> > Ideally all the filters that can be specified in the command line > > options are implemented so, what is the point of returning an error > > code if a filter is not implemented? > > As you said, 'ideally'. If some library tester not involved in the > library development wants to use the utility it would be nice to > report those warnings and error. > > I would rather put assertions in the switch cases for unimplemented > filters, or removing them from 'GNU_longOptions'. > > The primary function of pdf-filter is to serve as an utility, and not > as a test tool. I think that we should not change its interface.
I suggested this as an improvement from the end-user point of view. Removing the filter options until they are implemented works too. I just enclosed the "dead" code in pdf-filter with #if 0's, in the trunk.
