Max Bachmann

@gfairchild are there any plans to geht this merged? Otherwise I will close the PR.

Unless I have a special interest in a PR, I close them at some point around the 1 year mark. However feel free to merge this into a branch and...

Python destroys objects at some point after they are no longer referenced. In your case this means that when calling ```python parasail.sw_trace("asdf", "asdf", 11, 1, parasail.blosum62).cigar.seq ``` the object is...

I wouldn't call this a bug. It is clearly defined, that editdistance uses the hash value to compare two objects for equality. I do not really see a better way...

> What would be the problem with checking for object equality rather than checking for hash code equality? (Are you just concerned about speed?) Yes this implementation has the goal...

> There are a few things there I don't understand, possibly due to my own ignorance. Edit distance is said to be O(MN), where M and N are the lengths...

> this has to be done for every single element. In addition note that even though this equal comparision is only O(K), it has a large constant factor. It requires...

It is worth noting, that `Levenshtein.ratio` does not return the normalized Levenshtein distance which is calulated as: ``` 1 - lev_dist / max(len1, len2) ``` but returns the normalized InDel...

> I think this is also not correct though. It should be the length of the alignment path as denominator (which is slightly longer than the longest sequence). I am...

At least in my benchmarks this is largely dependent on the length of the input strings. Here is a comparision for different libraries using different string lengths. Both edlib and...