Dear Claus, glad I could help you. Some remarks: I won't force you to use any sort of OR mapper, it just has advantages. I found it mainly convinient with respect to code organisation - you can nicely organize your methods (like the stuff in SubmittingData, which should really go in DBMolecule, as it is in cdkweb). It is also nice to call getDBMolecule on a DBSpectrum and just get the spectrum related by foreign key. Obviously it also depends on the complexity of the db - if you have your structures as e. g. CML in a single table, you don't gain much, whereas when you split a molecule in a number of tables it might help. If you want to do this is hard to say. In a way it is easier to have it all in one table. On the other hand with my approach you are more flexible - we can for example model peak assignment as foreign key relationship betwenn shift and atom. > >>It then checks for duplicates via SMILES, returns the > > Ok, that makes the sql a lot easier than the fingerprints. Is that good > enough to use smiles for duplicate checking? are you aure to find them all? Given the smiles implemenatation is not buggy it should work for identity search (and therefore for duplicate checking). Substructure search is a different thing. Stefan
-- Stefan Kuhn M. A. Cologne University BioInformatics Center (http://www.cubic.uni-koeln.de) Zülpicher Str. 47, 50674 Cologne Tel: +49(0)221-470-7428 Fax: +49 (0) 221-470-7786 My public PGP key is available at http://pgp.mit.edu ------------------------------------------------------- This SF.net email is sponsored by: Splunk Inc. Do you grep through log files for problems? Stop! Download the new AJAX search engine that makes searching your log files as easy as surfing the web. DOWNLOAD SPLUNK! http://ads.osdn.com/?ad_idv37&alloc_id865&op=click _______________________________________________ Cdk-user mailing list [email protected] https://lists.sourceforge.net/lists/listinfo/cdk-user
