Re: [boost] RFC: Boost.XML API prototype in the sandbox

From: Stefan Seefeld (seefeld_at_[hidden])
Date: 2007-07-13 14:21:10

Phil Endecott wrote:

> Stefan Seefeld wrote:
>> Phil Endecott wrote:
>>> For example, I 
>>> might need to efficiently find all the <foo> nodes in a large tree.  
>>> How would I do that with your library?
>> You'd do an XPath query.
> 
> I'm pretty sure that libxml2 would visit every node in the tree to do 
> that.  With XSLT you can use the "key" mechanism to make this more 
> efficient, but XPath alone doesn't get you that.

I don't understand what you are saying. What does XPath not get me ?
I clearly can issue an xpath query, resulting in a nodeset filled with
all 'foo' elements of the document.

>>> I think that there are a lot of useful techniques that can't be done 
>>> using a libxml2 backend, including the symbol-table idea that I have 
>>> mentioned before and the whole business of pointer semantics i.e. deep 
>>> vs. shallow copy, copy-on-write and so on.
>> Right. But I don't see that as a limitation of my approach. Rather, I'd
>> use a different way to achieve that anyway.
> 
> Can you elaborate?

My point is that I wouldn't attach user-data to nodes by derivation, but
by some other, non-intrusive, means. Composition is often a much better
device than inheritance.

>> May I suggest that you search the boost ML archives to find more details.
> 
> E.g. these
> 
> http://thread.gmane.org/gmane.comp.lib.boost.devel/85985
> http://thread.gmane.org/gmane.comp.lib.boost.devel/86508
> http://thread.gmane.org/gmane.comp.lib.boost.devel/86872

Yep.

Regards,
                Stefan

-- 
      ...ich hab' noch einen Koffer in Berlin...