Flip easily between media sites. See http://iangilman.com/interchange/
- Node (for build system)
- Install the prerequisites.
- Download this repository.
- In your Terminal,
cd
into the repository directory. - Run
npm install
to load Grunt, the build system. - Run
grunt
once to make the bookmarklet. - Run
grunt server watch
for as long as you're developing; it'll automatically make the bookmarklet whenever you make changes, and it'll serve the files at http://localhost:8000/ for you. - If everything is working, you should be able to go to http://localhost:8000/js/bookmarklet.js and see a file full of code.
- In your browser, add a new bookmark to your bookmark bar, and give it this for the URL:
- javascript:var%20b=document.body;if(b&&!document.xmlVersion){void(z=document.createElement('script'));void(z.src='http://localhost:8000/js/bookmarklet.js');void(b.appendChild(z));}else{}
- Now go to a product page on, say, Amazon, and hit the bookmarklet in your bookmark bar.
Grunt automatically combines bookmarklet-base.js, jquery.js, and common.js into bookmarklet.js. It does the same for widget-base.js, jquery.js, and common.js, turning them into widget.js. In general you should only modify files in the src
folder; the js
folder is just for the built files.
Most of the interesting stuff is in common.js. At the top, there is a big block of sources (for reading off of websites) and destinations (for linking to websites).
Any additional questions, contact [email protected].
- Support http://dvd.netflix.com/Movie/Now-You-See-Me/70243464 URLs
- Goodreads is now blocking bookmarklets... can I get around this with an addon?
- http://www.worldcat.org/
- Explicit close box
- Watch for window size events and resize/reposition accordingly
- Add google analytics or some such for tracking
- Support Vdio for both in and out
- auto-show which sites have sources and destinations on the info page, based on the data in the script
- Have it be a link to an interchange page with the proper search, so it'll work in RSS feeds. Then override via JavaScript if possible for popup.
- live preview on creation page
- List sites in site error message
- Array of services
- Try to match Rdio urls better so they don't have to redirect
- When getting names from Rdio, make sure to remove underscores globally (now just getting first one)
- Wikipedia would be another destination
- Actually negotiate with SPL over the proper search while the box is up?
- Foreign characters (such as in bjork's name) don't work
- Support other biblio commons cities
- Get ISBN out of goodreads
- Better error reporting
- Pretty icons for the services
- More services
- Service plugin architecture
- Better info page
- Define our own link colors
- Better ISBN search for Amazon
- The Amazon mobile site doesn't have isbn, so we think it's music; we need better detection, or an "unknown" mode for that situation.