[vdr] Re: migration to 1.2.2

[Klaus Schmidinger <Klaus.Schmidinger@cadsoft.de>]

Rainer Zocholl wrote:
> 
> Klaus.Schmidinger@cadsoft.de(Klaus Schmidinger)  09.08.03 22:44
> 
> Once upon a time Klaus Schmidinger shaped the electrons to say...
> 
> >I believe I wrote a lot about all this in the file UPDATE-1.2.0:
> 
> at lot of stuff to read!

Well, if nobody is going to read this, maybe I can save the time
and effort and in the future just not write any more documentation ;-)

> >  In order for the "unique channel ID" to work, all channel
> >definitions now must  be unique with respect to the combination of
> >their Source, Frequency and SID   parameters.
> 
> There is no "name" mentioned...

Of course there isn't! Names are not unique.

> And:
> What's the problem if not?

Timers identify channels by their unique channel id.
If these ids are not unique, a timer wouldn't know
which channel to use - it can't know whether this was
done on purpose or just by accident.

> The i have one source with two names.
> But it's stil the same channel an satelite
> 
> What does it matter if i have
> a
> Neun Live:12480:v:S19.2E:27500:767:768:35:0:897:0:0:0
> and a
> tm3:12480:v:S19.2E:27500:767:768:35:0:897:0:0:0
> in my channels conf?

Simple: it doesn't work! Channel data has to be unique.
You can use the RID field to make them unique.

> (maybe led me to the thread were this was discussed)
> 
> >You may have to fix your
> >'channels.conf' manually if there are error   messages
> 
> what "error message" where would make seraching easier.
> I would expect that errror message on the OSD.
> But there is only "invald channel" regard less which
> key i press.
> 
> You know that the second and thir line belongs together.
> First I thought that this were 2 different errors:
> > vdr[3348]: loading /video/channels.conf
> > vdr[3348]: ERROR: channel data not unique!
> > vdr[3348]: ERROR: error in /video/channels.conf, line 133
> 
> maybe something like:
> > vdr[3348]: loading /video/channels.conf
> > vdr[3348]: ERROR(32): in /video/channels.conf, line 133 (32:channel data notunique!)
> 
> is more "human".

Just because it is in one line instead of two?
Gimme a break...!

> And:
> If VDR would have not loaded at all (because it actualy can't work
> in this state) it would be more likely that the first view is going
> into the logfile.
> (I looked first im /var/log/messages: there was no error
> listed!)

If there was no error listed, then everything must have been fine!
Are you sure it didn't say

vdr[3348]: loading /video/channels.conf
vdr[3348]: ERROR: channel data not unique!
vdr[3348]: ERROR: error in /video/channels.conf, line 133

in there?

> vdr[3348]: ERROR(99): vdr dies because of bad or missing channels.conf.

Ok, we can let VDR die if it fails to read that file.

> >in the log file when loading it. BE SURE TO MAKE A BACKUP COPY OF YOUR
> >'channels.conf' AND 'timers.conf' FILE BEFORE SWITCHING TO THIS
> >VERSION, AND CHECK   VERY CAREFULLY WHETHER YOUR TIMERS ARE STILL SET
> >TO THE RIGHT CHANNELS!
> 
> >So this shouldn't have come as such a big surprise... ;-)
> 
> For you who wrote it. yes. ;-)
> Try to go in the place of some one who only sees the "results"
> not the way to get them.

Documentation is there for the user to _read_.
After all, this is source code that you have to compile
yourself. You are actually _building_ your own settop box!
I'm trying to do my best to document everything that needs
to be taken into account when switching to a new version
of VDR, but I can't help you if you don't read that.

Klaus


-- 
Info:
To unsubscribe send a mail to ecartis@linuxtv.org with "unsubscribe vdr" as subject.