Re: Reserve_Capacity for Unbounded_String?

[Adam Beneschan <adam@irvine.com>]

On Jul 22, 12:54 pm, Maciej Sobczak <see.my.homep...@gmail.com> wrote:
> Why there is no Reserve_Capacity for Unbounded_String?
> Sounds like a natural idea to me.
>
> How to emulate it?
>
> The following seems to work fine:
>
> declare
>    S : Unbounded_String;
>    My_Capacity : constant := 1000;
> begin
>    S := To_Unbounded_String (My_Capacity);
>    Delete (Source => S, From => 1, Through => My_Capacity);
>    -- ...
> end;
>
> "Seems to work fine" means that after the above two operations the
> string is logically empty, but the subsequent appends run faster
> (which is actually the original motivation).
>
> Is this a reasonable approach if performance is of concern?

I sure wouldn't count on this.  I don't know what compiler you're
using or how it implements Unbounded_String; but just based on what I
know about how this package *might* be implemented and about the
suggested implementation in the Rationale, the performance
improvements you're seeing could be just an accident.  (One test: try
doing something like "Intp := new Integer;" after each append.  If the
Delete actually deallocates memory, and a piece of that memory could
be reused by a "new Integer" allocation, you may get very different
performance results than if the deallocated memory is never reused
except by new appends.  Here, though, I'm just guessing, since as I
said I have no idea how Unbounded_Strings is implemented.)

I'm with Jeff here: you may need an Unbounded_Strings package tailored
to your specific needs.  It just isn't possible to write a package
that will achieve the best performance for all programs that might
want to use it; even a simple idea like "Reserve_Capacity" could
introduce some implementation overhead that might be unwelcome in
programs that wouldn't benefit from that feature.

                           -- Adam