Open Bug 410737 Opened 15 years ago Updated 18 hours ago

Double download for saving files


(Firefox :: General, defect)

3.6 Branch
Windows XP





(Reporter: ehume, Unassigned)


(Whiteboard: DUPEME)

User-Agent:       Mozilla/5.0 (Windows; U; Windows NT 5.1; en-US; rv: Gecko/20071127 Firefox/
Build Identifier: Mozilla/5.0 (Windows; U; Windows NT 5.1; en-US; rv: Gecko/20071127 Firefox/

Saving a file that is already being displayed by Firefox requires a separate download for the filesave. IE does not seem to require a separate download.

Why not save the file from memory? 

Reproducible: Always

Steps to Reproduce:
1. View an image or a pdf.
(If the image or pdf is large, you will see that it takes a while to fully download for viewing in Firefox.)
2. Save it.
Actual Results:  
The file will have to be downloaded all over again. Thus a 25MB pdf becomes a 50MB combined viewing and saving download. 

Expected Results:  
IE behavior: download it once, save it from memory.

With pdf's one can specify that Foxit reader open the file outside Firefox. Then, when you want the file, you can use the Downloads Manager to open the containing folder to the temp directory, then cut the file from there and paste it where you want to keep it. OK, it's kludgy but it saves a second download.
This bug was originally reported on Firefox 2.x or older, which is no longer supported and will not be receiving any more updates. I strongly suggest that you update to Firefox 3.6.6 or later, update your plugins (flash, adobe, etc.), and retest in a new profile. If you still see the issue with the updated Firefox, please post here. Otherwise, please close as RESOLVED > WORKSFORME
Whiteboard: [CLOSEME 2010-07-30]
Version: unspecified → 2.0 Branch
This is still the way Mozilla saves files. 

Try it: go to

Open a date. Right-click on a picture. Select save. 

It reconnects to a server and downloads the pic from the server - and it's already on your machine. Ditto text files.

Interestingly, PDF files use the Adobe Reader save function, which seems to save from memory.

Check it out for yourself. This clunky method is so familiar to you you no longer see it for the bug that it is.
So you are thinking we should save from cache?
Whiteboard: [CLOSEME 2010-07-30]
Version: 2.0 Branch → 3.6 Branch
This has been a problem for a couple of years now. It happens in Windows and it happens in Linux and has happened ever since the beginning of 3.* as far as I can recall.

The problem is that firefox is re-issuing the request for the image when:
a) You right click to view image
b) You right click to save as

This is undesirable behaviour for both the client/user and the server.

For the client/user because:
a) The image is downloaded twice, wasting bandwidth.
b) The server may not respond the same way the second time and so the client/user may not get the desired image (for instance if the server checks the referrer it will respond with a web page complaining about hotlink protection)

For the server because:
a) The image is downloaded twice, wasting bandwidth, taking up request space, increasing traffic.

The only reason it may be undesirable is if a server does not want you to save an image for some reason. But if a server does not want you saving an image then it should not be making that image available on a public network.

When viewing an image and when saving an image already downloaded Firefox should _always_ download that image from local memory and _never_ re-issue the request. It _must_ already exist in local memory somewhere as it is being viewed in the browser.
Also, to really labour the point:

The preferred cache behaviour specified by the server is _irrelevant_ in this case. There should be no determination of whether the image should be re-requested.

When a user right clicks to view the image that means the user wants to take the image as it is presented to them and view is by itself, away from the page. That means taking the image held in memory and presenting it by itself. It has _nothing_ to do with making requests to a server.

When a user right clicks to save the image that means the user wants to take the image as it is presented to them and store it on their hard drive. That means taking the image held in memory and writing that data to a specified hard drive. It has _nothing_ to do with making requests to a server.
I am grateful that Richard has found related bugs. The earliest is from 2002. 

Some comments suggest that the second download window is actually showing progress of moving an image from cache to HD. I must admit that this had never occurred to me. If true, it would remove my original objection: the second true download from the Internet (thanks for clarifying, Richard). If true, it should either be labeled correctly, which would require yet another dialog; or proceed unremarked, which would present different problems.

It all depends on whether there is in fact a second download.
There is definitely a second download occurring.

This is evidenced by when sometimes the server will respond with text/html (a web page) that complains about hot-linking, or a substitute image that complains about hot-linking.

It can also be seen in a network monitor that bandwidth is being used.

You can even use the plugin Tamper Data to see that Firefox is re-issuing the request to the server.
Whiteboard: DUPEME
Also related is a bug report I filed a while ago and forgot about that someone just commented on.
Severity: normal → S3

I realize this is very old, but I'm not observing this behavior anymore on Firefox 111 64-bit on Windows 11 x86_64. Perhaps it's been fixed?

You need to log in before you can comment on or make changes to this bug.