Thread overview
[Issue 4530] New: Tidier function types
Jul 28, 2010
Stewart Gordon
Jul 29, 2010
Sobirari Muhomori
Jul 29, 2010
Stewart Gordon
Jun 15, 2011
yebblies
July 28, 2010
http://d.puremagic.com/issues/show_bug.cgi?id=4530

           Summary: Tidier function types
           Product: D
           Version: D2
          Platform: All
        OS/Version: All
            Status: NEW
          Severity: enhancement
          Priority: P2
         Component: DMD
        AssignedTo: nobody@puremagic.com
        ReportedBy: bearophile_hugs@eml.cc


--- Comment #0 from bearophile_hugs@eml.cc 2010-07-28 15:30:47 PDT ---
This is a valid D2 code (dmd 2.047):


import std.stdio;
void main() {
    string a();
}


Allowing/keeping ugly and error-prone function literals in D2 is bad. It's better for D2 to keep/allow only _one_ standard, clean and readable syntax for functions pointers (and one for delegates, if necessary). Keeping C function pointer syntax or something similar in D2 causes troubles.

-- 
Configure issuemail: http://d.puremagic.com/issues/userprefs.cgi?tab=email
------- You are receiving this mail because: -------
July 28, 2010
http://d.puremagic.com/issues/show_bug.cgi?id=4530


Justin Spahr-Summers <Justin.SpahrSummers@gmail.com> changed:

           What    |Removed                     |Added
----------------------------------------------------------------------------
                 CC|                            |Justin.SpahrSummers@gmail.c
                   |                            |om


--- Comment #1 from Justin Spahr-Summers <Justin.SpahrSummers@gmail.com> 2010-07-28 18:17:15 CDT ---
(In reply to comment #0)
> Allowing/keeping ugly and error-prone function literals in D2 is bad. It's better for D2 to keep/allow only _one_ standard, clean and readable syntax for functions pointers (and one for delegates, if necessary). Keeping C function pointer syntax or something similar in D2 causes troubles.

'a' is not a function literal... it's a declaration of a nested function. The C syntax for a function pointer would be:

string (*a)();

which I think is pretty unambiguous.

-- 
Configure issuemail: http://d.puremagic.com/issues/userprefs.cgi?tab=email
------- You are receiving this mail because: -------
July 28, 2010
http://d.puremagic.com/issues/show_bug.cgi?id=4530


Stewart Gordon <smjg@iname.com> changed:

           What    |Removed                     |Added
----------------------------------------------------------------------------
                 CC|                            |smjg@iname.com


--- Comment #2 from Stewart Gordon <smjg@iname.com> 2010-07-28 16:40:28 PDT ---
(In reply to comment #1)
> string (*a)();
> 
> which I think is pretty unambiguous.

It isn't.  Could be a function called string, which is being called with argument (*a), and the function returned from that call then called with an empty argument list.

D grammar aims to be unambiguous.  Removing C-style casts has already been a step in the right direction.  Removing C-style function pointer declarations would be another.

-- 
Configure issuemail: http://d.puremagic.com/issues/userprefs.cgi?tab=email
------- You are receiving this mail because: -------
July 29, 2010
http://d.puremagic.com/issues/show_bug.cgi?id=4530



--- Comment #3 from Sobirari Muhomori <dfj1esp02@sneakemail.com> 2010-07-29 09:21:49 PDT ---
Or a static opCall on the string type.
Or an extension method opCall(string b, int a){...}

-- 
Configure issuemail: http://d.puremagic.com/issues/userprefs.cgi?tab=email
------- You are receiving this mail because: -------
July 29, 2010
http://d.puremagic.com/issues/show_bug.cgi?id=4530



--- Comment #4 from Stewart Gordon <smjg@iname.com> 2010-07-29 14:51:27 PDT ---
But as far as the parser is concerned, that's still a function call.  It's the semantic analyser that identifies whether string is an actual function, a function pointer or delegate, a type on which a static opCall is defined, an object of a type that has an opCall, or whatever.

A function pointer declaration, on the other hand, is a syntactical form distinct from all of these.

-- 
Configure issuemail: http://d.puremagic.com/issues/userprefs.cgi?tab=email
------- You are receiving this mail because: -------
August 04, 2010
http://d.puremagic.com/issues/show_bug.cgi?id=4530



--- Comment #5 from bearophile_hugs@eml.cc 2010-08-03 18:13:23 PDT ---
See also enhancement request 4580

-- 
Configure issuemail: http://d.puremagic.com/issues/userprefs.cgi?tab=email
------- You are receiving this mail because: -------
June 15, 2011
http://d.puremagic.com/issues/show_bug.cgi?id=4530


yebblies <yebblies@gmail.com> changed:

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


--- Comment #6 from yebblies <yebblies@gmail.com> 2011-06-15 07:59:44 PDT ---
In the original comment, there is no 'ugly and error-prone function literal'.

There is a nested function declaration with no body.

import std.stdio;
void main() {
    string a();
    pragma(msg, typeof(&a));
}

Prints:
string delegate()

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