Last reviewed: 2012-2-16, IJM
This week [2012-2-6], I re-worked Java code for generating the PDFs so that it now generates blank spare copies of examinations according to what is specified in the database, and it includes the simplex/duplex specification in the file names of the output files. As a result, the PDF to PS conversion script no longer has simplex/duplex versions and no longer takes a "number of spares" parameter. This eliminates a potential source of error and also makes it easy to print all papers for a sitting all at once even if there are multiple examinations involved (e.g., OPD sittings).
Note: by "seating plan", I mean a plan of which seats are to be assigned and which are to be rush seats, based on how many candidates are registered for the examination.
After the above, there are all sorts of other things to do, but I think the above will address the main concern ISG has.
ST items are all linked from ST#69507.
Part of this work led to the notion that rush seating generalizes to all exam seating (i.e. unassigned means "all seats rush")
A beginning of documenting how the system thinks about examination candidates and seats is available at http://odyssey.uwaterloo.ca/teaching/ems/classify. After much thought I adopted consistent terminology (which also applies to the database field names) and found a way to classify candidates and seats according to very similar taxonomies. Although there is something for users to learn, I'm confident this won't need to be changed.