mikepersonick commented on a change in pull request #1557:
URL: https://github.com/apache/tinkerpop/pull/1557#discussion_r803720226
##########
File path: gremlin-dotnet/src/Gremlin.Net/Process/Traversal/P.cs
##########
@@ -85,79 +85,79 @@ public P Or(P otherPredicate)
public static P Between(params object[] args)
{
- var value = args.Length == 1 ? args[0] : args;
+ var value = args != null && args.Length == 1 ? args[0] : args;
return new P("between", value);
}
public static P Eq(params object[] args)
{
- var value = args.Length == 1 ? args[0] : args;
+ var value = args != null && args.Length == 1 ? args[0] : args;
return new P("eq", value);
}
public static P Gt(params object[] args)
{
- var value = args.Length == 1 ? args[0] : args;
+ var value = args != null && args.Length == 1 ? args[0] : args;
Review comment:
Florian, I've been working on a clearly defined semantics for `nulltype`
(and other types/values). This includes well defined behavior for comparisons
against it under a ternary boolean logics system. All described
[here](https://github.com/apache/tinkerpop/blob/1b2ed0c1c3dfd54dfe9cf95bb124571f9849d8a0/docs/src/dev/provider/gremlin-semantics.asciidoc#comparability-equality-orderability-and-equivalence)
and
[here](https://github.com/apache/tinkerpop/blob/1b2ed0c1c3dfd54dfe9cf95bb124571f9849d8a0/docs/src/dev/provider/gremlin-semantics.asciidoc#nulltype).
As such, `null` is a perfectly valid argument for the `P` predicates. It might
be evaluated to an `error` once evaluated, but it should not be disallowed as
an argument. .NET was the only language that was disallowing it up front, which
was making testing the semantics (and true error cases) impossible.
--
This is an automated message from the Apache Git Service.
To respond to the message, please log on to GitHub and use the
URL above to go to the specific comment.
To unsubscribe, e-mail: [email protected]
For queries about this service, please contact Infrastructure at:
[email protected]
