one file per hour

[mark@antistatic.org]

I recently stumbled upon Listener while looking for a way to record 
scanner audio on my Linux box. Currently I am using XCorder Lite on 
Windows.

Listener is nice, but I am wondering if anyone know of a way to use 
only one file per hour (XCorder has this option and it has been very 
useful to me).

From what I understand, the libsndfile API does not have the option of 
"appending" WAV data to an existing file, which is why listener makes 
lots of little files.

I have thought up a number of possibilities:

1) the cheap workaround: a utility that reads and combines all *.wav 
files per hour into one file. this can be done possibly with the 
"external script" option. anyone got one of these?

2) the impractical solution: wait for libsndfile to allow appending

3) the dangerous solution: re-open the file for that hour, pipe it into 
a new temp file, add the current sound data, close the temp file, 
rename the temp file for that hour. this could be dangerous because it 
uses a lot of cpu/memory.

4) the practical, but time consuming, solution: re-write parts of 
listener so that only one file per hour is open, then when there is 
data, write to it.
when the hour changes, close the file and open a new one.
(so basically, when there is silence, we dont close the file. when 
there is no silence, we dont open a new file, we just write.)

problems: if listener is killed we need to make sure to close the file 
properly, otherwise it may not be playable or we may lose the data 
altogether.

memory usage - in theory, we could use the full 60 minutes of audio.

I know #4 is probably not going to be done by the original writer 
because it changes a lot of stuff already worked into the program, so I 
may fiddle with it myself.

Anyone got any other ideas?

-Mark