Thorsten Behrens wrote:
Christian Lippka <[EMAIL PROTECTED]> writes:
I still thing that having a fresh start and using more scripted stuff
and not hard coding each effect is the way to go. But I'm easy persuaded
with good arguments here :)
I'm not - going that route leads to fragility and constant fixing
efforts. IMO, the only sensible way is to implement the basic SMIL
animation facilities using action script - which of course can also be
implemented iteratively. Having hard-coded effects poses the problem
of detecting them in your SMIL input, which is on the verge of
implementing AI ;-)
There are 2 sides to this: 1) what is the "most optimal" solution, and
2) what is the "most realistic" solution? The most optimal solution
involves creating an entire SMIL processing engine in ActionScript.
This is not even a C++ task, but an ActionScript task. If somebody
wanted to do this, it doesn't even really have anything to do with OOo.
You can do enough XML processing in ActionScript that you could just go
straight to the source. At this point in time, there is no point in
starting a new project like this without using ActionScript 3.0 (and
therefore targeting Flash 9). Despite the fact that this doesn't have
anything to do with C++ or OOo, a reasonable approach could be to "port"
OOo's C++ implementation to AS 3.0. Then "exporting" to Flash from OOo
would merely involve converting the static elements with appropriate
sprite structures for the SMIL engine to work with. Bundle in a
pre-compiled SMIL engine and the SMIL stream, and voila. For fun, add
in a high-level API to the SWF to allow seeking and animation triggering
(and perhaps others) so the outputs can be integrated into other systems
(like the one we use here).
I estimate that one of the SMIL experts on your team, if he or she also
knew a little ActionScript, could probably whip up a demo in 4-8 weeks.
Somebody who doesn't know much about SMIL or the OOo presentation engine
already, you would have to add a few weeks (at least) to learn about
that. Somebody who doesn't already know ActionScript to some basic
level will probably struggle--the free tools don't include a debugger
and the non-free tools are expensive (30 day trials are available
though). So, obviously, the perfect person to do this task is......
me. Sigh. Realistically, I don't think this is going to happen unless
a commercial entity decides they want to hire somebody for a few months
to make it happen.
So we come around to the "realistic" solution. The realistic solution
is to expand upon the work already done and push it as far as it goes.
My rough guess is that we can get to about 90% compatibility without
substantial structural changes. This is more than "good enough" for a
lot of people. Don't even go after the SMIL directly. Just use the
same code that the GUI uses to present stuff to the user. The upshot is
that this is a very incremental process and requires little to no
understanding of SMIL. If somebody can point us to the code that the
GUI uses so we can copy and paste (I don't think there's a high level
API for this), then 2 weeks of coding will produce impressive results.
After that, it will slowly accumulate corner cases over many months.
In my opinion, until somebody steps up and announces a willingness to
back the full-blown solution, there's not much point in worrying about
it. Do the cheap solution and let's move on with our lives to other
more important things.
Augustus
---------------------------------------------------------------------
To unsubscribe, e-mail: [EMAIL PROTECTED]
For additional commands, e-mail: [EMAIL PROTECTED]
