Re: Ann: Natools.Chunked_Strings, beta 1

[Natasha Kerensikova <lithiumcat@gmail.com>]

Hello,

On 2011-11-30, Yannick Duchêne <yannick_duchene@yahoo.fr> wrote:
> Le Tue, 29 Nov 2011 16:16:16 +0100, Natasha Kerensikova  
><lithiumcat@gmail.com> a écrit:
>> Chunked_String is a string container that exposes the same interface as
>> Unbounded_String
> I don't see the same interface as that of Ada.Strings.Unbounded in  
> Natools.Chunked_Strings.

Really? I did took all the subprogram names, with the same argument
names and order, and the same behavior. I only changed the type from
Unbounded_String to Chunked_String (obvisouly), and when a new
Chunked_String is created, I added Allocation_Unit and Chunk_Size with
default values, so that old calls to Ada.Strings.Unbounded subprogram
still work. Do you see something else?

>> http://fossil.instinctive.eu/natools/dir?ci=tip
> Dirty question: is your Natools.Chunked_Strings interface to be with-used  
> or withed ?

Same as Ada.Strings.Unbounded, really. I tend to with-rename, so I have
never tried how it feels when with-used. However I don't see anything
preventing it from being with-used.

> If you heard about static polymorphism, and you feel that's a good idea,  
> you may face some issues with the actual spec.

What kind of issues? Or rather, what issues that are not triggered
exactly the same way by Ada.Strings.Unbounded? I genuinely don't see
anything wrong...


Thanks for your help,
Natasha