wwwoffle-users  

Re: [WWWOFFLE-Users] making requests from firefox without creating windows

Micha
Wed, 15 Mar 2006 22:44:03 -0800


I found a solution for firefox.

1. You need the extension 'flashgot': 
Extras -> Extensions -> Download more
or <https://addons.mozilla.org/extensions/>

By case it's the top one anyway :)
You should install the #2, NoScript, right along, too.

2. Go via Extras -> Flashgot -> More options, and  
    add a new 'download manager': wwwoffle (no options)

3. Your done :)


-> You can mouse-hover over a link and type ctrl-F1 to load that 
     link with wwwoffle (without creating a window or tab).

-> You can mark part of the page and apply ctrl-F2 to load the 
     marked links.

-> You can hit ctl-F3 to load all links on that page.

It works both with wwwoffle offline (creating orders) 
or online (loading immediately). All these features are 
available via mouse context menu, too.

If you configure a button or hotkey for switching wwwoffle on/offline
then you can harvest download orders and load them later via another 
'fetch now' button. I use to do this via wmaker dockapps, but there
should be solutions for any kind of desktop, including keyboard hotkeys.

If you puzzle about firefox options for accessing external apps, mouse click 
behavbiour, tab behaviour,  key shortcuts or setting up the context menu, 
you may want to discover the Extensions->Preferences settings too.

I wonder if you can, for example, configure the 'Send Page to' extension 
to call a wwwoffle refresh/order/whatever page, with %s URI placeholder.
That could spare the javascript of any bookmarklets.

However, it depends on what extensions you install.
Me got:
tabprefs, flashgot, vrs, linkification, noscript, pdfdownload, diggler, 
live-http-headers, extended-context-menu, image zoom, nuke images,
web developer, DOM inspector, Checky, and some javascript stuff.
Hint: FasterFox and MouseGestures can cause problems.

It would be interesting to know if someone can come up with a solution for 
different browsers, though. Other ideas i ventilated include,

* The issue would be solved most easily if there would be a pooibility to
  launch wwwoffle in 'silent' mode, that is, any clicked link will be loaded or
  ordered in the background without leaving the actual page. I don't know if 
  this is possible, from the browser side. But if so, then i could configure a
  desktop icon to switch the wwwoffle mode. I could sitch into silent mode, 
  order (or fetch imeediately) some links, switch back, and continue browsing.

* wwwoffle daemon could insert a little icon (say, 12pixel radius disc) before 
   any link, that would order/fetch the link in background. Via href or submit.
   Would be optional, of course. Would uglyfy some pages a little, but OTOH 
   works on *any* browser :)

* Do the same by ModifyHTML? If it's not possible to fetch in background this 
   way, try to insert a [EMAIL PROTECTED]@URL into said icon link.
   That would mean to trick wwwoffle...maybe by combining anchor insertion with 
   a faked dontget-replacement ? All in all, sounds too complicated. 
   This was not about hacking trojans, somehow.
   Anyway, it would be desirable if ModifyHTML would provide some generic
   plcaeholders, like %L for the link URL .... then it would be very easy. 

* Alternativly, explore the possibility to insert a css ID or class which 
activates 
   Event-Handler code OnClick or something similar, which could turn out to
   slip through the javascript filter as always 'local' if the the css file 
would
   be stored in the local filesystem, on harddisk. 

* If the actual page URL is needed with JS disabled, then maybe it's possible to
   get that URL into a script by poking in wwwoffle lasttime ? Or from the 
browser
   history, provided it gets stored on file immediately (which i doubt).

* What happens if i say 'go offline' to a browser ? It requests links from the 
cache.
  What happens if that cache is configured to be 0 MB ? It always returns 
'sorry' message.
  Could this event trigger a custom command (like a script) ?
  For example, there must be a request to the filesystem. Basically it must be 
possible
  to catch this request, and extract the URL,  replacing the cache dir by 
something different. 
  Maybe rather kernel level though.
  It's just, there's this nice little button in the corner of galeon, and it's 
completely useless.

  (It really looks like a switch proxy button. I proposed to make it 
configurable to galeon 
  devel at least 2 years ago, but you know what happened. And I guess there's 
already a 
  firefox patch to switch proxy connection on the fly out there.)



   °
 /\/