platform independent kdeinit/klauncher (was KDE4's IPC)

[Thiago Macieira]

nf2 wrote:
>I think the design of kio-slaves is problematic anyway, because
>IO-slaves are single-threaded and can't handle multiple clients at once.

This is not a problem. That's how they were designed.

Besides, they *are* multi-threaded. At least the name resolution (DNS) 
happens in a multi-threaded manner.

>Therefore real session management is not possible (i guess that's why
>KIO can not write to zip or tar archives for instance).

That's not the reason.

And I don't understand what you mean by session management. It is working 
perfectly here, so you must be referring to something else as a problem.

>To achieve real session management and caching, io-slaves should be
>multi-threaded:

Nonsense. That can be accomplished quite well with multi-process 
single-threaded IOSlaves. The only drawback is that we assume that 
processes are cheap, which is not valid on Win32.

>1) Either a single daemon for all protocols,

a.k.a. "klauncher". The threads are called "kioslaves" and the simple fact 
is that they aren't threads, but full processes. There's no difference.

>2) One daemon per protocol or
>3) One daemon per remote user at server.
>
>IMHO such shortcomings should be corrected by writing new network
>transparency library (D-VFS, Common-VFS), perhaps before moving to
> KDE4...

I'll eagerly expect your code and patches to all ioslaves currently in 
existence in KDE.
-- 
  Thiago Macieira  -  thiago (AT) macieira.info - thiago (AT) kde.org
    PGP/GPG: 0x6EF45358; fingerprint:
    E067 918B B660 DBD1 105C  966C 33F5 F005 6EF4 5358

3. Ac seo woruld wearð geborod, swá se Scieppend cwæð "Gewurde Unix" and 
wundor fremede and him "Unix" genemned, þæt is se rihtendgesamnung.
-------------- next part --------------
A non-text attachment was scrubbed...
Name: not available
Type: application/pgp-signature
Size: 189 bytes
Desc: not available
URL: <http://mail.kde.org/pipermail/kde-core-devel/attachments/20051228/a1058c22/attachment.sig>