[CEDET-devel] Re[4]: [Ecb-list] sematic problem with templates

[Eric M. L. <er...@si...>]

Perhaps you fixed the C++ parser last time this template problem was
reported but forgot to check it in?

I can hope anyway. ;)
Eric

>>> "Berndl, Klaus" <kla...@sd...> seems to think that:
>Hi,
>
>but what i can really not understand why i get only the T*& 
>underlined in the code below ... all other template-parts are
>parsed and not red underlined - how could this be that the same parser
>results in different parsing-outputs????
>
>Eric, do you have an explanation for this?
>
>Thanks,
>Klaus
>
>I'm using
>Emacs  : GNU Emacs 21.3.1 (i386-msvc-nt5.1.2600)
> of 2003-03-28 on buffy
>Package: ECB: 2.23, semantic: 2.0beta1, eieio: 0.18beta1,
>         speedbar: 0.15beta1, JDEE: No JDEE
>
>Eric M. Ludlam wrote:
>> Hi,
>> 
>>   I do get some underlining under just the template part.  The C++
>> parser only handles the most rudimentary of template specifications.
>> 
>>   I would like someday to port the old GCC bison grammar and use that,
>> but haven't gotten around to finishing it.  Also I think there are
>> probably better ones around that may serve better as the one I have
>> is kind of old.
>> 
>> Eric
>> 
>>>>> "Berndl, Klaus" <kla...@sd...> seems to think that: Markus
>>>>> Gritsch wrote: Berndl, Klaus wrote: 
>>>>> 
>>>>> But in your code there is NO unmatched syntax - at least not with
>>>>> my semantic 2.0 (installed via cedet1.0betac).
>>>> 
>>>> Strange.  I also use cedet-1.0beta1c.tar.gz and there *are* red
>>>> underlines, see the attached screenshot.  However, if it is fixed in
>>>> the current CVS everything is fine.
>>> 
>>> Indeed strange... but the current CVS-ECB has nothing fixed concering
>>> parsing C++ - it has just fixed an annoying behavior of ECB...
>>> 
>>> But i do not really understand this... i'm using latest RELEASED
>>> cedet 
>>> 1.0 beta1c and my semantic/ECB parses this code without any problems.
>>> 
>>> But maybe Eric or David can give us a wise hint?!
>>> 
>>> For all who have not followed this thread, here is the code again:
>>> 
>>> Eric, how is this c++-code parsed with your semantic? Which parts of
>>> this code are red-underlined??
>>> 
>>> #include  <boost/smart_ptr.hpp>
>>> 
>>> namespace ico {
>>>   namespace doc {
>>> 
>>>      std::list<boost::shared_ptr<Document> >
>>> DocumentFactory::m_documentFactoryList; 
>>> 
>>>      void DocumentFactory::initFactories()
>>>      {
>>>         // body
>>>      }
>>> 
>>>      boost::shared_ptr<Document>
>>>         DocumentFactory::createFromFile(const char * filename) throw
>>>         (CouldNotOpenFileException, IllegalFormatException)      {
>>>      // body }
>>> 
>>>      bool DocumentFactory::isSupported(boost::shared_array<unsigned
>>>         char> p_buffer, unsigned int p_size)      { // body
>>>      }
>>> 
>>>      boost::shared_ptr<Document>
>>>        
>>>      DocumentFactory::createFromBuffer(boost::shared_array<unsigned
>>>         char> buffer, unsigned int size) throw
>>>      (IllegalFormatException) { // body }
>>> 
>>>      T*& elem(int i) {return static_cast<T*&>(Base::elem(i));}   }
>>> }
>>> 
>>> here is a screenshot which parts my semantic/cedet-installation can
>>> not 
>>> parse and therefore are red underlined:
>>> 
>>> 
>>> Ciao,
>>> klaus
>
>