sourceforge.net download problems

Jude DaShiell jdashiel at shellworld.net
Tue Apr 9 10:17:18 UTC 2013


wget has limited correction capacity when files don't completely 
download.  I'd like to find software which could take advantage of gpg 
md5sum and the like to check a local download instance to ensure 
correctness before giving up on the download but have been informed such 
software is an impossibility.

On Tue, 9 Apr 2013, Tony Baechler wrote:

> I posted this before, but perhaps it didn't go to the list.  I don't
> understand why you bother with wget or Lynx when there is a much simpler way.
> Actually, there are two simple ways.  One is to use Firefox, but that requires
> X and isn't practical.  The other is to use a good ftp client, like lftp or
> ncftp.  Just navigate to the project directory you want and download away.
> The only drawback is that it's a bit slower for people in the US.
> 
> ftp://ftp.heanet.ie/pub/sourceforge/
> 
> That path might not be correct, it might be under mirrors/ or similar.  Once
> you find it, change to the project directory, as follows:
> 
> cd n/nv/nvda
> cd m/ma/maximus
> 
> You can then download files as you normally would with ftp and no Javascript.
> I downloaded the latest NVDA that way.  It also is a lot easier than using
> their project browser to determine the file listings.
> 
> On 4/8/2013 12:17 PM, Martin McCormick wrote:
> > 	I beat the system a couple of nights ago when I made the
> > same discovery you describe when trying to download something
> > from sourceforge using lynx. I could tell it had actually
> > downloaded but lynx couldn't finish the process. I wonder what
> > non-existant problem somebody cured with this design?
> >
> > 	What I did was to cd to /tmp/and then find the temporary
> > directory lynx created and inside that was the temp file just
> > waiting for the last few steps to complete which now never
> > happens.
> >
> > 	I copied that file out in to my directory and it did
> > work but It has that sulphur smell of javascript.
> >
> >  It sounds like using wget is a much better solution.
> >
> > 	I had tried to use curl but not that hard so curl might
> > also work. Thanks for the tip.
> >
> >
> > Jude DaShiell writes:
> > > It used to be we could use a browser and download things from
> > > sourceforge.net and lynx even worked.  That's not the case any longer.
> > > All that happens now is you get 10 redirects and get thrown off the page.
> > > What does work though, is to copy the sourceforge.net url from what would
> > > be where the file was going to be pulled from and drop back to the local
> > > box and run wget -b and paste the url in.  Apparently wget is able to get
> > > these files and the bonus is the file names aren't mutilated as they are
> > > when a browser is used to download them from sourceforge.net.
> > >
> > > ---------------------------------------------------------------------------
> > > jude<jdashiel at shellworld.net>
> > > Microsoft, windows is accessible. why do blind people need screen readers?
> > >
> > > _______________________________________________
> > > Blinux-list mailing list
> > > Blinux-list at redhat.com
> > > https://www.redhat.com/mailman/listinfo/blinux-list
> > >
> > >
> >
> > _______________________________________________
> > Blinux-list mailing list
> > Blinux-list at redhat.com
> > https://www.redhat.com/mailman/listinfo/blinux-list
> 
> 

---------------------------------------------------------------------------
jude <jdashiel at shellworld.net>
Microsoft, windows is accessible. why do blind people need screen readers?




More information about the Blinux-list mailing list