Dear all, Yesterday I merged the contents of two branches in subversion into the trunk:
The first includes some changes to the build system that allow the RDKit to be built using the mingw system on windows. If you don't know what this means, it's probably not important to you, so don't worry about it. :-) The second change is more interesting (maybe): I've added two new molecular input format parsers: Mol2 and SLN. Given the chaos that the poorly defined atom types in the mol2 format make, we had to pick some "standard" definitions to use; we chose Corina atom types. Nik Stiefl, who wrote the parser and dealt with the mess that this format is, says it should also work reasonably well on output from Tripos's dbtranslate program. The SLN parser supports a subset of SLN: all molecule features should be supported, but only a subset of the query features are in place. I will do a feature matrix for the parser some time in the next couple of weeks so that it's clear what it can or cannot understand. Both parsers are relatively new, so I have no doubt that people will find problems and areas where they could be improved. Please report them in the bug tracker and, if possible, provide sample files that demonstrate the problems. Thanks, -greg
