www.digitalmars.com         C & C++   DMDScript  

digitalmars.D.bugs - [Issue 2274] New: all static if typeof seriously wrong

reply d-bugmail puremagic.com writes:
http://d.puremagic.com/issues/show_bug.cgi?id=2274

           Summary: all static if typeof seriously wrong
           Product: D
           Version: 2.017
          Platform: PC
        OS/Version: Linux
            Status: NEW
          Severity: major
          Priority: P2
         Component: DMD
        AssignedTo: bugzilla digitalmars.com
        ReportedBy: someanon yahoo.com


each of line 16, 17, 18 in attached file t.d 
when comment off, will generate an error.

line 16: // wrong: t.A:- t.B:+ t.C:- int:-
line 17: // wrong: t.A:- t.B:- t.C:- int:-
line 18: // wrong: t.A:- t.B:- t.C:- int:-

I wonder if anyone ever used static if typeof at all.


-- 
Aug 07 2008
next sibling parent d-bugmail puremagic.com writes:
http://d.puremagic.com/issues/show_bug.cgi?id=2274






Created an attachment (id=266)
 --> (http://d.puremagic.com/issues/attachment.cgi?id=266&action=view)
the source file


-- 
Aug 07 2008
prev sibling next sibling parent d-bugmail puremagic.com writes:
http://d.puremagic.com/issues/show_bug.cgi?id=2274


2korden gmail.com changed:

           What    |Removed                     |Added
----------------------------------------------------------------------------
             Status|NEW                         |RESOLVED
         Resolution|                            |DUPLICATE





no need for posting twise, you already reported an issue earlier.

*** This bug has been marked as a duplicate of 2154 ***


-- 
Aug 07 2008
prev sibling next sibling parent d-bugmail puremagic.com writes:
http://d.puremagic.com/issues/show_bug.cgi?id=2274


someanon yahoo.com changed:

           What    |Removed                     |Added
----------------------------------------------------------------------------
             Status|RESOLVED                    |REOPENED
         Resolution|DUPLICATE                   |





No, this is not DUPLICATE.

That one is that the compiler doesn't accept the code.

This one, is the binary generated behaves wrong.


-- 
Aug 07 2008
prev sibling next sibling parent d-bugmail puremagic.com writes:
http://d.puremagic.com/issues/show_bug.cgi?id=2274






I can't agree with you. The binary generated is perfectly valid. It never
crashes, it behaves just like intended. The fact alone that you expect
different result doesn't mean the executable is broken.

Your static if check is not passed, that's all.

If you expect to see the following output:
test.A:+ test.B:+ test.C:- int:-

then you should change your code. You are not allowed to access non-static data
via class name like this:
bool result = is(typeof(E.toString()) : string); // false unless toString is
static

or like this:
bool result = is(typeof(E.myInt) : int);

You have to create a class instance, first:
auto e = new E();
bool result = is(typeof(e.toString()) : string); // true for any class unless
toString is private

or, better, in one line:
bool result = is(typeof((new E()).toString()) : string);

Note that no allocation will take place in latter case since it is merely a
check.

Shall we consider an issue closed by now?


-- 
Aug 07 2008
prev sibling next sibling parent d-bugmail puremagic.com writes:
http://d.puremagic.com/issues/show_bug.cgi?id=2274







 I can't agree with you. The binary generated is perfectly valid. It never
 crashes, it behaves just like intended. The fact alone that you expect
 different result doesn't mean the executable is broken.
 
 Your static if check is not passed, that's all.
 
 If you expect to see the following output:
 test.A:+ test.B:+ test.C:- int:-
 
 then you should change your code. You are not allowed to access non-static data
 via class name like this:
 bool result = is(typeof(E.toString()) : string); // false unless toString is
 static
 
 or like this:
 bool result = is(typeof(E.myInt) : int);
 
 You have to create a class instance, first:
 auto e = new E();
 bool result = is(typeof(e.toString()) : string); // true for any class unless
 toString is private
 
 or, better, in one line:
 bool result = is(typeof((new E()).toString()) : string);
 
 Note that no allocation will take place in latter case since it is merely a
 check.
 
 Shall we consider an issue closed by now?
Thanks for your answer, but No. When you give an answer please test it: (with v2.017 on Linux) If I change the test code as you suggested: static if (is((new E()).toString()) : string) { // regardless of 'static' $ dmd t.d t.d(18): basic type expected, not ( t.d(18): found 'new' when expecting ')' t.d(18): found 'E' when expecting ')' t.d(18): found ')' when expecting ';' following 'statement' t.d(18): found ':' instead of statement t.d(20): Declaration expected, not 'else' t.d(23): Declaration expected, not 'return' t.d(24): unrecognized declaration BTW, I need static test, because inside '{}' I will do something with the method/attribute. --
Aug 07 2008
prev sibling next sibling parent d-bugmail puremagic.com writes:
http://d.puremagic.com/issues/show_bug.cgi?id=2274


someanon yahoo.com changed:

           What    |Removed                     |Added
----------------------------------------------------------------------------
             Status|REOPENED                    |RESOLVED
         Resolution|                            |FIXED





sorry I got it wrong: it should be static if (is(typeof((new E()).toString()) :
string)) {..}

Now the code compiles, and runs correctly.

Thanks for the reply.


-- 
Aug 07 2008
prev sibling next sibling parent d-bugmail puremagic.com writes:
http://d.puremagic.com/issues/show_bug.cgi?id=2274






Hold on, one more issue: in the original file

line 16: // wrong: t.A:- t.B:+ t.C:- int:-

Why an empty class B{} pass the test?  it has a static toString?


-- 
Aug 07 2008
prev sibling next sibling parent d-bugmail puremagic.com writes:
http://d.puremagic.com/issues/show_bug.cgi?id=2274






Yes, it inherits it from Object.


-- 
Aug 08 2008
prev sibling next sibling parent d-bugmail puremagic.com writes:
http://d.puremagic.com/issues/show_bug.cgi?id=2274







 Yes, it inherits it from Object.
 
/dmd/src/phobos/object.d class Object { void print(); string toString(); // you mean this one? hash_t toHash(); int opCmp(Object o); bool opEquals(Object o); final void notifyRegister(void delegate(Object) dg); final void notifyUnRegister(void delegate(Object) dg); static Object factory(string classname); } How it pass the static test in my original code? Shall we reopen the bug? ;-) --
Aug 08 2008
prev sibling next sibling parent d-bugmail puremagic.com writes:
http://d.puremagic.com/issues/show_bug.cgi?id=2274


someanon yahoo.com changed:

           What    |Removed                     |Added
----------------------------------------------------------------------------
             Status|RESOLVED                    |REOPENED
         Resolution|FIXED                       |





line 16: // wrong: t.A:- t.B:+ t.C:- int:-

Why an empty class B{} pass the test? 


-- 
Aug 10 2008
prev sibling parent d-bugmail puremagic.com writes:
http://d.puremagic.com/issues/show_bug.cgi?id=2274


Don <clugdbug yahoo.com.au> changed:

           What    |Removed                     |Added
----------------------------------------------------------------------------
             Status|REOPENED                    |RESOLVED
                 CC|                            |clugdbug yahoo.com.au
         Resolution|                            |INVALID




 line 16: // wrong: t.A:- t.B:+ t.C:- int:-
 
 Why an empty class B{} pass the test?
Because it has toString(). Try this code: --- import std.stdio; class B {} void main() { B b = new B; writefln(b.toString()); } --- There's no bug here. -- Configure issuemail: http://d.puremagic.com/issues/userprefs.cgi?tab=email ------- You are receiving this mail because: -------
Nov 23 2009