[cfe-dev] CXXRecordDecl Name

[Nikola Smiljanic]

I thought no error was reported? I'm not surprised that getName returns an
empty string for invalid code :)


On Wed, Jul 2, 2014 at 8:21 AM, Reid Kleckner <rnk at google.com> wrote:

> Why?  The AST is invalid, and Clang diagnoses it.
>
> $ cat t.cpp
> namespace N { };
> struct N::A { };
>
> $ clang -c t.cpp
> t.cpp:2:11: error: no struct named 'A' in namespace 'N'
> struct N::A { };
>
>
>
> On Tue, Jul 1, 2014 at 3:13 PM, Nikola Smiljanic <popizdeh at gmail.com>
> wrote:
>
>> Could you file a bug report for this Daniel?
>>
>>
>> On Wed, Jul 2, 2014 at 3:02 AM, Daniel Dilts <diltsman at gmail.com> wrote:
>>
>>> I found my answer in 9p11 in the standard.  Seems that without the
>>> forward declaration it is invalid.
>>>
>>> If a class-head-name contains a nested-name-specifier, the class-specifier
>>> shall refer to a class that was previously declared directly in the class
>>> or namespace to which the nested-name-specifier refers, or in an element of
>>> the inline namespace set of that namespace (i.e., not merely inherited or
>>> introduced by a using-declaration), and the class-specifier shall appear in
>>> a namespace enclosing the previous declaration.
>>>
>>>
>>> On Tue, Jul 1, 2014 at 9:21 AM, Daniel Dilts <diltsman at gmail.com> wrote:
>>>
>>>> If I have this code then everything works.  If I get rid of the forward
>>>> declaration, then any method returns an empty string.  Is the forward
>>>> declaration mandatory when defining the class like this?
>>>>
>>>> namespace N {
>>>>   class BaseClass1;
>>>> }
>>>> class N::BaseClass1 {
>>>> public:
>>>>   BaseClass1(int i);
>>>>   BaseClass1(const BaseClass1 &right);
>>>>   virtual ~BaseClass1();
>>>>   BaseClass1 &operator=(const BaseClass1 &right);
>>>> }
>>>>
>>>>
>>>> On Mon, Jun 30, 2014 at 3:35 PM, Reid Kleckner <rnk at google.com> wrote:
>>>>
>>>>> Try getDeclName(), getNameAsString(), or others.  getName() only works
>>>>> if the name is a simple identifier:
>>>>>
>>>>>   /// getName - Get the name of identifier for this declaration as a
>>>>> StringRef.
>>>>>   /// This requires that the declaration have a name and that it be a
>>>>> simple
>>>>>   /// identifier.
>>>>>   StringRef getName() const {
>>>>>     assert(Name.isIdentifier() && "Name is not a simple identifier");
>>>>>     return getIdentifier() ? getIdentifier()->getName() : "";
>>>>>   }
>>>>>
>>>>>
>>>>>
>>>>> On Mon, Jun 30, 2014 at 3:21 PM, Daniel Dilts <diltsman at gmail.com>
>>>>> wrote:
>>>>>
>>>>>> I am attempting to get the name of a CXXRecordDecl that I have
>>>>>> identified with an AST matcher.  I am using Class->getName().str() to get
>>>>>> the class name.
>>>>>>
>>>>>> namespace N {
>>>>>> class SomeNonInterface {
>>>>>>   virtual void func1() = 0;
>>>>>>   virtual void func2() {}
>>>>>> };
>>>>>> }
>>>>>>
>>>>>> It works great for the above code.  Gives me "SomeNonInterface" as
>>>>>> the name.
>>>>>>
>>>>>> class N::BaseClass1 {
>>>>>> public:
>>>>>>   BaseClass1(int i);
>>>>>>   BaseClass1(const BaseClass1 &right);
>>>>>>   virtual ~BaseClass1();
>>>>>>   BaseClass1 &operator=(const BaseClass1 &right);
>>>>>> };
>>>>>>
>>>>>> For the above code it gives me "" as the name.
>>>>>>
>>>>>> How would I get the class name so that it works in all cases, even if
>>>>>> the name in the declaration/definition is namespace qualified?  Or...did I
>>>>>> hit a case that isn't legal?
>>>>>>
>>>>>> _______________________________________________
>>>>>> cfe-dev mailing list
>>>>>> cfe-dev at cs.uiuc.edu
>>>>>> http://lists.cs.uiuc.edu/mailman/listinfo/cfe-dev
>>>>>>
>>>>>>
>>>>>
>>>>
>>>
>>> _______________________________________________
>>> cfe-dev mailing list
>>> cfe-dev at cs.uiuc.edu
>>> http://lists.cs.uiuc.edu/mailman/listinfo/cfe-dev
>>>
>>>
>>
>
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.llvm.org/pipermail/cfe-dev/attachments/20140702/c13771a1/attachment.html>