https://d.puremagic.com/issues/show_bug.cgi?id=8954


Andrej Mitrovic <andrej.mitrovich@gmail.com> changed:

           What    |Removed                     |Added
----------------------------------------------------------------------------
                 CC|                            |andrej.mitrovich@gmail.com


--- Comment #3 from Andrej Mitrovic <andrej.mitrovich@gmail.com> 2014-02-12 13:41:01 PST ---
Internal notes for other compiler devs:

Offending code for postblits in StructDeclaration::buildPostBlit:

-----
Loc loc = Loc();    // internal code should have no loc to prevent coverage
-----

Commenting that out fixes it for postblits.

For dtors it's in AggregateDeclaration::buildDtor(Scope *sc):

-----
Loc loc = Loc();    // internal code should have no loc to prevent coverage
-----

I think it can be fixed by adding an 'isInternal' field to the dtor/postblit classes and passing a valid Loc, then cov generation could check isInternal while other diagnostics code doesn't have to be touched. Will try this out.

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

https://d.puremagic.com/issues/show_bug.cgi?id=8954



--- Comment #4 from Andrej Mitrovic <andrej.mitrovich@gmail.com> 2014-02-12 13:46:16 PST ---
(In reply to comment #3)
> I think it can be fixed by adding an 'isInternal' field to the dtor/postblit classes and passing a valid Loc, then cov generation could check isInternal while other diagnostics code doesn't have to be touched. Will try this out.

Nevermind, it's actually CallExp which needs the Loc info. Still it's fixable.

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