Re: [boost] Re: Re: Re: [review] Formal review of the SmartContainerlibrarystarts today

From: John Torjo (john.lists_at_[hidden])
Date: 2004-10-06 09:43:32

> |
> | Another problem, IMO, is that client code will need to be more
> | complicated, in order to find out if a value is null or not.
> 
> I don't understand that. One of the objectives of a null object is that you
> don't have to
> worry about a value being null.

really?
I thought the purpose of null was to say "there's no object here. You 
can't do anything with me other than compare other pointers to me."

> | > | And, for each new (virtual) function you add, you should update the
> | > | null_object class - which I think can turn into a maintenance nightmare
> :(
> | >
> | > if your inheritance hierarchy contains N distinct classes, you would need
> to
> | > add N implementations of that virtual function anyway, right?
> |
> | Not necessary... I can add one implementation, and some of those
> | distinct N classes don't have to override it ;)
> 
> you could insert the null object class between the abstract base class and the
> leaf classes.

I've never seen a design like this, would never do something like this 
myself, and I think I would really dislike any design that exibits this.

> 
> | > | >
> | > | > ok, also if it means loss of optimality in for map/set with uinque
> keys?
> | > |
> | > | Would you please exemplify? Thanks.
> | >
> | > for example, consider
> | >
> | > boost::ptr_set<T> set;
> | > boost::ptr_vector<T> vec;
> | > ...
> | > // fill vec with many duplicates as defined by operator< on T
> | >
> | > for( 0..vec.size() - 1 )
> | > {
> | >     // #1
> | >     set.insert( vec[i] );
> | >     // #2
> | >     set.insert( allocate_clone( vec[i] ) );
> | > }
> | >
> | >  #1 can check if vec[i] already exists as defined by operator< on T.
> | >  If it does exists, there is no need to make a new clone. #2 always make a
> new
> | > clone
> | >  and might need to delete it afterwards
> |
> | I might be totally off-base here, but why not an insert_clone function?
> 
> That would be more explicit.
> 
> It might be worth to have the existing interface if users want to try changing
> from vector<string> to ptr_vector<string>
> just to see what happens.
> 

I don't like phrases like "hey, let's change from std::vector to 
std::map, just to see what happens" (I know I exagerated a bit). When 
you change something, you should have a reason, and know what you're 
doing (tradeoffs, etc.).

Best,
John

-- 
John Torjo
-- john_at_[hidden]
Contributing editor, C/C++ Users Journal
-- "Win32 GUI Generics" -- generics & GUI do mix, after all
-- http://www.torjo.com/win32gui/
-- v1.4 - save_dlg - true binding of your data to UI controls!
    + easily add validation rules (win32gui/examples/smart_dlg)