Richard Lithvall wrote:

I know this issue has been discussed before but I still have some questions.

Background. I receive some channels from more than one source and one of those sources carry a couple of days of EPG but suffers from low signal quality so I'd rather not use that source when recording/viewing. So I was thinking of exporting the EPG from that source and import it to my other sources EPG-data but found out that the <event id> and <table id> differs between these sources.

What sets the <event id> and <table id> and are those values calculable?

The <event id> is an arbitrary value. The only requirement is that it is unique within the given service.

The <table id> is simply the id of the table in which the event was found. Tables 0x4e and 0x4F are for present/following events, and 0x50..0x6F are for events in the future (the higher the table id, the further in th future).

If these values are not calculable I'd like to know if there is an counter part to <table id>=0, that is - always overwrite this event if conflicting data comes from the DVB stream (conflicting as in overlapping start time and duration).

If you send your "known to be good" data with table id 0x00, it should overwrite any existing events, provided either the event id is the same, or they start at the same time.

Klaus

Richard Lithvall wrote:

Klaus Schmidinger wrote:

...

...

What sets the <event id> and <table id> and are those values calculable?

The <event id> is an arbitrary value. The only requirement is that it is unique within the given service.

Hmm, does this value count when deciding an event's uniqueness?

I'm afraid I don't know what you mean. The event id has to be unique. Two events with the same event id are the _same_ event, no matter what the rest of their data is.

...

The <table id> is simply the id of the table in which the event was found. Tables 0x4e and 0x4F are for present/following events, and 0x50..0x6F are for events in the future (the higher the table id, the further in th future).

...

If these values are not calculable I'd like to know if there is an counter part to <table id>=0, that is - always overwrite this event if conflicting data comes from the DVB stream (conflicting as in overlapping start time and duration).

If you send your "known to be good" data with table id 0x00, it should overwrite any existing events, provided either the event id is the same, or they start at the same time.

OK, so if I send my "not known to be good" data with table id 0x50 and higher (or even 0x6F just to be sure), it would be overwritten by the providers EPG-data when they send their (conflicting) present/following events (0x4E/0x4F)?

Events from the data stream with table ids that are less than an existing event with the same event id or start time will overwrite the existing data.

Klaus

Richard Lithvall wrote:

Klaus Schmidinger wrote: ...

...

...

OK, so if I send my "not known to be good" data with table id 0x50 and higher (or even 0x6F just to be sure), it would be overwritten by the providers EPG-data when they send their (conflicting) present/following events (0x4E/0x4F)?

Events from the data stream with table ids that are less than an existing event with the same event id or start time will overwrite the existing data.

Sounds promising for my needs. To make sure I understand I'll try to summarize;

If I populate a channel with custom EPG-data and setting <table id> to 0x6F and the <event id> to some random made up unique value the EPG-data will always be overwritten by events coming from the providers data stream as long as the <table id> is less than 0x6F and start time are the same no matter what the <event id> are?

That's how it's supposed to be.

Klaus