Comment # 13
on bug 11860
from Guy Harris
...and "contains", "matches", and "bitwise_and", to name three *more*
relational operators.
Now, perhaps what we could do is make the set of relational operators be =, <,
<=, >, >=, "contains", "matches", and "bitwise_and", with *all* of them meaning
"there exists an instance of field {field} where...", and allow *any*
relational operator to be prefixed by "!", with "!{op}" meaning "for all
instances of {field}, the relation '{value of that instance of the field} {op}
{value}' is *false*".
That doesn't allow a "there exists an instance of {field} for which the
relation '{value of that instance of the field} is not equal to {value}", but
we could steal the <> operator from Pascal and have that mean "is not equal
to". (And, yes, you could do "{field} !<> {value}", which means "for all
instances of {field}, the value of that instance of the field is not unequal to
{value}", i.e. "all instances of {field} have the value {value}".
How many currently-used display filters will break with "!=" not being a
relational operator, but being an "invert the predicate" flag for the "="
operator, so that the display filters would have to be changed to use "<>", is
unknown.
You are receiving this mail because:
- You are watching all bug changes.