LR(1) vs. LALR(1)

I am interested in any empirical evidence of the difference between LR(1)
vs. LALR(1) machines for complete grammars written for "real" programming
languages.


Specifically, any studies which could provide data to support the answer to
the following two questions:


      Given a grammar,


      1. How many states in the LR(1) machine vs. the LALR(1)?
      2. What's the maximum number of LR(1) states which are merged into
            any single LALR(1) state?


Most of the literature I have come across give estimates like: the LR(1)
machine is twice the size of the LALR(1) machine, or, the ratio of LR(1)
configurations to LR(0) configurations is equal to the size of the token
set...


In addition to the forum, please direct any information and/or comments to:


georgev@fiu.edu


Thanks in advance. Vince.
--
Vincent A. George
School of Computer Science
Florida International University
Miami, FL, USA
Email: georgev@fiu.edu
--

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