Priority of EPG data - wrong?

[leecovuk]

Hello,

My comment here should be and needs to be confirmed by others.

In the latest and updated ViX image (and possibly previous ViX images), I think there is an issue regarding the priority of data in the EPG.

If EPG data has been obtained 'in advance' by importing it in CrossEPG or RyVix, it seems that data is not updated by 'live' data.
The point being, some existing data is quite likely to be (and often is) incorrect and needs to be over-written by any available live data derived in real time from the satellite.

I have no idea if the EPG settings in 'Customise' relate to this issue. Their purpose in relation to EPG plugins has always been and remains unclear to me. This is because there seems to be a lack of documentation and knowledge on the Enigma2 EPG, with any supplier's images.

If I use the PLi Beta images rather than ViX images, my perception is the priority and updating of the EPG data is correct.

Thank you,
Lee
UK

 

[andyblac]

Hello,

My comment here should be and needs to be confirmed by others.

In the latest and updated ViX image (and possibly previous ViX images), I think there is an issue regarding the priority of data in the EPG.

If EPG data has been obtained 'in advance' by importing it in CrossEPG or RyVix, it seems that data is not updated by 'live' data.
The point being, some existing data is quite likely to be (and often is) incorrect and needs to be over-written by any available live data derived in real time from the satellite.

I have no idea if the EPG settings in 'Customise' relate to this issue. Their purpose in relation to EPG plugins has always been and remains unclear to me. This is because there seems to be a lack of documentation and knowledge on the Enigma2 EPG, with any supplier's images.

If I use the PLi Beta images rather than ViX images, my perception is the priority and updating of the EPG data is correct.

Thank you,
Lee
UK

lol, well since vix 2.0 is Pli based i dont you what you are try to say ?

 

[leecovuk]

lol, well since vix 2.0 is Pli based i dont you what you are try to say ?

Presumably the way ViX has implemented or adjusted things isn't the same.
Maybe test what I find and see if I'm right. If not, there must be something unique to my set up and how it runs the ViX image.

 

[GiP]

lol, well since vix 2.0 is Pli based i dont you what you are try to say ?

I am not sure, but maybe it's the same thing I wrote about some time ago. He says "it seems that data is not updated by 'live' data", What I reported is that EPG can get messy when there is a change in the time of a program as new data is just added to old data by CrossEPG and, for example, you see in EPG Prog1 starting at 9:00 and ending at 9:30 and also Prog1 starting at 9:10 and ending at 9:40 this cause some problems with the visualization by VixEPG. This is a problem with CrossEPG and I reported it on the developers site (no answer yet...). What I do now is when this get to bothersome I clera all EPG data with a script and start over....
By the way, is there a simple way to add to some Vix menu this script so I don't have to ssh to the box?
GiP

 

[andyblac]

By the way, is there a simple way to add to some Vix menu this script so I don't have to ssh to the box?
GiP

you are wanting a menu option run a custom scripts. yeah i add the to the to do list.

 

[leecovuk]

Hello again,

There really does seem to be something wrong with how the EPG allows or doesn't allow live data to overwrite it. The problem in ViX seems to be similar to how the XML Import plugin works, which apparently is that EIT now / next doesn't overwrite existing EPG data. 'Doglover' said that on the PLi forums, regarding how the XML Import plugin works. (for some reason)

However I am not using the XML Import plugin (Ryvix) in ViX; I am using CrossEPG as pre-installed within the ViX image.
When I install and use the CrossEPG plugin in the PLi Betas, 'live' data usually does successfully overwrite existing EPG data to correct and update the current broadcast.

The E2 EPG system, with plugins or not, is obviously far from perfect. However I would imagine it should be key that any available live data can overwrite existing data. Otherwise, the EPG is often wrong and differs from now / next data which is shown correctly in the infobar.
I'm surprised that the beta testing apparently didn't notice this.

I already have all the EPGs turned on in 'Customise', which presumably should cause all available live data to be read.

Can anything be done, please, to correct the above?

Many thanks,
Lee
UK

 

[GiP]

This is not a VIx or PLI problem but a CrossEPG one.

I put a request (it's #73) for an easy way of cleaning the EPG data and start over on the CrossEPG site

http://code.google.com/p/crossepg/

but so far there was no interest in it. I wrote a script that does that but I have to ssh to the box and run it which is annoying.
Maybe you can post a request for an options to clear old data when new one is acquired (I will support this as it will solve the other problem too). If enough people show they are interested in solving the problem maybe they'll address it.

G

 

[yngvekl]

Yes, me wonder too... I'm using Ryvix epg, only Thor sattelite (0,8W). I update ryvix perhaps every other/third day. How can I get the live data coming from the sattelite to
overwrite what I've downloaded using Ryvix epg? Or is it doing so already? I think it's very often that what's being sent don't matches what's in the epg.

Sorry for bad english, I'm really, really tired!

/Yngve

 

[leecovuk]

This is not a VIx or PLI problem but a CrossEPG one.

However, as I explained, I get different results using the EPG and CrossEPG in the PLi betas. Not perfect, but somewhat better and different.

I can only guess at the causes of any problems because I don't know how the Enigma2 EPG or any plugins work. Is Enigma2 or its EPG system documented anywhere on the internet in English? I haven't found anything. I assume the developers such as Andy in ViX have to guess themselves.

As has been the case for a long time, we really need to understand how the EPG options work in Enigma2, and how plugins relate to that.
Deleting or clearing the EPG time and time again is of course not fixing the causes of any data issues.

Is it possible to 'look at' epg.dat and see what's in there? I don't mean 'browse' the EPG, but look at epg.dat and analyse its structure and contents. I haven't found anything to do that.

Lee
UK

 

[leecovuk]

I think it's very often that what's being sent don't matches what's in the epg.

Yes, this is the end result. It's obviously nice and helpful that people find and offer EPG data to import, but if live data cannot completely correct it then it is often pointless.

 

[GiP]

Deleting or clearing the EPG time and time again is of course not fixing the causes of any data issues.

Yes, definitely, the program that collects EPG should be smarter and not just add new data to the old even when there are contraddictions. But this is a complex job; an option to clear all data or to just delete old data in a channel when new infos are acquired would solve the problems while a more general solution is studied, and very easy to implement (at least the first one...).

G

G

 

[Rob van der Does]

But don't forget: "Live-EPG" data is not always more accurate then imported data. No script can be intelligent enough to decide which data is better.

 

[GiP]

But don't forget: "Live-EPG" data is not always more accurate then imported data. No script can be intelligent enough to decide which data is better.

True, but one OR the other is better then both, as the result is a mess. In my case I don't use any imported data, only sat-EPG data so more recent should always be better, and if imported data is compiled from sat-EPG it should be the same. As I said it's a complex job....

G

 

[yngvekl]

Can we get sat epg data from 0.8W (Thor) for more than now/next... like 5 days or something?

 

[GiP]

Can we get sat epg data from 0.8W (Thor) for more than now/next... like 5 days or something?

EPG does not depend from Sat but from provider/service, it depends on what channel and provider you want. But I don't have Thor so I'll let someone else tell you if what you need is there or not.....

G

 

[leecovuk]

an option to clear all data or to just delete old data in a channel when new infos are acquired would solve the problems while a more general solution is studied

Yes, true enough. Would that however be any different to us deleting epg.dat and restarting Engima2 then importing afresh?

I really don't know what to do regarding the EPG. I'm trying the Blackhole image at the moment which their details say has the best EPG system, however that too seems to be rather clunky with some data issues.
All the images seem to have EPG issues.

Does anybody know of an English speaker who develops Enigma2? Maybe he/she has the answers.

Ideally I don't want to import any data but let the receiver pick it up by itself. I suppose we all do.
Which are the channels which can be tuned in so that the receiver can then load full EPG listings? I assume there aren't many, hence the need to import from sources such as Rytec.
If there are any, how long do we need to stay tuned to that channel and how often?

Thanks,
Lee
UK

 

[leecovuk]

But don't forget: "Live-EPG" data is not always more accurate then imported data. No script can be intelligent enough to decide which data is better.

Sure, a script may struggle to differentiate, but I would say live data is always more accurate. How can it not be?
We surely need to allow all live data to dominate and overwrite any previous data.
If that has to screw up timers or planned recordings, so be it. Any recordings may have been ultimately fictitious anyway.

 

[Rob van der Does]

Sure, a script may struggle to differentiate, but I would say live data is always more accurate. How can it not be?

Unfortunately I must contradict that. I wish it were true.
The Dutch/Belgian providers hardly ever, if ever, update their EPG-data. But there are enough sources available (and in use by Rytec) that do. So again: if it is difficult, or even impossible for a human to decide which data is correct, a script will never be able to do so.

 

[leecovuk]

Unfortunately I must contradict that. I wish it were true.
The Dutch/Belgian providers hardly ever, if ever, update their EPG-data.

ok, I didn't realise that. I'm quite surprised at that. Presumably subscribers to Dutch/Belgian cable or satellite services get a different and correct EPG?

Regarding my comments about Blackhole; it appears that live data is actually overwriting previously imported data. I'll probably stick with Blackhole for a while and see if their EPG system really is the best. I'm using what I think is their latest image, version 1.4.5

If anybody wants to start a topic on this site or elsewhere, comparing their EPG experiences on different images, feel free to PM me the url of that topic. I'd be interested to read and contribute.

Lee
UK

 

[Rob van der Does]

ok, I didn't realise that. I'm quite surprised at that. Presumably subscribers to Dutch/Belgian cable or satellite services get a different and correct EPG?

Indeed: Dutch DVB-C and DVB-T subscribers usually get a good EPG (for 14 days). I am not sure about the Belgian situation.
Dutch/Belgian Satellite providers update the EPG only once a week (which is an improvement: until recently it was once every fortnight.......).