Fix Bugzilla 24582 - Detect unsafe cast(bool[])

[ntrel]

Deprecate in safe code.

[dlang-bot]

Thanks for your pull request and interest in making D better, @ntrel! 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
✓	24582	major	Detect unsafe `cast(bool[])`

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#16558"

[thewilsonator]

Dependant PR merged

[dkorpel]

Like I mentioned on bugzilla, I don't see how these casts are unsafe.

[ntrel]

@dkorpel I added a comment there with an example:
https://issues.dlang.org/show_bug.cgi?id=24582#c3

Only the runtime array cast is unsafe, the other 2 are OK - thanks. I need to update this pull to allow the literal cast, as well as change the error to a deprecation.

[RazvanN7]

Why is the cast being deprecated? Wouldn't it make more sense to simply fix the literal cast to create an array of 0 and 1s? For the sake of consistency, casting a literal or a variable shouldn't yield different results.

[dkorpel]

Why is the cast being deprecated?

It's not, it's disallowed in @safe code.

Wouldn't it make more sense to simply fix the literal cast to create an array of 0 and 1s?

That's what happens.

For the sake of consistency, casting a literal or a variable shouldn't yield different results.

Currently it does, but changing that behavior is a breaking change and out of scope for this bug report.

[ntrel]

Wouldn't it make more sense to simply fix the literal cast to create an array of 0 and 1s

That would go against the spec:

https://dlang.org/spec/expression.html#cast_array

The cast is done as a type paint

So the raw data is not modified.

[dkorpel]

No, as I posted on bugzilla, the byte contents is unchanged:

For the runtime cast yes, but the literal cast is different:

immutable c = cast(bool[]) [2, 4]; // literal cast applies to each element
pragma(msg, c); // [true, true]

[ntrel]

@dkorpel OK, sorry - so we are in agreement!

[thewilsonator]

(Not that you need to do this right away, but) can we link to/refer users to the spec section/other official documentation, as to why this is an unsafe operation? It was certainly not immediately obvious to me as to why this is the case.

[ntrel]

I've added a supplemental message for now.

[ntrel]

Tests now passing (other than 1 interop network request failure)