Klaus Schmidinger wrote: > Johannes Stezenbach wrote: > > > > BTW, I rewrote the TP list handling in scan.c, so TPs now are > > identified by frequency, not by some ids (which should be unique > > according to DVB standards, but aren't always...). > > This needs testing. > > Hmmm, VDR is going to switch from using the actual transponder frequencies > to using the transponder IDs in the next version for generating unique channel > IDs. If what you are saying is true, this wouldn't make much sense... > Or am I misunderstanding something here?
For IDs to be unique it takes some coordination work. Since broadcasters and sat/cable operators don't care enough, the IDs are sometimes not unique. Most STBs on the market don't care about those IDs, so there is no pressure on the broadcasters to cleanup their numbering. The only thing which is reliable is that the service_id is unique within *one* TP (the standard says "unique within one network"...). The frequency is also not absolutely reliable, because different NITs will list the same TP with slightly different frequencies (and possibly different IDs). is_same_transponder() in scan.c attempts to work around that. Sidenote: MHP requires unique IDs because they use dvb-locators (dvb://onid.tsid.service_id) for zapping... Johannes -- Info: To unsubscribe send a mail to [EMAIL PROTECTED] with "unsubscribe linux-dvb" as subject.
