[prev in list] [next in list] [prev in thread] [next in thread] 

List:       kde-devel
Subject:    Re: Size of 2.0 compared to 1.1.1 ?
From:       Matthias Ettrich <ettrich () troll ! no>
Date:       1999-11-02 20:32:44
[Download RAW message or body]


> It currently uses more, but not much. things like Unicode support make
> it more memory hungry and we can't avoid that (unless someone invents a
> technique that can store 16 bit values in 8 bit ;) ).

Using the 16Bit QString throughout KDE actually may even save some memory.
In KDE-1.x we often used char* and copied the same string all over the
place. A string that was loaded from a file, stored in some internal data
structure and displayed for example on a label or a listbox existed several
times in memory. Now, with QString being usable, there's just one single
copy of it, thoughout the whole application.

We could have done that with a good implicit-shared 8Bit stringclass as well,
but noone did (char* was just tooo practical for those of us with C
background;-)

Matthias

[prev in list] [next in list] [prev in thread] [next in thread]