* James Turner -- Tuesday 18 April 2006 17:22:
[runlevel groups]
Melchior is probably the best person to comment on how appropriate or
not these are for the Nasal scripts he's written.
I'm not aware of any Nasal module that really depends on a subsystem,
although I haven't looked with this in
Melchior FRANZ wrote:
* James Turner -- Tuesday 18 April 2006 17:22:
[runlevel groups]
Melchior is probably the best person to comment on how appropriate or
not these are for the Nasal scripts he's written.
I'm not aware of any Nasal module that really depends on a subsystem,
On 18 Apr 2006, at 10:07, Melchior FRANZ wrote:As you know, the reason for Nasal being the last of the subsystems is that it processes the files in $FG_ROOT/Nasal/ on initialization. And the idea was that these scripts should find all subsystems initialized, so that they can access some generated
* Melchior FRANZ -- Tuesday 18 April 2006 11:07:
Then Nasal files that *really* want to wait for a subsystem, could say
on_subsystem_init(environment, print_temperature);
Umm ... but what if it wants to wait for more than one subsystem. While
possible, that would quickly become disgusting.
On 18 Apr 2006, at 10:31, Melchior FRANZ wrote:Umm ... but what if it wants to wait for more than one subsystem. While possible, that would quickly become disgusting. Better real Nasal dependency resolution then. :-| I think adding more groups and keying off them is much easier - I don't have any
On 18 Apr 2006, at 11:36, Melchior FRANZ wrote:I think adding more groups and keying off them is much easier Sure. This wasn't meant as argument against your sublevel idea. (In my code I had added Nasal to the INIT group weeks ago, because this could have solved an exit problem that I had at
* Melchior FRANZ -- Tuesday 18 April 2006 12:49:
* James Turner -- Tuesday 18 April 2006 12:44:
I moved the 'end' part of NasalSys::init (running the
scripts) to postinit() to try and keep things working; I'm sure
that's not really the correct solution but it got things limping along.
* Melchior FRANZ -- Tuesday 18 April 2006 12:53:
* Melchior FRANZ -- Tuesday 18 April 2006 12:49:
* James Turner -- Tuesday 18 April 2006 12:44:
I moved the 'end' part of NasalSys::init (running the
scripts) to postinit()
It probably breaks joystick drivers [...]
Umm, no. The end
On 18 Apr 2006, at 11:53, Melchior FRANZ wrote:Umm, no. The "end part"?! This should be OK then. But it gets more and more fragile because of such manually resolved dependencies. Agreed - hence this discussion. Otherwise we'll end up with postpostpostinit(), or as I like to call it, 'run level 4'
9 matches
Mail list logo
