From: Hans Aberg (haberg@math.su.se)
Date: Sun Oct 14 2007 - 02:29:28 CDT
On 14 Oct 2007, at 02:54, Philippe Verdy wrote:
>> So you don't need this, I think. Essentially, what you are describing
>> is a new closure operator x$B!-(B, with the definition perhaps L
>> (x') =
>> {y $B":(B C*| y is a substring of a string in L(x) }. (To be a
>> "closure
>> operator", it should satisfy L(x'') = L('x).) Then your complement is
>> ~(x'), where L(~x) := C* \ L(x). But I am not sure of the usability
>> of this x' operator.
...
> Note also that searching matches with the Regexp "(a*)(a*)" in
> "aaaaa" will
> return 5 possible strings: "", "a", "aa", "aaa", "aaaaa".
So then you are merely finding all matches in the language L(x). It
is an interpretation and implementation of regular expression and the
languages they define.
It seems me you are constructing your own computer language, perhaps
with emphasis on getting more control over the from the language that
regular expressions generate.
Hans $B"r(Bberg
This archive was generated by hypermail 2.1.5 : Sun Oct 14 2007 - 02:30:44 CDT