To minimalize complexity, it would be nice to run a 2+ computer setup [1]
without having to bother with network filesystems like NFS, SMB or SSHFS.
According to my understanding of the locstor.accessRawAudioData XMLRPC
method url return value [2], I did some research (with an interactive
Python shell) and indeed was able to play the audioclip directly via this
URL instead of accessing it via the filesystem. I therefore supposed that
it is a client-side (scheduler and studio) constraint or design decision
and so grepped for "file:/" in the sources and found f.e. these files:
Now the questions:
-Does the "file://" prefix stored in the database and x-mitted as filename
in the return value of locstor.accessRawAudioDataOpen provoke the access
to the filesystem?
-What changes would be needed to make scheduler and studio using HTTP
instead of accessing the filesystem?
-What changes would be needed to extend scheduler and studio with some sort
of "knob" to switch between HTTP access ("streaming mode") and filesystem
use ("station mode")?
I hope not to strain your attention and eagerly await your response,
Greetings,
Fabiano