Autotimer not suporting ISO 8859-2 encoding?

[Ezee]

I was for first time trying the autotimer this being new to me since flashing the latest Vix image.
Working like a dream until I tried it on items of the EPG that are displayed with accents from the ISO 8859-2 table (Czech/Slovak). It is refusing me any service on items encoded as such. Everything is being displayed correctly in the EPG but not recognized bij the autotimer plugin. The plugin does provide the amazing choice of two encoding tables: unicode UTF-8 and the table for German accents ISO 8859-15.....does anybody know if there exists something like a patch to solve this?

 

[Ezee]

It seems that the encoding table is not included and just limited to the ones mentioned, which limits the span of users that can enjoy this plugin. Maybe there is somebody out there who is able and willing to edit the plugin and include more encoding tables then the poor number of two that it is containing now and make it at least suitable for all EU countries to use. Manualy adding this to every autotimer settings xml is possible but of course diminishes the atvantage of the plugin to the traditional timer into a marginal nil

 

[doglover]

Did you try to set it to utf-8?

Willy

 

[Ezee]

Yes that is the standard setting but this does not contain the specific accents for Czech/Slovak alphabet, it needs to actually contain the signs from the ISO 8859-2 set as for example german accents are just in the ISO 8859-15 set. I am basically surprised that encoding tables are still an issue anywhere, knowing that most commercial operating systems fully automated this and are containing any table imaginable. It reminds me of the times that DOS was having these problems long...loooong time ago. But I Guess that what is free does not reflect what the market demands but more what the (voluntary) programmers consider fun to do and put in.

 

[doglover]

I know that enigma2 (and even enigma1) understands all the ISO and utf-8 tables.
When I was compiling the XMLTV files I had to convert everything to the corresponding ISO table from utf-8. The xmltv importer in enigma1 does not understand utf-8. And since all the xmltv files (except the aribic and hebrew epg) are still compatible with enigma1 all the files are still in the ISO format.

If the format could be utf-8, the problem would not exists I guess.

Willy

 

[Ezee]

I would like to know what is misconstructed on the coding of the plugin that is has to be manually set to the correct character encoding table and on top of that is crazy limited like this.

If the EPG (Enigma2) understands the tables and perfectly displays any accented character then how is it possible to code a plugin for Enigma2 that cannot simply pick it up? The ambition to get a Plugin widely used badly misconceived here i guess....I think that a serious programmer will repair this easily.

Answers from the creator to much I did not get: the advise to manually ad the correct table into the xml.(imaging to have to do this with every single autotimer entry)...of course i tried this and even then it is not working......sad piece of junk if you ask me.

But then again it is for free so what can we expect to begin with..?..

 

[t0mas]

Hello Ezee,

I had the same problem, I was not able to use Autotimer when title contain national characters.
I must use partial title seach method and specify only part of the title that not contain any national character.

I also tried force code page 8859-2 to python files but then I found guide for code page iso6937 that I remebered that exist in my /usr/share/enigma2/encoding.conf related to Slovak & Czech EPG. I tryed use that iso code page instead of ISO8859-2 and it works ! A ono to funguje.

I found one guide that was for DM and I added there small modification for my VU+ DUO.
In Czech language you can find original guide (use google and part of python rows).

Based on info & files from user helour from other board:

1. Copy iso6937.py to /usr/lib/python2.6/encodings/
2. Replace file aliases.py in /usr/lib/python2.6/encodings/
3. Add row:
config.plugins.epgsearch.encoding=iso6937
to the /etc/enigma2/settings file

4. probably is also need modify EPGseach
On DM file /usr/lib/enigma2/python/Plugins/Bp/geminimain/confEPGSearch.py (untested by me)
row 66 modify to:
tmpstr = tmpstr.decode('UTF-8', 'replace').encode(config.plugins.epgsearch.encoding.value, 'replace')


On vu+ duo (VTI image) in /usr/lib/enigma2/python/Plugins/Extensions/EPGSearch/EPGSearch.py
row : searchString = searchString.decode('UTF-8', 'replace').encode(encoding, 'replace')
replace with:
searchString = searchString.decode('UTF-8', 'replace').encode(config.plugins.epgsearch.encoding.value, 'replace')


5. in file /usr/lib/enigma2/python/Plugins/Extensions/AutoTimer/AutoTimerEditor.py
replace row:
selection = ['UTF-8', 'ISO8859-15']
with row:
selection = ['UTF-8', 'ISO8859-15', 'iso6937']

6. reboot

Note: Files iso6937.py & aliases.py you can find in encoding_codec_iso6937.zip (search internet, that zip archive exist on several places).
NOTE: when you create new Autotimer with title that contain national characters please set manually, EPG encoding iso6937