Correction: This only happens on some links to PDF files, i.e., it gets saved without asking me. On other PDF links, it asks whether I want to open or save the file, as I prefer. I'm not sure what factors cause it to not ask me, but the case that caused me to post this was a link to a PDF file within an email, as read on the gmail webmail interface. If I hover the pointer over the link, all it shows me is the URL. If the webpage has other shenanigans that are responsible for me not being asked, I'm not sure how I would suss that out.

If you haven't already, you might also need to change the setting on the General panel of the Options page: under Downloads, "Always ask you where to save files." It's covered in this article: Startup, home page, tabs, and download settings.

Does that help?

What that change might reveal is that the download isn't specifically tagged by the server as an "application/pdf" file but is given some other designation that bypasses the PDF-specific setting.

Hi, jscher2000. Setting the parameter as you suggested caused FF to ask me where to save the PDF file, which is better than saving it to Downloads, but still doesn't ask me whether I want to simply open the file or the save it -- at least for the specific problem link that caused me to post. It's just a text URL in a gmail message, and if I hover over it, FF simply shows the URL on the lower left. At first, I wasn't sure whether the sender client mail reader converts it to a hyperlink or whether the gmail server on the recipient side does this. Or maybe neither does it, and it is FF doing it. I made gmail show me the unmolested source for the email message. The headers contain "Content-Type: multipart/alternative". The first part says:

Content-Type: text/plain; charset="UTF-8" Content-Transfer-Encoding: quoted-printable

The second part says:

Content-Type: text/html; charset="UTF-8" Content-Transfer-Encoding: quoted-printable

While the URL is plain text in the first part, it is a hyperlink in the 2nd part. The code is:

<a href=3D"http://www.someDomain.ca/Some/Path.pdf" target=3D"_b= lank">www.someDomain.ca/Some/Path.pdf</a>

Does that shed any more light on the reason for the behaviour?

Hmm, I get good results with an attachment, but a link... it really depends on the destination server. As shown in the attached screenshot, Firefox knows what to do with the file when the server indicates the content type is "application/pdf". But in other cases, you may not get an option to open it.

Note that the headers were viewed using the Network Monitor tool. When you click the download and the dialog appears, by that time the request/response should be recorded at the bottom of the Network Monitor and you can cancel the download.

Interesting. I'm not an IT guy, nor a developer, so the environment that you displayed is unfamiliar to me, but from the presence of the word "attachment", it looks like it's an email attachment rather than a URL in the email message (as suggested by your last post). Also, the qualifier "application/pdf" is something that I would normally associate with the metadata of email.

If I click on an *attached* PDF file, I do in fact get asked whether I want to save or open.