• d-bugmail puremagic.com (31/31) Sep 12 2008 http://d.puremagic.com/issues/show_bug.cgi?id=2358
• d-bugmail puremagic.com (14/14) Jan 22 2012 http://d.puremagic.com/issues/show_bug.cgi?id=2358

d-bugmail puremagic.com writes:

http://d.puremagic.com/issues/show_bug.cgi?id=2358

           Summary: offsetof inconsistent between structs and classes
           Product: D
           Version: 1.035
          Platform: PC
        OS/Version: Linux
            Status: NEW
          Keywords: rejects-valid, spec
          Severity: normal
          Priority: P2
         Component: DMD
        AssignedTo: bugzilla digitalmars.com
        ReportedBy: kamm-removethis incasoftware.de


For classes .offsetof works only on the instance and for structs it only works
on the declaration:

--
class C { int x; }
struct S { int x; }

void main() {
  S s;
  C c = new C;

  auto c1 = c.x.offsetof;
  auto c2 = C.x.offsetof; // error
  auto s1 = s.x.offsetof; // error
  auto s2 = S.x.offsetof;
}
--

For classes, the situation is documented. For structs, the spec merely says
that the offsetof property exists. I expect all combinations could me made
legal.


-- 

d-bugmail puremagic.com writes:

http://d.puremagic.com/issues/show_bug.cgi?id=2358


Walter Bright <bugzilla digitalmars.com> changed:

           What    |Removed                     |Added
----------------------------------------------------------------------------
             Status|NEW                         |RESOLVED
                 CC|                            |bugzilla digitalmars.com
         Resolution|                            |WORKSFORME



14:04:11 PST ---
auto s1 = s.x.offsetof; // error

works correctly now. The spec does document the current behavior. I hope to
remove the restriction on the class .offsetof in the future.

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