Hi There,
As part of a research project I have been developing a FUSE filesystem that exposes incomplete files as complete ones. The motivation behind it is to point any download program to save files through it, say a movie, and then other programs, say a movie player, could open the file and start playing it even if it has not been fully downloaded. In order for this to work we need to communicate with the download client to tell it which parts of the file are being requested by the player. As a prove of concept we are using bittorrent, and came across your project, Deluge.
What I'm trying to figure out is the most natural place for the Deluge to communicate with our FS (through a shared file or other mechanism) and find out what bytes are being requested, and for the appropriate torrent, set the chunk priority to the highest, so that the chunck will be downloaded and playback (read in general) can continue. As I said, this is for a proof of concept, so a quick dirty hack would be sufficient. Is there any part in the Deluge source that this would be easiest to integrate? It has to be a continuous thing, that checks the file often.
Does anyone have a good idea of the deluge architecture and where this fix might fit?
Thanks a lot
hacking a deluge plugin -- need help
hacking a deluge plugin -- need help
Last edited by andres on Wed May 13, 2009 10:39 pm, edited 1 time in total.
Re: hacking deluge
After a few hours of investigation I tried writing an implementation of what I wanted using a plugin. I followed the instructions here
http://dev.deluge-torrent.org/wiki/Deve ... i/Examples and also http://dev.deluge-torrent.org/wiki/Development/Plugins
I thought the core.py in my plugin was not being called, because the logs would not appear in the log file I specified in the command line. core.py:enable() was actually being called, only the logging is messed up. Does anyone know where this thing is logging to? I tried changing it, but only get silent failure. I know it is being called because I can print to stdout and I can create a directory from that function. I would love to know where those logs and the rest of the logs of deluge.core and basically everything that is not GTK UI are going to -- it's not the same file as we specify in the command line, that's for sure. Any help would be greatly appreciated!
I'm running it from the command line in the following way:
deluge -L debug -l /tmp/deluge.log -u gtk
~/.config/deluge/deluged.log is empty.
http://dev.deluge-torrent.org/wiki/Deve ... i/Examples and also http://dev.deluge-torrent.org/wiki/Development/Plugins
I thought the core.py in my plugin was not being called, because the logs would not appear in the log file I specified in the command line. core.py:enable() was actually being called, only the logging is messed up. Does anyone know where this thing is logging to? I tried changing it, but only get silent failure. I know it is being called because I can print to stdout and I can create a directory from that function. I would love to know where those logs and the rest of the logs of deluge.core and basically everything that is not GTK UI are going to -- it's not the same file as we specify in the command line, that's for sure. Any help would be greatly appreciated!
I'm running it from the command line in the following way:
deluge -L debug -l /tmp/deluge.log -u gtk
~/.config/deluge/deluged.log is empty.
Re: hacking a deluge plugin -- need help
If you leave off -l it just logs to the console.