http://d.puremagic.com/issues/show_bug.cgi?id=8213
Summary: Incorrect error message with pointer to ubyte[] and front Product: D Version: unspecified Platform: All OS/Version: All Status: NEW Severity: normal Priority: P2 Component: DMD AssignedTo: nob...@puremagic.com ReportedBy: jmdavisp...@gmx.com --- Comment #0 from Jonathan M Davis <jmdavisp...@gmx.com> 2012-06-09 02:21:47 PDT --- This code import std.array; struct S(R) { this(R* range) { _range = range; } auto front() { return _range.front; } R* _range; } void main() { ubyte[] buffer = [42]; auto s = S!(ubyte[])(&buffer); auto f = s.front; } results in this error: q.d(12): Error: no property 'front' for type 'ubyte[]' q.d(21): Error: template instance q.S!(ubyte[]) error instantiating I'm tempted to argue that this should just compile, since . automatically dereferences the type that it's on if it's a pointer, and the type in question (ubyte[]) has a front via std.array, but _range itself doesn't match front's parameters, because it's a pointer to a ubyte[] rather than a ubyte[], so I suspect that this _isn't_ supposed to work much as I'd like it to. If it _is_ supposed to work, however, then this obviously shows that it isn't currently and is therefore a bug. Regardless of the correct behavior, however, the error message is bad. It says that ubyte[] has no property front, which is not only not true, but the type which is failing is ubyte[]*, not ubyte[], so the type in the error message is wrong. Either this code needs to be fixed so that it compiles and works just fine (automatically dereferencing the pointer and using the type of the dereferenced pointer rather than the pointer itself when lookin for functions to use with UFCS), or the error needs to be fixed so that it gives the correct type. -- Configure issuemail: http://d.puremagic.com/issues/userprefs.cgi?tab=email ------- You are receiving this mail because: -------