I don't see much advantage to having this info in the restrain itself rather than a separate table (std::map or python dictionary) indexed by the restraint. The later is much more flexible since you can put structured data in it and speed is never going to be a concern for this sort of data.

On Nov 3, 2010, at 3:23 AM, Keren Lasker wrote:

> hi - > > How about adding a std::string variable to Restraint class named header. > It can serve as a place holder for useful data about the restraint > (papers it was described at etc). > Useful when dealing with restraints such as proteomics. > > Keren. > _______________________________________________ > IMP-dev mailing list > IMP-dev@salilab.org > https://salilab.org/mailman/listinfo/imp-dev