Thread overview
[Issue 4727] New: Passing std.concurrency.receive() a function of type Variant causes a static assert
Aug 26, 2010
Tim Keating
Aug 26, 2010
Tim Keating
Aug 26, 2010
Tim Keating
Aug 26, 2010
Tim Keating
Dec 10, 2011
klickverbot
August 26, 2010
http://d.puremagic.com/issues/show_bug.cgi?id=4727

           Summary: Passing std.concurrency.receive() a function of type
                    Variant causes a static assert
           Product: D
           Version: D2
          Platform: x86
        OS/Version: Mac OS X
            Status: NEW
          Severity: major
          Priority: P2
         Component: Phobos
        AssignedTo: nobody@puremagic.com
        ReportedBy: itsallaboutthedyo@gmail.com


--- Comment #0 from Tim Keating <itsallaboutthedyo@gmail.com> 2010-08-25 21:00:53 PDT ---
Example 13.6.2 from The D Programming Language (p. 405; trivial example of using a function that takes a single Variant type as a "catch all" message handler) does not compile for me. This is using DMD 2.048 and Phobos dated 8/10/10.

The compile error occurs in the static assert on line 385 of concurrency.d. As far as I can tell, the assert says "if we have only one param, and it's a variant, die." Seems to me it needs an additional clause, "... and this isn't the last element in the T array".

-- 
Configure issuemail: http://d.puremagic.com/issues/userprefs.cgi?tab=email
------- You are receiving this mail because: -------
August 26, 2010
http://d.puremagic.com/issues/show_bug.cgi?id=4727



--- Comment #1 from Tim Keating <itsallaboutthedyo@gmail.com> 2010-08-25 21:02:49 PDT ---
Shoot, that's what i < T.length does. Hmm, not sure why this is blowing up, then.

-- 
Configure issuemail: http://d.puremagic.com/issues/userprefs.cgi?tab=email
------- You are receiving this mail because: -------
August 26, 2010
http://d.puremagic.com/issues/show_bug.cgi?id=4727



--- Comment #2 from Tim Keating <itsallaboutthedyo@gmail.com> 2010-08-25 21:20:31 PDT ---
Ah, it's an off-by-one error. The index is 0-based, but T.length is 1-based. That expression should be:

if ( i < T.length-1 )

-- 
Configure issuemail: http://d.puremagic.com/issues/userprefs.cgi?tab=email
------- You are receiving this mail because: -------
August 26, 2010
http://d.puremagic.com/issues/show_bug.cgi?id=4727



--- Comment #3 from Tim Keating <itsallaboutthedyo@gmail.com> 2010-08-25 21:36:31 PDT ---
... and that if might want to be a static if also. I'm not sure why straight "if" wasn't working for me, but it was not successfully evaluating the comparison, so it would still get into the inner block when i == T.length-1 (confirmed by doing a static assert (i != T.length-1), which failed). Changing that to a static if seemed to make it work, however.

-- 
Configure issuemail: http://d.puremagic.com/issues/userprefs.cgi?tab=email
------- You are receiving this mail because: -------
December 10, 2011
http://d.puremagic.com/issues/show_bug.cgi?id=4727


klickverbot <code@klickverbot.at> changed:

           What    |Removed                     |Added
----------------------------------------------------------------------------
             Status|NEW                         |RESOLVED
                 CC|                            |code@klickverbot.at
         Resolution|                            |FIXED


--- Comment #4 from klickverbot <code@klickverbot.at> 2011-12-10 13:16:41 PST ---
Fixed in commit b78bc4c59d8d9c87f55659baf277fce4960de7db resp. since 2.053.

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