byacc vs bison?

[zenspider]

Are there any sticking issues for using bison over the current byacc?

ETA: (or vice versa)

[AllenOtis]

The byacc in the maglev source tree is a modified byacc .
See the README_maglev.txt , hich includes the following.
You would have to replicate this maglev-specific behavior in bison.

Allen

Maglev modifications include

Deleted struct YYSTACKDATA,
and changed the type of yystack cells , to allow integration with
garbage collector and so that state machine only has to maintain
one stack pointer. See YyStackElement and YyStackData in rubyom.hf .

Use a unified state table instead of multiple tables for better
C code generation in the state machine. C compiler should be able
to cache one pointer to start of the unified table.

Moved a bunch of definitions to rubyparser.h .

Moved yygrowstack implementation to the .y file .

On Thu, May 21, 2015 at 11:47 PM, Ryan Davis notifications@github.com
wrote:

Are there any sticking issues for using bison over the current byacc?

—
Reply to this email directly or view it on GitHub
#400.