On Sun, May 09, 2010 at 12:30:41PM +0400, Konstantin Tokarev wrote: > On Sun, 9 May 2010 10:16:13 +0200 > Michael Banck <mba...@gmx.net> wrote: > > > On Sun, May 09, 2010 at 11:19:14AM +0400, Konstantin Tokarev wrote: > > > > Yeah, that is pretty annoying; however, I almost always use an > > > > external optimizer anyway so haven't taken a close look at this. > > > > > > Out of curiosity: what optimizer? is it publicly available? Yes, > > > Molpro's internal is a holy crap for large molecule's, even authors > > > consider it. > > > > No, it's something hackish my Ph.D. advisor came up with some time > > ago. > > > > But I think it should be possible to reconstruct the geometry for > > > > each step, and this might even be some sort of "killer feature" > > > > for Molpro users. > > > > > > Why reconstruct if it's simplier to read cml? > > > > Because there is non-XML Molpro output as well. > > I don't understand why it's needed to parse plain text if XML is > always generated since 2008.1 (of course, if some looong key is not > specified in command line). IIRC, optional XML output was in 2006.1, > too.
Well, using 2006.1 without XML output is perfectly reasonable (and what I happen to do mostly). So I think it would be ok to accept patches which implement geometry optimization support for non-XML output, if somebody bothers to do the work. This is probably getting off-topic with respect to the initial problem, though. Michael ------------------------------------------------------------------------------ _______________________________________________ OpenBabel-Devel mailing list OpenBabel-Devel@lists.sourceforge.net https://lists.sourceforge.net/lists/listinfo/openbabel-devel
