v0.8.0
[0.8.0] - 2026-07-09
Added
- Added a pluggable multi-engine PDF -> asset extraction architecture selected
viaETL_ENGINE:pymupdf(default, fast, no models),pymupdf4llm
([pdf-plus], drop-in layout-aware upgrade),docling([docling],
MIT-licensed layout+table+formula+chart engine), andmineru([mineru],
highest-accuracy formula-to-LaTeX/table-to-HTML engine). All three resolve
Pillow>=12.2.0cleanly, unlike the still-disabledmarkerbackend. - Added a cross-platform Docling installer (
scripts/setup_docling.pyplus
.sh/.ps1wrappers) that provisions an isolated.venv-docling
interpreter and a subprocess bridge so the main environment never needs a
direct, version-sensitive Docling install. Seedocs/docling-setup.mdfor
the zero-configuration install path for CLI, VS Code, and MCP agents. - Added
StructuredPDFExtractor, a shared protocol implemented by
Marker/Docling/MinerU so structured engines are interchangeable behind
src/infrastructure/extractor_factory.pyand emit a Marker-compatible
result, reusing the existing ingestion pipeline without touching its logic.
Changed
DocumentService's structured-parser slot is now engine-agnostic
(StructuredPDFExtractorprotocol) while keeping its existing parameter
name and behavior for backward compatibility with existing callers/tests.
Full Changelog: v0.7.0...v0.8.0