Sinapsis Langchain Splitters

Templates for easy integration of LangChain text splitters with Sinapsis.

🐍 Installation • 🚀 Features • 📚 Usage example • 📙 Documentation • 🔍 License

The sinapsis-langchain-splitters module adds support for all the text splitters supported by LangChain

🐍 Installation

Example with uv:

  uv pip install sinapsis-langchain-splitters --extra-index-url https://pypi.sinapsis.tech

or with raw pip:

  pip install sinapsis-langchain-splitters --extra-index-url https://pypi.sinapsis.tech

[!IMPORTANT] The langchain readers templates may require extra dependencies. For development, we recommend installing the package with all the optional dependencies:

  uv pip install sinapsis-langchain-readers[all] --extra-index-url https://pypi.sinapsis.tech

[!IMPORTANT] Some langchain templates require additional system dependencies. Please refer to the official LangChain Document Loaders documentation for additional requirements.

🚀 Features

Templates Supported

The Sinapsis Langchain module provides wrapper templates for Langchain's community data loaders, making them seamlessly usable within Sinapsis.

[!NOTE] Each loader template supports one attribute:

• add_document_as_text_packet (bool, default: False): Whether to add the loaded document as a text packet. Other attributes can be dynamically assigned through the class initialization dictionary (class init attributes).

[!TIP] Use CLI command sinapsis info --all-template-names to show a list with all the available Template names installed with Sinapsis Langchain.

[!TIP] Use CLI command sinapsis info --example-template-config TEMPLATE_NAME to produce an example Agent config for the Template specified in TEMPLATE_NAME.

For example, for WikipediaLoaderWrapper use sinapsis info --example-template-config WikipediaLoaderWrapper to produce the following example config:

agent:
  name: my_test_agent
  description: Agent to split text using the RecursiveCharacterTextSplitter class from LangChain
templates:
- template_name: InputTemplate
  class_name: InputTemplate
  attributes: {}
- template_name: RecursiveCharacterTextSplitterWrapper
  class_name: RecursiveCharacterTextSplitterWrapper
  template_input: InputTemplate
  attributes:
    add_document_as_text_packet: false
    generic_key: WikipediaLoaderWrapper
    recursivecharactertextsplitter_init:
      separators: [" "]
      keep_separator: true
      is_separator_regex: false

A complete list of available document loader classes in LangChain can be found at: LangChain Text Splitters

📚 Usage example

The following example demonstrates how to use the RecursiveCharacterTextSplitterWrapper template to chunk the Documents from a WikipediaLoaderWrapper template.

agent:
  name: my_test_agent
  description: "Wikipedia loader example"

templates:

- template_name: InputTemplate
  class_name: InputTemplate
  attributes: {}

- template_name: WikipediaLoaderWrapper
  class_name: WikipediaLoaderWrapper
  template_input: InputTemplate
  attributes:
    add_document_as_text_packet: false
    wikipedialoader_init:
      query: GenAI
      lang: en
      load_max_docs: 1
      load_all_available_meta: false
      doc_content_chars_max: 4000
- template_name: InputTemplate
  class_name: WikipediaLoaderWrapper
  attributes: {}
- template_name: RecursiveCharacterTextSplitterWrapper
  class_name: RecursiveCharacterTextSplitterWrapper
  template_input: InputTemplate
  attributes:
    add_document_as_text_packet: false
    generic_key: WikipediaLoaderWrapper
    recursivecharactertextsplitter_init:
      separators: null
      keep_separator: true
      is_separator_regex: false

sinapsis run name_of_the_config.yml

📙 Documentation

Documentation for this and other sinapsis packages is available on the sinapsis website

Tutorials for different projects within sinapsis are available at sinapsis tutorials page

🔍 License

This project is licensed under the AGPLv3 license, which encourages open collaboration and sharing. For more details, please refer to the LICENSE file.

For commercial use, please refer to our official Sinapsis website for information on obtaining a commercial license.