www.digitalmars.com         C & C++   DMDScript  

digitalmars.D.bugs - [Issue 9429] New: Unclear runtime error in array assignment of misaligned types.

reply d-bugmail puremagic.com writes:
http://d.puremagic.com/issues/show_bug.cgi?id=9429

           Summary: Unclear runtime error in array assignment of
                    misaligned types.
           Product: D
           Version: D2
          Platform: All
        OS/Version: All
            Status: NEW
          Severity: normal
          Priority: P2
         Component: druntime
        AssignedTo: nobody puremagic.com
        ReportedBy: estewh gmail.com



Given some bug infested code:

struct hackit {
    float[3] mine;
    void opAssign(E)(in E val) if(isArray!E) {
        mine = cast(const float[]))(val.dup);
    }
}
void main() {
    hackit v;
    float[3] v1 = cast(const float[])([1.0, 2.0, 3.0]);    // OK, no error
    v = [1.0, 2.0, 3.0]; // Fails at runtime with error message below
}

This error is thrown at runtime: "object.Error: lengths don't match for array
copy, 3 = 6"

It is correct, although misleading. Both arrays are length 3 and I believe this
should be:

object.Error: array cast misalignment

Swapping the types in the above code gives the error: "object.Error array cast
misalignement"

-- 
Configure issuemail: http://d.puremagic.com/issues/userprefs.cgi?tab=email
------- You are receiving this mail because: -------
Jan 30 2013
parent d-bugmail puremagic.com writes:
http://d.puremagic.com/issues/show_bug.cgi?id=9429




Just some additional info:

The error itself *is* correct, I'm only reporting that the message is
misleading...if possible it would be nice to see it at compile time.

I tested on the following configurations:
OS: fedora 18 x86_64.
Compiler: DMD 2.061 64 bit
Binary produced: elf64-x86-64

OS: Windows 7 64 bit
Compiler: DMD 2.061 32 bit
Binary produced: 32 bit

The messages are virtually the same on both systems.

-- 
Configure issuemail: http://d.puremagic.com/issues/userprefs.cgi?tab=email
------- You are receiving this mail because: -------
Jan 30 2013