langchain-section-anything

A LangChain document loader for Section Anything: it finds the sections of a PDF (every heading, its level, its pages and its exact text) and returns one Document per section, with the heading path in the metadata. The default mode runs locally with rules only, calls no model and costs nothing.

Install

pip install langchain-section-anything
# latest from GitHub:
pip install "git+https://github.com/Omniwot/section-anything#subdirectory=packages/langchain-section-anything"

Python 3.11+. Depends on section-anything and langchain-core only.

Example

from langchain_section_anything import SectionAnythingLoader

loader = SectionAnythingLoader("paper.pdf")  # mode="free": rules only, no model, $0
docs = loader.load()  # or loader.lazy_load(), await loader.aload()

doc = docs[2]
print(doc.page_content[:59])         # '### 1.1 Scope\n\nSectioning splits a long document into parts'
print(doc.metadata["heading_path"])  # ['1 Introduction', '1.1 Scope']
print(doc.metadata["page_start"], doc.metadata["page_end"])  # 1 1

Sections vary in length; to cap chunk size, split the Documents further (for example with RecursiveCharacterTextSplitter.split_documents), which keeps the metadata on every chunk.

Options

SectionAnythingLoader(file_path, mode="free", *, include_heading=True, **opts)

• file_path: a path, or a list of paths.
• mode: free (bookmarks, table of contents and typography rules; nothing leaves the machine), quick (a small model reviews candidate heading lines, about $0.07 per 1,000 pages), standard (a model reads every line) or local:<ollama model>. Over the per-document cost cap (default $0.50) a run is downgraded (standard → quick → free) or refused.
• include_heading: start each Document with its heading as a Markdown line (## 1 Introduction). With False, sections with no text of their own (a heading directly followed by a subheading) are skipped.
• **opts go to section_anything.section: api_key, base_url, model, max_usd, prices, over_budget, max_pages, extract_timeout, progress, settings. Anything not given falls back to the SECTIONANY_* environment variables; see the library README. Keys are never logged or put in metadata.

Text and metadata

A Document's text is the section's own text, from its heading to the next heading, reflowed into paragraphs with running headers and footers removed (the same text as section-anything --body). Text before the first heading is a Document of its own (flag front_matter).

heading_path and flags are lists. For a vector store that accepts only scalar metadata, join them (" > ".join(...)) or drop them with filter_complex_metadata from langchain_community.vectorstores.utils before indexing.

Licence

Apache-2.0. See LICENSE and NOTICE.