Added the ability for custom Include and Exclude UDAs. If multiple UDAs are given the last applies (with exception).

[JonathanILevi]

I added some feature I need of a project of mine. I hope you will accept them up stream.

You can now give custom Include or Exclude UDAs:

Grouped!(Includer!CustomInclude, Excluder!CustomInclude, Includer!Another).serialize(data);

enum A;
enum B;
class Data {
	@A int a;
	@Exclude @B int b;
}
serialize(data); // `a` included;
Grouped!(Excluder!A).serialize(data); // returns [];
Grouped!(Includer!B).serialize(data); // `a` and `b` included.

The last of stacked UDAs is used (with exception to @DecodeOnly and @EncodeOnly):

class Data {
	int a; // included
@Exclude:
	int b; // excluded
	@Include int c; // included.
@Condition(cond):
	int d; // if condition
	@Include int e; // included
	@Exclude int f; // excluded
	@Condition(cond) int g; // if second condition (first is ignored)
@DecodeOnly:
	// Nothing will ever be encoded (only decoded).  `@DecodeOnly` and `@EncodeOnly` override.
	@Exclude int h; // alway excluded
	@Exclude @Include; // only decoded
}

[codecov-io]

Codecov Report

Merging #3 into master will not change coverage.
The diff coverage is 100%.

@@          Coverage Diff          @@
##           master     #3   +/-   ##
=====================================
  Coverage     100%   100%           
=====================================
  Files           2      2           
  Lines         139    193   +54     
=====================================
+ Hits          139    193   +54

Impacted Files	Coverage Δ
src/xserial/attribute.d	100% <ø> (ø)	⬆️
src/xserial/serial.d	100% <100%> (ø)	⬆️

---

Continue to review full report at Codecov.

Legend - Click here to learn more
Δ = absolute <relative> (impact), ø = not affected, ? = missing data
Powered by Codecov. Last update 39214ea...b62c480. Read the comment docs.

[Kripth]

Thank you again for the contribution. Just one thing, can you update the readme with the new functionalities?

[JonathanILevi]

I refactored Grouped to Group I think I like it better.

[JonathanILevi]

@Kripth I can do that, though, I realized that I think I should rework the current handling of stacked groups.

enum GroupA;
enum GroupB;
struct Test {
@Exclude: // Exlcude everything by default
    @GroupA @GroupB ubyte c; // included in both
}
assert(Group!(Includer!GroupA,Excluder!GroupB).serialize(Test(1,2,3))==[1]); // Group A without any B
assert(Group!(Includer!GroupB,Excluder!GroupA).serialize(Test(1,2,3))==[2,3]); // All of B even c because B is given after A for c;

Stacked groups should not look at the last group UDA given to the value but look at the order given in the Group template. The second assert should yield [2].