arq5x / lumpy-sv

lumpy: a general probabilistic framework for structural variant discovery
MIT License
307 stars 119 forks source link

There seems to be some bugs in the scripts l_sort.py&l_merge.py #339

Open MaoSihong opened 4 years ago

MaoSihong commented 4 years ago

Actually, these two scripts both are not python2/3 compatible:

l_sort.py use the code line as "sort(cmp = )" however the cmp para is already canceled in python3(at least it's not availabel in 3.6)

l_sort.py&l_merge.py both use code like 'print(i,end = '')' . 'end' para is not used in python2 to take the place of "\n". Just use "print(i)," is OK!

MaoSihong commented 4 years ago

well,I found another coding error in l_merge.py that is doing deletion while traversal the items in dict. BUT the right way is to traversal the dict.keys() instead to avoid the dict crash. by the way if we want to use sort(cmp=,) in python3 we can import functools and then assign it to the para ‘key’

ryanlayer commented 4 years ago

Thanks for digging through and finding the bug!

The latest wrapper for running lumpy, including merging multiple runs, is smoove

https://github.com/brentp/smoove

Please give it a try

On Jun 24, 2020, at 10:26 PM, MaoSihong notifications@github.com wrote:

 well,I found another coding error in l_merge.py that is doing deletion while traversal the items in dict. BUT the right way is to traversal the dict.keys() instead to avoid the dict crash. by the way if we want to use sort(cmp=,) in python3 we can import functools and then assign it to the para ‘key’

— You are receiving this because you are subscribed to this thread. Reply to this email directly, view it on GitHub, or unsubscribe.