pdftotextMost PDF redaction is a black rectangle drawn over text that is still in the file. It looks right, it passes review, and the number comes straight back out of any parser. This endpoint rasterises instead, so there is nothing underneath to recover.
Talk to our technical team → — for integration questions, a security review or a volume estimate.
A PDF stores two separate things: the text, as data, and the
appearance, as drawing instructions. Almost every tool that offers to redact a PDF adds a
filled rectangle to the second one. The characters underneath are untouched. They are still
selectable, still copyable, still returned by pdftotext and by every parser your
counterparty might run.
This is not a hypothetical. It is the single most common way a redacted document leaks, precisely because the failure is invisible: the file looks correct on every screen it is opened on, so it clears review and gets sent.
pdftotext returns it.pdftotext redacted.pdf - and grep for the number. That takes ten seconds, it is
the only test that settles the question, and it is the one an auditor will run. Do it against
whatever you are using today before you take anyone's word for it — ours included.It is a real trade and worth stating plainly rather than burying.
The trade is unavoidable, not an implementation shortcut. A PDF that is still searchable in the redacted region is a PDF where the redacted text is still present. You cannot have both, and any vendor offering both is offering the annotation.
The identifier is rarely only where you first saw it. A loan file
repeats the Aadhaar number in an annexure; an e-Aadhaar carries it on both sides of a two-page
print; a signature page nobody scrolls to holds the PAN again. Every page is processed, and
X-Redacted-Count reports the merged region count across the document.
A count of zero means nothing was found and the PDF is returned unchanged and fully readable. On an unclassified document that is often correct — but it is also what a document too poor to read returns, and that is the one you least want forwarded. Branch on it rather than assuming it.
For a folder rather than a file, see bulk processing: a ZIP out with a manifest naming every document that came back unchanged. For the Aadhaar-specific masked form on a multi-page e-Aadhaar, see Aadhaar PDF masking.
A PDF holds its text as data and its appearance as drawing instructions. Adding a filled rectangle changes only the appearance. The characters underneath are still in the file, still selectable, and still returned by pdftotext or any parser. The document looks redacted and is not.
The output is images in a PDF container, so text is no longer selectable or searchable anywhere in the document, and file size usually rises. That is the trade: a searchable PDF is one where the redacted text is also still there.
Yes. Every page is processed, which matters because the identifier is rarely only on page one — a loan file repeats the Aadhaar number in an annexure and a signature page that nobody checks.
Run pdftotext over the output and grep for the number. That is the test that matters, it takes one command, and it is the one an auditor will run. X-Redacted-Count tells you how many regions were covered.
Tell us how many documents a month you process and what your peak looks like. We will size a plan, provision a key with a test quota, and answer a security questionnaire if you need one.