Scanning of SIGNS

[Till Mossakowski <till@informatik.uni-bremen.de>]

Dear friends,

during the creation of example specs by Markus, the following
question arose: is 

	name[c1[c2]]

a legal ID?
The parsers in Paris and Amsterdam do accept it,
but our parser in Bremen recognizes "]]" as a SIGNS, and this
lexical scanning prevents the above string from being
recognized as an ID. A similar problem occurs with
"]=", "{}:" "]:" etc. which are recognized as *one*
lexical symbol due to the rule on page
C-10 "The next lexical symbol recognized is as
long as possible".
Obviously, this rule is not entirely appropriate
(unless we require the above to be written
as "name[c1[c2] ]").
How should the rules for lexical analysis be modified?

Greetings,
Till