We investigate the complexity of a variety of normal-form transformations f
or extended context-free grammars, where by extended we mean that the set o
f right-hand sides for each nonterminal in such a grammar is a regular set.
The study is motivated by the implementation project GraMa which will prov
ide a C++ toolkit for the symbolic manipulation of context-free objects jus
t as Grail does for regular objects. Our results generalize known complexit
y bounds for context-free grammars but do so in nontrivial ways. Specifical
ly, we introduce a new representation scheme for extended context-free gram
mars (the symbol-threaded expression forest), a new normal form for these g
rammars (dot normal form) and new regular expression algorithms. (C) 2001 E
lsevier Science B.V. All rights reserved.