Bug 8316 - Confusing diagnostic for code that misuses conversion operators

Description:

Last confirmed:

Opened: 2002-10-22 21:36

Consider the following code fragment:
struct X {
  char &operator[](char i);
  operator char *();
  operator const char *() const;
};

void f(X &x)
{
  x[0] += 1;
}

Obviously this is bad practice, since x[0] should be interpreted too ways.  The compiler ought to tell the programmer that he/she is doing something wrong.

The compiler does do that.  However, the diagnostic is so confusing that a user who would make this mistake is unlikely to be able to figure out what the compiler is saying.

With 3.1 and 3.2, the output is:
bash-2.05$ /usr/local/gcc-3.1/bin/g++ -c foo.cc
foo.cc: In function `void f(X&)':
foo.cc:11: choosing `char& X::operator[](char)' over `operator[]'
foo.cc:11:   because worst conversion for the former is better than worst 
   conversion for the latter
foo.cc:11: choosing `char& X::operator[](char)' over `operator[]'
foo.cc:11:   because worst conversion for the former is better than worst 
   conversion for the latter

Note that (1) the warning is repeated; and (2) it doesn't get to the heart of the matter and tell the user that it's a bad idea to provide both operator* and operator[].  A user is unlikely to be able to understand the problem from reading this message.

With 3.3 the behavior is exactly the same, except that 'warning' is replaced by 'error'.  This is not an improvement.

Release:
3.1, 3.2, pre-3.3 TOT

Environment:
Linux

------- Comment #1 From Matt Austern 2002-10-23 04:29 -------

From: austern@apple.com
To: gcc-gnats@gcc.gnu.org
Cc:  
Subject: c++/8316: Confusing diagnostic for code that misuses conversion operators
Date: 23 Oct 2002 04:29:40 -0000

 >Number:         8316
 >Category:       c++
 >Synopsis:       Confusing diagnostic for code that misuses conversion operators
 >Confidential:   no
 >Severity:       serious
 >Priority:       medium
 >Responsible:    unassigned
 >State:          open
 >Class:          sw-bug
 >Submitter-Id:   net
 >Arrival-Date:   Tue Oct 22 21:36:00 PDT 2002
 >Closed-Date:
 >Last-Modified:
 >Originator:     austern@apple.com
 >Release:        3.1, 3.2, pre-3.3 TOT
 >Organization:
 >Environment:
 Linux
 >Description:
 Consider the following code fragment:
 struct X {
   char &operator[](char i);
   operator char *();
   operator const char *() const;
 };
 
 void f(X &x)
 {
   x[0] += 1;
 }
 
 Obviously this is bad practice, since x[0] should be interpreted too ways.  The compiler ought to tell the programmer that he/she is doing something wrong.
 
 The compiler does do that.  However, the diagnostic is so confusing that a user who would make this mistake is unlikely to be able to figure out what the compiler is saying.
 
 With 3.1 and 3.2, the output is:
 bash-2.05$ /usr/local/gcc-3.1/bin/g++ -c foo.cc
 foo.cc: In function `void f(X&)':
 foo.cc:11: choosing `char& X::operator[](char)' over `operator[]'
 foo.cc:11:   because worst conversion for the former is better than worst 
    conversion for the latter
 foo.cc:11: choosing `char& X::operator[](char)' over `operator[]'
 foo.cc:11:   because worst conversion for the former is better than worst 
    conversion for the latter
 
 Note that (1) the warning is repeated; and (2) it doesn't get to the heart of the matter and tell the user that it's a bad idea to provide both operator* and operator[].  A user is unlikely to be able to understand the problem from reading this message.
 
 With 3.3 the behavior is exactly the same, except that 'warning' is replaced by 'error'.  This is not an improvement.
 >How-To-Repeat:
 
 >Fix:
 
 >Release-Note:
 >Audit-Trail:
 >Unformatted:

------- Comment #2 From Wolfgang Bangerth 2002-12-19 17:16 -------

State-Changed-From-To: open->analyzed
State-Changed-Why: Confirmed.
    
    Apart from the unnecessary duplication of error messages
    (where does this come from, this is not even a template
    which would be processed in two phases?), I can only
    think of telling that
      X::operator[] (char)
    is chosen over
      ::operator[] (whatever)
    (note the :: at the beginning). Matt, what better message
    do you have in mind? I think it is not so obvious
    what to say here.
    
    W.
    
    (Ah, and having an index operator that takes a character
    instead of an integer is an abuse as well :-)

------- Comment #6 From Jason Merrill 2003-03-19 18:19 -------

From: jason@gcc.gnu.org
To: gcc-gnats@gcc.gnu.org
Cc:  
Subject: c++/8316
Date: 19 Mar 2003 18:19:53 -0000

 CVSROOT:	/cvs/gcc
 Module name:	gcc
 Changes by:	jason@gcc.gnu.org	2003-03-19 18:19:53
 
 Modified files:
 	gcc/cp         : ChangeLog call.c 
 
 Log message:
 	PR c++/8316, c++/9315, c++/10136
 	* call.c (print_z_candidate): Split out from...
 	(print_z_candidiates): ...here.
 	(joust): Use it.
 
 Patches:
 http://gcc.gnu.org/cgi-bin/cvsweb.cgi/gcc/gcc/cp/ChangeLog.diff?cvsroot=gcc&r1=1.3284&r2=1.3285
 http://gcc.gnu.org/cgi-bin/cvsweb.cgi/gcc/gcc/cp/call.c.diff?cvsroot=gcc&r1=1.369&r2=1.370

------- Comment #8 From Jason Merrill 2003-03-21 07:12 -------

From: jason@gcc.gnu.org
To: gcc-gnats@gcc.gnu.org
Cc:  
Subject: c++/8316
Date: 21 Mar 2003 07:12:50 -0000

 CVSROOT:	/cvs/gcc
 Module name:	gcc
 Changes by:	jason@gcc.gnu.org	2003-03-21 07:12:50
 
 Modified files:
 	gcc/testsuite/g++.old-deja/g++.ext: overload1.C 
 Added files:
 	gcc/testsuite/g++.dg/overload: new1.C 
 Removed files:
 	gcc/testsuite/g++.old-deja/g++.other: new2.C 
 
 Log message:
 	PR c++/8316, c++/9315, c++/10136
 	* call.c (print_z_candidate): Split out from...
 	(print_z_candidiates): ...here.
 	(joust): Use it.
 
 Patches:
 http://gcc.gnu.org/cgi-bin/cvsweb.cgi/gcc/gcc/testsuite/g++.dg/overload/new1.C.diff?cvsroot=gcc&r1=NONE&r2=1.1
 http://gcc.gnu.org/cgi-bin/cvsweb.cgi/gcc/gcc/testsuite/g++.old-deja/g++.ext/overload1.C.diff?cvsroot=gcc&r1=1.1&r2=1.2
 http://gcc.gnu.org/cgi-bin/cvsweb.cgi/gcc/gcc/testsuite/g++.old-deja/g++.other/new2.C.diff?cvsroot=gcc&r1=1.4&r2=NONE

------- Comment #10 From ext.zxmultran021@astrium.eads.net 2004-09-21 16:15 -------

Hello,

I use g++ 3.3.3 to compile my application, and I encountered a problem :

src/OpePreparator.cc:2075: error: ISO C++ says that `int 
   Ptr<T>::operator!=(const T*) const [with T = C_OPL::Header_att]' and `
   operator!=' are ambiguous even though the worst conversion for the former is 
   better than the worst conversion for the latter

It seems that this kind of problem is known in the GCC database as the 13089 
bug (duplicate of 8316)

Can you confirm me in which release this bug is fixed ?

Thank you a lot

Bug 8316 depends on:			Show dependency tree Show dependency graph
Bug 8316 blocks:			Show dependency tree Show dependency graph

GCC Bugzilla Bug 8316

Confusing diagnostic for code that misuses conversion operators

Last modified: 2004-09-21 16:15:29

Known to work:
Known to fail: