Re: Ambiguous call to overloaded function

On 12/01/2017 06:47 PM, Larry Evans wrote:
> On 11/24/2017 01:01 PM, Michael Powell wrote:
>> On Fri, Nov 24, 2017 at 1:03 PM, Michael Powell 
>> <[email protected]> wrote:
>>> I'm not positive, this may be a similar situation as with:
>>>
>>> http://boost.2283326.n4.nabble.com/Spirit-X3-employee-cpp-compile-error-td4687704.html 
>>>
>>>
>>> Related to VS2015 Update 3 support.
>>>
>>> With solutions that seem to revolve around necessity of 
>>> BOOST_SPIRIT_DEFINE (?).
>>
>> And if I just take the error on its face, just what is the difference
>> between Attribute& and ActualAttribute&, other than in name only?
>>
>>> Can someone clarify?
>>>
>>> Thank you.
>>>
>>> Cheers,
>>>
>>> Michael Powell
>>>
> Hi Michael,
> 
> I think ( I'm not real sure) that  Attribute is the Attribute as
> declared in the rule, but the ActualAttribute is the actual type
> of the attribute passed to the parse function. AFAICT, the
> attribute transform done in the call_rule_definition here:
> 
> https://github.com/boostorg/spirit/blob/develop/include/boost/spirit/home/x3/nonterminal/detail/rule.hpp#L297 
> 
> 
> converts the ActualAttribute to the Attribute required by the rule.
> 
> HOWEVER, that conclusion is just based on my many hours examining the
> code

Actually, just looking at the rule<...>::parse function code:

https://github.com/boostorg/spirit/blob/develop/include/boost/spirit/home/x3/nonterminal/rule.hpp#L57

and then the function called from there:

https://github.com/boostorg/spirit/blob/develop/include/boost/spirit/home/x3/nonterminal/detail/rule.hpp#L297

confirms the conclusion that Attribute is the rule attribute whereas
ActualAttribute is what is actually passed to the parse function for
rule_definition.

If you're wondering how might they become different, then you might
have to look at how sequence attributes are handled.  In particular,
the parse_sequence function in:

https://github.com/boostorg/spirit/blob/develop/include/boost/spirit/home/x3/operator/detail/sequence.hpp

IIUC, for a sequence, the attribute is converted to a fusion iterator
and then increments of that are passed to the various parts of the
sequence, and I think the ActualAttribute might be one of these
fusion iterators actually pointing to the rule attribute.

At least that's my *partial* understanding of how the ActualAtrribute
can be different than Attribute.

HTH.

-regards,
Larry



------------------------------------------------------------------------------
Check out the vibrant tech community on one of the world's most
engaging tech sites, Slashdot.org! http://sdm.link/slashdot
_______________________________________________
Spirit-general mailing list
[email protected]
https://lists.sourceforge.net/lists/listinfo/spirit-general