Date/time support for Postgres. Year/month/day operations on Columns.

[radeusgd]

Pull Request Description

Closes #6115

Important Notes

Checklist

Please ensure that the following checklist has been satisfied before submitting the PR:

• The documentation has been updated, if necessary.
• Screenshots/screencasts have been attached, if there are any visual changes. For interactive or animated visual changes, a screencast is preferred.
• All code follows the
  Scala,
  Java,
  and
  Rust
  style guides. In case you are using a language not listed above, follow the Rust style guide.
• All code has been tested:
  • Unit tests have been written where possible.
  • If GUI codebase was changed, the GUI was tested when built using ./run ide build.

[GregoryTravis]

Why does this always fail?

[radeusgd]

Because it is called on an error value and a non-error was expected.

This is due to a slightly weird calling convention.

If I call a method on a dataflow error, by default it is not called but the dataflow error is "propagated" - so it is just returned as-is. So - if I define an extension method on Any, it will resolve on any value, but it will return dataflow errors unchanged if called on it. In fact you can call (Error.throw "X").sgdjkkgjkdgjkdfkjgfjk and this will be a valid call and will return the Error "X" - that is because if we are holding an error, there's no way to tell what "would be" instead of it if there wasn't an error and what methods could that have - so all methods are allowed and they just act as identity.

But sometimes, I don't want the call to just fall through. To avoid this, I can explicitly add an extension method on the Error type. Then such methods will take precedence on error instances and allow me to customize the behaviour.

So in our Test suite, it is customary to define every extension method both on Any and on Error. That way if a value returns an Error, we can get an information that a test failed. It is very important because if you do not define it on Error, then a call (Error.throw "X").should_equal "Y" would fall-through and return the dataflow error - but if it's result is nowhere used - that will just be silently ignored - so a test will pass even if it's wrong - that's very bad for a test suite.

[radeusgd]

Other date/time types do not have this.

Having it breaks the to_text invocation on type.

I.e. Time_Of_Day.to_text now works. With that shadow definition it wasn't working properly.

It's a bit of a workaround to #1538

[jdunkerley]

Couple of nits. But LGTM