Should we rename the allow_bad_kmers option in consume to make more intuitive?
This opt is passed to SeqToHashes as the force argument.
Behaviour seems to be:
True - Skip over bad kmers, do not hash them, do not raise error.
False - hash/count all kmers up to the first non-DNA character then raise an error and stop processing the input sequence.
I would expect allow_bad_kmers = True to hash all kmers including those with non-dna characters, and False to skip the bad kmers and carry on with processing.
Should we rename the allow_bad_kmers option in
consume
to make more intuitive?This opt is passed to SeqToHashes as the
force
argument.Behaviour seems to be:
True - Skip over bad kmers, do not hash them, do not raise error. False - hash/count all kmers up to the first non-DNA character then raise an error and stop processing the input sequence.
I would expect allow_bad_kmers = True to hash all kmers including those with non-dna characters, and False to skip the bad kmers and carry on with processing.
Suggest change option to
skip_bad_kmers