v2.113.0

moeru-ai/airiv2.113.0Jul 14, 2026by docling-ops[bot]

AI Summary

This release enhances document parsing capabilities by adding chart extraction for PowerPoint and Excel, exposing cloud storage backends (GCS, Azure, Google Drive), and improving formatting stability for Word and Markdown documents.

Key Highlights

  • Parse native PowerPoint charts as classified pictures
  • Expose GCS, Azure Blob, and Google Drive source/target types
  • Fix MS Word paragraph formatting when trailing run is whitespace-only
  • Stop duplicating table cells in the Markdown backend
  • Add request timeout to model download to avoid indefinite hangs

New Features

  • PPTX chart parsing
  • Cloud storage backends (GCS, Azure, Google Drive)
  • Lazy import of pypdfium2 in pdf_outline

Full Release Notes

### Feature

* **pptx:** Parse native PowerPoint charts as classified pictures with data ([#3794](https://github.com/docling-project/docling/issues/3794)) ([`3b5891c`](https://github.com/docling-project/docling/commit/3b5891c81c3b6f250cea53534d84a8b2f8303b7e))
* **service:** Expose GCS, Azure Blob, and Google Drive source/target types ([#3795](https://github.com/docling-project/docling/issues/3795)) ([`e1e2fbc`](https://github.com/docling-project/docling/commit/e1e2fbc755ed489d549c3d54e56e83aa41e795cc))

### Fix

* **msword:** Keep paragraph formatting when a trailing run is whitespace-only ([#3799](https://github.com/docling-project/docling/issues/3799)) ([`1db1c32`](https://github.com/docling-project/docling/commit/1db1c323387980557381c45209d043b69059d7c1))
* **asciidoc:** Drop phantom cell from trailing-pipe table rows ([#3787](https://github.com/docling-project/docling/issues/3787)) ([`f976796`](https://github.com/docling-project/docling/commit/f976796580bf0071f10a53696166f3495948cce0))
* **md:** Stop duplicating table cells in the Markdown backend ([#3781](https://github.com/docling-project/docling/issues/3781)) ([`6cfdc5f`](https://github.com/docling-project/docling/commit/6cfdc5f39a73f4f1b30666308ca641c7b0f28399))
* Import pypdfium2 lazily in pdf_outline ([#3796](https://github.com/docling-project/docling/issues/3796)) ([`06e9c89`](https://github.com/docling-project/docling/commit/06e9c8990930f19c0de56fee95a163411cfa8bed))
* **msword:** Preserve cells and merges in DOCX tables with gridBefore rows ([#3745](https://github.com/docling-project/docling/issues/3745)) ([`1e197a0`](https://github.com/docling-project/docling/commit/1e197a0912e4fdbd6b6e55fb3039b313e182cd5c))
* Chain original exception in ConversionError for error classification ([#3757](https://github.com/docling-project/docling/issues/3757)) ([`5f2a3b6`](https://github.com/docling-project/docling/commit/5f2a3b628ba2cd6ece77ce1c0bdfd37c494dc6ad))
* **utils:** Add request timeout to model download to avoid indefinite hangs ([#3784](https://github.com/docling-project/docling/issues/3784)) ([`e2329d9`](https://github.com/docling-project/docling/commit/e2329d986e549184ceb91c4b2d80cc05c6b8d620))