OK, I try to improve my question: If you open a PDF in PDF.js with 20 pages and you click 'select all' and copy and then you paste it into Word, only 4 pages are pasted i.e. the PDF is not fully recognized. How can I force PDF.js to recognize the full document?

Some PDFs are not rendered in their entirety when loaded, but only as you continue to scroll down. It seems that scanned books and other highly dense materials are the examples I see most. This could be some kind of optimization to be able to let you start reading rather than wait for the actual amount of time it would take to fully render the PDF. But that's just a guess.

jscher - your are still after many years the BEST!!! My problem is: no matter which PDF - they all are not fully recognized - rendered yes, but I cannot copy and paste all pages (only 4 pages are pasted - please try it) - the root problem for me is: I would like to SEARCH PDFs like HTML files and yes I use a legacy add-on with cyberfox 52, HUGO SEARCH IN ALL TABS with which I can search in 50 or 100 tabs and therefore it is important that the PDFs are not only rendered but completely searchable - Attention: I verified the PDF.js behavior with the current Firefox version, so Cyberfox is not the issue!!!

It is unfortunately not true that PDF.js just takes some time to render the full PDF - it only renders he pages you call up and no more pages - I could easily wait 5 or 10 minutes until all pages are rendered, unfortunately PDF.js does not render all pages for searching over time and actually that's what I am asking for: How can I force PDF.js to render all pages - it would be great!

jscher thank you for your help you gave me whenever I had a problem with Firefox, maybe you know a special solution this time too.?

Hmm, I tested on a 5-page PDF generated by printing from a blog. If I open the file, select all, copy without scrolling at all, I get two pages when I paste. If I scroll to the end, select all, copy then I get all five pages when I paste.

With a 10-page, 2-column article, the results are quite variable and extremely messy.

I don't know a solution to the progressive rendering/search problem. Extensions are not allowed to run in the PDF viewer (to automatically scroll the document for you to force it to render).

Thank you very much for helping and confirming the situation. It would be a good thing if the PDF.js developers would fix this.

You could file a bug report at https://bugzilla.mozilla.org/

Thanks, yes, but this issue is known since many years and development did not address it.