[Top][All Lists]

[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

Re: [Gnash-dev] Enhancement suggestion for GNASH

From: Dimitri Turbiner
Subject: Re: [Gnash-dev] Enhancement suggestion for GNASH
Date: Wed, 31 Oct 2007 03:59:01 -0400
User-agent: Icedove (X11/20071020)

Several solutions, from nice and simple, to hairy...

1) use xwd to grab the contents of the window in which flash is being rendered or of an Xvfb server in which you launched the player (this should make it easier if you want custom resolutions, etc). If your Xserver doesn't redraw a window if there is some other window on top of it, then be sure to enable backing store. pipe the output to xwdtopnm

then pipe that into mplayer or mencoder specifying the right raw image format (widht, height, format, size of the pnm header you should skip, etc). You can grab the audio at the same time specifying the right options.

This approach should work for the macromedia flash plugin running in a firefox window as well.

2) Edit the sourcefile of the fb renderer code in gnash to output to a different fb file and then cat that file, pipe it through sed to get rid of whatever you don't want and then pipe into mencoder.

3) Edit that same fb code to dump to stdout the content of the frame it just received from the rendering backend (agg, I believe). Again, pipe to mplayer.

4) If you're really desperate, just send me the flash files and I'll do it for you in a couple of minutes since I've already got everything setup.

Good luck!


strk wrote:
On Wed, Oct 31, 2007 at 09:46:49AM +0800, dengdd wrote:
Hi, all

We can get the memory buffers of frame when the Gnash renderering  and
convert it to other video-file formats.
At present there are some problems when we do it.

1:some flash movies need interactived. difference interactives show
difference frames.
2:video-file we converted without optimization, so it's size maybe big.

3:How to handle the sound ? the same way as frame buffers?   when need
sound,  get the sound is memory buffers?

One option is to implement a special renderer and sound_handler and possibily
even a specialized GUI for this kind of things.

Relying on buffers would mean duplicating the code for each 
or binding to a specific one, at which point you can just clone and modify to 
rather then output.


Gnash-dev mailing list

reply via email to

[Prev in Thread] Current Thread [Next in Thread]