Re: Explicitly specializing std::min() on VC++ 2005 Express Edition

Matthias Hofmann wrote:
"Tom Widmer [VC++ MVP]" <tom_usenet@xxxxxxxxxxx> schrieb im Newsbeitrag news:OJCC7GlkHHA.1340@xxxxxxxxxxxxxxxxxxxxxxx
Matthias Hofmann wrote:
"Tom Widmer [VC++ MVP]" <tom_usenet@xxxxxxxxxxx> schrieb im Newsbeitrag news:uCq1XvVkHHA.4188@xxxxxxxxxxxxxxxxxxxxxxx

The problem is that VC2005 is ignoring 5.16/4 (about ?: ):
"If the second and third operands are lvalues and have the same type, the result is of that type and is an lvalue."

Instead, it is applying the lvalue-to-rvalue conversion (e.g. the array-to-pointer conversion) to the parts of the ?: expression, and thus trying to return a pointer where an array is expected.
That's good news, I first thought that there was a bug in my code.
It's been reported and fixed apparently:
https://connect.microsoft.com/VisualStudio/feedback/ViewFeedback.aspx?FeedbackID=101496

I have just downloaded and installed Service Pack 1 for VC++ 2005 Express Edition, which I found here:
http://www.microsoft.com/downloads/details.aspx?displaylang=en&FamilyID=7b0b0339-613a-46e6-ab4d-080d4d4a8c4e

The file I downloaded is named 'VS80sp1-KB926748-X86-INTL.exe'. Unfortunately, it did not fix the bug, which shouldn't be a surprise, as the report whose link you provided does not address any issues related to the conditional operator.

It doesn't? Did you look at the steps to reproduce? The specific error mentioned there is:
copy_t& D = (epsilon>0) ? (A) : (B); // this does not work

where copy_t is an array type (2D I think), which is essentially the problem you were having, wasn't it? It's possible they've fixed a special case of the general error. A simple test case is:

int main()
{
int a[1];
int (&aref)[1] = 1 ? a : a;
}

That should compile I think, but doesn't on my VC2005, no SP.

Tom
.

Relevant Pages

  • Re: Explicitly specializing std::min() on VC++ 2005 Express Edition
    ... "If the second and third operands are lvalues and have the same type, ... it is applying the lvalue-to-rvalue conversion (e.g. the ... I first thought that there was a bug in my code. ... report whose link you provided does not address any issues related to the ...
    (microsoft.public.vc.language)
  • Re: arrays, pointers
    ... According to the standard, they're both lvalues of type int, and the ... references, and the type of aand *here is int, not int&. ...
    (microsoft.public.vc.language)
  • Re: C questions
    ... richard@xxxxxxxxxxxxxxx (Richard Tobin) writes: ... int foo ... lvalues just as array names are. ... There is a real reason: an lvalue designates an object, ...
    (comp.lang.c)
  • Re: address of functions return value
    ... lvalues refer to locations in memory. ... register int r; ... If p eventually holds the address of an int object in memory, then *p will be an lvalue. ...
    (comp.lang.c)
  • Re: address of functions return value
    ... lvalues refer to locations in memory. ... lvalues refer to objects. ... register int r; ... If p eventually holds the address of an int object in memory, ...
    (comp.lang.c)
Loading