http://d.puremagic.com/issues/show_bug.cgi?id=4891

           Summary: Assignment from non-pure function to pure function
                    pointer compiles when it shouldn't
           Product: D
           Version: unspecified
          Platform: Other
        OS/Version: Linux
            Status: NEW
          Severity: normal
          Priority: P2
         Component: DMD
        AssignedTo: nobody@puremagic.com
        ReportedBy: jmdavisProg@gmx.com


--- Comment #0 from Jonathan M Davis <jmdavisProg@gmx.com> 2010-09-18 16:07:33 PDT ---
Take this program

struct S
{
public:

    this(int function(int) pure func)
    {
        _func = func;
    }

    @property int function(int) pure func()
    {
        return func;
    }

private:
    int function(int) pure _func;
}

int add1(int num)
{
    return num + 1;
}

void main()
{
    auto s = S(&add1);
}


The construction of s should not compile because add1() is not a pure function.
Granted, it could be pure if I declared it as such, but I didn't. And the
compiler isn't figuring out that add1() could be pure and letting it through
because of that because if I add a writeln() call to add1(), it still compiles.

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

http://d.puremagic.com/issues/show_bug.cgi?id=4891



--- Comment #1 from Jonathan M Davis <jmdavisProg@gmx.com> 2010-09-18 17:15:25 PDT ---
Oh, there's an error in my example - func return func instead of _func, resulting infinite recursion and therefore a segfault. I could have removed that function for the bug report anyway.

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

http://d.puremagic.com/issues/show_bug.cgi?id=4891


bearophile_hugs@eml.cc changed:

           What    |Removed                     |Added
----------------------------------------------------------------------------
                 CC|                            |bearophile_hugs@eml.cc


--- Comment #2 from bearophile_hugs@eml.cc 2010-09-19 18:04:25 PDT ---
See bug 3833

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

http://d.puremagic.com/issues/show_bug.cgi?id=4891


yebblies <yebblies@gmail.com> changed:

           What    |Removed                     |Added
----------------------------------------------------------------------------
             Status|NEW                         |RESOLVED
                 CC|                            |yebblies@gmail.com
         Resolution|                            |DUPLICATE


--- Comment #3 from yebblies <yebblies@gmail.com> 2011-06-06 19:33:40 PDT ---
This is a dupe of 3797, the purity of a function pointer (among other things) is not checked when performing implicit conversions.

With dmd pull 88 you get the following error from dmd:
testx.d(26): Error: constructor testx.S.this (int function(int) pure func) is
no
t callable using argument types (int function(int num))
testx.d(26): Error: cannot implicitly convert expression (& add1) of type int
fu
nction(int num) to int function(int) pure

*** This issue has been marked as a duplicate of issue 3797 ***

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