When using enumerations where the values itself contain a comma, the rules generated by a GDST are unexpected, as the "contains in" operator splits those values in the enumerations. Example enumeration:
Note: see the 'new york, boston' sample.
The code generated will be:
Basically "paris" and "new york,boston" will be treated by the DSL parser as 3 strings in the DRL generation and will produce someting simiular to
p : person( city in ( "paris", "new york", "boston" ) )
But what the customer expects is the following
p : person( city in ( "paris", "new york,boston" ) )