Screengrab! is a GPL3 piece of code, just so you know.

I’m also working on a new release that should resolve a few of the long-standing complaints that people have with Screengrab! and will add some nifty new features.

See the Google Code issues page for a list of what I’m working on for the new release.

This isn’t exhaustive, and none of it is guaranteed to happen by the next release.

The biggest changes going in are under the covers and involve using Java AND the Canvas to always accurately capture what is in the page.

I embarked on a big code cleanup and rework expedition, which consumed my passion and tired me out, so I haven’t really looked at this in a while. I’ll get back to those little things that people want soon.

However, here is an increment of the version counter. Expect another one when Firefox 3.0 comes out.

It’s currently waiting for approval, but it should be out there soon. Good luck!

Head over to BookSmarts to watch the work very slowly progress!

The canvas grab is thanks to me stumbling on this page a couple of years ago and then waiting for the feature to be fully implemented in the browser. The motivation came from my main competitors Pearl Crescent Page Saver (their grabber is pretty good - but you have to pay for the selection box). who had been previously capturing images using some very clever native libraries to do their work. I’m not that smart, so I was using the language I know: Java. My ego decided that I couldn’t lose the Firefox screengrabbing race so I got back to work.

(There’s also another extension called Snapper that does this now. It’s good for quick shots.)

The copy feature was inspired by all of the people who said they wanted it as a feature. You would not believe how hard it was to do. I’m not even doing it right, it’s a hack. I still don’t know how to put image data on the clipboard using the interfaces provided.

Anyway, here’s how the hack works…

• I grab the page and get the dataUrl from the canvas.
• I load the dataUrl into a hidden image in my extension chrome (it’s in the main overlay).
• I then trick Firefox into thinking that the last thing the user clicked on in the browser was this image that they can’t possibly see (so that it thinks the context menu would be using it).
• Then I invoke the command that drives the “Copy Image” context menu item.

(There’s also a couple of timers in there while I wait for Firefox to get with the program)

If you think that’s hacky, one day I’ll explain how I get Java to take those shots and scroll the page. It’s CRAZY!

It’s coming. I found a bug getting the visible portion using Canvas while making this site, so I just need to squash it.

I also need to add my cool new icon to everything, incorporate some locales and get the Firefox extension guys to put it up on their site. I’m going to stop hosting the download myself.

This is it. It even has its own domain. Amazing.