Agreed.
Beside, a mkdir -p in the launch script of the PND will be way faster than having pnd_run.sh crawl the PND directory structure and create the missing directories on each PND start.

Cheers,
sebt3


2012/8/29 skeezix <skeezix@skeleton.org>
On Wed, 29 Aug 2012, Neelix wrote:

# Just a thought,  but would there be any practicality to adjusting the part of libpnd that creates the appdata directory if it doesn't exist to also scan the PND and manually recreate its directory structure in  the new appdata
# directory?    This may alleviate the issue of having to have the run scripts manually create any necessary subdirectories.
# Thoughts?

        Thought about that in the distant past, and the idea was shot down
mostly because.. you may have some wild-ass directory structure in the
PND, but only read it; in general, the changes you intend to make usually
aren't really widespread, so why create an enormous directory structure on
SD, when chances are you only need to make one or two.

        And it didn't seem like there would be any way to predict what
would be changed.. ie: can only leave it for the subject matter expert,
the author of the pnd.

        The way aufs works is pretty annoying there (Self defeating?), but
not sure theres a great automatic workaround. (Maybe aufs has had
improvements over time, maybe theres new options that make it work better?
Someone should check that :P)

                jeff

--
If everyone would put barbecue sauce on their food, there would be no war.
_______________________________________________
Firmware-dev mailing list
Firmware-dev@openpandora.org
http://openpandora.org/cgi-bin/mailman/listinfo/firmware-dev