[Biojava-l] Newick/Nexus processing of non-binary trees

[Tiago Antão]

Hi,

> My suggestion: for somebody else to verify my findings. I might be
> doing something stupidly wrong. Maybe things are correct. Just a
> simple tree like (1,2,3) (as long as it is not binary) - should expose
> the problem.
>

Has nobody answered here is my take:

1. The error reported probably exists
2. Most probably nobody is using the parser (as it only supports binary trees).

In this light, changing the API should not be a problem at all.

I would not mind correcting the problem (I have already corrected the
previous 2 ones in my local version).
I would suggest removing the call to the unweighted graph. Reasons:
1. A weighted version is enough. If branch lengths are not specified,
then weights could be set to 0. There there would not be a decrease in
functionality.
2. Severely reducing the size of the code is important. Clearly the
code is not much maintained (and I am not offering to maintain it in
the long run, just putting it in good shape) and not much used.
Therefore a smaller, more easy to manage code base makes even more
sense.

If you accept a solution along these lines. I would correct all the
bugs and also include test code (which is also missing).



-- 
"The hottest places in hell are reserved for those who, in times of
moral crisis, maintain a neutrality." - Dante