[Bug c++/40793] "Error: no matching function for call to XYZ" doesn't display function-template-arguments

[Bug c++/40793] "Error: no matching function for call to XYZ" doesn't display function-template-arguments

[pinskia at gcc dot gnu.org]

http://gcc.gnu.org/bugzilla/show_bug.cgi?id=40793

--- Comment #9 from Andrew Pinski <pinskia at gcc dot gnu.org> 2010-11-30 23:52:05 UTC ---
So with the example in comment #5, we get:
t.cc: In function ‘void f()’:
t.cc:5:31: error: no matching function for call to ‘staticPrint()’
t.cc:2:18: note: candidate is: template<class T, T t> void staticPrint()

--- CUT ---
Is that good enough?

[Bug c++/40793] "Error: no matching function for call to XYZ" doesn't display function-template-arguments

[manu at gcc dot gnu.org]

http://gcc.gnu.org/bugzilla/show_bug.cgi?id=40793

--- Comment #10 from Manuel López-Ibáñez <manu at gcc dot gnu.org> 2010-12-01 00:13:52 UTC ---
(In reply to comment #9)
> So with the example in comment #5, we get:
> t.cc: In function ‘void f()’:
> t.cc:5:31: error: no matching function for call to ‘staticPrint()’
> t.cc:2:18: note: candidate is: template<class T, T t> void staticPrint()
> 
> --- CUT ---
> Is that good enough?

I don't think so. The problem is that looking at the error, one cannot say what
are the template arguments (which may be the result of something else). The
desired output is mentioned in comment #5.

Printing the candidates is of course an improvement.

[Bug c++/40793] "Error: no matching function for call to XYZ" doesn't display function-template-arguments

[jonathan.sd24 at yahoo dot de]

http://gcc.gnu.org/bugzilla/show_bug.cgi?id=40793

--- Comment #11 from Jonathan Schmidt-Dominé <jonathan.sd24 at yahoo dot de> 2011-03-23 16:29:31 UTC ---
Couldn't this patch be accepted? It looks very nice…

[Bug c++/40793] "Error: no matching function for call to XYZ" doesn't display function-template-arguments

[manu at gcc dot gnu.org]

http://gcc.gnu.org/bugzilla/show_bug.cgi?id=40793

--- Comment #12 from Manuel López-Ibáñez <manu at gcc dot gnu.org> 2011-03-23 17:17:37 UTC ---
(In reply to comment #11)
> Couldn't this patch be accepted? It looks very nice…

Unfortunately, I don't have enough free time to test this patch and submit it
properly. Even if I had some time in the future for GCC, there are other things
I would like to fix first. I am not even sure it is correct in the general
case. So Jonathan or anyone else, please feel free to take it through
contribution process: http://gcc.gnu.org/contribute.html

[Bug c++/40793] "Error: no matching function for call to XYZ" doesn't display function-template-arguments

[paolo.carlini at oracle dot com]

http://gcc.gnu.org/bugzilla/show_bug.cgi?id=40793

Paolo Carlini <paolo.carlini at oracle dot com> changed:

           What    |Removed                     |Added
----------------------------------------------------------------------------
                 CC|gcc-bugs at gcc dot gnu.org |

--- Comment #13 from Paolo Carlini <paolo.carlini at oracle dot com> 2011-09-29 10:42:52 UTC ---
Manuel, is this the same as PR39813? (the latter should get a better Summary)

[Bug c++/40793] "Error: no matching function for call to XYZ" doesn't display function-template-arguments

[paolo.carlini at oracle dot com]

http://gcc.gnu.org/bugzilla/show_bug.cgi?id=40793

--- Comment #14 from Paolo Carlini <paolo.carlini at oracle dot com> 2011-09-29 10:45:40 UTC ---
Never mind, I don't think it's the same, but 39813 is also pretty weird, why
error.c sees the member template as a friend it's a mystery to me...

[Bug c++/40793] "Error: no matching function for call to XYZ" doesn't display function-template-arguments

[manu at gcc dot gnu.org]

http://gcc.gnu.org/bugzilla/show_bug.cgi?id=40793

--- Comment #15 from Manuel López-Ibáñez <manu at gcc dot gnu.org> 2011-09-29 12:13:17 UTC ---
(In reply to comment #14)
> Never mind, I don't think it's the same, but 39813 is also pretty weird, why
> error.c sees the member template as a friend it's a mystery to me...

Paolo, feel free to take the patch in comment #7 and updated as needed.
However, I am not very happy with it because it uses %E and that means we
reconstruct an expression, which sometimes leads to something that does not
resemble the original source code at all. It would be neat to have a way to
print the template arguments without using %E.

Also, I am not sure what is the current output after Nathan improved the
candidate diagnostics.

[Bug c++/40793] "Error: no matching function for call to XYZ" doesn't display function-template-arguments

[paolo.carlini at oracle dot com]

http://gcc.gnu.org/bugzilla/show_bug.cgi?id=40793

--- Comment #16 from Paolo Carlini <paolo.carlini at oracle dot com> 2011-09-29 12:20:51 UTC ---
Thanks. Now in mainline we say this:

40793.C: In function ‘void f()’:
40793.C:5:31: error: no matching function for call to ‘staticPrint()’
40793.C:5:31: note: candidate is:
40793.C:2:6: note: template<class T, T t> void staticPrint()
40793.C:2:6: note:   template argument deduction/substitution failed:
40793.C:5:31: error: ‘double’ is not a valid type for a template constant
parameter

In 4.6.x we don't produce the last note and error message. What do you think?

[Bug c++/40793] "Error: no matching function for call to XYZ" doesn't display function-template-arguments

[manu at gcc dot gnu.org]

http://gcc.gnu.org/bugzilla/show_bug.cgi?id=40793

--- Comment #17 from Manuel López-Ibáñez <manu at gcc dot gnu.org> 2011-09-29 14:35:14 UTC ---
(In reply to comment #16)
> Thanks. Now in mainline we say this:
> 
> 40793.C: In function ‘void f()’:
> 40793.C:5:31: error: no matching function for call to ‘staticPrint()’
> 40793.C:5:31: note: candidate is:
> 40793.C:2:6: note: template<class T, T t> void staticPrint()
> 40793.C:2:6: note:   template argument deduction/substitution failed:
> 40793.C:5:31: error: ‘double’ is not a valid type for a template constant
> parameter
> 
> In 4.6.x we don't produce the last note and error message. What do you think?

Well, I think it is even better than Clang now. 

The locations are a bit off, the first error should point to 5.1 and the last
one to 5.13, but that was not the point of the original submitter and I am not
volunteering to do the work, so if you are happy, close it as fixed.

[Bug c++/40793] "Error: no matching function for call to XYZ" doesn't display function-template-arguments

[paolo.carlini at oracle dot com]

http://gcc.gnu.org/bugzilla/show_bug.cgi?id=40793

Paolo Carlini <paolo.carlini at oracle dot com> changed:

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

--- Comment #18 from Paolo Carlini <paolo.carlini at oracle dot com> 2011-09-29 14:52:27 UTC ---
Agreed. I think that about locations we have other PRs, and, arguably, the
issues are well known. Fixed in 4.7.

[Bug c++/40793] "Error: no matching function for call to XYZ" doesn't display function-template-arguments

[bangerth at gmail dot com]

------- Comment #8 from bangerth at gmail dot com  2010-07-01 21:38 -------
I think that would already be an improvement.


-- 


http://gcc.gnu.org/bugzilla/show_bug.cgi?id=40793

[Bug c++/40793] "Error: no matching function for call to XYZ" doesn't display function-template-arguments

[manu at gcc dot gnu dot org]

[-- Warning: decoded text below may be mangled, UTF-8 assumed --]
[-- Attachment #1: Type: text/plain, Size: 1643 bytes --]



------- Comment #7 from manu at gcc dot gnu dot org  2010-07-01 21:31 -------
With the patch below, we print this:

pr40793.C:5:31: error: no matching function for call to ‘staticPrint<double,
"text">()’
pr40793.C:2:18: note: candidate is: template<class T, T t> void staticPrint()

but I am not sure if it is possible to print the type of "text". Any ideas?


Index: call.c
===================================================================
--- call.c      (revision 161658)
+++ call.c      (working copy)
@@ -3247,12 +3247,19 @@ build_new_function_call (tree fn, VEC(tr
        {
          if (!any_viable_p && candidates && ! candidates->next
              && (TREE_CODE (candidates->fn) == FUNCTION_DECL))
            return cp_build_function_call_vec (candidates->fn, args, complain);
          if (TREE_CODE (fn) == TEMPLATE_ID_EXPR)
-           fn = TREE_OPERAND (fn, 0);
-         if (!any_viable_p)
+           {
+             if (!any_viable_p)
+               error ("no matching function for call to %<%E(%A)%>",
+                      fn, build_tree_list_vec (*args));
+             else
+               error ("call of overloaded %<%E(%A)%> is ambiguous",
+                      fn, build_tree_list_vec (*args));
+           }
+         else if (!any_viable_p)
            error ("no matching function for call to %<%D(%A)%>",
                   DECL_NAME (OVL_CURRENT (fn)), build_tree_list_vec (*args));
          else
            error ("call of overloaded %<%D(%A)%> is ambiguous",
                   DECL_NAME (OVL_CURRENT (fn)), build_tree_list_vec (*args));


-- 


http://gcc.gnu.org/bugzilla/show_bug.cgi?id=40793

[Bug c++/40793] "Error: no matching function for call to XYZ" doesn't display function-template-arguments

[m-matti-a dot lehtonen at iki dot fi]

------- Comment #6 from m-matti-a dot lehtonen at iki dot fi  2010-07-01 08:58 -------
I have spend recently many, Many hours with heavily template based code (using
CGAL - http://www.cgal.org/) until I figured, why my code failed.

Telling input types and candidate list of all mismatched functions would be
more then helpful than current way of reporting error.


-- 


http://gcc.gnu.org/bugzilla/show_bug.cgi?id=40793

[Bug c++/40793] "Error: no matching function for call to XYZ" doesn't display function-template-arguments

[bangerth at gmail dot com]

------- Comment #5 from bangerth at gmail dot com  2009-08-25 14:25 -------
Jonathan, the point everyone is trying to make is this: since no function
or function template matches the call, all the compiler could possibly
do is list all declarations of the name staticPrint() or none, but of 
course either choice is equally good.

But I think I understand what he means. In this example
----------------------
template<typename T, T t>
void staticPrint();

void f() {
  staticPrint<double, "text">();
}
----------------------
He would like the error message to
say
  no matching function for call to 
  'staticPrint<double,(const char[5])"text")>()'
instead of
  no matching function for call to 'staticPrint()'
The reason being that it can indeed sometimes be hard to figure out what
template type the compiler is trying to match if they result from some
template trickery rather than being explicitly specified as done in the
example above.

I agree with this.

W.


-- 

bangerth at gmail dot com changed:

           What    |Removed                     |Added
----------------------------------------------------------------------------
                 CC|                            |bangerth at gmail dot com
             Status|UNCONFIRMED                 |NEW
     Ever Confirmed|0                           |1
           Keywords|                            |diagnostic
   Last reconfirmed|0000-00-00 00:00:00         |2009-08-25 14:25:09
               date|                            |


http://gcc.gnu.org/bugzilla/show_bug.cgi?id=40793

[Bug c++/40793] "Error: no matching function for call to XYZ" doesn't display function-template-arguments

[jonathan dot sd24 at yahoo dot de]

------- Comment #4 from jonathan dot sd24 at yahoo dot de  2009-07-29 21:05 -------
Template-Parameters can be results of long calculations or typedefs with
several layers of indirection.


-- 


http://gcc.gnu.org/bugzilla/show_bug.cgi?id=40793

[Bug c++/40793] "Error: no matching function for call to XYZ" doesn't display function-template-arguments

[jwakely dot gcc at gmail dot com]

------- Comment #3 from jwakely dot gcc at gmail dot com  2009-07-21 17:45 -------
It tells you the location of the call. If you've specified the template
arguments at that location then you can see what they are, I don't see how
putting them in the diagnostic will save hours if you could find out the same
info in seconds by looking at the line.

There's no candidate list because there is no match for the template arguments
you used. The empty candidate list tells you something by itself.


-- 


http://gcc.gnu.org/bugzilla/show_bug.cgi?id=40793

[Bug c++/40793] "Error: no matching function for call to XYZ" doesn't display function-template-arguments

[jonathan dot sd24 at yahoo dot de]

------- Comment #2 from jonathan dot sd24 at yahoo dot de  2009-07-18 10:02 -------
When I explicitly specify the template-arguments g++ should know it.
And there was no candidate-list.

For example:
template<typename T, T t>
void staticPrint()
{
  cout << t << endl;
}

Now I call:
staticPrint<double, "text">();
I get:
"no matching function for call to void staticPrint()"
No candidates... And I would like to know, what's wrong with the
template-arguments.
Beleive me: You can spent hours with searching for the reason of such an error.


-- 

jonathan dot sd24 at yahoo dot de changed:

           What    |Removed                     |Added
----------------------------------------------------------------------------
             Status|RESOLVED                    |UNCONFIRMED
         Resolution|WONTFIX                     |


http://gcc.gnu.org/bugzilla/show_bug.cgi?id=40793

[Bug c++/40793] "Error: no matching function for call to XYZ" doesn't display function-template-arguments

[rguenth at gcc dot gnu dot org]

------- Comment #1 from rguenth at gcc dot gnu dot org  2009-07-18 08:59 -------
But it doens't know it looks for a template function.  It should display the
candidate list from which you can see what happens.

This is how name-lookup and overload resolution works...


-- 

rguenth at gcc dot gnu dot org changed:

           What    |Removed                     |Added
----------------------------------------------------------------------------
             Status|UNCONFIRMED                 |RESOLVED
         Resolution|                            |WONTFIX


http://gcc.gnu.org/bugzilla/show_bug.cgi?id=40793