gziped files

[jsgounot]

Hi,

would it be possible to allow parsnp to handle gzipped compressed files? This would save some processing and disk space when running parsnp on a lot of genomes that are already compressed.

I initially thought I could update the main python script, but it seems that a change needs to be made in the core algorithm as well. During that time, I noticed that the sequence lengths are miscalculated multiple times for multi-fasta files, like here, where other headers are used in the reference length calculation.

Regards,
JS