Tools / PDF to JSON
PDF to JSON
Every page as data: its text, its tables, its size.
- Your file is uploaded to do this
- Deleted 24 hours after processing
- What we do with it
Choose a PDF
or drop them here
How it works
- Add the PDF.
- Choose JSON for everything, or plain text for just the words.
- Download it.
A document as data
Copying out of a viewer gives you a wall of text with the structure thrown away. This keeps the structure.
You get text, tables and page geometry
Enough to feed a script, fill a sheet, or compare two versions of the same document.
Page size and position come with it
Which is what lets you tell a header from a body paragraph without guessing.
Plain text is there when you only want the words
Choose it rather than parsing JSON you do not need.
Diffing two versions works best on the text output
The structure is useful; for spotting what changed, the words are easier to read.
Where this is the wrong tool. This uploads. The file is sent over an encrypted connection, read on our servers and deleted after 24 hours. It reads the text a PDF already holds: a scanned document has none, and needs recognition first.
When the browser is not enough
Scanly for Android
Take document-scanning workflows to your Android phone with Scanly. See the Google Play listing for current features, requirements and data-processing information.
Questions
What is in the JSON?
One object per page with its number, its width and height in points, its text and its characters, plus any tables found on it as rows of cells. Tick the box and you also get every word with the coordinates of its box, which is what you need to line data up with the page it came from.
Will it read a scan?
No. This tool requires an existing text layer. Image-only PDFs are uploaded before the server can reject them. For plain text from a scan, use PDF to JPG and then Image to Text. That workflow does not create a searchable PDF for this tool; use a separate searchable-PDF OCR workflow first.
How good is the table detection?
Good on a real table, whether it is ruled or just cleanly aligned, and no use at all on columns of prose that happen to line up. Check what comes back before you rely on it.
Does my file leave my device?
Yes. The parser that does this runs on our servers. Your upload is deleted 24 hours later, and no account is needed.