def __init__()

in chunking/chunkers/langchain_chunker.py [0:0]

• 15 lines of code
• 1 McCabe index (conditional complexity)

    def __init__(self, data):
        """
        Initializes the TextChunker with the given data and sets up chunking parameters from environment variables.
        
        Args:
            data (str): The document content to be chunked.
        """
        super().__init__(data)
        self.max_chunk_size = int(os.getenv("NUM_TOKENS", "2048"))
        self.minimum_chunk_size = int(os.getenv("MIN_CHUNK_SIZE", "100"))
        self.token_overlap = int(os.getenv("TOKEN_OVERLAP", "100"))
        self.supported_formats = {
            "md": "markdown",
            "txt": "text",
            "html": "html",
            "shtml": "html",
            "htm": "html",
            "py": "python",
            "csv": "csv",
            "xml": "xml"
        }