Back to Home / #mythtv / 2008 / 05 / Prev Day | Next Day
#mythtv IRC Logs for 2008-05-15

---Logopened Thu May 15 00:00:33 2008
00:01-!-foxbuntu [] has joined #mythtv
00:36-!-jhulst [n=jhulst@unaffiliated/jhulst] has joined #mythtv
01:36-!-adante [] has quit [Remote closed the connection]
01:36-!-adante [] has joined #mythtv
01:47<leprechau>brb....kernel update...
01:50-!-leprechau [i=charlie@] has quit ["Leaving"]
01:56-!-leprechau [i=charlie@] has joined #mythtv
02:10-!-arie [] has joined #mythtv
02:11-!-arie [] has left #mythtv []
02:12-!-gnome42 [] has quit [Remote closed the connection]
02:28-!-EtherNet [] has joined #mythtv
02:30-!-nsaspook [] has quit [Remote closed the connection]
02:32-!-feiner [] has quit [Read error: 110 (Connection timed out)]
02:43-!-EtherNet_ [] has quit [Read error: 110 (Connection timed out)]
02:44-!-jhulst [n=jhulst@unaffiliated/jhulst] has quit ["Konversation terminated!"]
02:45-!-feiner [] has joined #mythtv
02:54-!-EtherNet [] has quit [Read error: 110 (Connection timed out)]
03:07-!-dekar1 [] has joined #mythtv
03:11-!-dekarl [] has quit [Read error: 104 (Connection reset by peer)]
03:18-!-joobie [n=joobie@] has quit ["This computer has gone to sleep"]
03:30-!-stuarta [n=stuarta@unaffiliated/stuarta] has quit ["leaving"]
03:32-!-stuarta [n=stuarta@unaffiliated/stuarta] has joined #mythtv
03:48-!-mattwire [] has joined #mythtv
03:51-!-xris [] has quit []
03:55-!-jk1joel [] has quit [Remote closed the connection]
04:00-!-mattwire [] has quit [Remote closed the connection]
04:01<janneg>gbee: at least [17332] is wrong. numRowsAffected is correct for non SELECT queries
04:01<janneg>I haven't looked at the other commit yet
04:21<janneg>gbee: the mythplugin commits are correct
04:22<gbee>janneg: didn't mean to change any non-select queries
04:24<gbee>ahh, missed the new query - only saw the SELECT above it
04:43-!-Enmity [] has joined #MythTV
04:49-!-Enmity [] has left #MythTV ["Leaving"]
05:04<janneg>gbee: no problem, it's fixed. I was suspecious since I already converted all numRowsAffected for SELECTs in mythtv
06:15<janneg>do I understand #5364 correctly? does it suggest that we are already doing?
06:28-!-dagar [] has quit [Read error: 110 (Connection timed out)]
06:28<gbee>that's the way I read it, we're already doing that
06:29<gbee>I think it's likely this guy converted his database to UTF8 already and that's the cause behind any problems he saw with the conversion
06:37-!-Anduin [] has quit [Read error: 104 (Connection reset by peer)]
06:41-!-dagar [] has joined #mythtv
06:43<jarle>Trying to debug a segfault in mythtv-setup I get this output: I'm following the guide at, is this info outdated?
06:45-!-MrGandalf [] has quit ["blah"]
06:51<gbee>jarle: move "-x gdbcommands" before "mythtv-setup"
06:51<jarle>gbee: so the howto should be updated then?
06:51<gbee>and if that doesn't work use "--args mythtv-setup"
06:52<gbee>jarle: generally what you've typed should work, I'm not sure why it's not but the howto could probably be made bullet proof
06:52<jarle>gbee: the move solution did not work...
06:53<gbee>ahh, remove the "-l myth.log" bit from gdbcommands
06:54<jarle>gbee: yeah, that's where I thought the problem was...
06:54<gbee>mythtv-setup doesn't accept -l
06:54<gbee>it probably should
06:54<jarle>gbee: so this arg is supposed to be used together with the backend ony, right?
06:56<jarle>gbee: I suppose mythtv-setup will not accept -v record,channel,siparser either?
06:56<gbee>jarle: some of the options parsed in gdbcommands are only really useful for the frontend or backend, but aside from the "-l myth.log" bit the instructions should work for any mythtv application
06:56<gbee>jarle: no it will accept -v just fine
07:00<jarle>gbee: seems to work now.... but now I can't get it to segfault :)
07:02<jarle>I get a lot of output from ATSCStreamData::HandleTables(): Unknown table 0x46, should this function be used for dvb-s scanning also?
07:05<gbee>no .... at least the code has changed a fair bit since I last worked on it but I doubt anyone would create name a class ATSCStreamData if it also operated on DVB-S data
07:06<gbee>janneg or danielk22 would know for sure
07:07<stuarta>there's a lot of commonality in that code
07:07<jarle>gbee: especially since I live in Europe and has never ever used ATSC on my system...
07:07<stuarta>hence sometimes it comes out of ATSC even when dvb is being used
07:08<jarle>stuarta: oki, guess a lot of code could be re-used when it comes to digital tv..
07:08<gbee>stuarta: true ... though if that's the case they should share a base class where the common methods reside
07:09<stuarta>there is, i think that's just an oddity which could do with fixing
07:11<gbee>jarle: yeah, there is a lot of shared code, I'm just honestly suprised to see that we're using an ATSC class, or at least one named as such on DVB
07:13<jarle>gbee: too many US-based developers maybe? :)
07:13<jarle>is ATSC used outside of US, BTW?
07:15<purserj>only in a few countries
07:17<stuarta>japan maybe
07:18-!-joobie [] has joined #mythtv
07:23-!-TelnetManta [] has quit ["Ex-Chat"]
07:25-!-joobie [] has quit ["This computer has gone to sleep"]
07:31-!-joobie [] has joined #mythtv
07:46<janneg>we have MPEGStreamData and DVBStreamData too, each handling tables for it's standard
07:48<stuarta>mpegstreamdata is kinda the base due to the commonality's between mpeg, atsc, dvb
07:49<janneg>afaik all or at least some goes through all classes for scanning to detect channels of each standard which is assumed unknown
07:49<stuarta>sounds about right
07:50-!-joobie [] has quit ["This computer has gone to sleep"]
07:53-!-joobie [] has joined #mythtv
07:57-!-famicom [] has quit ["Leaving"]
08:01<gbee>janneg: makes sense I guess
08:01-!-joobie [] has quit ["This computer has gone to sleep"]
08:02-!-famicom [] has joined #mythtv
08:09-!-joobie [] has joined #mythtv
08:10-!-joobie [] has quit [Client Quit]
08:23-!-TelnetManta [n=benwilli@] has joined #mythtv
08:49<danielk22>ATSC is used outside the Americas, but that's not why this is used for satellite scanning. It's used because when scanning we always run all data through all the stream data classes (unless there is a specific reason not to), this allows us to detect the standard being used for a particular channel. In Europe it's likely to be either MPEG or DVB, and in the US, MPEG, DVB, or ATSC. (And in Japan I bet you could see MPEG, DVB, ATSC, and ISDB). End user
08:50<stuarta>the joy of standards
08:51<danielk22>heh, there is a ticket for the mythtv channel scanner because it doesn't allow more than one standard to be used per transport!
08:51<stuarta>now that's a nasty concept
08:56<gbee>not the fast time I've asked but I forgot to write down the answer last time - where do I want to set the breakpoint for qassert?
08:59<gbee>found it,
09:00-!-danielk22 [] has quit ["Leaving."]
09:01<gbee>or not
09:04<gbee>qFatal did it
09:19-!-melunko_ [n=hmelo@] has joined #mythtv
09:30-!-Cardoe [n=Cardoe@gentoo/developer/Cardoe] has joined #mythtv
09:42-!-gbee [] has quit [Read error: 110 (Connection timed out)]
09:53-!-foxbuntu [] has quit ["Leaving"]
10:03-!-gbee [] has joined #mythtv
10:13-!-MrGandalf [] has joined #mythtv
10:15-!-bbeattie [n=bbeattie@] has joined #mythtv
10:16-!-Loto [n=Loto@xbmc/user/Loto] has quit []
10:16<GreyFoxx>cust: I use recording groups on mine. So my wifes stuff, my daughters and mine are all in seperate groups
10:17-!-Loto [n=Loto@xbmc/user/Loto] has joined #mythtv
10:27-!-foxbuntu [] has joined #mythtv
10:37-!-renato [n=renato@] has quit ["Ex-Chat"]
10:37-!-renato_ [n=renato@] has joined #mythtv
10:48-!-stuarta_ [] has joined #mythtv
10:53-!-stuarta [n=stuarta@unaffiliated/stuarta] has quit [Read error: 110 (Connection timed out)]
10:54-!-stuarta_ is now known as stuarta
10:58-!-MrGandalf [] has left #mythtv []
10:59-!-danielk22 [] has joined #mythtv
11:03-!-MrGandalf [] has joined #mythtv
11:04<danielk22>mrg: this segfault, it happened with 0.21 or trunk? backend during eit scanning i presume?
11:04<MrGandalf>it's just shy of 0.21
11:04<MrGandalf>yes, active eit scan
11:05<MrGandalf>I get one of these a day on average (or used to before I added a cache)
11:05<MrGandalf>just wondering if it's a Qt bug or Myth stomping over itself.
11:05<danielk22>my guess with just that info is that mysql disconnected.
11:06-!-f|at|ine [] has joined #mythtv
11:06<MrGandalf>I have the connection timeout to 1 year
11:06<danielk22>hmmm, then I have no idea.
11:06<MrGandalf>for a given tuner, active eit is single threaded, correct?
11:07<danielk22>it should be
11:07<MrGandalf>oh wait.. this happened when a second tuner started recording and likely passive eit kicked in as well
11:07<danielk22>But stuarta is more expert on that code right now than me.
11:07<MrGandalf>wondering if maybe a lock would help
11:08<MrGandalf>or a .tryLock() in a short loop
11:08<MrGandalf>well, thanks for looking :)
11:08<danielk22>The code was originally written to require very little locking by being careful about not sharing certain strings. But I think this was changed at some point to save memory or some such.
11:09<danielk22>that completely changed the locking theory for the classes, and i don't understand how it currently works.
11:09<MrGandalf>I see
11:10<stuarta>it does do locks around all the DB access
11:10<MrGandalf>maybe I'll ping on stuarta then or just remove my caching hack and put in a lock
11:10<stuarta>and anytime it manipulates the cache
11:10<MrGandalf>ah, there he is
11:10*stuarta waves
11:11<danielk22>stuarta, but how does it make sure that all the QStrings are only used in one thread at a time?
11:11<MrGandalf>specifically EITHelper::get_chan_id_from_db()
11:11*stuarta browses some code
11:12<danielk22>It's those scary qstrA = qstrB that I worry about.
11:13<danielk22>I wanted to avoid QDeepCopy<> since we are processing so much data that the mallocs would really cause problems.
11:13<stuarta>there's a lot of local string stuff going on
11:13<stuarta>most of which fall out of context at the end of a fixup
11:15<danielk22>stuarta: yep those things should be safe since once fixup is done the qstring is passed of to a different thread, and the two threads never touch the same string.
11:15<danielk22>(at the same time)
11:16<MrGandalf>danielk22: for a given tuner..
11:16<MrGandalf>but what about two or more tuners calling the same fixup at the same time?
11:16<danielk22>the fixups are re-entrant, no?
11:17<danielk22>if they use QRegExp classes, those are protected by a lock.
11:18<stuarta>much use is made of QRegExp's
11:18<danielk22>hmm, was eitfixup originally used on a per tuner basis, and it is now shared?
11:19<stuarta>no the cache is shared
11:19<stuarta>static actually
11:19<stuarta>it's still per tuner
11:19<janneg>no, each eithelper has it's own fixup
11:19<danielk22>ok, cuz I'm looking at it now, and eitfixup is so not threadsafe :)
11:19<danielk22>or re-entrant even :)
11:19<MrGandalf>but wouldn't get_chan_id_from_db() get called before any fixup using QRegExpt() is called?
11:19<stuarta>its not meant to be
11:20<MrGandalf>I see
11:20<stuarta>each eitfixup runs in it's own thread against it's own data from the card it is associated with
11:20<janneg>yes, get_chan_id_from_db is called before the fixups
11:21<janneg>fixup is called on the complete DBEvents
11:21-!-bbeattie [n=bbeattie@] has left #mythtv []
11:21<danielk22>eitcache doesn't even have qstrings.. probably nothing to worry about there..
11:22<stuarta>it's nice and clean in that way :)
11:22-!-reynaldo_ [] has joined #mythtv
11:23<danielk22>hmm, and get_chan_id_from_db really shouldn't be called very often at all since it's result is cached.
11:23<MrGandalf>danielk22: it's cached where?
11:23<janneg>srv_to_chanid[key] = chanid;
11:24<MrGandalf>damn, didn't see that
11:25<MrGandalf>I added a cache directly to get_chan_id_from_db() and I can tell you it takes several minutes for that cache to become complete (for me)
11:25<danielk22>still once it does...
11:25<MrGandalf>once it does, a segfaultdue to this never occurs
11:26<danielk22>basically after one complete transport scan at say 5 minutes per transport it will have the full map.
11:26<MrGandalf>it's within that 10-15 minutes or so..
11:26<janneg>MrGandalf: that depends only on the number of channels and distribution of event
11:26<MrGandalf>janneg: yes, and you know how many channels my EIT goes through..
11:27<janneg>Dish uses a single transponder for events of all channels, right?
11:27<MrGandalf>~2600 channels
11:27<MrGandalf>9 days worth of events
11:28<janneg>I estimate over 1 million events
11:29<MrGandalf>my program table has just shy of 1 million, but it has other things as wel..
11:29<MrGandalf>I'd say 750-800
11:29<janneg>if the events are grouped by channel it would take a while to complete the cache
11:30<MrGandalf>gotta go.. bbl
11:30<janneg>50 events per channel and day is probably a little high
11:34-!-reynaldo [] has quit [Read error: 110 (Connection timed out)]
11:38<danielk22>prolly, 8-48, avg 30
11:42-!-zaheer_ [] has joined #mythtv
11:44-!-beandog [n=steve@gentoo/developer/beandog] has joined #mythtv
11:44-!-zaheerm [] has quit [Nick collision from services.]
11:44-!-zaheer_ is now known as zaheerm
11:50-!-rwalker [n=chatzill@] has joined #mythtv
11:51-!-jk1joel [] has joined #mythtv
11:58-!-Chutt [] has joined #mythtv
11:59-!-xris [n=xris@] has joined #mythtv
12:02-!-gnome42 [] has joined #mythtv
12:19-!-zaheerm [] has quit [Read error: 110 (Connection timed out)]
12:28-!-jamesd [] has quit ["Leaving"]
12:33<MrGandalf>stuarta: so, how should I go about this then? Should I put a lock around get_chan_id_from_db() or change out the QString in get_chan_id_from_db()?
12:46-!-Andreaz [] has joined #mythtv
12:46<Andreaz>Hello Friends... Could it be that mythwelcome is not reacting on lirc-keypresses?
12:47<MrGandalf>seek topic
12:49<Andreaz>Sorry, false channel.. :(
12:54<stuarta>MrGandalf: if that's your suspect then stick some QMutexLocker's in there
12:58<MrGandalf>stuarta: already started :) figured that would be a sure bet.. thanks
12:59<MrGandalf>well, not QMutexLocker().. if they are indeed hitting at the same time, they can deadlock each other
12:59-!-zaheerm [] has joined #mythtv
12:59<MrGandalf>I'll put a tryLock() in a short loop waiting 100 usec between each try
13:03-!-Andreaz [] has quit [SendQ exceeded]
13:32<MrGandalf>stuarta: you probably don't want a ticket and a patch for this, do you?
13:36-!-Anduin [] has joined #mythtv
13:58-!-Chutt [] has quit [Remote closed the connection]
14:00*xris reminds people about the group for mythtv:
14:04-!-robthebob [] has joined #mythtv
14:40-!-richards [] has joined #mythtv
15:38<Cardoe>what's the big hub bub with LinkedIn anyway..
15:38<laga>it's web 2.0!
15:40-!-kormoc [] has joined #mythtv
15:42<Cardoe>So I just logged into LinkedIn after like almost a year of not using it
15:43<Cardoe>in the upper right there's 3 "people you might know"
15:43<Cardoe>One of the guys is the domain squatter/hacker that stole a domain of mine a few years back
15:45<laga>well, you know him.
15:56<kormoc>Cardoe, if you want to link to me, let me know :P
16:01<Cardoe>laga: but how did it know!?
16:01<gbee>it's all seeing and all knowing .... be afraid
16:02<MrGandalf>they've been spying on you..
16:02<laga>you should have put on your tinfoil hat
16:02<MrGandalf>don't be paranoid just because people are after you..
16:03<gbee>all these web 2.0 things are basically there to data mine for advertising purposes
16:03<laga>gbee: which is quite sad as some of them are very handy.
16:03<beandog>never have so many people given up so much personal info willingly
16:06-!-clever [] has quit [Read error: 104 (Connection reset by peer)]
16:09-!-TelnetManta [n=benwilli@] has quit ["Ex-Chat"]
16:21-!-reynaldo_ [] has quit [Read error: 110 (Connection timed out)]
16:25-!-MrGandalf [] has quit ["home"]
16:27-!-lsobral [n=sobral@] has quit ["Ex-Chat"]
16:30-!-reynaldo [] has joined #mythtv
16:30-!-lsobral [n=sobral@] has joined #mythtv
16:40-!-Andreaz [] has joined #mythtv
16:41-!-slow [] has joined #mythtv
16:57-!-Andreaz [] has quit [SendQ exceeded]
17:01-!-rwalker [n=chatzill@] has quit ["ChatZilla [Firefox]"]
17:28-!-foxbuntu [] has quit ["Leaving"]
17:55-!-clever [] has joined #mythtv
18:00-!-Cardoe [n=Cardoe@gentoo/developer/Cardoe] has quit ["Leaving"]
18:11-!-Loto [n=Loto@xbmc/user/Loto] has quit [Remote closed the connection]
18:12-!-Loto [n=Loto@xbmc/user/Loto] has joined #mythtv
18:13-!-richards [] has quit ["bye all"]
18:15-!-MrGandalf [] has joined #mythtv
18:30-!-czth_ [n=dbrobins@nat/microsoft/x-9f1fa7690c3ea683] has joined #mythtv
18:45-!-robthebob [] has quit [Read error: 113 (No route to host)]
18:46-!-czth [n=dbrobins@nat/microsoft/x-844613f66d69aa49] has quit [Read error: 110 (Connection timed out)]
18:50-!-joobie [n=joobie@] has joined #mythtv
18:54-!-briand [] has quit [Read error: 110 (Connection timed out)]
18:54-!-briand [] has joined #mythtv
19:07-!-jafa [] has quit [Read error: 104 (Connection reset by peer)]
19:09-!-jafa [] has joined #mythtv
19:24-!-beandog [n=steve@gentoo/developer/beandog] has quit ["Leaving"]
19:33-!-adante_ [] has joined #mythtv
19:38-!-jamesd [] has joined #mythtv
19:48-!-adante [] has quit [Read error: 101 (Network is unreachable)]
19:48-!-adante_ is now known as adante
19:52-!-turbo [] has joined #mythtv
19:53-!-briand [] has quit [Read error: 110 (Connection timed out)]
20:02-!-gbee [] has quit [Read error: 110 (Connection timed out)]
20:03-!-gbee [] has joined #mythtv
20:08-!-danielk22 is now known as danielk_dinner
20:10-!-turbo [] has quit [Read error: 110 (Connection timed out)]
20:15-!-turbo [] has joined #mythtv
20:18-!-xris [n=xris@] has quit []
20:35-!-mattwire [] has joined #mythtv
20:35-!-mattwire [] has quit [Client Quit]
20:44-!-xris [] has joined #mythtv
20:55-!-briand [] has joined #mythtv
20:56-!-turbo [] has quit [Read error: 110 (Connection timed out)]
21:16-!-Ethald [] has joined #mythtv
21:17<Ethald>what is a good tv tuner card for me to test mythtv out with?
21:19-!-Ethald [] has left #mythtv ["Leaving"]
21:24-!-davilla [] has joined #mythtv
21:27-!-davilla_ [] has joined #mythtv
21:27-!-davilla_ [] has left #mythtv ["Leaving"]
21:27-!-davilla [] has left #mythtv ["Leaving"]
21:28-!-davilla [] has joined #mythtv
21:29-!-davilla [] has quit [Client Quit]
21:29-!-davilla [] has joined #mythtv
21:29-!-davilla [] has quit [Remote closed the connection]
21:29-!-davilla [] has joined #mythtv
21:30-!-davilla [] has left #mythtv ["Leaving"]
21:31-!-davilla_ [] has joined #mythtv
21:52-!-davilla_ [] has quit ["Leaving"]
22:02-!-davilla [] has joined #mythtv
22:03-!-davilla [] has left #mythtv ["Leaving"]
22:09-!-davilla [] has joined #mythtv
22:11-!-davilla [] has left #mythtv ["Leaving"]
22:35-!-davilla [] has joined #mythtv
22:38-!-joobie [n=joobie@] has quit [Read error: 110 (Connection timed out)]
22:39-!-davilla [] has left #mythtv ["Leaving"]
22:50-!-jafa [] has quit ["Leaving."]
22:55-!-turbo [] has joined #mythtv
22:56-!-briand [] has quit [Read error: 110 (Connection timed out)]
23:13-!-turbo [] has quit [Read error: 110 (Connection timed out)]
23:19-!-turbo [] has joined #mythtv
23:28-!-jmk_ [] has joined #mythtv
23:31-!-jmk [] has quit [Read error: 110 (Connection timed out)]
23:37-!-otwin_ [n=otwin@] has joined #mythtv
23:49-!-otwin [n=otwin@] has quit [Read error: 110 (Connection timed out)]
23:49-!-otwin_ is now known as otwin
23:49-!-famicom [] has quit [Read error: 110 (Connection timed out)]
23:54-!-davilla [] has joined #mythtv
23:55-!-davilla [] has left #mythtv ["Leaving"]
---Logclosed Fri May 16 00:00:08 2008