images, audio, etc. within
+    messages).
+
+    **Key distinction:**
+
+    - **Documents** (this module): For **data retrieval and processing workflows**
+        - Vector stores, retrievers, RAG pipelines
+        - Text chunking, embedding, and semantic search
+        - Example: Chunks of a PDF stored in a vector database
+
+    - **Content Blocks** (`messages.content`): For **LLM conversational I/O**
+        - Multimodal message content sent to/from models
+        - Tool calls, reasoning, citations within chat
+        - Example: An image sent to a vision model in a chat message (via
+            [`ImageContentBlock`][langchain.messages.ImageContentBlock])
+
+    While both can represent similar data types (text, files), they serve different
+    architectural purposes in LangChain applications.
 """
 
 from typing import TYPE_CHECKING

libs/core/langchain_core/documents/base.py

@@ -1,4 +1,16 @@
-"""Base classes for media and documents."""
+"""Base classes for media and documents.
+
+This module contains core abstractions for **data retrieval and processing workflows**:
+
+- `BaseMedia`: Base class providing `id` and `metadata` fields
+- `Blob`: Raw data loading (files, binary data) - used by document loaders
+- `Document`: Text content for retrieval (RAG, vector stores, semantic search)
+
+!!! note "Not for LLM chat messages"
+    These classes are for data processing pipelines, not LLM I/O. For multimodal
+    content in chat messages (images, audio in conversations), see
+    `langchain.messages` content blocks instead.
+"""
 
 from __future__ import annotations
 
@@ -19,15 +31,13 @@
 
 
 class BaseMedia(Serializable):
-    """Use to represent media content.
-
-    Media objects can be used to represent raw data, such as text or binary data.
+    """Base class for content used in retrieval and data processing workflows.
 
-    LangChain Media objects allow associating metadata and an optional identifier
-    with the content.
+    Provides common fields for content that needs to be stored, indexed, or searched.
 
-    The presence of an ID and metadata make it easier to store, index, and search
-    over the content in a structured way.
+    !!! note
+        For multimodal content in **chat messages** (images, audio sent to/from LLMs),
+        use `langchain.messages` content blocks instead.
     """
 
     # The ID field is optional at the moment.
@@ -45,61 +55,60 @@ class BaseMedia(Serializable):
 
 
 class Blob(BaseMedia):
-    """Blob represents raw data by either reference or value.
+    """Raw data abstraction for document loading and file processing.
 
-    Provides an interface to materialize the blob in different representations, and
-    help to decouple the development of data loaders from the downstream parsing of
-    the raw data.
+    Represents raw bytes or text, either in-memory or by file reference. Used
+    primarily by document loaders to decouple data loading from parsing.
 
     Inspired by [Mozilla's `Blob`](https://developer.mozilla.org/en-US/docs/Web/API/Blob)
 
-    Example: Initialize a blob from in-memory data
+    ???+ example "Initialize a blob from in-memory data"
 
-    ```python
-    from langchain_core.documents import Blob
+        ```python
+        from langchain_core.documents import Blob
 
-    blob = Blob.from_data("Hello, world!")
+        blob = Blob.from_data("Hello, world!")
 
-    # Read the blob as a string
-    print(blob.as_string())
+        # Read the blob as a string
+        print(blob.as_string())
 
-    # Read the blob as bytes
-    print(blob.as_bytes())
+        # Read the blob as bytes
+        print(blob.as_bytes())
 
-    # Read the blob as a byte stream
-    with blob.as_bytes_io() as f:
-        print(f.read())
-    ```
+        # Read the blob as a byte stream
+        with blob.as_bytes_io() as f:
+            print(f.read())
+        ```
 
-    Example: Load from memory and specify mime-type and metadata
+    ??? example "Load from memory and specify MIME type and metadata"
 
-    ```python
-    from langchain_core.documents import Blob
+        ```python
+        from langchain_core.documents import Blob
 
-    blob = Blob.from_data(
-        data="Hello, world!",
-        mime_type="text/plain",
-        metadata={"source": "https://example.com"},
-    )
-    ```
+        blob = Blob.from_data(
+            data="Hello, world!",
+            mime_type="text/plain",
+            metadata={"source": "https://example.com"},
+        )
+        ```
 
-    Example: Load the blob from a file
+    ??? example "Load the blob from a file"
 
-    ```python
-    from langchain_core.documents import Blob
+        ```python
+        from langchain_core.documents import Blob
 
-    blob = Blob.from_path("path/to/file.txt")
+        blob = Blob.from_path("path/to/file.txt")
 
-    # Read the blob as a string
-    print(blob.as_string())
+        # Read the blob as a string
+        print(blob.as_string())
 
-    # Read the blob as bytes
-    print(blob.as_bytes())
+        # Read the blob as bytes
+        print(blob.as_bytes())
 
-    # Read the blob as a byte stream
-    with blob.as_bytes_io() as f:
-        print(f.read())
-    ```
+        # Read the blob as a byte stream
+        with blob.as_bytes_io() as f:
+            print(f.read())
+        ```
     """
 
     data: bytes | str | None = None
@@ -213,7 +222,7 @@ def from_path(
             encoding: Encoding to use if decoding the bytes into a string
             mime_type: If provided, will be set as the MIME type of the data
             guess_type: If `True`, the MIME type will be guessed from the file
-                extension, if a mime-type was not provided
+                extension, if a MIME type was not provided
             metadata: Metadata to associate with the `Blob`
 
         Returns:
@@ -274,6 +283,10 @@ def __repr__(self) -> str:
 class Document(BaseMedia):
     """Class for storing a piece of text and associated metadata.
 
+    !!! note
+        `Document` is for **retrieval workflows**, not chat I/O. For sending text
+        to an LLM in a conversation, use message types from `langchain.messages`.
+
     Example:
         ```python
         from langchain_core.documents import Document

libs/core/langchain_core/documents/compressor.py

@@ -21,14 +21,14 @@ class BaseDocumentCompressor(BaseModel, ABC):
 
     This abstraction is primarily used for post-processing of retrieved documents.
 
-    Documents matching a given query are first retrieved.
+    `Document` objects matching a given query are first retrieved.
 
     Then the list of documents can be further processed.
 
     For example, one could re-rank the retrieved documents using an LLM.
 
     !!! note
-        Users should favor using a RunnableLambda instead of sub-classing from this
+        Users should favor using a `RunnableLambda` instead of sub-classing from this
         interface.
 
     """
@@ -43,9 +43,9 @@ def compress_documents(
         """Compress retrieved documents given the query context.
 
         Args:
-            documents: The retrieved documents.
+            documents: The retrieved `Document` objects.
             query: The query context.
-            callbacks: Optional callbacks to run during compression.
+            callbacks: Optional `Callbacks` to run during compression.
 
         Returns:
             The compressed documents.
@@ -61,9 +61,9 @@ async def acompress_documents(
         """Async compress retrieved documents given the query context.
 
         Args:
-            documents: The retrieved documents.
+            documents: The retrieved `Document` objects.
             query: The query context.
-            callbacks: Optional callbacks to run during compression.
+            callbacks: Optional `Callbacks` to run during compression.
 
         Returns:
             The compressed documents.

libs/core/langchain_core/documents/transformers.py

@@ -16,8 +16,8 @@
 class BaseDocumentTransformer(ABC):
     """Abstract base class for document transformation.
 
-    A document transformation takes a sequence of Documents and returns a
-    sequence of transformed Documents.
+    A document transformation takes a sequence of `Document` objects and returns a
+    sequence of transformed `Document` objects.
 
     Example:
         ```python

libs/core/langchain_core/embeddings/fake.py

@@ -18,7 +18,7 @@ class FakeEmbeddings(Embeddings, BaseModel):
 
     This embedding model creates embeddings by sampling from a normal distribution.
 
-    !!! warning
+    !!! danger "Toy model"
         Do not use this outside of testing, as it is not a real embedding model.
 
     Instantiate:
@@ -73,7 +73,7 @@ class DeterministicFakeEmbedding(Embeddings, BaseModel):
     This embedding model creates embeddings by sampling from a normal distribution
     with a seed based on the hash of the text.
 
-    !!! warning
+    !!! danger "Toy model"
         Do not use this outside of testing, as it is not a real embedding model.
 
     Instantiate:

libs/core/langchain_core/language_models/__init__.py

@@ -6,12 +6,13 @@
 **Chat models**
 
 Language models that use a sequence of messages as inputs and return chat messages
-as outputs (as opposed to using plain text). Chat models support the assignment of
-distinct roles to conversation messages, helping to distinguish messages from the AI,
-users, and instructions such as system messages.
+as outputs (as opposed to using plain text).
 
-The key abstraction for chat models is `BaseChatModel`. Implementations
-should inherit from this class.
+Chat models support the assignment of distinct roles to conversation messages, helping
+to distinguish messages from the AI, users, and instructions such as system messages.
+
+The key abstraction for chat models is `BaseChatModel`. Implementations should inherit
+from this class.
 
 See existing [chat model integrations](https://docs.langchain.com/oss/python/integrations/chat).
 

libs/core/langchain_core/language_models/fake_chat_models.py

@@ -1,4 +1,4 @@
-"""Fake chat model for testing purposes."""
+"""Fake chat models for testing purposes."""
 
 import asyncio
 import re

libs/core/langchain_core/language_models/llms.py

@@ -1,4 +1,7 @@
-"""Base interface for large language models to expose."""
+"""Base interface for traditional large language models (LLMs) to expose.
+
+These are traditionally older models (newer models generally are chat models).
+"""
 
 from __future__ import annotations
 

libs/core/langchain_core/load/serializable.py

@@ -97,11 +97,14 @@ class Serializable(BaseModel, ABC):
         by default. This is to prevent accidental serialization of objects that should
         not be serialized.
     - `get_lc_namespace`: Get the namespace of the LangChain object.
+
         During deserialization, this namespace is used to identify
         the correct class to instantiate.
+
         Please see the `Reviver` class in `langchain_core.load.load` for more details.
         During deserialization an additional mapping is handle classes that have moved
         or been renamed across package versions.
+
     - `lc_secrets`: A map of constructor argument names to secret ids.
     - `lc_attributes`: List of additional attribute names that should be included
         as part of the serialized representation.