Re: [Boost-users] [serialization] no RTTI data error

Subject: Re: [Boost-users] [serialization] no RTTI data error
From: Noah Roberts (roberts.noah_at_[hidden])
Date: 2010-06-22 16:14:40

In article <MPG.268a99cfa5838af4989687_at_[hidden]>, 
roberts.noah_at_[hidden] says...
> 
> I'm hoping someone can help me narrow down what I might be looking for.  
> Already spent a lot of time trying to get somewhere and still trying to 
> step through but using VS2010 and it's slower than ANYTHING ELSE THAT 
> EXISTS!!
> 
> The problem is that when I load my file I get an, "Access violation - no 
> RTTI data," error.  Yes, RTTI is turned on and the application actually 
> relies upon it very heavily throughout its operation (acyclic visitor is 
> a major component).  The problem seems to have just cropped up and is 
> within code that has previously passed tests.  
> 
> Best I can do at this point is describe the problem and hope someone has 
> an idea of what to look for because I've been inch by inch stepping 
> through serialization template code with a debugger that takes 30 
> seconds each damn step.  I won't be done for hours. (anyone know how to 
> speed this bitch up?)
> 
> The problem seems to involve a type that I'm using 
> save/load_construct_data on.  There is a collection of these objects 
> that is stored and loaded previously in the file and then I try to load 
> a vector of objects that have shared pointers to these previously loaded 
> objects on an individual basis.
> 
> It was working fine and then I attached more data after the vector.  Now 
> it throws this exception when I try to load the pointer; before it 
> reaches the new code that is otherwise entirely unrelated (it's just an 
> object that happens to be stored in the same parent).
> 
> If and when I get more information I'll post but for now I'm hoping 
> someone can save me some time and frustration.  This debugger is the 
> slowest damn thing I've ever seen in my life.

I "fixed" this problem (and it was the use of typeid deep inside of the 
shared_helper reset() function) by changing the order that I saved 
things in a containing object.  The two types are completely unrelated 
except that they are both contained by the same thing.  The only thing I 
can think of is that the object I added after these broken ones used 
BOOST_CLASS_EXPORT to register and the broken objects are registered at 
the beginning with ar.template register<type>().  Is there some weird 
way that these two methods can poorly interact or something?

-- 
http://crazyeddiecpp.blogspot.com/