[cfe-commits] r147202 - in /cfe/trunk: include/clang/Basic/DiagnosticGroups.td include/clang/Basic/DiagnosticParseKinds.td include/clang/Parse/Parser.h lib/Parse/ParseStmt.cpp test/Parser/warn-dangling-else.cpp

[Chad Rosier]

No problem.  I missed the initial discussion, do I was just looking for clarification.

Thanks David, Chandler, Nico, etc.



On Jan 16, 2012, at 7:26 PM, Nico Weber <thakis at chromium.org> wrote:

> Hi Chad,
> 
> On Mon, Jan 16, 2012 at 7:03 PM, Chad Rosier <mcrosier at apple.com> wrote:
>> Hi Nico,
>> I get a warning with the following test case.
>> 
>> int foo(int a, int b, int c) {
>>    int rval = 0;
>>    if (a != b)
>>        if (b == c) {
>>            rval = 0;
>>        } else {
>>            rval = 1;
>>        }
>>    return rval;
>> }
>> 
>> mcrosier$ clang -O3 t.c -Wall -c -o /dev/null
>> t.c:6:11: warning: add explicit braces to avoid dangling else [-Wdangling-else]
>>        } else {
>>          ^
>> 1 warning generated.
>> 
>> Is this expected behavior?  This code seems rather explicit to me.
> 
> This was discussed when I proposed the warning:
> http://lists.cs.uiuc.edu/pipermail/cfe-commits/Week-of-Mon-20111219/050474.html
> 
> Richard and Eli both preferred the warning to fire in this case,
> Chandler was against it.
> 
> I'm happy with adding any tweak we agree on (say "don't warn if the
> else starts on the same line the previous if starts" or something like
> that).
> 
> How often do you run into this in practice?
> 
> Nico