glm syntax

[Jason Stover]

I was looking at the q2c documentation so I can write the GLM syntax.
It says this:

     grammar-rules ::= command-name opt-prefix : subcommands .
     command-name ::= ID
                  ::= STRING
     opt-prefix ::=
                ::= ( ID )
     subcommands ::= subcommand
                 ::= subcommands ; subcommand


...so, after the command name, I enter a list of subcommand names.
The problem is that GLM's syntax differs from that of other procedures.
It looks like this:

   GLM dependent variable [BY factor list [WITH covariate list]] ...

so the first token after the command name is a variable name. I'm not
sure how to write the q2c rules for that. 

I thought about adding an optional VARIABLE subcommand, then assuming
if it isn't there that the first token goes with the unused
subcommand, but this approach seems like a kludge.

Does this mean I need to write a custom parser? I hope not. 

I don't know much about lexer and q2c, so the answer to my questions
might be 'RTFM'. That's okay, but please mention which FM to R.

-Jason