10 replies to this topic

[EnoSat]

I saw some interesting additions in SignalFinder.

Would it be possible to add them to the SDG?

 

 

Read TID,NID,POS

 screenshot_20220709194420.jpg   80.73KB   7 downloads

 

Show SDT

 screenshot_20220709194438.jpg   56.44KB   7 downloads

[Huevos]

That's my mod. OpenPLi were never interested in the implementation.

[WanWizard]

Never seen it before myself, can't imagine what the objections would have been?

[Huevos]

1) Littlesat didn't like that it is not built in to the C++ code.

2) Littlesat didn't like that it polls.

3) Littlesat didn't like that it uses AutoBouquetsMaker's dvbreader.so file.

4) Maybe other things, I can't remember.

 

The class just drops in without touching the base SatFinder class.

 

https://github.com/O...in.py#L594-L926

 

And if the reader is not available it just drops back to the base class.

		if dvbreader_available:
			session.openWithCallback(boundFunction(SatfinderCallback, close), SatfinderExtra)
		else:
			session.openWithCallback(boundFunction(SatfinderCallback, close), Satfinder)

Oh, and obviously needs the widgets in the skin.

 

The original version used something similar to AutoDiseqc.py, but was just too slow and clunky with continual spinners every time it changed frequency and inaccurate results.

 

Was just a personal project to fill my needs.

[WanWizard]

I think 1 is related to 3, and I guess he had a point there, you don't want Enigma to have a dependency on ABM, 99% of OpenPLi users don't use that.

[littlesat]

Indeed the feature is Nice… but the implementation is not that optimal. It was long time ago I did evaluate it. All code that can check a transponder is already available in enigma2… when eg you scan.

There are more implementation I did comented as I world prefer to have it implemented in a better structured method.

Autobouquets maker lakes from to sides. The way it scanners and the way it fills the bouquets…. But the devs made it with what they could do with their knowledge and I do not have the time to show how it can be Made better to demonstratie it.

It should be startend to find a way to use the Scanning code we have in enigma2 and/or add/adapt the ‘api’ instead of using an additional binary… so I prefer To have the functionality that we have in the so file moven to the standaard enigma2 binary code.

Edited by littlesat, 15 July 2022 - 21:01.

[Huevos]

@Littlesat, The ONID/TSID code is available in AutoDiseqc.py but it just does not work properly for repeated retunes.

 

I'm not aware of any code in the binary that reads the orbital position from the transport stream.

 

All this mod does is use the *.so file, nothing to do with ABM bouquets. Just reads from the transport stream.

 

Converting that code to the enigma2 binary? What is the point? OpenPLi uses other *.so files for things not built in to the binary, e.g. bitrate. I think that needing everything built in to the binary is a philosophy that is not based on good logic. In the case of Fastscan and CableScan it is obvious that it is clearly wrong.

 

Anyway, the reader is built in to OpenViX. For OpenPLi, not so simple.

Edited by Huevos, 16 July 2022 - 14:11.

[littlesat]

Bitrate is a plugin… but unless this I also think this can be integrated. And you point indeed to the autodiseqc part… why not adapt it?
The code was made in a way the developers of it could do at that moment, but it was never implemented in a structured/proper or better say optimal way. I tried to get it to a better direction but never succeed with it…

[WanWizard]

The problem is not the binary in itself, the problem is that enigma will need a depdency on ABM to get access to that binary, which means ABM will have to be present in every OpenPLi image. Which we don't want.

[Huevos]

That file is so useful I made it a standalone package a few years ago.

https://github.com/Huevos/dvbreader/

[littlesat]

Just to give you the real pain I have with this.

 

At the moment abm was created I was not in it because I prefered to get fastscan extended and preferably also cablescan integrated so there is one kind of scan tool.

 

aActually Cablescan and fastscan demonstrates how you should do it, including filling the userbouquets and lamedb without manipulating the files and reload them with extra parsers etc.

but the code needs to made for a big part in cpp… and the developers of abm did made it in python because as far I understood could not create it easily in cpp… except for the small cpp code that was still mandatory…

 

As long nobody has time or energy to make it properly it will never be implemented better. So we have to live with it as it is.

 

there is no need to defend the so method is good or wrong. It is more about time and energy someone needs and can spend into it to do it really right.

 

for me it has low priority on ‘my personal scrum board’… at this moment I do not have the time and energy to do it due to circumstances.

Edited by littlesat, 17 July 2022 - 08:33.