Excalibur — извлечение таблиц из PDF через веб-интерфейс

★ 1.8K

Excalibur is a web interface for extracting tabular data from PDFs, built on the Camelot library (Python 3). Reach for it when you need to pull tables out of a PDF into a structured form (CSV/Excel) through a convenient interface rather than programmatically: upload a document, select pages and tables, and get data for further processing. IMPORTANT: it works only with text-based PDFs where you can select the text with the mouse, not with scans (scans need OCR). It requires ghostscript installed, installs via pip, and initializes its own metadata database. Its focus is extracting tables specifically from text-based PDFs via a web UI, not full OCR of scans (OCRmyPDF/PaddleOCR for that) or whole-document layout analysis (huridocs): a narrow "tables from PDF → data" job. It is useful to analysts and anyone who regularly receives reports with tables in PDF.