Back to Home / #mythtv / 2004 / 08 / Prev Day | Next Day
#mythtv IRC Logs for 2004-08-20

---Logopened Fri Aug 20 00:00:09 2004
00:16thor_ any reason I can't go ahead and change mythui stuff to build as a library?
00:23--- User: *** Morph_ is now known as Morph
00:26Chutt i'm not quite ready for that, yet
00:27thor_ k
00:27thor_ just fiddling with mfe stuff, good time to force myself into mythui stuff
00:29--- User: *** bline_ is now known as bline
00:55rkulagow_ i'm guessing if we're running the nv driver then this is to be expected?
00:55rkulagow_ collect2: ld returned 1 exit status
00:55--- <<-- Agrajag- [] has quit ("My damn controlling terminal disappeared!")
00:56rkulagow_ usr/bin/ld: cannot find -lXvMCNVIDIA
00:56--- ---> dcstimm [] has joined #mythtv
00:59rkulagow_ chutt: here? why is XVMC now the default?
00:59thor_ I think that got in there by mistake
01:00rkulagow_ ah. but then this?
01:00rkulagow_ make[2]: Entering directory `/home/mythtv/mythtv/programs/mythepg'
01:00rkulagow_ g++ -c -pipe -Wall -W -g -D_REENTRANT -D_GNU_SOURCE -D_FILE_OFFSET_BITS=64 -DPREFIX=\"/usr/local\" -DMMX -DCONFIG_VIDEO4LINUX -DUSING_IVTV -DLCD_DEVICE -DUSING_OSS -DUSING_OPENGL_VSYNC -DQT_THREAD_SUPPORT -DQT_SHARED -DQT_NO_DEBUG -I/usr/share/qt3/mkspecs/default -I. -I../../libs -I../../libs/libmyth -I/usr/local/include -I/usr/include/qt3 -o main.o main.cpp
01:00rkulagow_ g++ -o mythepg main.o -L/usr/share/qt3/lib -L/usr/X11R6/lib -L../../libs/libmyth -L../../libs/libmythtv -L../../libs/libavcodec -L../../libs/libavformat -lmythtv-0.15 -lmythavformat-0.15 -lmythavcodec-0.15 -lmyth-0.15 -lfreetype -lmp3lame -L/usr/X11R6/lib -lXinerama -lXv -lXxf86vm -lGL -lGLU -lqt-mt -lXext -lX11 -lm -lpthread
01:00rkulagow_ ../../libs/libmythtv/ undefined reference to `VideoOutputXvMC::VideoOutputXvMC[in-charge]()'
01:00rkulagow_ collect2: ld returned 1 exit status
01:01rkulagow_ i've make distcleaned and restarted compile in case the define screwed up something along the way
01:01thor_ I dunno, I'm just parroting something I saw earlier where someone said xvmc seems to have sneaked into and Chutt said oops
01:01thor_ or something to that effect
01:01rkulagow_ ah. also, what's this about nipples?
01:01Beirdo mmmm. nipplevideo
01:02thor_ milkshakes, showers, swedish girlfriends
01:02thor_ it all fits
01:02rkulagow_ perfect sense then.
01:04Captain_Murdo| rkulagow_: did you take xvmc out of
01:04rkulagow_ on my local copy, yes
01:05rkulagow_ and now i'll see if it still complains in the mythepg directory
01:05Captain_Murdo| I had to remove videooutbase.o and recompile to get it to work even after I commented out xvmc and reran qmake. was quicker than make distclean. :)
01:05Captain_Murdo| no xvmc here either.
01:05rkulagow_ ah, but mebbe i've got a supercomputer in the house now?
01:06rkulagow_ nah. couldn't afford the electricity.
01:06rkulagow_ although it's cool to lounge on an old cray
01:06Captain_Murdo| just did a "cvs update" after having waited about 2 weeks and ccache saved a lot of time, but something caused globalsettings.cpp to have to recompile. :(
01:06rkulagow_ ick. that's the xbox killer
01:07rkulagow_ (when i tried self-compiling on an xbox)
01:07rkulagow_ killer in the sense that after a few hours i said, "this may not finish in my lifetime", not as in it fell over
01:07Captain_Murdo| yeah, my main frontend is still running an older distro so I have to compile a copy on it and it's only a celeron 466.
01:08Captain_Murdo| big incentive to upgrade that box...
01:08rkulagow_ cau decode 1080i on that thing? :)
01:09* Captain_Murd thinks cool, getting rid of those at()'s in uitypes.cpp has let me compile it on that older gcc.
01:09Captain_Murdo| haven't tried. usually only record at 352x480 (1/2 D1 res) on my M179 cards.
01:09Captain_Murdo| has a i810 chipset and using an external VGA->composite converter (don't have SVideo on my ancient TV)
01:10Captain_Murdo| takes a while to startup playing sometimes on longer recordings but usually it chugs along fine. I think the delay in starting is because it's having to load the positionmap info.
01:11rkulagow_ i think i'm kidding about 1080i. didn't someone just say they needed a 2.6Ghz to do 1920x1080i HDTV?
01:13Captain_Murdo| :)
01:13rkulagow_ that's a lot of horsepower for TV.
01:14rkulagow_ i never did get around to testing my P166 to see if it'd decode a 352x240 MPEG-2 stream. that way lies madness!
01:16Captain_Murdo| could throw a DXR3 card in there. ;)
01:17rkulagow_ feh. don't care enough about it. and besides, that's cheating.
01:18* Captain_Murd was hoping he could pawn his off on somebody.
01:18rkulagow_ that's fun:
01:18rkulagow_ QGList::takeNode: Corrupted node
01:18rkulagow_ Program received signal SIGSEGV, Segmentation fault.
01:18rkulagow_ [Switching to Thread 81926 (LWP 21889)]
01:18rkulagow_ 0x41513c77 in mallopt () from /lib/
01:19--- <<-- snowblind [] has quit ()
01:19rkulagow_ 2004-08-20 00:17:45 Started Commercial Flagging for 'Sealab 2021' recorded from channel 1047 at Wed Aug 18 23:12:00 2004
01:19rkulagow_ 2004-08-20 00:17:45 Started Commercial Flagging for 'Friends' recorded from channel 1009 at Wed Aug 18 21:59:00 2004
01:19rkulagow_ 2004-08-20 00:17:45 Started Commercial Flagging for 'Cheap Seats' recorded from channel 1055 at Thu Aug 19 21:30:00 2004
01:19rkulagow_ 2004-08-20 00:17:45 Started Commercial Flagging for 'The Daily Show With Jon Stewart' recorded from channel 1042 at Wed Aug 18 21:59:00 2004
01:19rkulagow_ 2004-08-20 00:17:45 Started Commercial Flagging for 'Sealab 2021' recorded from channel 1047 at Wed Aug 18 23:14:00 2004
01:19rkulagow_ (then the error)
01:25Captain_Murdo| 5 shows? you just restart the backend?
01:25rkulagow_ yes
01:25Chutt *cough* broken *cough* *cough*
01:25rkulagow_ i think a few of them are 0-byters from a previous crash
01:29Captain_Murdo| I rewrote the way it's firing off commercial flagger jobs so I can merge the queues for transcoding and comm flagging. seems to be working ok here with the comm flagging stuff. trying to update my production systems so I can give it a workout before proceeding. My backend has been crashing every so often usually right after it's finished a recording, so I've been wondering if it was the newer comm detect code I added post 0.15.1
01:29Captain_Murdo| n able to get anything out of gdb worthwihle though.
01:31Captain_Murdo| anyway, gotta go to bed, dentists appointment in a few hours. :(
01:31thor_ ewww
01:32Captain_Murdo| yeah :( he's ok, but occasionally you get an assistant who works on your teath like she used to run jackhammers on a construction site.
01:32thor_ double ewww
01:34* Captain_Murd goes to bed and leaves current CVS compiling on his slow frontend box.
01:35dcstimm hey guys what version of ivtv do you guys run?
01:42--- <<-- dcstimm [] has quit (Read error: 104 (Connection reset by peer))
01:47--- <<-- rkulagow_ [] has quit (Read error: 113 (No route to host))
01:50--- <<-- Jayakumar [~Jayakumar@] has quit (Connection reset by peer)
01:50--- ---> Jayakumar_ [~Jayakumar@] has joined #mythtv
02:05--- Netsplit <-> quits: CanadaBoy00
02:06--- Netsplit over, joins: CanadaBoy00
02:17Chutt neat.
02:17Chutt one small adjustment to the bob display
02:18Chutt and it's damn near perfect on the tv out =)
02:18thor_ which TV out?
02:18thor_ 350?
02:18Chutt nvidia
02:18thor_ sweeeeet
02:18Chutt it really looks nice
02:19--- ---> Agrajag- [] has joined #mythtv
02:19Chutt no osd degradation, either, like you get on a monitor
02:19thor_ exceedingly sweet
02:19Chutt but, doesn't look like this tweak works on a monitor
02:20--- ---> linagee_ [] has joined #mythtv
02:23Chutt thor, if you want to play with mythui and lib-ize it, go ahead =)
02:24thor_ k
02:24Chutt i'm just looking at my schedule, and i'm going to be a tad busy next few weeks
02:24Chutt wouldn't want to hold you back from doing anything
02:24thor_ heh
02:24thor_ now that I finally have some time
02:25thor_ other than lib'izing stuff, I'll try not to mess with it too much
02:25thor_ (i just need a few widgets for mfe)
02:27Chutt there's a bug or two in there, too
02:27Chutt the menu doesn't draw right for 'blue', for instance
02:27thor_ ah
02:27thor_ ok
02:27Chutt works for gant, though
02:27Chutt probably something really dumb
02:28thor_ ok
02:28--- ---> _kch_ [] has joined #mythtv
02:29thor_ bah, sleep
02:31--- ---> linagee [] has joined #mythtv
02:31Chutt sleep?
02:31Chutt what's that?
02:31thor_ heh
02:36--- <--- mchou [] has left #mythtv ()
02:41Chutt oh yeah, i said i'd look at the internal player
02:41Chutt blah!
02:46--- <<-- linagee_ [] has quit (Client Quit)
03:00Chutt hrm
03:00--- <<-- CanadaBoy00 [] has quit (Read error: 104 (Connection reset by peer))
03:00Chutt the osd menu doesn't quite look as good
03:00Chutt those one pixel high lines don't work well on a tv :(
03:12--- ---> linagee_ [] has joined #mythtv
03:25Chutt that wasn't so bad.
03:29--- <<-- linagee [] has quit (Client Quit)
03:52--- <<-- mdz [] has quit (Read error: 60 (Operation timed out))
03:52--- ---> mdz [] has joined #mythtv
03:54--- ---> choenig [] has joined #mythtv
04:11--- ---> gfiend [] has joined #mythtv
04:26--- <<-- choenig [] has quit (Remote closed the connection)
04:30--- <<-- GeckoFiend [] has quit (Read error: 110 (Connection timed out))
04:34--- ---> choenig [] has joined #mythtv
04:45--- <<-- newbie123424 [] has quit ("Using KVIrc 3.0.1 'System Virtue'")
04:46--- ---> CanadaBoy [] has joined #mythtv
05:34--- ---> mithro [] has joined #mythtv
05:41gfiend heh app locked in the plugin. never even considered that.
05:48--- User: *** gfiend is now known as GecjkoFiend
05:48--- User: *** GecjkoFiend is now known as GeckoFiend
05:57--- User: *** Netslayer is now known as Netslayer|KDE-Co
05:57--- User: *** Netslayer|KDE-Co is now known as Netslayer|KDE
06:00--- ---> Lad [] has joined #mythtv
06:01--- <<-- choenig [] has quit ("wuuuusch...")
06:01--- <--- Lad [] has left #mythtv ("Kopete 0.8.3 :")
06:01--- ---> thies__ [] has joined #mythtv
06:03--- <<-- thies_ [] has quit (Read error: 54 (Connection reset by peer))
06:05--- User: *** Netslayer|KDE is now known as Netslayer|ZZZ
06:05--- User: *** Netslayer|ZZZ is now known as Netslayer|zzzz
06:11--- ---> lisadaveh_ [] has joined #mythtv
06:12--- <<-- lisadaveh_ [] has quit (Client Quit)
06:35--- ---> lisadaveh_ [] has joined #mythtv
06:35--- <<-- lisadaveh_ [] has quit (Client Quit)
06:40--- <<-- mdz [] has quit (Read error: 60 (Operation timed out))
06:46--- <<-- lisadaveh [] has quit (Remote closed the connection)
07:25--- ---> mdz [] has joined #mythtv
07:30--- <<-- _kch_ [] has quit ("Terminando cliente")
07:33--- ---> onyxsoft_ [] has joined #mythtv
08:04--- ---> o_c [] has joined #mythtv
08:22--- <<-- o_cee [] has quit (Read error: 110 (Connection timed out))
09:15--- ---> duckeo [] has joined #mythtv
09:25--- <<-- BleedingE [] has quit ("changing servers")
09:26--- ---> BleedingE [] has joined #mythtv
09:32--- ---> pats76 [~pats76@] has joined #mythtv
09:38pats76 i was just given an ati tv wonder ve. can anyone give me there opinion on how this will work on a pII-400 system with 256 mb ram. i would only be recording and then playing back on my tv
09:48--- <<-- mdz [] has quit (Read error: 113 (No route to host))
09:48--- ---> mdz [] has joined #mythtv
10:07--- ---> dc_ [] has joined #mythtv
10:08--- ---> dc__ [] has joined #mythtv
10:08--- <<-- dc__ [] has quit (Client Quit)
10:11stoffel pats76, /topic please
10:36--- Channel: pats76 changed the topic of #mythtv to: help
10:36--- <<-- pats76 [~pats76@] has quit ("Chatzilla 0.9.64b [Mozilla rv:1.7/20040626]")
10:43--- Channel: stoffel changed the topic of #mythtv to: This is not a support channel. Go to #mythtv-users
10:45kvandivo that pats76.. He's a funny one...
10:47--- ---> johnback [~jackback@] has joined #mythtv
10:47--- <--- johnback [~jackback@] has left #mythtv ()
10:51GreyFoxx Hehe I hadn't noticed that this channel wasn't +t :)
11:01--- <<-- MythJames [] has quit ("Chatzilla 0.9.64f [Mozilla rv:1.7/20040707]")
11:10--- ---> mecraw_ [~lmarlow@] has joined #mythtv
11:21--- ---> schultmc [] has joined #mythtv
11:23--- ---> choenig [] has joined #mythtv
11:24--- <<-- choenig [] has quit (Remote closed the connection)
11:36--- ---> choenig [] has joined #mythtv
11:37--- ---> nrh1703 [] has joined #mythtv
11:38--- ---> jf100 [] has joined #mythtv
11:49--- <<-- DJ_nixbox [] has quit (Read error: 110 (Connection timed out))
11:52--- ---> DJ_nixbox [] has joined #mythtv
12:02--- ---> sfr_ [] has joined #mythtv
12:08--- <<-- stoffel [] has quit (Read error: 110 (Connection timed out))
12:19--- ---> vagrant [] has joined #mythtv
12:25--- User: *** sfr_ is now known as stoffel
12:33--- ---> linagee [] has joined #mythtv
12:34--- ---> steelep [~signwatch@] has joined #mythtv
12:39--- <<-- o_c [] has quit ("Leaving")
12:42--- <<-- Rince [] has quit (Read error: 110 (Connection timed out))
12:48--- <<-- linagee_ [] has quit (Read error: 113 (No route to host))
13:05--- ---> Rince [] has joined #mythtv
13:32--- <<-- linagee [] has quit (Client Quit)
13:40--- ---> snowblind [] has joined #mythtv
13:48--- <<-- thies__ [] has quit ("your momma is so fat, she has other small fat women orbiting her")
13:50--- <<-- jf100 [] has quit (Read error: 110 (Connection timed out))
14:06--- ---> linagee [] has joined #mythtv
14:21--- ---> o_cee [] has joined #mythtv
14:22o_cee oh yeah
14:22thor_ eh?
14:23o_cee Chutt: you had an Aria too right? what kind of fucked up backplate thingies where you put expansion cards is that? very strange solution, heh
14:23* o_cee waves at thor
14:23* thor_ donates milkshake supplies to o_cee
14:24o_cee i'd _really_ need that now, yeah :)
14:24o_cee i did see your comment last night as well, thanks
14:24o_cee :)
14:24thor_ comment?
14:25o_cee yepp
14:25o_cee showers.. actuallly, she's in the shower right now
14:25thor_ ah
14:25thor_ heh
14:25o_cee that's something for you maybe
14:26thor_ heh, possibly ... java just confuses me though
14:26o_cee and no, it's no porn
14:26thor_ anyway, I just commited the stuff to make mfd work with iTunes 4.6
14:26thor_ back to mfe
14:26o_cee cool
14:27o_cee been reading java at the university for a year now, time for c++ now when i start again
14:27thor_ seems an odd order to do it in ...
14:28thor_ "Now that you understand OO, we'll add pointers and take away garbadge collection"
14:28thor_ err, garbage
14:28o_cee dunno,, this c++ is optional
14:28o_cee java isn't
14:28thor_ ah
14:29o_cee math... or c++... hmm.. even MOOORE math.. or some fun c+.. hard decision
14:30thor_ well, advanced math from a good prof can be amazing stuff
14:31o_cee already got enough math... heh
14:31kvandivo tsk tsk tsk
14:31--- <<-- Rince [] has quit ("leaving")
14:32Chutt and advanced math from a prof that can't speak the language is, well, incomprehensible
14:32Chutt and _all_ math profs can't speak english. it's a rule.
14:32thor_ we cee the lineman funkin vedy vedy useful
14:33kvandivo i remember for partial diff eq i had a native english speaker, but he might as well have been talking ancient greek
14:33o_cee Chutt: heh, i know what you mean, last guy we had.. well, took some time hear what he talked about :P
14:34--- ---> Rince [] has joined #mythtv
14:34--- <<-- anduin [~awithers@] has quit (Nick collision from services.)
14:34o_cee Chutt: the back of your aria, where you put exxpansion slots.. do you really need to remove those two extra screwsm and the four screws that holds the cards every time?
14:34Chutt yeah
14:34Chutt i just left that big shield off
14:35--- ---> anduin [~awithers@] has joined #mythtv
14:35o_cee how extremely stupid, heh..
14:36Chutt eh
14:36o_cee of antec, not you
14:36o_cee :)
14:36Chutt it's not like you mess around with that very often
14:36o_cee true, but it's still a really dumb way of doing it
14:37o_cee had to remove the psu to get the cooler in place as well..
14:38Chutt right, but that's easy :p
14:39o_cee yeah, but when i put it back i had to bend the plate on top of the psu.. but i got it there finally :P
14:40Chutt didn't have to bend anything, here
14:41o_cee was really tight
14:44--- <<-- Rroet64 [~rroet@] has quit ("leaving")
14:47--- ---> Rroet [] has joined #mythtv
15:00--- <<-- vagrant [] has quit (Read error: 60 (Operation timed out))
15:04--- ---> AridWork [] has joined #mythtv
15:12o_cee anyone else who had any objections on my mythnews patch?
15:27--- Netsplit <-> quits: onyxsoft, dopester, jams, CanadaBoy, AridWork, Rince, pigeon, schultmc, Sir-Al, taco
15:28--- <<-- choenig [] has quit ("wuuuusch...")
15:28--- Netsplit over, joins: AridWork, Rince, schultmc, CanadaBoy, jams, onyxsoft, dopester, taco, Sir-Al, pigeon
15:30--- ---> choenig [] has joined #mythtv
15:30--- <<-- Netslayer|zzzz [] has quit (Remote closed the connection)
15:33--- ---> Netslayer [] has joined #mythtv
15:35--- <<-- Netslayer [] has quit (Remote closed the connection)
15:36--- ---> Netslayer [] has joined #mythtv
15:50--- <<-- linagee [] has quit (Client Quit)
15:59--- ---> linuxwhor1 [] has joined #mythtv
15:59--- <<-- linuxwhore [] has quit (Read error: 113 (No route to host))
16:03--- ---> gfiend [] has joined #mythtv
16:07gfiend chutt around?
16:08Chutt yup
16:08--- <<-- GeckoFiend [] has quit (Nick collision from services.)
16:09--- User: *** gfiend is now known as GeckoFiend
16:10GeckoFiend You comment about seeking via the internal player got me wondering. Would it be possible to use mythcommflag to build a markup table for other files (assuming it was modified to handle non ProgramInfo types of things)?
16:12--- <<-- Netslayer [] has quit (Remote closed the connection)
16:13Chutt sure
16:13Chutt it just takes time to do
16:16GeckoFiend I was thinking the meta data editor could check to see if they had specified "Internal" and ask them if they wanted to generate a markup table then spin off a mythcommflag in the background if they said yes. Maybe a --noflag option for mythcommflag so it just rebuilds the markup table.
16:18--- ---> Netslayer [] has joined #mythtv
16:25Chutt Sure
16:25Chutt that'd work
16:26Chutt i don't think it'd be too terribly difficult to improve performance in the non-seektable case, though
16:32--- <<-- onyxsoft_ [] has quit ("ChatZilla 0.9.61 [Mozilla rv:1.7.1/20040707]")
16:34--- <<-- choenig [] has quit (Remote closed the connection)
16:43--- ---> kvirc [] has joined #mythtv
17:07--- <<-- jams [] has quit (Read error: 110 (Connection timed out))
17:10--- ---> jams [] has joined #mythtv
17:11Chutt geckofiend, how about native dvd playback? =)
17:15--- <<-- mithro [] has quit (Read error: 60 (Operation timed out))
17:21Chutt i can't decide where to run my sql server
17:22Chutt on the nfs server, or on the master backend
17:22Chutt ah well
17:28--- <<-- AridWork [] has quit (Read error: 104 (Connection reset by peer))
17:38GeckoFiend Chutt I had thought of that as well. What I'd like to do is this: 1) Get full support for playing videos just like recordings (seeking etc) 2) Get the internal player to handle all the formats it should. (I've many of my files *try* to play but fail) 3) Native DVD playback.
17:44Chutt sounds good.
17:45Chutt native dvd playback would be nice
17:45Chutt no crappy osd
17:45Morph scrambled tv channel playback would be nice.
17:46Chutt heh
17:46* Morph checks
17:46Morph still illegal though..
17:47Chutt does your cable company still broadcast premium channels over analog?
17:47jams i'm pretty sure you can watch on the scrambled channel's you want :)
17:47Morph no.
17:47Chutt comcast here requires digital for almost everything
17:47Morph so does adelphia.
17:48thor_ core HBO and Showtime are analog here
17:48jams so does time warner. over analog they scramble tv-land and mplex
17:58--- <<-- steelep [~signwatch@] has quit ("me is bugging out")
17:59--- ---> choenig [] has joined #mythtv
18:02--- <<-- choenig [] has quit (Remote closed the connection)
18:06--- ---> choenig [] has joined #mythtv
18:32--- <<-- duckeo [] has quit (Read error: 104 (Connection reset by peer))
18:32--- ---> duckeo [] has joined #mythtv
18:32thor_ o_cee, if you're playing with ourTunes, mfd works with it just fine, but ourTunes will not talk to it if it's running on the same host
18:43o_cee allright
18:44thor_ (this is, as far as I can surmize, the fault of ourTunes)
18:49Morph what is ourtunes?
18:49thor_ java thing for connection with iTunes
18:49Morph ah..
18:52Morph ah cool.
18:52Morph and it works with mfd?
18:53thor_ yup
18:53Morph nice.
18:59--- <<-- jeffpc [] has quit (Remote closed the connection)
19:06o_cee Chutt: did you get my gant update the other day?
19:07--- <<-- schultmc [] has quit ("Client exiting")
19:18o_cee also got some .ts updates i'd like to apply, halfway done with it
19:33--- <<-- choenig [] has quit (Remote closed the connection)
19:33--- ---> cmorgan [] has joined #mythtv
19:38--- ---> jeffpc [] has joined #mythtv
19:49--- <<-- m0j0_ [~m0j0@] has quit ("Leaving")
19:54o_cee now bruce beat me to updating it.. bah
19:55--- <<-- bbeattie [~bbeattie@] has quit ("Client exiting")
19:56o_cee different thing anyway though
19:57--- ---> prophecy__ [] has joined #mythtv
19:58GeckoFiend Thor around?
19:58GeckoFiend erm thor_ even ;)
20:04thor_ am now
20:04thor_ GeckoFiend even ;)
20:07GeckoFiend You feel like dealing with Xavier? Maybe you'll have better luck with him. I've apparently hurt his feelings or something.
20:08GeckoFiend ok this just wrong
20:09thor_ heh
20:10thor_ I can respond to his last messages if you like
20:11thor_ ah, you already did
20:11thor_ if he comes back upset, I'll smooth feathers
20:12GeckoFiend he just needs to submit his feature. He kept sending in a bunch of revisions with different fixes but they always had extar crap in them
20:12thor_ and try and get him to produce just the useful bit
20:12thor_ hmmm ... din dins
20:12thor_ bbl
20:26Captain_Murdo| GeckoFiend: mythcommflag doesn't do anything special to rebuild the positionmap, it only tells the decoder to save the positionmap to the DB if one doesn't exist. A function to rebuild the positionmap would be almost as simple as something like this: if (!hasFullPositionMap) { while (!eof) GetFrame(1,true); decoder->SetPositionMap(); } if the decoder doesn't have a full position map then go through the whole file f
20:26Captain_Murdo| then when done tell the decoder to save the positionmap to the DB. You could potentially even set a flag so the decoder wouldn't decode, it would just seek which would speed things up even more.
20:27Captain_Murdo| seek meaning seek to the next frame based upon the size in the frame header, not seek as in positionmap seek. :)
20:29Captain_Murdo| I had a patch once for mythcommflag that would fixup the seektable in a nuppel file, but never put it in CVS. It made mythcommflag actually open the file and read the packet headers so it could quickly scan every packet in the file and rebuild the seektable for a damaged file for instance.
20:38--- <<-- mecraw_ [~lmarlow@] has quit ("Trillian (")
20:51GeckoFiend Captain_Murdoch that's good to know thanks
21:12--- ---> rkulagow_ [] has joined #mythtv
21:13rkulagow_ chutt: here?
21:17rkulagow_ captain_murdoch: here?
21:22Chutt rkulagow, hey
21:22Chutt i'm here now
21:23rkulagow_ hey.
21:23rkulagow_ did you glean anything from that backtrace i mailed you? i'm still at gdb
21:23Chutt not really
21:24Chutt i think it's a qt issue, but kinda brought on by the whole 'start 5 commflags at once' bug
21:24Chutt you can exit gdb, though, i don't think there's anything more useful to get from it
21:24rkulagow_ ok, i'm out.
21:25rkulagow_ i don't remember if there's a "max" parameter for the commflag stuff.
21:26rkulagow_ i'll check in a second
21:34--- <<-- o_cee [] has quit ("Leaving")
22:00Captain_Murdo| Chutt, you think it's the fact that there are 5 or a bug in the code somewhere or what?
22:11--- <<-- kvirc [] has quit ("Using KVIrc 3.0.1 'System Virtue'")
22:18Chutt i just think it should be a linear process
22:18Chutt but i don't know
22:19Chutt could investigate to make _sure_ that all those db connections are unique
22:24Captain_Murdo| QString name = QString("commercial%1%2").arg(getpid()).arg(rand()); MythSqlDatabase *commthread_db = new MythSqlDatabase(name);
22:25Captain_Murdo| I rewrote all that anyway to use a jobqueue table. hooking up a status page for it now on the status screen.
22:26Captain_Murdo| so eventually you'll be able to see a list of what's comm flagging, what's transcoding, any other jobs running, and what's queued up.
22:26rkulagow_ hrmm. i hope i don't have a flaky backend. got a segfault agin while mythbackend was generating the pix for mythweb
22:26rkulagow_ Program received signal SIGSEGV, Segmentation fault.
22:26rkulagow_ [Switching to Thread 213006 (LWP 25099)]
22:26rkulagow_ 0x41519a97 in memcpy () from /lib/
22:27Captain_Murdo| this will allow setting a max jobs also, so you could set it to only do 2 jobs at a time or run all sequentially. parallel jobs would only run if they were for independent recordings, so it wouldn't parallize transcoding and comm detection for the same recording at the same time.
22:28Captain_Murdo| doing the same thing or are you also referring to the earlier one when it was running the commflags.
22:29rkulagow_ nope, i ran mythcommflag manually (since i know it goes one at a time) to empty the queue, then started mythbackend. seemed to be peachy, then access the "recorded programs" in mythweb which then started generating all the thumbnails (it'd been a while.)
22:30rkulagow_ it did about 50-60 and then segfaulted
22:33Captain_Murdo| hmm.
22:33rkulagow_ i'm stil at gdb if that helps
22:34Captain_Murdo| anything useful in the backtrace?
22:36rkulagow_ i don't know how to grok gdb, so i'm just the monkey in front of the keyboard at this point
22:38Captain_Murdo| you can email me the top of the backtrace from "thread apply all bt full" if you want and I'll take a look.
22:46--- ---> bobnvic [] has joined #mythtv
22:46--- <<-- cmorgan [] has quit (Remote closed the connection)
22:51--- ---> cmorgan [] has joined #mythtv
22:53--- <<-- snowblind [] has quit ()
23:07--- ---> ipso [] has joined #mythtv
23:52* Captain_Murd ventures into the land of myth's gui and tries to make the status screen's content box scrollable.
---Logclosed Sat Aug 21 00:00:45 2004