Hi Greg,
Thanks - it works with smart pointers but I got another problem: To display structure I need to put string MDLMol into the control. If I call compute2DCoords - it raised an exception but if do not - all atoms in one dot and you cannot see it at all The same is if I try to read call sanitizeMol. Looking like problem in Kekilizer because some of my fragments can contain atoms and bonds taken from the aromatic ring (and it must be so). Simple fragments like chain fragments displayed Ok because they are not causing exception in compute2Dcoords. What can you suggest? Regards, Evgueni > -----Original Message----- > From: Greg Landrum [mailto:greg.land...@gmail.com] > Sent: 26 June 2009 19:45 > To: Evgueni Kolossov > Subject: Re: Fragmentation > > Hi Evgueni, > > storing molecules in an std::vector just isn't going to work. Use > smart pointers (RDKit::ROMOL_SPTR) instead. > > On Fri, Jun 26, 2009 at 10:50 AM, Evgueni > Kolossov wrote: >> Hi Greg, >> >> No, the solution you suggested is not working. >> I have found the problem but I need your help to sort it out: >> >> I am storing fragments in the std::vector<RDKit::ROMol> m_vFrags >> Looking like near all operation with this vector causing the problem >> with operator '=': >> m_vFrags.clear() ->problem >> >> RDKit::ROMol *mol; >> mol = RDKit::SmilesToMol(strSmi); >> m_vFrags.push_back(*mol) ->problem >> >> etc. >> >> What you can suggest? >> >> Regards, >> Evgueni
