# Changelog of PDF Tables to JSON with Cell Coordinates + RAG Chunks (OCR) (`landoz/pdf-to-json-tables`) Actor

- **URL**: https://apify.com/landoz/pdf-to-json-tables/changelog.md
- **Full Actor documentation**: https://apify.com/landoz/pdf-to-json-tables.md

## Changelog

### 0.1.5 (2026-10-23)

- Every chunk now carries `citation` (`{"type": "page", "page": 3, "label": "p. 3"}`), the family chunk schema shared with transcribe-to-json-chunks, which emits the same chunks with time citations. Additive change: existing chunk fields are unchanged.
- Chunking moved to the shared `python/family_chunker.py`; output is identical apart from the new field.

### 0.1 (2026-10-23)

- First release: PDF URLs in, structured JSON out.
- Tables as rows of cells with page number, bounding box and optional per-cell boxes; ruled tables by lines, unruled tables by text alignment with quality filters.
- Page text, Markdown (font-size headings, in-place tables) and sentence-aligned RAG chunks with page, charStart and charEnd.
- Optional OCR (off, auto, always) with Tesseract, English model.
- Pay-per-event pricing: one page-processed event per page, one extra ocr-page event per OCR page.