Disallow assigning to struct rvalue

[ntrel]

When assignment would call an operator overload.
Fixes #21507.

Note: This does not disallow calling e.g. opAssign on an rvalue, only actual assignment.

[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

Your PR doesn't reference any Bugzilla issue.

If your PR contains non-trivial changes, please reference a Bugzilla issue or create a manual changelog.

⚠️⚠️⚠️ Warnings ⚠️⚠️⚠️

• In preparation for migrating from Bugzilla to GitHub Issues, the issue reference syntax has changed. Please add the word "Bugzilla" to issue references. For example, Fix Bugzilla Issue 12345 or Fix Bugzilla 12345.(Reminder: the edit needs to be done in the Git commit message, not the GitHub pull request.)

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#21717"

[ibuclaw]

What other rvalues could appear here? Does it make sense to only restrict structs?

[ntrel]

The check in opOverloadBinaryAssign would get hit by:

    int[] a;
    a.length += 1;

It would also get hit for a class C (new C) += 4;. That should be prevented too, but I don't want to do that in this pull.

[kinke]

rvalues cannot be assigned to by definition, correct?

[ntrel]

I snipped the buildkite failures below. The libdparse one is fixed upstream. The other projects are failing due to one of these dependency failures.

    Building vibe-d:redis ~master: building configuration [vibe-d-redis-test-library]
redis/vibe/db/redis/idioms.d(74,8): Error: cannot assign to struct rvalue `this.opIndex(id)`
            this[id] = args;
                ^

phobos:
../dmd/generated/linux/release/64/dmd -conf= -I../dmd/druntime/import  -w -preview=dip1000 -preview=dtorfields -preview=fieldwise -m64 -fPIC -O -release -unittest -version=StdUnittest -c -ofgenerated/linux/release/64/unittest/std/container/array.o std/container/array.d
make[2]: *** [Makefile:407: generated/linux/release/64/unittest/std/container/array.o] Segmentation fault (core dumped)

ae:
net/ietf/message.d(347,7): Error: cannot assign to struct rvalue `list(this.author, this.authorEmail)`
        list(author, authorEmail) = decodeAuthor("From" in headers ? decodeRfc1522(headers["From"]) : null);
            ^
net/ietf/message.d(350,8): Error: cannot assign to struct rvalue `list(null, this.authorEmail)`
            list(null, authorEmail) = decodeAuthor(decodeRfc1522(headers["Reply-To"].findSplit(", ")[0]));
                ^
net/smtp/client.d(159,7): Error: cannot assign to struct rvalue `list(codeStr, null, line)`
        list(codeStr, null, line) = line.findSplit(" ");
            ^

Compiling parse tester... ../src/dparse/parser.d(8981): Error: cannot modify struct rvalue `checked(this.suppressMessages.back())`
            ++suppressMessages.back.checked;
                                   ^

../.dub/packages/mir-algorithm/3.22.4/mir-algorithm/source/mir/bignum/fixed.d(966,23): Error: cannot assign to struct rvalue `a.view()`
    size_t overflow = a.view *= b;
                      ^
../.dub/packages/mir-algorithm/3.22.4/mir-algorithm/source/mir/bignum/fixed.d(1063,61): Error: cannot assign to struct rvalue `view.topLeastSignificantPart(1LU)`
    view.coefficients[$ - 1] += view.topLeastSignificantPart(a.data.length) += c.view;
                                                            ^

[ntrel]

rvalues cannot be assigned to by definition, correct?

Currently dmd allows it for a struct overloading assignment.

[ntrel]

• I've asked for a new release with the libdparse fix, which was a real bug detected by this PR. The test runners are failing mainly due to that AFAICT, so they need that version update.
• Now the error only fires when the struct is not nested, so this doesn't break ae.utils.array.list any more.
• I've made a PR for mir-algorithm, which calls opOpAssign directly. For that one, mir actually uses the result of opOpAssign and throws away the rvalue assignment.
• I think that just leaves the vibe-d:redis one, which I'll look at.
• Given the fixes needed, this should probably be a deprecation/error in an edition.

[ntrel]

Ideally this would catch the libdparse case but not the mir ones. I'll have a look next week.

[ntrel]

Closing in favour of the other PR.