This is a small independent project. Replies can take a few days, and occasionally a message goes unanswered for longer than it should. Saying so up front beats an auto-reply promising a twenty-four-hour turnaround nobody here keeps.
Please don't attach documents. Email isn't a secure channel, and there's no processing pipeline on this side that could use the file even if it were, because the tools only exist inside your own browser. Describe the problem instead; if the file itself is genuinely needed to diagnose it, it'll be asked for separately.
Before you write: three things that cover most cases
- Extraction came back empty
- The file has no text layer: it's a page image, which usually means a scan. The text extractor can still read it — tick “OCR scanned pages (English)”, which downloads a recognition engine of about 7 MB the first time you use it. It handles printed English; handwriting and poor scans come back wrong. Every other tool on the site reads text that's already in the file.
- It asks for a password and won't continue
- That password has to be one you already have. Nothing here recovers or bypasses it, and no amount of email changes that.
- A large file stalled or produced nothing
- Everything runs in one browser tab, so the ceiling is the memory that tab can get. Splitting a very long document into chunks and processing them one at a time usually works where the whole file didn't.
If none of those fit, it's probably a genuine bug, and worth reporting.
What makes a report reproducible
"It didn't work" can't be acted on. This can:
- Which tool, and roughly what you were doing with it.
- Browser and version, operating system.
chrome://versionhas both in copyable form. - The approximate size and page count of the file. Not the file itself.
- What you expected, and what happened instead, including the exact wording of any error.
- Whether it still happens with the network switched off. That one answer halves where the problem can be.
Anything about a possible security weakness is welcome at the same address. Please mark it clearly in the subject line.
What nobody here can do
- Recover your document. No file is ever uploaded, so no copy exists to restore. Whatever you downloaded is the only copy.
- Retrieve an output you closed without saving. Results are built in the tab and given to you once.
- Take a file by email and process it for you. Doing so would need the kind of server whose absence is the point.
- Tell you whether your use is lawful. That depends on the document and on you.
Everything else is fair game, including "I don't understand why it does it this way", which is usually the most useful kind of question, because it tends to turn into better wording on a page.