hashcat Forum

For wordlists generate with expander it's easy to generate a lot of short candidates with combinator attack. That's a waste of time if one knows there won't be any plains of say, less than length 8. I don't necessarily want to take out all length 1 or length 2 words, because they are useful when combine with length 7, length 6, etc.

Allow specifying minimum and maximum length of plains to be attempted with combinator or hybrid attacks.

I know it can be done by splitting up the wordlists by length and scripting it, but that becomes tedious if there are a lot of different wordlists to try.

This would be especially helpful for autocrack, since the masks are picked at random right now. I guess I could program around it by selecting the left side first, then choosing a right side that works well with the chosen left side...

This is not possible, at least not for the fast algorithms like MD5 because of how the combinator engine and the rule engine is designed. That works for the slow algorithms like WPA and md5crypt. For the fast algortihm, all i could do it so check the length of each word in the dictionay for a specific length and sorting out all non-matching. But from how I understood your suggest, this wont help. What you think?

I was afraid it might take just as long to check length as to generate the hash for fast algorithms.

I can't think of a simple way to implement such a feature w/o checking the candidate length just before computing the hash. Any other way I can think of would be complicated to explain to users.