fix issue 20915 - add support for is(identifier == version)

[ghost]

This fills the small gap that existed because __traits(allMembers) could return the identifiers of the VersionSymbols defined in the source but there was no way to identify them as such.

If accepted spec PR will follow.

[dlang-bot]

Thanks for your pull request and interest in making D better, @NilsLankila! We are looking forward to reviewing it, and you should be hearing from a maintainer soon.
Please verify that your PR follows this checklist:

• My PR is fully covered with tests (you can see the coverage diff by visiting the details link of the codecov check)
• My PR is as minimal as possible (smaller, focused PRs are easier to review than big ones)
• I have provided a detailed rationale explaining my changes
• New or modified functions have Ddoc comments (with Params: and Returns:)

Please see CONTRIBUTING.md for more information.

---

If you have addressed all reviews or aren't sure how to proceed, don't hesitate to ping us with a simple comment.

Bugzilla references

Auto-close	Bugzilla	Severity	Description
✓	20915	enhancement	__traits(allMembers) results include custom version identifiers, which is unusable

Testing this PR locally

If you don't have a local development environment setup, you can use Digger to test this PR:

dub run digger -- build "master + dmd#11255"

[Geod24]

Hum... Should allMembers really return locally-defined versions ? Is there any use case for it ?

[ibuclaw]

Bar the small implementation note. Please read up on the last 20 years history of why version (!ident) has been rejected time and time again.

[ghost]

Please read up on the last 20 years history of why version (!ident) has been rejected time and time again.

I'm aware of that but this is not an attempt to introduce fragments of version expressions sneakily.
Any attempt would clearly be visible in the parser.

[ibuclaw]

I'm aware of that but this is not an attempt to introduce fragments of version expressions sneakily.
Any attempt would clearly be visible in the parser.

But what arguable difference would there be between the rejected version (linux || OSX) versus static if (is(linux == version) || is(OSX == version))? They are functionally identical.

[ghost]

indeed, incredible, I didn't realize at all that the change allowed version operators, although the syntax is a bit verbose.

[ghost]

But what arguable difference would there be between the rejected version (linux || OSX) versus static if (is(linux == version) || is(OSX == version))? They are functionally identical.

People already abuse 'enum + static if' everywhere to work around the absence of version expressions. If allMembers returns versions, there must exist a predicate to filter them out.

[ibuclaw]

indeed, incredible, I didn't realize at all that the change allowed version operators, although the syntax is a bit verbose.

Well I'd probably let @WalterBright weigh in, but my feeling is that you should not be including predefined symbols in the is(ident == version) match.

What about @Geod24's question? Is there a use case for allMembers returning versions?

[WalterBright]

Various versions of this proposal (algebraic combinations of versions) have come up before. We are not doing this. The version system in D is specifically designed to discourage this sort of programming. Besides, if you want to know if a version is a version:

version (isthisaversion)
    pragma(msg, "isthisaversion is indeed a version");
else
    pragma(msg, "never heard of isthisaversion");

[WalterBright]

__traits(allMembers) must not return version ids or debug ids.

[adamdruppe]

I've actually been wanting some kind of available version detection to help document possible compile time options.

I'll probably work it into my doc generator though since it has its own parser and can look for both version=X and version(X) and at least generate a list.

[ghost]

Various versions of this proposal (algebraic combinations of versions) have come up before. We are not doing this. The version system in D is specifically designed to discourage this sort of programming. Besides, if you want to know if a version is a version:

version (isthisaversion)
    pragma(msg, "isthisaversion is indeed a version");
else
    pragma(msg, "never heard of isthisaversion");

O, gosh! This was not a proposal for algebraic combinations at all! We are well aware of your aversion to those.

Having an ability to introspect into version identifiers is a completely different and potentially useful feature (Adam has a use for it). The fact that the version-detecting predicate can be abused is completely irrelevant (as I said, people are already abusing 'static if' to work around your ban on version algebra).

Side note: language design based mainly on preexisting use cases and personal experience means another dead language in the near future.

[ghost]

there's a callback_API now in DMD if you need things that don't change the semantic...

[ghost]

@adamdruppe see 2efa524. You can propose more callbacks if you need.

[ghost]

@maxsamukha, it wasn't sure, not directly ;]

[WalterBright]

language design based mainly on preexisting use cases and personal experience means another dead language in the near future.

Indeed I do have 40 years of experience with the stinky pile of poo that always results from algebraic combinations of versions.