Re: Question re the (non-)equivalence of Z -> z and Z -> z e (e the empty string)

"SLK Parsers" <parsersinc@earthlink.net>
3 Sep 2004 12:31:59 -0400

          From comp.compilers

Related articles
[3 earlier articles]
Re: Question re the (non-)equivalence of Z -> z and Z -> z e (e the em dhalitsky@cumulativeinquiry.com (2004-08-11)
Re: Question re the (non-)equivalence of Z -> z and Z -> z e (e the em cppljevans@cox-internet.com (Larry Evans) (2004-08-11)
Re: Question re the (non-)equivalence of Z -> z and Z -> z e (e the em kwheinri@bsr2.uwaterloo.ca (Kenn Heinrich) (2004-08-13)
Re: Question re the (non-)equivalence of Z -> z and Z -> z e (e the em dhalitsky@cumulativeinquiry.com (2004-08-13)
Re: Question re the (non-)equivalence of Z -> z and Z -> z e (e the em parsersinc@earthlink.net (SLK Parsers) (2004-08-15)
Re: Question re the (non-)equivalence of Z -> z and Z -> z e (e the em dhalitsky@cumulativeinquiry.com (2004-08-23)
Re: Question re the (non-)equivalence of Z -> z and Z -> z e (e the em parsersinc@earthlink.net (SLK Parsers) (2004-09-03)
| List of all articles for this month |

From: "SLK Parsers" <parsersinc@earthlink.net>
Newsgroups: comp.compilers
Date: 3 Sep 2004 12:31:59 -0400
Organization: SLK Parser Generator
References: 04-08-046 04-08-102 04-08-110
Keywords: theory
Posted-Date: 03 Sep 2004 12:31:59 EDT

> Does your use of the example mean that you think it is OK (in
> language theoretic examples anyway) to think of epsilon (the emptry
> string symbol) as being an element of Vt which can appear to the
> right of the arrow in productions ?


Not in Vt. Nonterminals and actions can also appear on the rhs of a
production. My point was that it is not incorrect to explicitly say that
there is nothing between the symbols by putting the empty string there.
Redundant and hard to read, but not incorrect.


A better phrasing might be that the LL(0) conflict is on the imaginary, or
implied empty string at the beginning of every production. Since k=0, the
conflict is on nothing, but I needed to formalize this in a way that was
consistent with the cases where k>0.


SLK Parser Generator: http://home.earthlink.net/~slkpg/


Post a followup to this message

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