langchain-jostack-mdparse 0.1.0

A LangChain integration for jostack-mdparse

langchain-jostack-mdparse

A LangChain Document Loader for jostack-mdparse.

Installation

pip install langchain-jostack-mdparse

Quick Start

from langchain_jostack_mdparse import MarkdownExtractLoader

# Load a single file
loader = MarkdownExtractLoader("README.md")
docs = loader.load()

# Load with options
loader = MarkdownExtractLoader(
    "docs/guide.md",
    heading_level="1,2",
    strip_html=True,
    split_sections=True,
)
docs = loader.load()

# Load multiple files
loader = MarkdownExtractLoader(["a.md", "b.md"])
docs = loader.load()

Parameters Reference

Parameter	Type	Default	Description
file_path	str | Path | List	—	(Required) Path to Markdown file(s)
split_sections	bool	True	Split output into per-section Documents

| format | str | "text" | Output format: json, text, html | | quiet | bool | False | Suppress console logging | | heading_level | str | None | Filter by heading levels (comma-separated) | | sections | str | None | Extract sections by heading text (comma-separated) | | include_frontmatter | bool | True | Include YAML/TOML frontmatter | | strip_html | bool | False | Strip inline HTML tags | | include_code_blocks | bool | True | Include fenced code blocks | | include_toc | bool | False | Add generated table of contents | | flatten_lists | bool | False | Flatten nested lists | | section_separator | str | None | Separator between sections | | normalize_links | bool | False | Convert relative links to absolute |

Development

pip install -e .
make test
make lint

License

Apache-2.0