`Floating point exception (core dumped)`

[subwaystation]

Hi there,

when running

wfmash pangenome.fa.gz pangenome.fa.gz --threads 2

I get

[mashmap] MashMap v3.1.1
[mashmap] Reference = [pangenome.fa.gz]
[mashmap] Query = [pangenome.fa.gz]
[mashmap] Kmer size = 19
[mashmap] Sketch size = 298
[mashmap] Segment length = 5000 (read split allowed)
[mashmap] Block length min = 25000
[mashmap] Chaining gap max = 20000
[mashmap] Mappings per segment = 1
[mashmap] Percentage identity threshold = 90%
[mashmap] Do not skip self mappings
[mashmap] Hypergeometric filter w/ delta = 0 and confidence 0.999
[mashmap] Mapping output file = /home/heumos/git/test-datasets/data/pangenomics/homo_sapiens/wfmash-QrP6qP
[mashmap] Filter mode = 1 (1 = map, 2 = one-to-one, 3 = none)
[mashmap] Execution threads  = 2
[mashmap::skch::Sketch::build] minmer windows picked from reference = 0
[mashmap::skch::Sketch::index] unique minmers = 0
[mashmap::skch::Sketch::computeFreqHist] No minmers.
[wfmash::map] time spent computing the reference index: 0.00405686 sec
[mashmap::skch::Map::mapQuery] mapped  0.00% @ 0.00e+00 bp/s elapsed: 00:00:00:00 remain: 00:00:00:00Floating point exception (core dumped)

The file can be found at https://github.com/nf-core/test-datasets/blob/modules/data/pangenomics/homo_sapiens/pangenome.fa.gz.

Thanks for any feedback!

[subwaystation]

Alright, the problem is that if I set a -s smaller then the sequences in the FASTA, this happens. If I set e.g. -s 500, then this does not happen.

[ekg]

Might be worth throwing a warning rather than a segfault.