Skip to main content

runic-rag-docling

Optional Docling adapters that implement the runic.rag file-oriented ports DocumentParser and DocumentChunker. Docling parses PDF/DOCX/PPTX/XLSX/HTML/images structure-aware (layout, tables, headings) and chunks the structured document directly — no Markdown re-parse. The heavy Docling dependency lives only in this add-on; the runic.rag core stays light and imports no Docling.

Install

In-process (default, heavy — pulls torch):

uv add 'runic-rag-docling[local]'

Lightweight client against a docling-serve instance:

uv add 'runic-rag-docling[server]'

Usage

Variant A — inject DoclingChunker via the GraphRAG constructor

Wire the Docling chunker as the document_chunker; ingest_document then parses and chunks the original structure-aware, while ingest_text keeps the built-in ParagraphChunker.

from runic.rag import GraphRAG, ParagraphChunker, RagSettings  # plus the other ports
from runic_rag_docling import DoclingChunker, DoclingSettings

rag = GraphRAG(
    store,
    ontology=ontology,
    chunker=ParagraphChunker(settings),
    document_chunker=DoclingChunker(DoclingSettings(mode="local")),
    # extractor, embedder, resolver, retrievers, reranker, synthesizer, settings ...
)
rag.bootstrap_schema()
report = rag.ingest_document("whitepaper.pdf")  # Docling parses + chunks the original

Runnable: examples/docling_explicit_wiring.py wires the full stack by hand (no build_graphrag).

Variant B — build_graphrag one-liner (default stack + Docling)

from runic.rag import load_settings
from runic_rag_docling import DoclingSettings, build_graphrag

rag = build_graphrag(
    load_settings(),
    DoclingSettings(mode="server", server_url="http://localhost:5001"),
)
rag.bootstrap_schema()
report = rag.ingest_document("whitepaper.pdf")

For parse-only behavior, wire DoclingParser as the document_parser (keeps the core Chunker).

Configuration

DoclingSettings reads from the environment with the RUNIC_DOCLING_ prefix (e.g. RUNIC_DOCLING_MODE=server, RUNIC_DOCLING_SERVER_URL=..., RUNIC_DOCLING_API_KEY=..., RUNIC_DOCLING_MAX_TOKENS=512).

See docs/rag/docling.md for the full guide.

Download files

Download the file for your platform. If you're not sure which to choose, learn more about installing packages.

Source Distribution

runic_rag_docling-0.2.2.tar.gz (9.7 kB view details)

Uploaded Source

Built Distribution

If you're not sure about the file name format, learn more about wheel file names.

runic_rag_docling-0.2.2-py3-none-any.whl (11.9 kB view details)

Uploaded Python 3

File details

Details for the file runic_rag_docling-0.2.2.tar.gz.

File metadata

  • Download URL: runic_rag_docling-0.2.2.tar.gz
  • Upload date:
  • Size: 9.7 kB
  • Tags: Source
  • Uploaded using Trusted Publishing? No
  • Uploaded via: uv/0.12.1 {"installer":{"name":"uv","version":"0.12.1","subcommand":["publish"]},"python":null,"implementation":{"name":null,"version":null},"distro":{"name":"macOS","version":null,"id":null,"libc":null},"system":{"name":null,"release":null},"cpu":null,"openssl_version":null,"setuptools_version":null,"rustc_version":null,"ci":null}

File hashes

Hashes for runic_rag_docling-0.2.2.tar.gz
Algorithm Hash digest
SHA256 83a64260f6710f127e874410287f2f2e5f8728bce1d3aa6ed276b563f2090684
MD5 1bed7251679a955b8294b284c3e65824
BLAKE2b-256 6a8d45b807a1608956edeb84f3a985db84a4d29fc7efb8cc4c5347a55a4ba0d4

See more details on using hashes here.

File details

Details for the file runic_rag_docling-0.2.2-py3-none-any.whl.

File metadata

  • Download URL: runic_rag_docling-0.2.2-py3-none-any.whl
  • Upload date:
  • Size: 11.9 kB
  • Tags: Python 3
  • Uploaded using Trusted Publishing? No
  • Uploaded via: uv/0.12.1 {"installer":{"name":"uv","version":"0.12.1","subcommand":["publish"]},"python":null,"implementation":{"name":null,"version":null},"distro":{"name":"macOS","version":null,"id":null,"libc":null},"system":{"name":null,"release":null},"cpu":null,"openssl_version":null,"setuptools_version":null,"rustc_version":null,"ci":null}

File hashes

Hashes for runic_rag_docling-0.2.2-py3-none-any.whl
Algorithm Hash digest
SHA256 62c8cfe11c34181e4220410be3eafc38946367c7b8784ec17a999d1053348660
MD5 24052072b9cf517763eed193616c9804
BLAKE2b-256 5bc810e864ff142b809f509337f014834bc5f38dee632d6ced7689200305a627

See more details on using hashes here.

Release history Release notifications | RSS feed

This release

0.2.2 This release

2 files

0.2.1

2 files

0.2.0

2 files

Anthropic, PBC Visionary sponsor Bloomberg Visionary sponsor Hudson River Trading Visionary sponsor Meta Visionary sponsor NVIDIA Visionary sponsor Microsoft Sustainability sponsor Depot Continuous Integration AWS Cloud computing and Security Sponsor Datadog Monitoring Fastly CDN Google Download Analytics Sentry Error logging StatusPage Status page