Tomita (GLR) Parser Devlopment Question

hi,


I have gotten pretty Far in developing my GLR parser System. I Just
Have one question relating to optimizing. Right Now, When I Reduce, I
have to traverse backwards along the parse-tree vertices to get to the
next state to "goto" from. Is There A Standard Way Of doing this
That'S Faster? It Seems to me like the only faster way is to have
separate stacks for each active path, but that ruins the whole point
of the packed parse-forest. Plus The code in the back of the tomita
Book Is In A Language I Can't understand. I'll keep trying to
decipher it while i wait for a response to this post.


By the way, if there is a group specifically geared towards parsers
and/or development, I would be very grateful for that information.


thanks in advance
peace,
dave
[Parsing is about the most popular topic in comp.compilers. See the
archives. Tomita parsers come up now and then. -John]

Return to the comp.compilers page.
Search the comp.compilers archives again.