Hi,I wanted to take a look at the parser anyways, so I took the opportunity.
As far as i see this, the newline is just a minor part of the problem. I think the bigger issue here is parsing inner node labels. I attached a patch that fixes the problem, at least for inner nodes with label and inner nodes with label and weights. Wikipedia states that Newick allows leaves without any labels, but in case of phylogenetic trees I think one can safely ignore this, though the parser should maybe throw an exception.
If you are interested I also updated the unit test. hope it helps, cheers, -thasso On Dec 15, 2009, at 12:43 , Tiago Antão wrote:
2009/12/15 Richard Holland <[email protected]>:Hi there.I believe the code used to be able to parse this kind of tree, but TIago recently rewrote it so I'm no longer certain.Tiago - your new code doesn't seem to be coping with the insertion of a newline at random points in the Tree string. I think you might need to modify your tokenize() method to handle this better? Could you also add a unit test using this particular tree.Yep, will take care of this over the weekend. Maybe before, but no promises._______________________________________________ Biojava-l mailing list - [email protected] http://lists.open-bio.org/mailman/listinfo/biojava-l
TreesBlock.patch
Description: Binary data
_______________________________________________ Biojava-l mailing list - [email protected] http://lists.open-bio.org/mailman/listinfo/biojava-l
