|Q: regarding regular grammars ... email@example.com (Michael Roach) (1997-11-24)|
|Re: Q: regarding regular grammars ... firstname.lastname@example.org (Henry Spencer) (1997-11-28)|
|Re: Q: regarding regular grammars ... email@example.com (Karsten Nyblad) (1997-11-29)|
|Re: Q: regarding regular grammars ... firstname.lastname@example.org (Jos A. Horsmeier) (1997-11-29)|
|Re: Q: regarding regular grammars ... email@example.com (Henry Spencer) (1997-11-30)|
|Re: Q: regarding regular grammars ... firstname.lastname@example.org (1997-12-05)|
|From:||email@example.com (A Johnstone)|
|Date:||5 Dec 1997 01:13:52 -0500|
|Organization:||Royal Holloway, University of London|
Michael Roach wrote:
: > I thought regular expressions were regular grammars, is that assumption
: > wrong? And if so, could you explain the differences or point me to some
: > references. Thanks.
Karsten Nyblad (firstname.lastname@example.org) wrote:
: Yes, it is wrong. Regular grammar are a superset of context free
: (CFG). Even the class of langauges of CFG's is larger than the class of
: languages of regular expressions.
I think Karsten has misread the original question - Michael was
talking about regular grammars not context free grammars. Contrary to
Karsten's claim, regular grammars are a subset of CFG not a
superset. It is easy to convert from a RE to a RG grammar as long as
the RE doesn't use extended notions like variables (these things are
not really RE's, formally).
I find the Parsing Theory twin volume set recommended by Karsten good
for some theory but less good for everyday queries. The most readable
reference I know is Dick Grunne's book on parsing published a few
years ago by Ellis Horwood and now sadly out of print, but obtainable
from your library, perhaps? Personally I find the treatment in the
dragon book repetetive and disorganised.
Dr Adrian Johnstone, Senior Lecturer in Computing, Computer Science Dep,
Royal Holloway, University of London, Egham, Surrey, TW20 0EX, England.
Email email@example.com Tel: 01784 443425 Fax: 01784 439786
Return to the
Search the comp.compilers archives again.