I recently needed to make a screenshot of an entire web page, i.e. not just the part the fits on the screen (or in the viewport), but the whole thing from top to bottom.

Somehow I missed the possibility that there might be a zillion browser extensions for just this purpose. From writing features (acceptance tests) with Cucumber I’m familiar with PhantomJS, a “headless”, scriptable browser based on WebKit.

Here’s how it works:

$ webshot --help
Usage: webshot [OPTIONS]... URL IMAGE_FILE

Options:
          -j|--jquery           Inject jQuery into the page.
         +-e|--exec «value»     Execute script in page context.
         +-f|--file «value»     Execute script file in page context.
          -v|--viewport «value» Viewport size WxH; default 1280x800.

          (* mandatory option)
          (+ repeatable option)

Luckily, I didn’t have to write the commandline parsing myself. But it was fun to figure out how Deferreds might be implemented.