50 replies to this topic

[littlesat]

From an EPG source (xml source or provider?)

[WanWizard]

That was the question, yes. EPG readers should convert to UTF-8, if not, we might have to fix something. If XMLTV, then that provider needs to fix something.

 

Which is why I asked:

 

Autotimers are created either manually (you type  in the title and search criteria), or you create it from an EPG entry.

 

Assuming you did the last, I was wondering where the EPG data came from.

 

Did you receive it OTA from the channel provider ( in which case I would like to know provider and channel as we might have an EPG reader issue ) or from EPG Import ( in which case the data in the xmltv import was incorrect ).

[klorinczi]

And the question still remains: where did the ISO8859-15 text in the autotimer come from?

Did you see my screenshots?
I showed how it is possible to set ISO8859-15 encoding in Autotimer.

[littlesat]

What is the source from the epg and as far I understood it should always be utf-8

[klorinczi]

EPG is likely UTF-8, coming from EIT.

I changed encoding in Autotimer plugin to ISO8859-15 accidentally.

Edited by klorinczi, 24 November 2024 - 11:52.

[WanWizard]

The point is:

 

You only need to change that if the timer you created was in ISO8859-15, and the EPG you want to match against is ISO8859-15. In all other cases, changing that setting is pointless.

 

Given the fact you had a crash when it was set to UTF-8, it suggests that something wasn't UTF-8, but ISO8859-15.

 

The question asked several times was "what exactly was not in UTF-8"? So we need to know if your provider is to blame, or we need to fix something.

[Dimitrij]

 

Given the fact you had a crash when it was set to UTF-8, it suggests that something wasn't UTF-8, but ISO8859-15.

No, and again no!

 

 

 

1)See screenshot 1 and 2  https://forums.openp...dpost&p=1652751

 

I changed encoding in Autotimer plugin to ISO8859-15 accidentally.

2)run autotimer and see result in python3

File "/usr/lib/enigma2/python/Plugins/Extensions/AutoTimer/AutoTimer.py", line 322, in parseTimer

builtins.AttributeError: 'str' object has no attribute 'decode' 

'str' object has no attribute 'decode'  --> match.decode('UTF-8')

 

 

old code

		if timer.encoding != 'UTF-8':
			try:
				match = match.decode('UTF-8').encode(timer.encoding)
			except UnicodeDecodeError:
				pass

timer.encoding = "EPG encoding"(        default = timer.encoding
        selection = ['UTF-8', 'ISO8859-15']
        if default not in selection:
            selection.append(default)
        self.encoding = NoSave(ConfigSelection(choices=selection, default=default))

Edited by Dimitrij, 25 November 2024 - 06:25.

[littlesat]

A real solution should do all encodings and not just 2… or even ensure. But how does that source get iso8859-15 in the first place…

[klorinczi]

Given the fact you had a crash when it was set to UTF-8, it suggests that something wasn't UTF-8, but ISO8859-15.

 

You wrote wrongly.

 

The inverse is true:

I had a crash when it was set to ISO8859-15. It suggests that something wasn't ISO8859-15, but UTF-8.

This caused crash in Autotimer plugin.

 

Autotimer should handle if wrong encoding is set, like ISO8859-15 is set in Autotimer, while UTF-8 is the EIT EPG encoding.

[littlesat]

The inverse is true

nd what if you ensure it is always UTF-8 when you compare?

Again.... what was the source/reason/story behind you got ISO8859-15 in the first place?

[WanWizard]

 

Given the fact you had a crash when it was set to UTF-8, it suggests that something wasn't UTF-8, but ISO8859-15.

 

You wrote wrongly.

 

Ok, now I get it. Thanks.