MiracleBox Prem Twin HD - 2@DVB-T2 + Xtrend et8000 - 5(incl. 2 different USBs)@DVB-T2[terrestrial - UK Freeview HD, Sandy Heath] - LAN/USB-stick/HDD
Unfortunately I cannot take a screenshot from the 6.0 image of the same problem right now, because currently there is no problematic events in any of channels.
The description for it was just empty in the screen. It was like in the below screenshot, which I have edited manually:
screen_snapshot_fw_6_0_006_edited.jpg
Could it be so that truncating UTF-8 encoded Unicode text in the end may cut in the middle of a UTF-8 encoded code-point and cause corruption?
I was seeing, for example, Benalmádena until I changed the encoding table to ISO6937, when the correct text Benalmádena was stored.
@Joe_90 worked out what the encoding should have been, although I don't know how.
IIRC - because ISO6937 was specifically mentioned in the Freeview/Freesat D Book as one of the acceptable encodings.
EDIT - here's the reference in "D" book:
It goes on to list UCS-2, UTF-8 and ISO 6937 encodings and the name for each character.Code:F.1 Introduction Table F-1 lists the displayable characters that shall be implemented by all decoders conforming to this specification. Note that the UK MHEG-5 profile specifies several characters that are not defined within ISO/IEC 6937 [42] on which the DVB SI specifications are based. These characters may be broadcast in, and shall be displayed correctly within MHEG-5 applications; there is no guarantee that they will be displayed correctly within SI delivered text. The characters in the shaded cells are shown only to emphasise gaps.
Last edited by Joe_90; 14-02-22 at 11:26.
GB Quad Plus, Mut@nt HD51, AX HD61, 80cm dish and Supreme Dark motor. Sony STR-DN 1060, Sony UHP-H1 Bluray, Odroid N2+ (CoreElec), Monitor Audio Bronze 5.1 speakers
ccs (14-02-22)
This is what the description probably should look like (6 extra characters at the end).....
Code:Naisleijonien kesäleiri 2020 oli Anni Keisalalle isku vasten kasvoja - ja tarpeellinen sellainen. Nuorena hän pärjäsi lahjoillaan, mutta enää se ei riittänyt. Hänen oli otettava itseään niskasta kiinni. Keisala teki käänteentekevän päätökse
That's not what I'm trying to say.
What I meant was that everything in the EPG cache should be utf-8 (whether it's the "correct" text is another matter).
That way you won't get an error when searching/extracting from it.
So anything that updates the EPG cache should only put valid utf-8 into it, which appears to not be the case currently.
MiracleBox Prem Twin HD - 2@DVB-T2 + Xtrend et8000 - 5(incl. 2 different USBs)@DVB-T2[terrestrial - UK Freeview HD, Sandy Heath] - LAN/USB-stick/HDD
Code:root@vuultimo4k:~# echo Naisleijonien kesäleiri 2020 oli Anni Keisalalle isku vasten kasvoja - ja tarpeellinen sellainen. Nuorena hän pärjäsi lahjoillaan, mutta enää se ei riittänyt. Hänen oli otettava itseään niskasta kiinni. Keisala teki käänteentekevän päätökse|od -h 0000000 614e 7369 656c 6a69 6e6f 6569 206e 656b 0000020 c373 6ca4 6965 6972 3220 3230 2030 6c6f 0000040 2069 6e41 696e 4b20 6965 6173 616c 6c6c 0000060 2065 7369 756b 7620 7361 6574 206e 616b 0000100 7673 6a6f 2061 202d 616a 7420 7261 6570 0000120 6c65 696c 656e 206e 6573 6c6c 6961 656e 0000140 2e6e 4e20 6f75 6572 616e 6820 a4c3 206e 0000160 c370 72a4 c36a 73a4 2069 616c 6a68 696f 0000200 6c6c 6161 2c6e 6d20 7475 6174 6520 c36e 0000220 c3a4 20a4 6573 6520 2069 6972 7469 c374 0000240 6ea4 7479 202e c348 6ea4 6e65 6f20 696c 0000260 6f20 6574 7474 7661 2061 7469 6573 a4c3 0000300 a4c3 206e 696e 6b73 7361 6174 6b20 6969 0000320 6e6e 2e69 4b20 6965 6173 616c 7420 6b65 0000340 2069 c36b c3a4 6ea4 6574 6e65 6574 656b 0000360 c376 6ea4 7020 a4c3 a4c3 c374 6bb6 6573 0000400 000a 0000401 root@vuultimo4k:~# root@vuultimo4k:~# echo Naisleijonien kesäleiri 2020 oli Anni Keisalalle isku vasten kasvoja - ja tarpeellinen sellainen. Nuorena hän pärjäsi lahjoillaan, mutta enää se ei riittänyt. Hänen oli otettava itseään niskasta kiinni. Keisala teki käänteentekevän päätökse|od -c 0000000 N a i s l e i j o n i e n k e 0000020 s 303 244 l e i r i 2 0 2 0 o l 0000040 i A n n i K e i s a l a l l 0000060 e i s k u v a s t e n k a 0000100 s v o j a - j a t a r p e 0000120 e l l i n e n s e l l a i n e 0000140 n . N u o r e n a h 303 244 n 0000160 p 303 244 r j 303 244 s i l a h j o i 0000200 l l a a n , m u t t a e n 303 0000220 244 303 244 s e e i r i i t t 303 0000240 244 n y t . H 303 244 n e n o l i 0000260 o t e t t a v a i t s e 303 244 0000300 303 244 n n i s k a s t a k i i 0000320 n n i . K e i s a l a t e k 0000340 i k 303 244 303 244 n t e e n t e k e 0000360 v 303 244 n p 303 244 303 244 t 303 266 k s e 0000400 \n 0000401 root@vuultimo4k:~# root@vuultimo4k:~# echo Naisleijonien kesäleiri 2020 oli Anni Keisalalle isku vasten kasvoja - ja tarpeellinen sellainen. Nuorena hän pärjäsi lahjoillaan, mutta enää se ei riittänyt. Hänen oli otettava itseään niskasta kiinni. Keisala teki käänteentekevän päätökse|wc 1 32 257
Last edited by ccs; 14-02-22 at 13:52.
I have noticed now this same issue with some Finnish channels. For some reason there is something wrong at EPG descriptions and autotimer fails. But there is also an issue with recording; if I manually add timer, the system will keep rebooting during while recording the program. Do You have any known fixes or workarounds for this issue?
Gigablue Quad 4K & UE 4K
.........FBC Tuners:
------------------> GT-Sat unicable LNB to 1.5M dish(28.2E)
------------------> Gigablue unicable LNB to 80 cm dish(19.2E)
.......................> FBC & DVB-S2X into 90cm dish (27.5W) Opticum robust Unicable LNB
AX HD61, Edision Osmio 4K+, Zgemma H9Combo, Octagon SF8008 , gbtrio4k, h9se using unicable ports
Zgemma H9 C/S into Giga4K
I am typically using AutoTimer to search some text in a title. Log is displaying for example "[eEPGCache] lookup events with 'Simpsonit' in title (ignore case)". I am just wondering why an illegal description is causing the issue, although the code should not care about the description text at all? Would it be possible to fix at least the title search case?
Gigablue Quad 4K & UE 4K
.........FBC Tuners:
------------------> GT-Sat unicable LNB to 1.5M dish(28.2E)
------------------> Gigablue unicable LNB to 80 cm dish(19.2E)
.......................> FBC & DVB-S2X into 90cm dish (27.5W) Opticum robust Unicable LNB
AX HD61, Edision Osmio 4K+, Zgemma H9Combo, Octagon SF8008 , gbtrio4k, h9se using unicable ports
Zgemma H9 C/S into Giga4K
The debug log attached (SW 6.2.011). The exception is the same as in the first post.
Code:< 196.0866> 18:49:10.1041 [eEPGCache] lookup events with 'Eräelämää' in title (ignore case)< 196.1279> 18:49:10.1454 [Task] >>> Error: [<Components.Task.FailedPostcondition object at 0x6917d418>] < 196.1295> 18:49:10.1471 [Task] job Components.Task.Job name=Automaattiajastus #tasks=27 completed with [<Components.Task.FailedPostcondition object at 0x6917d418>] in Components.Task.Task name=Eräelämää_4 < 196.1317> 18:49:10.1493 [Task] unrecoverable task failure < 196.1322> 18:49:10.1498 Automaattiajastus Virhe: [Failure instance: Traceback: <class 'SystemError'>: <built-in function eEPGCache_search> returned a result with an exception set /usr/lib/python3.10/threading.py:1009:_bootstrap_inner /usr/lib/python3.10/threading.py:946:run /usr/lib/python3.10/site-packages/twisted/_threads/_threadworker.py:47:work /usr/lib/python3.10/site-packages/twisted/_threads/_team.py:182:doWork --- <exception caught here> --- /usr/lib/python3.10/site-packages/twisted/python/threadpool.py:244:inContext /usr/lib/python3.10/site-packages/twisted/python/threadpool.py:260:<lambda> /usr/lib/python3.10/site-packages/twisted/python/context.py:117:callWithContext /usr/lib/python3.10/site-packages/twisted/python/context.py:82:callWithContext /usr/lib/enigma2/python/Plugins/Extensions/AutoTimer/AutoTimer.py:344:JobStart /usr/lib/enigma2/python/Plugins/Extensions/AutoTimer/AutoTimer.py:450:parseTimer
So you need to go through you channels and find which one has junk in the epg.
Yes. I have a python script, which goes through all channels, and try to get epg data by requesting "http://<ip>/api/epgservice?sRef=<sref>&endTime=<endtime>". In case of failure, it tries to find the longest possible request without errors. And yes, in that way I can find the program, which one has junk in the epg, and I can create an autotimer for it to reproduce the issue.