This is the input processing, which starts at rnews (aka cunbatch).  It
invokes newsspool, which actually puts the stuff where it belongs (peeling
off the silly and unnecessary "#! cunbatch" header, if any, as it goes).

Newsrun should be run regularly; it unspools the stuff and feeds it into
the relay subsystem for processing.  Newsrunning can be used to turn this
function on and off if you want to avoid news processing during busy hours.
There is a configuration option to arrange for newsrun to be run after
each newsspool -- this is expensive but cuts down latency.

"make r" builds everything and runs a full regression test on most everything.
 NameMtimeSizeType
0 Makefile 1994-01-28 00:35:28Z 4.5 KiB application/octet-stream
1 README 1990-09-11 18:59:59Z 654 B text/plain; charset=utf-8
2 bdecode 1994-01-28 01:04:35Z 44 KiB application/octet-stream
3 bdecode.c 1991-11-05 15:18:23Z 2.4 KiB text/x-csrc
4 c7decode 1994-01-28 01:04:35Z 35 KiB application/octet-stream
5 c7decode.c 1989-02-22 19:57:18Z 3.2 KiB text/x-csrc
6 cunbatch 1994-01-28 01:00:48Z 1.5 KiB application/octet-stream
7 newsrun 1994-01-28 00:35:29Z 3.6 KiB application/octet-stream
8 newsrunning 1994-01-28 00:35:29Z 516 B application/octet-stream
9 newsspool 1994-01-28 01:04:34Z 51 KiB application/octet-stream
10 newsspool.c 1991-11-05 15:13:34Z 5.9 KiB text/x-csrc
11 recenews 1994-01-28 00:35:30Z 245 B application/octet-stream
12 recpnews 1994-01-28 00:35:31Z 285 B application/octet-stream
13 rnews 1994-01-28 00:35:32Z 1.5 KiB application/octet-stream