Thought I'd also have a go at translating ch-1.pl and ch-2.pl into C..

Both C versions produce identical (non-debugging and debugging)
output to the Perl originals.

ch-2.c is considerably more complex than ch-2.pl, ch-2.pl built ideal data
structures in a typical Perlish fashion and used them idiomatically to focus
on the problem.  Instead, ch-2.c only builds one main data structure - a sorted
list of strings (i.e. a set) - and then ch-2.c declares slists and arrays of
slists.  As ever, it's the memory management in C that's the killer.  However,
ch-2.c works correctly (as far as I can tell) and doesn't leak any memory (after
quite a lot of work to make that happen:-)).

These C versions use some of my regular support modules:
- a command-line argument processing module args.[ch],
- plus a CSV splitting module csvsplit.[ch] to split
  a single argument up
- plus (new for challenge 209) the module slist.[ch] that stores the sorted
  list of strings that can be used like a simple set