I have tried to be consistent with naming, particularly during the recent reorganisation in preparation for going open source. However, at this stage the project was around 4 years old and there are some exceptions to the rules.
The following table should explain all the font usage in this manual:
|Perl program or package||perlgp-run.pl||normal font, with trailing .pl or .pm|
|File or directory name||`results/best.pl'||normal font, in single quotes||see text|
|Environment variable||PERLGP_LIB||normal capitals, no $|
|Public object method||loadSet()||typewriter font, trailing (), no underscores, capitalised words, except first word, which is usually a verb||Fitness(), Age(), mutation operators|
|Private object method||_init_tree()||typewriter font, leading underscore, usually all lowercase with underscores between words|
|Mutation operator||copy_subtree()||typewriter font, lowercase with underscores, trailing ()|
|Other Perl functions||eval()||typewriter font, trailing ()|
|Object attribute||AlarmTime||typewriter font, no underscores, capitalised words||don't forget the get and set forms: $self->KeepBest() and $self->KeepBest(5)|
|Data types||NUMBER||small capitals|
Finally, some subroutines and methods take a hash table as their argument. This is a way of having ``named'' arguments, which don't have to be given in any particular order (and can be left out if not required). In the manual, this is presented as:
Be extremely careful typing these argument names, because if you spell them wrong, it will probably silently ignore it.
The type called ATTRIBUTE-HASH is just a hash where the keys are valid Attribute names, and the values are the values you want them to have.
Non absolute file names are always relative to the current ``experiment directory'', which is the directory which contains the Algorithm.pm, Individual.pm and so on.