[Fsatom] Re: CML and macromolecules

[Joerg K. Wegner]

Hi all,

 > The problem
 > with all conversion programs is that you are constrained by the data
 > structures they provide. Not their fault...

As surely known: JOELib can handle all kind of data structures you like 
using the reflection mechanism under Java... C++ can NOT support such a 
mechanism, or am i wrong, because the factoy pattern must ba always hard 
coded ?

http://www-ra.informatik.uni-tuebingen.de/software/joelib/index.html

E.g.:
1. Implement your DataResult class
2. Tell JOELib that these entries are known (e.g. in SDF format) in 
joelib/data/plain/extKnownResults.txt, where you can use a name or 
regular expressions.
3. If you will access a descriptor/dataStructure 
(mol.getData("myData",parse=true)) JOELib can parse the text if it knows 
the type.

Here is the list of supported PLAIN IO-types:
http://www-ra.informatik.uni-tuebingen.de/software/joelib/tutorial/functionalities/mol-io.html

BTW, i not the time to implement such things :-)
Please use Bug and Support tracking system:
http://sourceforge.net/tracker/?group_id=39708&atid=425969
http://sourceforge.net/tracker/?group_id=39708&atid=425970

Regards, Joerg

> Cc: gmx-developers@gromacs.org, h.rzepa@imperial.ac.uk, fsatom@tddft.org
> From: Geoff Hutchison <hutchisn@chem.northwestern.edu>
> Subject: Re: [Fsatom] Re: CML and macromolecules
> Date: Sun, 26 Oct 2003 01:26:17 -0600
> To: Peter Murray-Rust <pm286@cam.ac.uk>
> 
> 
> On Oct 19, 2003, at 6:15 PM, Peter Murray-Rust wrote:
> 
>>>The remainder (mainly REMARK stuff) is most likely less interesting. 
>>>If
>>>CML would provide the stuff listed above we don't have to worry about
>>>reading pdb files anymore, and we could even use the bond information
>>>that babel provides.
>>
>>I have provided core functionality only for Babel. I do not know what 
>>additional data structure they provide for PDB-like elements. If it's 
>>easy I can probably hack it. I doubt they support BIso. The problem 
>>with all conversion programs is that you are constrained by the data 
>>structures they provide. Not their fault...
> 
> 
> One advantage with an open conversion program is that you can add other 
> data structures (or suggest that they be added).
> 
> Open Babel has pages for bugs and for feature requests. ;-)
> http://openbabel.sourceforge.net/
> 
> Since you can add per-molecule, per-residue, per-bond, or per-atom 
> custom information, it's a matter of agreeing to data format and coding 
> the OBGeneric class. Something like the process of extending CML, 
> though of course there's some C++ involved.
> 
> Certainly if you feel there's something inadequate in Open Babel, we'd 
> rather you complain bitterly about it than to just mumble so that we 
> never know what's wrong. In particular with conversion programs, some 
> of the developers are coding formats that we don't use much (if at 
> all).
> 
> Cheers,
> -Geoff
> 
> --
> -Geoff Hutchison		<hutchisn@chem.northwestern.edu>
> Ratner/Marks Groups	(847) 491-3295
> Northwestern Chemistry	http://www.chem.northwestern.edu/

-- 
Dipl. Chem. Joerg K. Wegner
Center of Bioinformatics Tuebingen (ZBIT)
Department of Computer Architecture
Univ. Tuebingen, Sand 1, D-72076 Tuebingen, Germany
Phone: (+49/0) 7071 29 78970
Fax: (+49/0) 7071 29 5091
E-Mail: mailto:wegnerj@informatik.uni-tuebingen.de
WWW:    http://www-ra.informatik.uni-tuebingen.de
--
Never mistake motion for action.
                          E. Hemingway