yes, ? wasn't easy due to java and data munging would get messy as
we map between things.
So we just left it as any valid java identifier, but using the $
prefix as a coding convention make it easier to differentiate
fieldnames and bindings
Person( age : age )
Person( age == age )
The above looks a little confusing compared to:
Person( $age : age )
Person( age == $age )
Davide wants to enforce the $ prefix so that parsing can be easier.
Mark
On 24/09/2010 00:31, Michael Neale wrote:
The $name: Pattern thing I am convinced is to do with
Mark's prior history of being abused by perl ;)
But the real reason is we wanted to use ?name: Pattern() -
using "?" like the clips lineage of languages - but IIRC even
ilog allows that. We wanted our labels to be compatible with
java source code - where $variable is a valid name (although no
one actually uses it) and ?name is not.
FWIW: in Perl, there are both operators
as well (|| and 'or'). However, they are
*not* exactly the same. Although they can be
used in any context to render a boolean
expression, their priority makes the
difference. Taken from official
documentation (http://bit.ly/dgw4GT):
Low precedence "and", "or", "xor" were
introduced to permit "Perl poetry", or, more
seriously, to
permit control flow using a logical
expression, especially after function calls
without parentheses.
see Naples or die; # same as: see(Napes)
|| die(); but not: see(Naples || die() );
No way this makes any sense in Drools.
-W
Binary "or"
returns the logical disjunction of the
two surrounding
expressions. It's equivalent to ||
except for the very low precedence.
This makes it useful for control flow.
Nonetheless, it must be taken into
account that the distinction makes sense
for a Perl programmer. For a rules-writing
guy (or girl) perhaps the distinction is
extremely obscure.