Mozilla/5.0 (X11; U; Linux x86_64; en-GB; rv:1.9.2) Gecko/20100305 Unity/1.9.2-0.1 (2010) Firefox/3.6.3

Background

I want to ease downloading from Rapidshare. Using the Skipscreen extension, I can queue up a bunch of tabs and then just go to bed. However, when any file is downloaded, I am always prompted by Firefox to decide what to do with the file.

Thankfully the download begins in the background and is saved to my temporary folder. So when I wake and I see a mess of dialogs, I don't actually have to wait for any downloads to happen. This is, however, a bit risky. If I lose power, for example, that effort is somewhat wasted. At best I'd have to have manual intervention to rescue the files.

So instead, I don't want to be prompted at all. I always want the files saved to disk with no prompt.

Failed steps

Provided so that others don't have to go through the mess that I did.

I renamed ~/.mozilla/firefox/profiles/$profile/mimeTypes.rdf and restarted Firefox.

Edit > Preferences > General > Downloads: "(o) Save files to" (selected a folder)

about:config and reset these:

browser.download.dir
browser.download.downloadDir
browser.download.folderList
browser.download.lastDir
browser.download.useDownloadDir

• browser.helperApps.alwaysAsk.force true
• browser.helperApps.neverAsk.saveToDisk and browser.helperApps.neverAsk.openFile
  • Failed:
  • Failed: application/force-download,application/rar,application/x-rar
  • Failed: file
  • Failed: *

I've heard that the above items are just copied/converted into mimeTypes.rdf. I undid these changes and deleted my mimeTypes.rdf again and re-tested.

References

• [1] File types and download actions
• [2] Managing file types
• [3] Unable to download or save files

• [4] Bug 315536 - "do this automatically for files like this from now on" often grayed out
  • I can't imagine how this could be considered fixed.

• [5] Bug 331259 - checking "do this automatically for files like this from now on" doesn't have any effect on future downloads of same file type

• [6] Bug 250914 - Clicking on WMV file opens What should Firefox do? dialog

• [7] Open in Browser - This extension allows you to open the document directly in browser. You can also change the mime type of a document being viewed.
  • This is a workaround extension until [8] and [9] are fixed.

Unsolvable

I learned that Firefox always prompts to save because Rapidshare does not have proper MIME types working. There is NO solution for Firefox. They refuse to work with files based on extension. Bastards.

Why?

: [10] Starting in Firefox 3 and SeaMonkey 2, the "Opening " dialog will also appear if you have selected "Always ask" as the action for the file's content type in your (Helper) Applications options/preferences.

Rapidshare Premium Account

One solution is to use a download helper with a Rapidshare premium account. You'd log into Rapidshare using your browser, and then use the helper. Some examples:

• [11] DownThemAll
  • In Firefox, you need to "Enable 3rd party cookies"
  • See also [12]
  • I've used this addon extensively, and it does actually work for non-premium accounts for Rapidshare, but it cannot deal with the timer on multiple files.

• [13] Rapidfire

• Various other Firefox extensions may exist, and lots of download tools probably do. Any number of them would probably be able to queue a list of Rapidshare links and save them to disk without prompting.

[https://addons.mozilla.org/en-US/firefox/addon/8376 RDown]

This no longer works.

Feature requests:

• It does currently remember the list between sessions. However, if Firefox is killed for any reason the list will be reverted to its previous state. To solve this, save the list after each user edit.
• Remember the window position and size.
• When processing an item, keep that URL as selectable for copying. Just ignore any user changes (reset it to what's being processed when any keystroke is detected)
• Add another text area. When items pass through the queue, deposit them in the new "completed" field. Currently the file is totally forgotten and the user needs to use a text file to remember all the filenames. This feature is important for files which (for whatever reason) fail to download. For example, RDown does not gracefully Rapidshare server downtime.

I think it's expecting to have one item per line. I can either copy directly if I'm given text links, or I can copy links in bulk using the Snap Links Plus extension. Unfortunately this extension copies them space-separated. I do a simple search-and-replace from   into \n with using Geany, with regular expressions and escape sequences both selected.

Note that Multi Links seems to be a good replacement for Snap Links because it can properly copy one-per-line, but I found a horrifying issue with it which would need to be solved before I could use or recommend it.