PDF to CSV Reads Scanned Tables

PDF to CSV now extracts tables from scanned PDFs instead of failing with a no data error. The new OcrQuality parameter adds the enhanced OCR engine for difficult scans.

Fixes

• PDF to CSV - a scanned table returned error 5004, "No data to extract from PDF file", even with EnableOcr set to Scanned or All. Recognized pages now keep their column layout, so their tables come out as CSV like tables in text PDFs.
• PDF to CSV - a cell holding a quote now has its quotes doubled, so every CSV reader splits the row correctly. Cells holding the delimiter or a line break are quoted as before.
• PDF to TXT and PowerPoint to TXT - with IncludeFormatting set to true, recognized pages now keep their layout too, as pages with their own text already did.

Parameters

• OcrQuality (collection, default fast) - new on PDF to CSV. fast processes pages quickly, enhanced reads low quality scans, photos, small print and rotated text much better, at the cost of slower processing.

Try it out

• PDF to CSV
• API sandbox: pdf-to-csv