Printing a webpage to PDF often produces incomplete PDFs

I know that printing webpages to PDF is a difficult thing and that results cannot be guaranteed unless a webpage has special CSS or is very simple, so I don't expect miracles. But I have to raise this problem because saving to PDF is an important method for preserving a webpage's information for offline reading, which is safer and more private than online reading and gives access to information when there is no Internet connection. Other methods such as saving a webpage (CTRL+S) are also unreliable, for example sometimes the saved webpage via this method is just a Cloudflare warning instead of the page I intended. I think it is important that TorBrowser users have a reliable way to save information for offline usage and at the moment this is not true.

In my experience, printing a webpage to PDF produces an incomplete PDF most of the time in different ways. "Incomplete" means that most of the webpage content is not included, not that one or two elements did not appear. This is happening for a long time on many versions of TorBrowser over the past few years.

The most common result for me is a PDF exactly three pages long. The first page and last page will be blank except for the Header and Footer information (URL ect), and the second page will usually be just the top portion of the webpage, but it will stop after that, even if the page is very long.

Sometimes the PDF will be the content of a different webpage but from the same site, which is very odd. This doesn't happen often.

To work around this problem I try using Reader mode but the results there are unpredictable too. Sometimes Reader mode does not load images. Sometimes Reader mode will ignore some paragraphs of the original page before I even print to PDF. Sometimes different paragraphs or sections of webpages will be missing when I print from Reader mode and other paragraphs or sections will be missing if I print from the non-Reader mode. This is difficult to spot unless you read the entire page in Reader mode before printing to PDF, and then you have to check the PDF carefully after you create it.

The Print to PDF option is so unreliable that I now always have to open every PDF I print from TorBrowser to check it for missing information before I can be sure it is acceptable. Sometimes it is obvious just from the progress bar - if it only creates a one page PDF on a long webpage, for example, or if it creates the common 3-page PDF. This is time-consuming. Sometimes now I just copy and paste the text of a page into a text document, but this loses most of the formatting.

I suggest two things. First there should be a warning message for every user each time they print to PDF to warn them to check the output. Second there should be some effort towards making print to PDF more reliable, perhaps by encouraging websites to include the necessary CSS.

I will include an example of incomplete PDFs the next time it happens on pages it is safe for me to share.