stats.sh buffer overruns?

[colindaven]

Hello,

thanks for maintaining this on github now!

I've been using bbmap stats.sh for quite some time. It works great for contigs as intended (at least on <1million contigs).

Since I like it so much, I tried using it on before and after filtering nanopore read comparisons.

When running on 1-2million reads or more I notice stats getting very wacky and disagreeing with other read and total sequence counting tools

I'll try to find a public read set example. But is there an obvious reason this might be happening (eg using int not long etc in counting functions ?)

Thanks,
Colin