Adam Sutton [Sat, 11 Jan 2014 21:03:01 +0000 (21:03 +0000)]
linuxdvb: some simplications to EN50494 support
I've removed some of the bounds checking, not a bad thing to have it, but
in general we work on the principle that if the user really is trying hard
enough to pass broken values, then so be it!
Also correct some spelling mistakes and change UI refs to "Unicable" rather
than the standard "EN50494".
Adam Sutton [Sat, 11 Jan 2014 00:21:08 +0000 (00:21 +0000)]
linuxdvb: minor tweak to adapter UUID generation
This overcomes issues with multi-frontend tuners that change the
tuner name based on the selected frontend. We simply force the selection
of the first available frontend, which should keep things constant.
build: Add missing include for Dl_info on FreeBSD which is only used
when execinfo is enabled.
src/trap.c:144: error: 'Dl_info' undeclared (first use in this function)
src/trap.c:144: error: expected ';' before 'dli'
cc1: warnings being treated as errors
src/trap.c:187: warning: implicit declaration of function 'dladdr'
src/trap.c:187: error: 'dli' undeclared (first use in this function)
freebsd: Assume all devices on FreeBSD are USB 2.0 per default because
FreeBSD does not have sysfs to read the device speed and type properly.
The good thing is that there are no PCI devices that support V4L right
now so all supported devices are from webcamd which only supports USB
devices so that default is always correct.
Adam Sutton [Thu, 2 Jan 2014 21:29:14 +0000 (21:29 +0000)]
mpegts: fix broken input handling for inputs with >1 active services..
This affects IPTV where there is only 1 input, if 2 services have the same PIDs
then both services were receiving each others packets and causing a mess!
I think later on I should really remove the transport list from the input and
have it on the mux? I think I did it this way to keep some things simpler, but
it's caused confusion here.
Adam Sutton [Sat, 21 Dec 2013 22:06:15 +0000 (22:06 +0000)]
mpegts: update to the way init scan table completion is handled
Should be a bit better at handling tables that may or may not exist.
We have an initial timeout after which we check if data for any tables
has been received but incomplete.
This is still far from ideal, really what is needed is to make this
all more async based on when we sync etc... but this will do for now.
Adam Sutton [Wed, 18 Dec 2013 18:53:02 +0000 (18:53 +0000)]
capmt: attempt to avoid confusion over validity of configuration
UI now has 3 possible colours, red - not working, orange - not connected
(but will probably work, used for new non-persistent mode 2) and green -
actively connected.
Adam Sutton [Wed, 4 Dec 2013 23:28:29 +0000 (23:28 +0000)]
tvhlog: stop full debug by default if --trace specified
Often I want to enable trace (and debug) for a specific set of subsys,
but I had it configured to enable ALL debug if any debug related options
were spec'd (for compat). But this was a PITA and I kept doing
--trace blah --debug none just to get around it!
Adam Sutton [Wed, 4 Dec 2013 22:38:44 +0000 (22:38 +0000)]
subscriptions: reverse EPG and initscan priorities
For networks with extended EIT, having EPG take precdence causes chaos
during init scan. The reality is that forcing an EPG scan to stop because
someone wants to scan a new mux is probably the right thing to do anyway.
Adam Sutton [Wed, 4 Dec 2013 22:17:22 +0000 (22:17 +0000)]
tsfile: cmdline flag to specify ATSC input
Previously all standard DVB and ATSC tables were installed, this
would cause the initscan to time out. We try and be more clever
with determining the type, but this will do for now.
The code allows the option to be per mux (file), but the cmdline
is currently limited to a global setting, which is likely to be
more than enough.
The commit fixes simultaneous channel decryption.
We are enumerating services based on the generated seq value.
In fact it should be elementary stream PID, but currently
we can leave it until PMT code rework.
The CA_SET_PID and CA_SET_DESCR from OSCam was always set with
seq=0 leading to setting all keys always to first subscribed
service.
Adam Sutton [Mon, 2 Dec 2013 13:30:59 +0000 (13:30 +0000)]
imagecache: migrate configuration to new API
I started to do this as there was a possible issue with imagecache as a result
of updates. However this has thus far not been proved, but still a useful
update. Hopefully not too broken!
capmt: add simplified OSCam dvbapi protocol as new capmt mode
The new mode uses one domain socket connection for all subscribed
services. The CA_SET_PID and CA_SET_DESCR are received over same
socket connection instead of one UDP socket per adapter.
The minimum OSCam version for use with this mode is svn rev 9063.