47 replies to this topic

[Lost in Space]

Done

[cdr]

hi gutemine

did test your last update openepg on DMM CVS don't work saying download fail  

thaks for your effort 

[Lost in Space]

For support of the plugin you have to ask in OoZooN or DMM board as this is here the Rytec section. But download failed typically happens when you satellites.xml has 28.2 and you load with 28.4 profile or vice versa. Simply try the other one and it will work.

And the ipk should work also with PLi images.

Edited by gutemine, 9 May 2016 - 11:15.

[Rob van der Does]

Better delete 28.4 completely as that is not an official position (28.2 is).

[doglover]

The same applies for the use of:

 

0.9 W and 1 W instead of the official 0.8 W for Thor

and

4.9 E and 5.0 E instead of the official 4.8 E for Viasat

 

Using these wrong values in the satellites.xml file creates a lot of problems.

 

Willy

Edited by doglover, 9 May 2016 - 12:46.

[Lost in Space]

Probably this mess is already the root cause why all other EPG tools failed.

[Abu Baniaz]

Who decides what is official?

The data on the stream for Thor says it is 1w, not 0.8w.

BTW,
Thanks for your updated program Gutemine, unfortunately it did not work for me using 28.2.

Sent from my HTC One M9PLUS using Forum Fiend v1.3.3.

[Lost in Space]

I tested the ipk on my dm7080 with OpenATV Image and it works nicely. Check if the settings match what the satellites.xml file says. Because the worst case is xml with one version, channels from settinngs with the other one and then in Tuner setup one with a 50% chance to be wrong.

 

Then it helps ONLY to get a proper satellites.xml with 282, do a factory reset to get rid of all channels and setup tuners correctly and scan with this setting,

 

And as I said in the actual Plugin threads - if you post only "doesn't work" and ask for fixes  your will not get any support. We need clean settings and enigma2 logs in case it then fails.

Edited by gutemine, 9 May 2016 - 19:29.

[Abu Baniaz]

I flashed OpenATV 5.3 20160509 ( Technomate 3T Combo)

I completely removed the channel data, including the one that they force on people

I trimmed Satellites.xml so that it only contains 28.2 entry. Attached

Manually scanned 28.2 instead of using ABM or AB 28.2. Lamedb attached

I installed "enigma2-plugin-extensions-openepg_1.4_r3_mips32el.ipk", (I had version r1 before)

 

It succeeded on one attempt, failed on another (line 507).

 

 

Note:

As the plugin depends on CrossEPG downloader, it is worth pointing out that OpenPLI removed CrossEPG downloader by this commit, because apparently, non-mips builds fail as a result.

https://github.com/E...4f0b3898537b837

 

The OE-A version builds for sh4, arm and obviously mips. Perhaps what is useful in the OE-A build process can be utilised?

Attached Files

•  for pli forum.zip   36.6KB   5 downloads

Edited by Abu Baniaz, 9 May 2016 - 20:53.

[display name]

http://www.ses.com/f...ch_by=satellite

 

http://www.telenorsa...atellite-fleet/

 

 

 

[display name]

V1.4_r3 also OK

[Rob van der Does]

The data on the stream for Thor says it is 1w, not 0.8w.

Then that data is wrong, simple as that.

[Abu Baniaz]

Then that data is wrong, simple as that.

Thanks for clarifying. Looks like the SES and Telenor websites are wrong too then.

Edited by Abu Baniaz, 9 May 2016 - 22:25.

[Lost in Space]

There was a good reason to remove CrossEPG.

 

Even I decided to build the binaries for the OpenEPG Plugin from the public sources and as all the recent changes seem to be in other areas they worked nicely.

 

The only binary without sources is the ones for MovieStar, but they promised to make then public when finished.

 

Anyway, now that somebody confirmed that it even works for UK EPG when you have a clean setup and don't mix settings and satellites.xml without knowing what you are doing, maybe more people will continue testing.

 

As I said already , OpenEPG should now run nicely also on OpenPLI Images and even DMM OE 2.0 Images. If you put in sh4 versions of the few binaries it will probably run on this plattform.

 

BUT my key development plattform are Dreamboxes running DreamOS, so don't expect too much.

 

I simply didn't like the argument that it is not possible to make it work also on other plattforms. Check the epgdat.py - it is less than 1 page of code to support also the old way of loading EPG and only a handfull lines of code to even get the epg.dat way of loading that OE 2.0 needs.

 

So the whole argument falls apart and definitely NOT on my side of the discussion.

 

But this is OT here, if you want to contribute to OpenEPG testing or development there are other places ...

 

gutemine

Edited by gutemine, 9 May 2016 - 22:42.

[Lost in Space]

Just regarding your log - you are using the load epg.dat setting to yes, this is mainly for OE 2.0 Dreambox images, as this is the only way they support to get external EPG. OpenATV and OpenPLi based images includes the Oudeis Patch which allows to load EPG event data directly into running enigma2, so this settings is not needed and should not be used.

 

I added this settings mainly to be able to test and debug this variant of EPG loading too, but normally it is set to false and should stay that way. This is probably the reason why your load works sometime and sometimes not.

 

PS: On OE 2.0 Images or Images without Oudeis patch this settings is always yes and not even shown, because epg.dat loading is there the only choice. And off course on DreamOS this setting is not available either, as it is useless when you can load into the epg.db database.

Edited by gutemine, 9 May 2016 - 23:21.

[Abu Baniaz]

With all due respect, failure issue has nothing to do with people who don't know what they are doing messing around with satellites.xml. In essence, all I did was remove the other satellites.

<  7348.814> [OPENEPG] found no free tuner
<  7348.815> [OPENEPG] download failed

On this occasion (this current log is from openvix), this was on a setup with one fixed 28.2 dish and a separate motorised dish.

Attached Files

•  Enigma2-2016-05-09_23-16-56.log   62.35KB   0 downloads

Edited by Abu Baniaz, 9 May 2016 - 23:30.

[Abu Baniaz]

I disabled the load by /media/hdd/epg.dat option and still got a failure. line 830

I hope the log helps improve things.

 Enigma2-2016-05-09_23-33-02.log   57.72KB   0 downloads

[Rob van der Does]

 

Then that data is wrong, simple as that.

Thanks for clarifying. Looks like the SES and Telenor websites are wrong too then.

 

Yes, they probably are. See (a.o.) http://www.itu.int/o...ub0=Select&ie=y

and more in general http://www.itu.int/ITU-R/go/space/en

 

Not sure why this is done, but whatever those values are won't change the actual allocated orbital positions

[Lost in Space]

The CrossEPG binaries aren NOT very good in handling complex SAT setups, therefore I added the zapping to the servicereference of the Transponder which is briadcasting the EPG Data in the Plugin.

 

BUT this means it will only work for simple DiseqC 1.0 A/B(and maybe C/D) switching like I have also in my own configuration, If you have a complex setup like a rotor for the dish you will need to swtch to another channel on the needed SAT before doing the EPG download in Plugin, so that it only needs to change tuning to the relevant Transponder (a Radio one in case of Sky UK), but it will not do any complex DiseqC 1,1 switching or Rotor movements. Especialyl if you have different setups for multiple tuners this is a source of suffering (but the behaviour is the same for CrossEPG). At then moment also demux1 on non-DreamOS is hardcoded, which means the FIRST tuner of the box should be able to receive the SAT position you need fpr loading the EPG or it wil lfail because it cann't tune and hence can't download.

 

I hope this explaination helps.

 

This is what testing is all about ...

Edited by gutemine, 10 May 2016 - 20:20.

[Abu Baniaz]

Thank you for the explanation.

The behaviour is not the same in the CrossEPG downloader (I am speaking with regards to the OE-A version). It handles complex setups very well. Feel free to adopt what you find useful in the OE-A version.