[PATCH] fix mapping of chassitype to system.formfactor

[Danny Kukawka]

On Wednesday 10 May 2006 20:04, David Zeuthen wrote:
[...]
> No, I don't think you need that. Btw, property_atomic_update_*() only
> ensures that updating multiple properties are sent out as a single dbus
> PropertyChanged changed. It doesn't really "lock" anything, it just
> queues up changes for the dbus interface. Which is kinda useless for
> coldplug as we haven't even claimed the org.freedesktop.Hal service
> before all coldplug is done (to prevent sending out lots of signals when
> adding devices in coldplug).

Okay, good to know, I miss-interpreted the function. 

[...]
> I think the problem you were seeing was related to the missing "== 0" in
> the strcmp.

I attached a new version of the patch. This should fix it, without the 
property_atomic_update_*() calls and should also complete skip the hole 
mapping if the value of system.formfactor was already set in a other place 
(this is only a little better performance, but it's better ;-) ).

Btw. The patch look bigger than it really is ...

Danny

 ChangeLog            |   11 +++
 hald/linux2/osspec.c |  123 ++++++++++++++++++++++---------------------
 2 files changed, 75 insertions(+), 59 deletions(-)
-------------- next part --------------
A non-text attachment was scrubbed...
Name: hal-fix-map_system.formfactor_again_v4.diff
Type: text/x-diff
Size: 7387 bytes
Desc: not available
Url : http://lists.freedesktop.org/archives/hal/attachments/20060510/3e8c2dbc/hal-fix-map_system.formfactor_again_v4.bin