TG Edit-Pdf
TG Edit-Pdf
Advertisement

Help & Security Center

This is the support and trust hub for pdf-edit.tech, a browser-based PDF toolkit built and maintained by TakshaGroups. If you have a question about how your files are handled, why a tool behaved the way it did, or how to work with the API, you'll find a clear answer below. The tools run on React 18 with a Spring Boot backend that uses Apache PDFBox 3.x, plus LibreOffice and ocrmypdf for conversion and OCR.

Is my data safe?

Yes, and here is exactly how the privacy model works so you can judge it for yourself rather than take our word for it.

  • Everything moves over HTTPS. Your file is encrypted in transit between your browser and the server.
  • Processing happens in memory. When you upload a file, it is held in server memory only for as long as it takes to produce your result. PDFBox reads the bytes, performs the operation, and writes the output.
  • Files are deleted immediately after the output is generated. Nothing is written to a long-term store, archived, or kept for analytics. Once your download is produced and the request finishes, the input and the in-memory copies are gone.
  • There are no accounts. You don't sign up, log in, or create a profile, so there is no user record tied to your files.
  • No file storage, period. We don't build a library of your documents. Each request stands alone.

A couple of operations don't even reach the server. When you merge PDFs and images, your browser counts pages locally with pdf.js before anything is uploaded, and the create a PDF flow assembles your document client-side. The less that leaves your machine, the better.

A practical note on passwords: protect applies AES encryption to your PDF, and unlock removes a password from a file you can already open — it is not a cracking tool. If you've forgotten the password, we can't recover it, and that's by design.

File size & format limits

Two hard limits apply to every upload:

  • 100 MB per file
  • 500 MB per request

If your file is larger than 100 MB, the upload will be rejected before processing. For very large PDFs, run compress first, then feed the smaller result into the tool you actually wanted.

Formats accepted depend on the tool:

A word on what edit does: it stamps a transparent overlay onto pages so you can annotate, highlight, or cover content (redaction-as-overlay). It is not a true text editor that re-flows the original words, so set expectations accordingly.

Why did my conversion or tool fail?

Most failures come down to a handful of causes. Here's how to diagnose them.

  • The PDF is corrupt or encrypted. If a file won't open in a normal reader, our tools can't process it either. For password-protected files, run unlock first (you'll need the password), then retry your tool.
  • You're converting a scanned PDF to Word. PDF to Word is best-effort for text-based PDFs — it extracts real text. A scanned document is just images of text and contains no extractable characters. Run OCR first to add a real text layer, then convert.
  • The file is too large. Anything over 100 MB per file or 500 MB per request is rejected. Compress or split first.
  • Unsupported file type. Feeding a tool a format it doesn't accept (for example, a spreadsheet into a PDF-only tool) will fail. Check the format list above.

Conversion tools that lean on external software — office to PDF, PDF to Word, PDF to PowerPoint (LibreOffice), and OCR plus PDF/A (ocrmypdf, which uses Tesseract and Ghostscript) — are the most likely to be picky about input quality. Layout fidelity is best-effort; complex documents may shift. If a conversion produces an odd result, try simplifying the source document.

If compression didn't shrink your file much, check the mode. Compress offers smart (recompresses embedded images while keeping text selectable) and rasterize (flattens each page to an image for maximum shrink, after which text is no longer selectable). Smart is the safe default; rasterize trades searchable text for size.

Supported languages

The interface is available in 18 languages. You can switch with the language selector, or by adding ?lang=xx to any URL. The site remembers your choice for next time.

OCR is separate from the UI language. When you run OCR, you choose the language of the text inside the document so Tesseract recognizes the right alphabet and characters. The default is English (eng). If your document is in another language, select it before running OCR — picking the wrong language is a common reason OCR output looks like gibberish.

Using the API

The same operations are available through a REST API under /api/pdf on pdf-edit.tech. It's rate limited to keep the service fair for everyone:

  • 60 requests per minute per IP by default.
  • 600 requests per minute if you authenticate with an API key, sent as an X-API-Key header.

Exceeding the limit returns HTTP 429; an unrecognized key returns 401. The same 100 MB / 500 MB size limits apply to API calls.

To request an API key, email takshagps@gmail.com with a short note about what you're building and your expected volume. Keys are issued on request.

Still need help?

If something here didn't answer your question, write to us at takshagps@gmail.com. To get a fast, useful reply, include:

  • The tool you used (for example, PDF to Word or Compress) and the URL.
  • What you expected and what actually happened, including any error message or HTTP status code.
  • The kind of file involved — its rough size, whether it's a scanned or digital PDF, and whether it's password-protected. Please don't email confidential documents; describe the problem instead.
  • Your browser and operating system if the issue looks display-related.

pdf-edit.tech is a project of TakshaGroups. We read every message.

Advertisement
Advertisement
Advertisement