This can rely on WS Export to perform the task. However, WS Export seems more tailored to export proofread text. Instead, I'm asking to create a function to allow for the download of a pdf consisting of scanned images.

Isn't this just downloading the file from archive.org (https://archive.org/details/popularsciencemo02newy) or wikicommons. Why would this be part of wikisource and wsexport when we are talking original images, it seems out of scope.

No, it's downloading a subset of that file. A journal volume can be hundreds of pages, while an article can be less than one.

It's in scope for several reasons:

1. Even if a proofread text exists, users may want to download the original article.
2. This would be a great way to attract help/sponsoring from academic institutions.
3. It would make the generation of a TOC even more useful.
4. It would a major step forwards towards making Wikisource a viable alternative to extremely expensive databases that monetize the PD.

It is still nothing to do with WS-EXPORT, which does nothing with scans, and nothing particularly to do with Wikisource that is reproducing CORRECTED text. WS does just uses the images for proofreading, nothing else. If it is a task then if it belongs anywhere, then it belongs at Commons related to the images.

I would not be encouraging us to extract PDFs of DJVUs/PDFs that have the uncorrected text and tack them them beside each other in an export of an article. The download function at this time has been about producing digital books for reading, not trying to provide for academia for something that has been twice proofread of an older work. If your wish is bigger than that, then I think that it needs to be rescoped.

Commons is not the right place because this would require two major components that only Wikisource has both:

1. The labeling of page ranges as belonging to a specific article. On Wikisource, this is done through the generation of a TOC. Wikisource does this on the Index ns.
2. The ability to generate a PDF with images. Wikisource does this already for proofread text.

In essence, this would read the TOC entry, send the page range to a WS Export, and offer it as a download.

The relevant section is "pages index="Austen - Sense and Sensibility, vol. I, 1811.djvu" from=9 to=20"

As I said earlier, this is an immature proposal where you are not certain what you are wanting, or how to achieve it.

1. It is not a pdf of the original scan.
2. the rationale for the proposal is not backed by a community, nor factually supported, nor is particularly pertinent for the vast bulk of the works that the Wikisources prepare, and the latest example for Austen's book is not a journal.