Why does it say no text could be read from the document?

You upload a scan or a photo of a supplier invoice, purchase order, or delivery note, and instead of a filled-in draft VegaERP reports that it could not read any text from the document. This is not a bug: the document genuinely contains no text VegaERP can read, and the message says so instead of leaving you with an empty draft that looks broken.

What the screen says

V tem dokumentu ni bilo mogoče prebrati besedila, zato ni bilo izpolnjeno nobeno polje. Če gre za skeniran dokument ali fotografijo, naložite ostrejšo kopijo. Sicer PDF znova shranite iz programa, ki ga je pripravil, ali dokument vnesite ročno. (“No readable text was found in this document, so no fields could be filled in. If it is a scan or a photo, upload a sharper copy. Otherwise save the PDF again from the program that produced it, or enter the document by hand.”)

The same message appears in all three places where VegaERP reads documents automatically:

• Prejeti računi (“Received invoices”) → Uvozi (“Import”)
• Naročilnice (“Purchase orders”) → Uvozi (“Import”)
• Prejete dobavnice (“Received delivery notes”) → Uvozi (“Import”)

That file’s row in the import batch is marked Napaka (“Error”), with the message shown underneath. Other files in the same batch process independently and are not held up by it.

Why this happens

VegaERP first checks whether the document already carries a real text layer, the kind a PDF exported straight from another accounting program has. If there is too little of it, it treats the file as a scan and runs text recognition. The message appears when even that finds no readable text anywhere in the document:

• A photo of a photo, or a badly lit shot. Re-photographing an image (for example a photo of a screen showing a scan) loses the sharpness text recognition needs.
• A scan or image with no text layer at all. A purely image-based PDF (paper scanned without OCR at the time) is just a picture to a computer, no matter how legible it is to the eye.

Both end the same way: without this message, a draft with every field blank would look like a broken import, when in fact there was genuinely nothing to read.

A password-protected PDF, a damaged PDF and a PDF with more pages than can be read at once do not show this message. VegaERP recognises those files when it opens them, and the import row shows a message that names the problem and what to do: save an unlocked copy without a password, save the PDF again from the program that produced it, or split it into smaller files.

What to do

1. Upload a sharper copy. Re-scan at a higher resolution, or photograph the document directly, without an intermediate screen or a shadow across the page.
2. If the document originates from another program (for example an export from a previous accounting system), export it again from there. A PDF that was “printed” to an image format, or forwarded on as a scan of a scan, loses its text layer even though it looks identical.
3. If a better copy of the document isn’t available, enter it by hand. Open Nov prejeti račun (“New received invoice”), or the equivalent “new” action on a purchase order or delivery note, and type in the figures from the document yourself; you can still attach the uploaded file as supporting evidence.

4. You can remove a row that is stuck in error from the batch list, or click Ponovi (“Retry”) to try again if you replaced the file behind the same row in the meantime.

If you can’t fix this yourself

If the document is legible to the eye, at a good resolution, and has no password, but the message keeps appearing, contact support and attach the exact file you tried to upload.