Hello Carson,
classical fragmentation as described in RFC5050 is not implemented yet.
But the issue you described sounds like a job for the dtnstream tool. To use dtnstream you need a sender and a receiver. The sender writes the data to sent into the standard output pipe and this is piped into the dtnstream application. For a DTN webradio this looks like this.
$ wget -O - http://webradio.url/audio.mp3 | dtnstream -s radio-station -d dtn://receiver-node/webradio
Additionally, you could specify the size of each chunk "-c" and the lifetime of the bundles "-l". The parameter "-s" defines the application part of the source EID of the sender or the destination EID of the receiver.
On the receiver side, you just need to pipe the received stream into a player application.
$ dtnstream -s webradio | vlc -
The parameter "-t" sets a timeout to skip lost bundles after a cuple of seconds and "-w" tells the application to wait until the first chunk was received. By default the app takes the first received chunk as starting point.
Kind regards, Johannes
Am 23.01.2012 17:13, schrieb Carson Dunbar:
Hello,
I am currently working on my own type of "fragmentation" for IBR-DTN.Did you implement any fragmentation code and if so, where was that code located? I am trying to split a bundle's stream into smaller packets and I'm not sure how to do it with any other file type than a plain text files. Is it possible to adjust the stream so that they are reading with ios::binary enabled? Thank you.
Carson Dunbar