From: Egon W. <ego...@gm...> - 2008-11-06 16:15:24
|
Christoph, I think the RingSet code originates from you... what do you think about the below questions of Rajarshi? I have no clue... I'm all in favor of consistent behaviour but no idea on how much code actually depends the current inconsistent behavior... Moreover, this close to the 1.2 release, I'd rather not change the functional API... On Mon, Sep 29, 2008 at 2:43 PM, Rajarshi Guha <rg...@in...> wrote: > It looks like the add method of IRingSet does not allow you to add > rings from a RingSet that are already in the set. > > There is no add method that allows you to add a single Ring object to > the set. > > It looks like you're supposed to use addAtomContainer from the > superclass. However this method does *not* check for duplicates. > > So two questions > > 1. Is that the intended behavior (i.e., allow duplicates) for > IAtomContainerSet.addAtomContainer()? > > 2. If one is to add a IRing to a IRingSet via addAtomContainer then > this behavior does not match the behavior provided by IRingSet.add() > > Solution - either implement a method, add(IRing) in IRingSet that > prevents addition of duplicates. Or else, update addAtomContainer to > preevent addition of duplicates So, the question is: is this a bug or a intended feature? Egon -- ---- http://chem-bla-ics.blogspot.com/ |