Jump to content


Photo

Bug in the new debugging system


  • Please log in to reply
4 replies to this topic

#1 Taapat

  • PLi® Core member
  • 2,345 posts

+121
Excellent

Posted 12 February 2016 - 18:15

Has anyone tried the new debug system without timestaps in log?

When I disabled timestaps in a debug log (like simply commenting code from the 92-96 row, which is the same as used the flag _DBGFLG_NOTIME), https://github.com/O.../eerror.cpp#L92
then I see a strange enigma freezing when in log show some python variables with print.
 
I first this discovered on my spark sh4 receiver, but the same is on my vu+ solo2.

The problem I have with my plugin youtube: https://github.com/T...-plugin-youtube

Everything works well until in the log with print in python code need display variable with a link to the video: https://github.com/T...uTubeUi.py#L490
Enigma freez with spinner and nothing further happens.
In the log there is no error, log stops, this link is not displayed in log, and on the screen is a spinner.
 

If I remove the print of these variable from plugin, then enigma freezes when the link to the playable file is displayed in log from enigma components.
This link is nothing special, just a link to the actual video on youtube.
If I use timestaps, then everything works without any problems.
 
Simply text print or other variables displaying also work.

Before implementation of the new debugging system, log without timestaps also worked for me.
 

Why the new debugging system does not work normally in all cases without timestaps?



Re: Bug in the new debugging system #2 MiLo

  • PLi® Core member
  • 14,054 posts

+298
Excellent

Posted 12 February 2016 - 18:46

Is there a "%" sign in the string?

Looks like there's a bug in the Python log output, it shouldn't pass the string as format string to the log function.
Real musicians never die - they just decompose

Re: Bug in the new debugging system #3 Taapat

  • PLi® Core member
  • 2,345 posts

+121
Excellent

Posted 12 February 2016 - 18:58

Yes, the link is for example as follows:

https://r1---sn-bavc5ajvh-gpme.googlevideo.com/videoplayback?itag=22&upn=6CT6qUsHPUo&fexp=9416126%2C9419445%2C9420452%2C9421522%2C9422596%2C9422971%2C9423293%2C9423661%2C9423662%2C9423819%2C9424724%2C9424815%2C9425394%2C9425791%2C9427976&source=youtube&pl=17&dur=229.923&sver=3&mt=1455299507&mv=m&id=o-AAGFO14ESUbJAA5riRXvtbtFMyGs16T4jwY8B2l_tM9h&ms=au&sparams=dur%2Cid%2Cinitcwndbps%2Cip%2Cipbits%2Citag%2Clmt%2Cmime%2Cmm%2Cmn%2Cms%2Cmv%2Cpl%2Cratebypass%2Crequiressl%2Csource%2Cupn%2Cexpire&mn=sn-bavc5ajvh-gpme&ip=159.148.29.37&key=yt6&initcwndbps=4388750&lmt=1417369095585262&mm=31&expire=1455321317&ipbits=0&ratebypass=yes&mime=video%2Fmp4&requiressl=yes&signature=C454CEC5447FF02BFF59E18990AFFF266CFDC3F8.A5C53E09E26733D16264F5E95B5E27986404F14A

I see that log with timestaps also don't have print with this link, but there is no enigma freezes.

In log is only timestap.


Edited by Taapat, 12 February 2016 - 19:00.


Re: Bug in the new debugging system #4 MiLo

  • PLi® Core member
  • 14,054 posts

+298
Excellent

Posted 12 February 2016 - 19:26

Just luck if it works - because the stack happens to look a bit different, it might get away with it.

Have a fix ready, just a bit of testing needed here...
Real musicians never die - they just decompose

Re: Bug in the new debugging system #5 Taapat

  • PLi® Core member
  • 2,345 posts

+121
Excellent

Posted 12 February 2016 - 19:59

Thank you, this fix the problem: https://github.com/O...541b3bc4f6f298f


Edited by Taapat, 12 February 2016 - 20:00.



2 user(s) are reading this topic

0 members, 2 guests, 0 anonymous users