Compiling a build:
You will need to target the 10.7 SDK because Liten depends on a deprecated API calls.
The original project was started by building off of some sample code. The core algorithm of finding duplicates was built from this Python code: http://code.google.com/p/liten/
Essentially, an md5 checksum is only performed after the size of the file matches a key in a hash. This makes for a pretty efficient tree walk.
You can download the application without compiling it here:
- Updating the deprecated API calls to 10.8
- Fixing various UI bugs
- Getting a better sort in the UI for human readable duplicates.
- Would love to do a folder merge.
- Better bulk delete