Fons Adriaensen wrote:
> On Mon, Apr 23, 2007 at 04:22:28PM +0200, Pieter
Palmers wrote:
>
>> Or lets fix it at 4.
>
> That should be enough - even to generate lots of
confusion.
The confusion argument is a good one.
>
> Why do we need names of up to 256 chars ? In text mode
nobody
> wants to type them. In an GUI they are a real pain to
display.
> Even with just five useful bits per char, 32 chars
gives you
> more names than there are atoms in the known universe.
We are
> not talking of a file system here that could hold
zillions
> of files - just ports in an audio system.
>
The previous calculation was more to illustrate that even in
worst case,
things are not that bad.
BTW: from jack/port.h:
> #define JACK_PORT_NAME_SIZE 256
> <...>
> #define JACK_CLIENT_NAME_SIZE 33
> <...>
>
> typedef struct _jack_port_shared {
> <...>
> char
name[JACK_CLIENT_NAME_SIZE+JACK_PORT_NAME_SIZE];
> <...>
> } jack_port_shared_t;
Meaning that if we want to ensure complete backwards
compatibility,
we'll have to keep the 256 character port names. Or we get 4
aliases
'for free' if we truncate the port names to 64 characters.
What I would consider more realistic:
* port names of 32 chars
* 2 aliases (= one for the old name, one for the alias)
Giving a lot less overhead.
My feeling is that if you have a setup where you actually
use 1000
ports, this 2 Meg is not going to make the difference. Don't
have any
numbers on that though.
Greets,
Pieter
------------------------------------------------------------
-------------
This SF.net email is sponsored by DB2 Express
Download DB2 Express C - the FREE version of DB2 express and
take
control of your XML. No limits. Just data. Click to get it
now.
http://sourcefor
ge.net/powerbar/db2/
_______________________________________________
Jackit-devel mailing list
Jackit-devel lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/jackit-dev
el
|