Sascha Steinbiss
Sascha Steinbiss
Could you probably narrow down your input file `copia_LTR.sorted.gff3` to an offending feature and make it available to us for debugging? We will need to reproduce the issue before we...
I see, thanks Gordon. It is not urgent, I was trying to help someone extracting these data and was baffled that this didn't work. I think that for overlapping top-level...
That's what I meant: ``` ----------- ---------- gene ----------------- ===========| |===========| intergenic ``` So between the two overlapping genes there would be no intergenic feature. Maybe we both meant the...
Sure, I can imagine many of these cases currently found in the format strings are remains of a search/replace run during the transition from `%lu`. So I am OK with...
Looks like something to do with pseudo node generation in nodes with `Parent`s set but with no `ID`s. The GenomeTools parser is trying to be smart here but seems to...
The node to be removed in the offending call to `gt_queue_remove` is very weird: a feature node with no line number, no attributes and no type. ``` (lldb) p *(GtFeatureNode*)...
I haven't looked at it for quite some time, but IIRC you don't need to run it twice. The detection is done on the inverted repeat structure, which is the...
Nice! I will have a look at it tomorrow.
Thanks for the initial work!
Due to this problem Ruby binding tests are currently disabled on Mac OS X. They must be re-enabled when this problem is fixed.