Re: --yacc and token #define's

[Joel E. Denny]

On Sat, 10 Jun 2006, Joel E. Denny wrote:

> The idea of removing token #define's in the absence of --yacc has come up 
> a few times over the past year.

I need a little guidance on this test case:

  113: regression.at:57   Early token definitions

It tests whether tokens are defined before user prologues.  I thought I'd 
split it into two test cases: testing the enum without --yacc, and testing 
the #define with --yacc.  I suppose I should copy the GCJ comment to both.

I see that Bison guarantees that the tokens are defined before the 
pre-prologue, but the original test case only checks that they're defined 
before the post-prologue.  Is there a reason for that?

Joel