This header is constructed by Symfony (the framework used by this tool). This issue was reported a couple of years ago, and the response was that the unquoted form is correct.

Are you seeing an error with a particular browser?

@Samwilson I'm not seeing any error.

I was/am trying to pull the filename out of that header and found that it wasn't quoted, which is in opposition to what MDN says. If it's "not a bug" then I defer to the Symfony devs. Thanks for the reply.

Yeah, I agree with you in that it seems slightly odd, but if Symfony devs are happy then I think it's fine.

What are you trying to do? It seems possibly odd to read the filename out when to get the request you must be submitting what is more or less that same filename (with subpage slashes replaced). We might be able to do something else to help you.

The use case is a piece of extreme laziness on my part where I want to drag the "Download EPUB" link into a program with XDND to download it and display it (thus avoiding oh-so-arduously having to download it, save it, then drag the downloaded file). This means I only have the URL and whatever headers the response comes with to work with.

However, the idea was not to limit it to understanding ws-export links like https://wsexport.wmflabs.org/?format=epub-3&lang=en&page=Page_title and allow it to more universally extract the filename from things like content-disposition and then fall back on the URL.

Which then made me look closer at the content-disposition field because parsing that is non-trivial.