CXX11 patch to warn if 'override' is missing on overriding virtual function

[Douglas Gregor]

LGTM

> On Sep 26, 2014, at 4:10 PM, jahanian <fjahanian at apple.com> wrote:
> 
> 
> On Sep 26, 2014, at 3:37 PM, Douglas Gregor <dgregor at apple.com <mailto:dgregor at apple.com>> wrote:
> 
>> 
>>> On Sep 26, 2014, at 3:03 PM, jahanian <fjahanian at apple.com <mailto:fjahanian at apple.com>> wrote:
>>> 
>>> 
>>> On Sep 25, 2014, at 11:51 AM, Argyrios Kyrtzidis <kyrtzidis at apple.com <mailto:kyrtzidis at apple.com>> wrote:
>>> 
>>>> 
>>>>> On Sep 25, 2014, at 11:24 AM, Douglas Gregor <dgregor at apple.com <mailto:dgregor at apple.com>> wrote:
>>>>> 
>>>>>  I’d feel a lot better if some part of the warning could be on by default. For example, if you’ve uttered “override” at least once in a class, it makes sense to warn-by-default about any other overrides in that class that weren’t marked as “override”, because you’re being locally inconsistent. Or maybe you can expand that heuristic out to a file-level granularity (which matches better for the null point constant warning) and still be on-by-default.
>>>> 
>>>> This seems like a great idea to me!
>>>> For the 'override' I much prefer if it is class specific to make it less of a burden as an “always on” warning. We could have the checking done at the end of the class definition.
>>>> 
>>> 
>>> Here is the patch. Warning is on by default. Number of new warnings on clang tests is greatly reduced but there are still some.
>> 
>> +def warn_function_marked_not_override_overriding : Warning <
>> +  "%0 is not marked 'override' but overrides a member functions">,
>> +  InGroup<CXX11WarnOverrideMethod>;
>> 
>> “a member functions” shouldn’t be plural. Also, perhaps we should turn this around:
>> 
>> 	“%0 overrides a member function but is not marked ‘override’”
>> 
>> because that puts the context of the problem before the problem.
>> 
>> +  if (HasMethodWithOverrideControl) {
>> +    // At list one method has the 'override' control declared.
>> +    // Diagnose all other overridden methods which do not have 'override' specified on them.
>> +    for (auto *M : Record->methods())
>> 
>> “At list” -> “At least”.
>> 
>> Also, this means we’ll be taking two passes over the methods if any “override” is present, even though we won’t often warn here. How about extending this:
>> 
>> +      if (M->hasAttr<OverrideAttr>())
>> +        HasMethodWithOverrideControl = true;
>> 
>> with
>> 
>> 	else if (M->begin_overridden_methods() != M->end_overridden_methods())
>> 	  HasOverridingMethodWithoutOverrideControl = true;
>> 
>> and we only do this second pass when we know we’re going to warn, e.g., if HasMethodWithOverrideControl && HasOverridingMethodWithoutOverrideControl?
> 
> Thanks for quick review. Here is the updated patch.
> 	
> <override-patch.txt>
> 
> - Fariborz
>> 
>> 	- Doug

-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.llvm.org/pipermail/cfe-commits/attachments/20140926/3abe4584/attachment.html>