sumtype assignment inconsistency with other types

[Eliyaan]

Describe the bug

When assigning a sumtype a to the variable b you expect both objects to be different (copied like with struct, and arrays that tell you to either do an explicit copy or explicit non copy), but for the moment, the 2 variables are referencing the same memory. (see the discussion on discord https://discord.com/channels/592103645835821068/592294828432424960/1443361143539110018)

Reproduction Steps

struct Aa {
mut:
    a int
}

struct Bb {
mut:
    a int
}

type Sum = Aa | Bb

a := Sum(Aa{32})
mut b := a
b.a = 47
println(a.a)
println(b.a)

Expected Behavior

32
47

Current Behavior

47
47

Possible Solution

No response

Additional Information/Context

No response

V version

V 0.4.12 85ff886

Environment details (OS name and version, etc.)

V full version	V 0.4.12 bf7f43f2fe7a62502d79136c574b5b7c3376e5bf.85ff886
OS	linux, Linux version 6.13.12-200.fc41.x86_64 (mockbuild@9778ff19adb64779af61fc903691b7f0) (gcc (GCC) 14.2.1 20250110 (Red Hat 14.2.1-7), GNU ld version 2.43.1-5.fc41) #1 SMP PREEMPT_DYNAMIC Sun Apr 20 15:52:43 UTC 2025
Processor	4 cpus, 64bit, little endian, Intel(R) Pentium(R) Gold 7505 @ 2.00GHz
Memory	7.06GB/15.34GB
V executable	/home/void/v/v
V last modified time	2025-11-26 17:18:11
V home dir	OK, value: /home/void/v
VMODULES	OK, value: /home/void/.vmodules
VTMP	OK, value: /tmp/v_1000
Current working dir	OK, value: /home/void
Git version	git version 2.49.0
V git status	weekly.2025.29-713-g85ff8864
.git/config present	true
cc version	cc (GCC) 14.3.1 20251022 (Red Hat 14.3.1-4)
gcc version	gcc (GCC) 14.3.1 20251022 (Red Hat 14.3.1-4)
clang version	N/A
tcc version	tcc version 0.9.28rc 2025-02-13 HEAD@f8bd136d (x86_64 Linux)
tcc git status	thirdparty-linux-amd64 696c1d84
emcc version	N/A
glibc version	ldd (GNU libc) 2.40

[!NOTE] You can use the 👍 reaction to increase the issue's priority for developers.

Please note that only the 👍 reaction to the issue itself counts as a vote. Other reactions and those to comments will not be taken into account.

[jorgeluismireles]

I didn't know b.a = 47 was possible when all variants of the sum-type have a field named a. Though a and b are different objects, seems field a is a single slot in memory:

println('a=${u64(&a)} a.a=${u64(&a.a)}')
println('b=${u64(&b)} b.a=${u64(&b.a)}')

a=140721487784600 a.a=22902148415472
b=140721487784536 b.a=22902148415472

a.a and b.a has the same address. Am I doing this correctly?

[Eliyaan]

I think a and b are pointers to the same struct so the address of a and b are different but the address they store are the same