Netscape/Mozilla problems downloading dmg.gz files

ddldreaming · 01-21-2002, 04:42 PM

I have a minor problem in both Netscape 6.2.1 and the latest Mozilla builds: clicking on a link to a *.dmg.gz file will display the file's contents in the browser window, rather than downloading it to disk. While I can easily work around it by right[control]-clicking on the link and selecting "Save Link As...", doing so often assigns the wrong filename -- and it can often be hard to tell in advance what the actual filename is, since many links are CGI redirections. (FWIW, if a file has a single extension (e.g. "file.gz"), Mozilla handles it fine.)

I've tried setting up helper applications for this extension in the browser (using application/x-gzip, application/x-zip-compressed and application/octet-stream as MIME types), and I've messed with the mimetypes.rdf XML file directly, both without success. In testing it on my local webservers, I've also modified /etc/httpd/mime.types to try and force the server to send the file with the appropriate MIME type -- no dice (but I may not be doing it right).

My dad has the same problem with a brand-spankin'-new install of Netscape for OS X, so I know it's not just me. (Actually, in googling around I came across scattered other references to similar problems on other platforms too.) I was hoping to figure it out and post a nice tip to www.macosxhints.com, but I'm stumped. Any guesses?

thanks,
-d.d-

Benad · 01-21-2002, 05:13 PM

Control-click on the link and choose "Save As...".

I also think that download support in Mozilla/Netscape6 is very flaky. I use the command-line program "curl" from my downloads (get it from Fink, but I'm sure there are some shareware or free GUI applications that are made for that.

Also, iCab is pretty nice with downloads.

- Benad

tomsinclair · 01-22-2002, 12:45 AM

I get that as well with NS 6.2

Usually, I Ctrl-click on the link and select Copy Link Location then paste it into the command line like so:

curl -O <pasted URL>

This usually works, except with URLs that use redirection and have a lot of bogus characters like & in them.

I haven't tried the Save As yet, though. That's a good one.

phramok · 04-01-2005, 09:03 PM

Wow, the Curl package is really powerful. Thanks to you guys for bringing it to my attention.

Downloading from redirects is really a problem with all of my browser's. Just I wanted to download a Chinese music file from this address:

http://www.shanghaining.com/download...Hrap/PP928.mp3

Maybe I'm just stupid, but I can't stop my browsers from opening this up in Quicktime every time I click on the link, when of course what I really want to do is download it.

At first, Curl still did not seem to work; running

curl -O http://www.shanghaining.com/download...Hrap/PP928.mp3

causes it to open a connection and immediately close it again. But! running curl in verbose mode:

curl -V -O http://www.shanghaining.com/download...Hrap/PP928.mp3

yields the real location: http://download.shanghaining.com/SHrap/hqzd.mp3

Finally, I got the file! (Too bad it sucks!)

01-21-2002, 04:42 PM	#1
ddldreaming Prospect Join Date: Jan 2002 Posts: 3	Netscape/Mozilla problems downloading dmg.gz files I have a minor problem in both Netscape 6.2.1 and the latest Mozilla builds: clicking on a link to a *.dmg.gz file will display the file's contents in the browser window, rather than downloading it to disk. While I can easily work around it by right[control]-clicking on the link and selecting "Save Link As...", doing so often assigns the wrong filename -- and it can often be hard to tell in advance what the actual filename is, since many links are CGI redirections. (FWIW, if a file has a single extension (e.g. "file.gz"), Mozilla handles it fine.) I've tried setting up helper applications for this extension in the browser (using application/x-gzip, application/x-zip-compressed and application/octet-stream as MIME types), and I've messed with the mimetypes.rdf XML file directly, both without success. In testing it on my local webservers, I've also modified /etc/httpd/mime.types to try and force the server to send the file with the appropriate MIME type -- no dice (but I may not be doing it right). My dad has the same problem with a brand-spankin'-new install of Netscape for OS X, so I know it's not just me. (Actually, in googling around I came across scattered other references to similar problems on other platforms too.) I was hoping to figure it out and post a nice tip to www.macosxhints.com, but I'm stumped. Any guesses? thanks, -d.d-

01-21-2002, 05:13 PM	#2
Benad Prospect Join Date: Jan 2002 Location: Montreal, Canada Posts: 38	Maybe... Control-click on the link and choose "Save As...". I also think that download support in Mozilla/Netscape6 is very flaky. I use the command-line program "curl" from my downloads (get it from Fink, but I'm sure there are some shareware or free GUI applications that are made for that. Also, iCab is pretty nice with downloads. - Benad

Thread Tools
Show Printable Version Email this Page
Display Modes	Rate This Thread
Linear Mode Switch to Hybrid Mode Switch to Threaded Mode	Rate This Thread:

01-22-2002, 12:45 AM	#3
tomsinclair Prospect Join Date: Jan 2002 Posts: 7	I get that as well with NS 6.2 Usually, I Ctrl-click on the link and select Copy Link Location then paste it into the command line like so: curl -O <pasted URL> This usually works, except with URLs that use redirection and have a lot of bogus characters like & in them. I haven't tried the Save As yet, though. That's a good one.

04-01-2005, 09:03 PM	#4
phramok Prospect Join Date: Mar 2005 Posts: 15	Wow, the Curl package is really powerful. Thanks to you guys for bringing it to my attention. Downloading from redirects is really a problem with all of my browser's. Just I wanted to download a Chinese music file from this address: http://www.shanghaining.com/download...Hrap/PP928.mp3 Maybe I'm just stupid, but I can't stop my browsers from opening this up in Quicktime every time I click on the link, when of course what I really want to do is download it. At first, Curl still did not seem to work; running curl -O http://www.shanghaining.com/download...Hrap/PP928.mp3 causes it to open a connection and immediately close it again. But! running curl in verbose mode: curl -V -O http://www.shanghaining.com/download...Hrap/PP928.mp3 yields the real location: http://download.shanghaining.com/SHrap/hqzd.mp3 Finally, I got the file! (Too bad it sucks!)