I see no reason why that sort of encryption would be crack-able, if correctly implemented:
Since both SHA-1 and AES-256 are impossible to crack (though they both have some vulnerabilities - see Wikipedia), and the key encrypted with SHA-1 is created using a salt (supposedly created using a PRG), this should be fail-safe (although i can see no improvement over simple AES-256).
Even if several vulnerabilities would be found to this algorithm, it would still be robust enough to encrypt data nowadays with no chance of cracking (and no viable brute-forcing).
Edit:
Should someone manage to crack this, please go ahead and try cracking AES and SHA-1, though :). There should be a lot more money than $200.000 for those. ;)