qpdf | PDF viewer widget for Qt | Document Editor library

As already assumed in a comment to the question, the PDF file is not encrypted at all.

Please check lines 1841 - 3258

The lines 1841 - 3258 are part of a stream from line 1739 (OTTO...) to 3258 and contain an embedded OpenType font, compare the preceding stream dictionary

I think flattening the PDF might help you to make it un-editable in case your target is

• Just the form fields then you might use this from the PDF-LIB github repo

• The entire PDF then, see if pdf-flatten package helps for Node.js

I added these lines to my .travis.yml and passed my travis build:

I think \code{\link[dplyr:filter]} should be \code{\link[dplyr]{filter}}, see https://cran.r-project.org/doc/manuals/r-release/R-exts.html#Cross_002dreferences . That should (?) solve some of your problems, don't know if it will solve all of them.

A somewhat brutish way of handling this that can be updated to subset multiple files by only modifying the bigfile and output = then re-running it all. Will update once I make a proper function for it.

If you do R CMD build --help you will see the two lines

This is an anticlimactic solution, but as it turns out I should have used the actual miniz instead of the single-header mirror which is miniz-cpp. This single-header library is using a severely outdated version of the library from 2017 and simply couldn't read the data correctly.

When using the actual miniz, the test passed and everything worked. My code was 100% correct, it just used the wrong library.

I'm sure sure if there are different options/flags that I could use to achieve decompression of the file without invalidating the signature.

If anybody knows a way to decompress without invalidating signatures please let me know!

That is impossible.

A digital signature integrated in a PDF (at least an interoperable one) signs all the bytes of that file (except the placeholder containing the signature itself). Decompressing obviously changes those bytes, so the signature is invalidated.

Consequentially, you can also not spoof those bytes without invalidating the signature.

What you can try, though, is

• change content in the placeholder (which can be larger than the signature and be filled with zeros) or
• add content at the end as an incremental update.

Also you can try techniques as presented on the PDF Insecurity site. Current viewer versions should notice these techniques but probably they don't....

The problem is probably a corrupt PDF file. QPDF is indeed able to work around that. That's why I would recommend using the pikepdf library instead of PyPDF2 in this case. Pikepdf is based on QPDF. First install pikepdf (for example using pip or your package manager) and then try this code: