[Issue 5709] New: template instantiation fails with extra non-mutable class parameter
d-bugmail at puremagic.com
d-bugmail at puremagic.com
Sat Mar 5 18:38:22 PST 2011
http://d.puremagic.com/issues/show_bug.cgi?id=5709
Summary: template instantiation fails with extra non-mutable
class parameter
Product: D
Version: unspecified
Platform: All
OS/Version: Linux
Status: NEW
Severity: normal
Priority: P2
Component: DMD
AssignedTo: nobody at puremagic.com
ReportedBy: jmdavisProg at gmx.com
--- Comment #0 from Jonathan M Davis <jmdavisProg at gmx.com> 2011-03-05 18:35:26 PST ---
This program fails to compile
import std.traits;
class C
{
}
void func1(S)(in S str, immutable C a = null)
if(isSomeString!S)
{
func2(str, a);
}
void func2(S)(in S str, immutable C a = null)
if(isSomeString!S)
{
}
void main()
{
func1("hello world");
func2("hello world");
}
giving this error:
q.d(10): Error: template q.func2(S) if (isSomeString!(S)) does not match any
function template declaration
q.d(10): Error: template q.func2(S) if (isSomeString!(S)) cannot deduce
template function from argument types
!()(const(immutable(char)[]),immutable(C))
q.d(20): Error: template instance q.func1!(string) error instantiating
Removing the template constraints has no effect. Thus far, anything that I
change immutable C a to anything other than a const C, _does_ compile, but it
fails to compile as long as it's const or immutable. The fact that it's a
default parameter seems to have no effect. If pass it a new C() instead of
null, that has no effect. If I change the func2 call in func1 to func2!S, then
it compiles just fine. But for some reason, when the second class parameter is
there and it's not mutable, the template instantiation fails to determine the
correct type of S for the inner instantiation, and you're forced to tell it.
So, this bug is just plain weird, and it _can_ be gotten around, but it's
_definitely_ a bug.
--
Configure issuemail: http://d.puremagic.com/issues/userprefs.cgi?tab=email
------- You are receiving this mail because: -------
More information about the Digitalmars-d-bugs
mailing list