Back to Home / #mythtv / 2008 / 07 / Prev Day | Next Day
#mythtv IRC Logs for 2008-07-01

---Logopened Tue Jul 01 00:00:54 2008
00:03-!-SqWaw [] has quit [Nick collision from services.]
00:03-!-Joegsd [] has quit ["( :: NoNameScript 4.21 )"]
00:03-!-SqWaw [] has joined #mythtv
00:05-!-reynaldo [] has quit [Read error: 113 (No route to host)]
00:06-!-jpabq_hs [] has quit ["Leaving"]
00:28-!-foxbuntu [] has quit ["Leaving"]
00:30-!-gnome42 [] has quit [Remote closed the connection]
00:42-!-dragonian [] has quit [Read error: 110 (Connection timed out)]
01:03-!-joobie [] has joined #mythtv
01:32-!-ahbritto [] has quit [Client Quit]
01:44-!-|gunni| [] has joined #mythtv
02:01-!-_gunni_ [] has quit [Read error: 110 (Connection timed out)]
02:10-!-SqWaw` [] has joined #mythtv
02:15-!-SqWaw [] has quit [Read error: 60 (Operation timed out)]
02:22-!-dragonian [] has joined #mythtv
02:25-!-stoffel [] has joined #mythtv
02:26-!-stoffel [] has quit [Client Quit]
02:31-!-edannenbe [] has joined #mythtv
02:37<jjwin2k1>Can I put my project ( on the page?
02:57-!-dekarl1 [] has joined #mythtv
03:15-!-dekarl1 [] has left #mythtv []
03:16-!-cskowronnek [] has joined #mythtv
03:16-!-reynaldo [] has joined #mythtv
03:22-!-noisymim1 [] has joined #mythtv
03:22-!-noisymim1 [] has quit [Read error: 104 (Connection reset by peer)]
03:25-!-noisymime [n=josh@nat/ibm/x-36cae08c5b72fcee] has quit [Read error: 110 (Connection timed out)]
03:26-!-ahbritto [] has joined #mythtv
03:32-!-poptix [] has quit [Read error: 110 (Connection timed out)]
03:38-!-Netsplit <-> quits: iamlindoro, tris, jams, Captain_Murdoch, The_Ball, clever, PointyPumper, Tanthrix
03:38-!-Netsplit over, joins: PointyPumper, tris, Tanthrix, jams, Captain_Murdoch, iamlindoro, clever, The_Ball
03:39-!-tris [] has quit [SendQ exceeded]
03:41-!-tris [] has joined #mythtv
03:48-!-reynaldo [] has quit [Read error: 110 (Connection timed out)]
04:56-!-jjwin2k1 [] has left #mythtv []
05:02-!-grokky [] has joined #mythtv
05:10-!-Pumpernick [i=Pintlezz@] has joined #mythtv
05:10-!-PointyPumper [i=Pintlezz@] has quit [Read error: 104 (Connection reset by peer)]
05:11-!-poptix [i=poptix@] has joined #mythtv
05:37<gbee>Chutt: nice, hard to be too enthusiastic about it though because now I've probably got to resolve a bunch of conflicts :p
05:38-!-grokky [] has quit []
05:50<gbee>not too bad actually, at this point my themedmenu patch has so many conflicts with changes made in the last week that it's easier to start again, but that's the only big casualty
05:53-!-joobie_ [] has joined #mythtv
05:56-!-joobie [] has quit [Read error: 104 (Connection reset by peer)]
05:57-!-joobie [] has joined #mythtv
06:03-!-MaverickTech [] has joined #mythtv
06:07-!-grokky [] has joined #mythtv
06:16-!-joobie_ [] has quit [Read error: 110 (Connection timed out)]
06:17-!-grokky [] has quit []
06:22-!-Mad93 [n=Marc@] has joined #mythtv
06:34-!-streamtrade [n=jsass@] has quit [Read error: 110 (Connection timed out)]
06:35-!-Dibblah [] has quit [Read error: 104 (Connection reset by peer)]
07:01-!-MrGandalf [] has quit ["work"]
07:02-!-Mad93 [n=Marc@] has quit ["KVIrc 3.2.6 Anomalies"]
07:12<Chutt>gbee, it was really mostly minor stuff
07:16<gbee>mythui would be because janneg and I went through and fixed most of the more obvious stuff, thanks for finishing it :)
07:18<gbee>libmyth and libmythtv will probably present a few challenges ;)
07:24<Chutt>ilbmyth's going to be a pain
07:24<Chutt>most of the widget code
07:24<Chutt>libmythupnp shouldn't be too bad
07:27-!-grokky [] has joined #mythtv
07:30-!-HReadren [] has joined #mythtv
07:33<gbee>if I can step up the speed of the mythui port then we might just be able to reduce or even eliminate the widget code in libmyth entirely
07:36<gbee>all depends on the timeline I guess, if you want to remove qt3support from libmyth in the three months or so then converting over the widget code is a necessary job
07:38-!-MavT [] has joined #mythtv
07:40<Chutt>i was going to see how hard it'd be
07:41<gbee>no harm in that
07:41<Chutt>but, i'm doin libmythupnp first
07:45<gbee>any reason I shouldn't start removing widgets from uitypes which are no longer used? (UIAnimatedImageType for starters)
07:48<Chutt>go for it
07:53<gbee>428 lines gone
07:54<Chutt>cool. =)
07:57-!-MaverickTech [] has quit [Read error: 113 (No route to host)]
08:05<Chutt>are you still rewriting themedmenu?
08:11<gbee>well about to start again anyway, merging in changes from the last week is more fiddly than just going back to the beginning
08:15<gbee>don't be, it needed doing and I don't think starting from scratch is a bad idea really
08:18<Chutt>yeah, i got a good chunk of the way through the library reorg and started over
08:18<gbee>seemed easier when I did it for the first time six months ago, the second attempt I started two weeks ago was clumsy and I was changing my mind about things a fair bit
08:32<danielk_Zzzzz>FYI mythtv trunk doesn't seem to need any changes for intrepid, but gcc does. A compiler bug prevents it from compiling dbcheck.cpp with debugging symbols or with optimization.
08:32-!-danielk_Zzzzz is now known as danielk22
08:32<Chutt>that's amusing
08:33<Chutt>stuff compiles really fast with -O0, i've noticed
08:33<Chutt>and reaaaaly slowly with -O3
08:33<Chutt>with gcc 4.2.3
08:33<danielk22>I think all those patches Erik did the trick :)
08:34<gbee>danielk22: I've had to build dbcheck with -O0 since upgrading to gcc 4.3.0
08:34<danielk22>yeah, gcc 4.3 does some extra optimizing.
08:34<danielk22>10% speedup with some things.
08:34<Chutt>i don't understand what there is to optimize in dbcheck.cpp
08:34<Chutt>the entire file is strings.
08:34<gbee>if I don't then it chews up memory and never seems to finish
08:35<Chutt>maybe we should change them from QString's to const char*'s
08:35<danielk22>Yeah, I quadrupled the xen memory allocation from 256m to 1g to try to compile it.
08:37<danielk22>I'm going to submit a bug report, lets see if they can fix it. They probably use something memory and time inefficient to construct constants in 4.3
08:37<gbee>even with 2Gb available it isn't enough, eventually the OOM-killer starts so I can't leave it compiling unattended anymore
08:37<Chutt>danielk22, hrm
08:37<Chutt>i'm willing to bet it's all the QString constructors
08:37<Chutt>not that i have gcc4.3 to test against.. =)
08:42-!-Pumpernick [i=Pintlezz@] has quit [Read error: 104 (Connection reset by peer)]
08:44-!-PointyPumper [i=Pintlezz@] has joined #mythtv
08:55-!-joobie [] has quit ["This computer has gone to sleep"]
09:01-!-grokky [] has quit []
09:06-!-SqWaw` [] has quit ["( :: NoNameScript 4.21 )"]
09:08-!-Cardoe [n=Cardoe@gentoo/developer/Cardoe] has joined #mythtv
09:25<gbee>ack, wondered why the progress dialog wasn't working for the prescaling popup and it's because we don't parse the theme before the scaling
09:29<gbee>yeah, wasn't thinking ;)
09:29<clever>prescale just those images, scale on the fly, use qt:P
09:32<danielk22>If anyone has anything to add to the bug report...
09:50<danielk22>anyone know of a way to make qmake output a compilation target specifically for dbcheck.o ?
09:50<clever>danielk22: you could just qmake -win32 or other and grep it out of the makefile
09:51<danielk22>clever: I'm looking for something automated..
09:52<clever>why would you need to do that?
09:52<danielk22>so that I can commit it
09:53<danielk22>so that mythtv can be compiled with gcc 4.3.1 without needing to hand edit Makefiles
10:00-!-jjwin2k1 [] has joined #mythtv
10:06-!-jgarvey [] has joined #mythtv
10:48-!-JoeBorn [] has joined #mythtv
11:02-!-Dibblah [] has joined #mythtv
11:07<danielk22>I found it.. QMAKE_EXTRA_UNIX_TARGETS -- this could also be used to add icc support, there are really only a few files in ffmpeg that don't work with icc...
11:33-!-foxbuntu [] has joined #mythtv
11:41-!-iamlindoro__ [n=iamlindo@] has joined #mythtv
11:47-!-JoeBorn [] has quit ["Leaving"]
11:55-!-cskowronnek [] has quit ["Ex-Chat"]
11:57<gbee>any benchmarks around comparing ICC against GCC for recent versions?
11:58-!-kvandivo [] has joined #mythtv
11:58-!-MasseR [] has joined #mythtv
11:59-!-MasseR [] has left #mythtv []
11:59-!-xris [n=xris@] has joined #mythtv
11:59-!-edannenbe [] has quit [Remote closed the connection]
12:00-!-edannenbe [] has joined #mythtv
12:01<gbee>btw, if would be nice if we can find a way to use tune/cpu values if the version of GCC supports them, without complicating configure too much e.g. k8-sse3, core2
12:05<clever>ive used a whole toolchain&kernel from a real 486
12:05<clever>and it ran blazing fast on a c2d with no problems at all
12:06<clever>though it could probly go even faster if i built it properly for c2
12:06<danielk22>gbee: I found some a couple weeks ago, gcc 4.3.1 was about 10% faster than 4.2.x, and a recent icc was about 25% faster than 4.2.x .. but those are pretty meaningless, we would need to test with MythTV. ffmpeg might not be optimize well because of the assembly; you might need to disable that to get a faster binary with icc.
12:07<gbee>I was curious to see how much 4.3 had closed the gap
12:08<danielk22>I've also had a program run 10x faster compiled with icc than gcc. But that program was written with SoA rather than AoS for the layout of the data, which vectorizes very well.
12:09<danielk22>gcc was inserting lots of unneeded loads and stores; while icc figured out that the processor had more than two registers..
12:13-!-xris [n=xris@] has quit []
12:14-!-iamlindoro___ [n=iamlindo@] has joined #mythtv
12:15-!-iamlindoro__ [n=iamlindo@] has quit [Read error: 104 (Connection reset by peer)]
12:16<Chutt>did we ever do hostname lookups?
12:17-!-mattwire [] has joined #mythtv
12:19<gbee>AFAIK no, but localhost was an exception
12:20<Chutt>i didn't see any code in there
12:21-!-xris [n=xris@] has joined #mythtv
12:24-!-beandog [n=steve@gentoo/developer/beandog] has joined #mythtv
12:32-!-edannenbe [] has quit [Remote closed the connection]
12:33-!-gnome42 [] has joined #mythtv
12:34-!-jmk [n=jmk@] has joined #mythtv
12:52<gbee>still tempted to have dialogs, if not screens, add themselves to the stack as it would save a bunch of repetition
12:54<Chutt>see, part of the reasoning behind that was having dialogs get created in a separate thread
12:55<Chutt>and having generic support in the base code to create that thread, throw up a busy dialog
12:55<Chutt>get the new dialog, etc
12:56<gbee>not quite following, but if you say there is a good reason then I'll take your word :)
12:57<Chutt>so say there's a screen that takes a while to show up
12:57<Chutt>like, oh, the recorded programs dialog
12:57<Chutt>right now it creates the dialog in the ui thread
12:58<Chutt>blocks everything
12:58<Chutt>so, if we thread off the dialog creation instead
12:58<Chutt>it wouldn't block the ui, and we could put up a busy dialog
12:59<Chutt>and that _could_ be done generically in libmythui
12:59<Chutt>that's why the Init() function for dialogs is supposed to do all the creation work, and not the constructor like the old ui does
13:00<gbee>Create() but I see your point
13:01-!-andreax [] has joined #mythtv
13:10-!-_packetscan is now known as packetscan
13:17<clever>i just had an idea
13:17<clever>allowing livetv over upnp
13:18<clever>if you could just advertise every channel as a streaming video feed
13:18<clever>might be alot of work but it could allow livetv on a wider range of players/systems
13:18<Chutt>doesn't upnp require the file size?
13:19<clever>for streaming files proly
13:19<Chutt>and there'd be issues with the way we split programs into separate files, i think
13:19<clever>with winamp i can tune into 'channels'
13:19<clever>it isnt broken up into seperate files per show
13:19<clever>i just start pulling the data and i get the whatever was on when i started
13:20<Chutt>mythtv breaks livetv into separate recordings per show
13:20<clever>yes internaly it does
13:20<Chutt>externally it does :p
13:20<clever>what about those network encoder boxes?
13:20<Chutt>the recorder still breaks them into separate recordings per show
13:20<Chutt>they're separate files
13:20<clever>do they break it up into seperate shows or just provider 1 digital stream with on/off/tune commands?
13:21<gbee>I don't think upnp is ever going to replace the MythProtocol for livetv or recordings, but I believe someone has already said they intend to add livetv support to upnp
13:21<GreyFoxx>the upnp:videoBroadcast element can define channels and such , but it's the seperate files that are the issue
13:22<GreyFoxx>you can't really even define a playlist since it's "future" recordings whcih haven't happened yet
13:22<clever>with simple remuxing you can probly merge the current and next 'files' into a single stream
13:22<clever>as you pipe it out to the upnp client
13:23<clever>so the client doesnt even know it changed files
13:23<clever>and ive seen headers in the radio feeds ive listened to, that inform the player of the current track/song
13:23<clever>which get sent mid stream
13:24<GreyFoxx>Sure, but it will you can "proxy" the seperate recordings into a single stream which is about the only way it would work with myth's livetv
13:24<GreyFoxx>damn my english gets worse when Ive had no sleep
13:24<clever>i know
13:24<clever>i was awake about 29 hours a few days ago
13:25<clever>using that hack you could make a seperate stream for every channel you can tune
13:25<clever>posibly masking out those which are behind ocupied tuners
13:25<clever>the only problem is the conflicts it may force into things
13:26<clever>and the lack of a ui to ask what to do like mythfrontend has
13:27<GreyFoxx>Don't ask
13:27<GreyFoxx>just cut the stream and allow the other schedule to take it
13:27<GreyFoxx>and insert "Tuner needed for recording" frame over and over :)
13:28<clever>could make an option to set a default prio for the upnp channels
13:28<clever>so you can put the schedules on either side of it
13:28<clever>and just default the upnp channels to -20
13:29<clever>so it does look posibly, just alot of 'fun' to write up and get it to cooperate with every device:P
13:30<Chutt>oh, sure it's possible
13:35-!-bkero [n=bkero@osuosl/staff/bkero] has quit ["Lost terminal"]
13:38-!-bkero [n=bkero@osuosl/staff/bkero] has joined #mythtv
13:38-!-naeo- [n=asl@] has joined #mythtv
13:38<clever>technicaly i could put a master backend on the moon:P
13:38<clever>now the feds cant bust me for storing 300gig of ilegal music on it
13:38<clever>but it wouldnt be cheap:P
13:38<laga>AFAIK, there is an american flag on the moon
13:39<clever>where do you go to get a search warrant for the moon?
13:39<clever>i'll just plant a canadian flag on my 'property' after i buy it and plant a server up there:P
13:40-!-PointyPumper [i=Pintlezz@] has quit [Read error: 104 (Connection reset by peer)]
13:40-!-stoffel [] has joined #mythtv
13:41-!-Netsplit <-> quits: fish_, naeo, iamlindoro_
13:44-!-fish__ [] has joined #mythtv
13:44-!-Netsplit over, joins: naeo, fish_, iamlindoro_
13:44-!-fish_ [] has quit [Success]
13:45-!-beandog [n=steve@gentoo/developer/beandog] has quit [Read error: 110 (Connection timed out)]
13:45-!-beandog [n=steve@gentoo/developer/beandog] has joined #mythtv
13:53-!-CDev-m [] has joined #mythtv
13:54<CDev-m>clever: upnp supports live broadcasts, I just haven't coded it yet.
13:57<CDev-m>I'm hoping I will have time to work on mythtv again (been over a year now) in another couple of months.
13:57-!-naeo [n=asl@] has quit [Connection timed out]
13:57<CDev-m>In the mean time, I still try to read this channel and the dev list.
13:59-!-CDev-m [] has quit [Read error: 104 (Connection reset by peer)]
14:01-!-CDev-m [] has joined #mythtv
14:03-!-CDev-m [] has quit [Read error: 104 (Connection reset by peer)]
14:07-!-|gunni| [] has quit [Read error: 104 (Connection reset by peer)]
14:07-!-_gunni_ [] has joined #mythtv
14:41-!-Anduin [] has quit [Read error: 104 (Connection reset by peer)]
14:46-!-renato_ [n=renato@] has quit ["Ex-Chat"]
14:55-!-renato [n=renato@] has joined #mythtv
14:57-!-czth__ [n=dbrobins@nat/microsoft/x-3a7e3b252447b0bb] has quit ["Leaving"]
14:58-!-czth [n=dbrobins@nat/microsoft/x-95a8f4c865454008] has joined #mythtv
15:00-!-czth_ [n=dbrobins@nat/microsoft/x-2101c2f77028ed7b] has joined #mythtv
15:01-!-CDev [] has quit [Read error: 54 (Connection reset by peer)]
15:02-!-stoth [] has joined #mythtv
15:05<Chutt>danielk22, in your posted dbcheck-full.cpp - it's pulling in some qt3 headers?
15:06<danielk22>this is the 0.21 dbcheck.cpp
15:06<Chutt>there's 39,228 lines of code before it gets to the actual dbcheck.cpp
15:06<danielk22>heh, is that unusual ?
15:06<Chutt>it seems like a lot.
15:07<danielk22>headers pull in other headers, etc, etc.. i just don't know what is typical these days
15:07<Chutt>i know, but it still seems like a lot
15:08-!-Cardoe [n=Cardoe@gentoo/developer/Cardoe] has quit [Remote closed the connection]
15:10<danielk22>tv_play-full.cpp is 108,937 lines...
15:11<Chutt>tv_play.cpp also takes a very long time to compile
15:11<danielk22>ok, dvbrecorder-full.cpp is 87,777 lines and it is not so difficult to compile
15:12-!-Cardoe [n=Cardoe@gentoo/developer/Cardoe] has joined #mythtv
15:13<danielk22>util-xv-full.cpp is 50,135 lines.. I think maybe dbcheck.cpp is not so bad in that regard.
15:14<Chutt>so i changed the const QString updates[] -> const char *updates[]
15:14<Chutt>for gcc 4.2.whatever
15:14<Chutt>compile time went from 50 seconds, down to 15
15:14<danielk22>are we using any special characters or other justification for QString?
15:14<Chutt>didn't test to see if it'd actually work, but...
15:15<Chutt>oh, only thing is the termination string is ""
15:15<danielk22>i mean i'd like the compiler to be fixed.. but
15:15<danielk22>so change it to NULL :) problem solved! heh
15:16<Chutt>i don't have 4.3 to test with
15:16<danielk22> -- and I'll give it a try
15:16<Chutt>just in vim, do a search and replace
15:16<Chutt>that's all i did
15:16<Chutt>(ie, i doubt it'll actually _function_)
15:17<Chutt>but that should be minor
15:17<danielk22>heh, ok i'll look at it a little bit later...
15:17-!-czth [n=dbrobins@nat/microsoft/x-95a8f4c865454008] has quit [Connection timed out]
15:17<Chutt>:set nomagic
15:17<Chutt>:%s/const QString updates[]/const char *updates[]/g
15:27-!-jjwin2k1 [] has left #mythtv []
15:29-!-beandog [n=steve@gentoo/developer/beandog] has quit [Read error: 110 (Connection timed out)]
15:30-!-beandog [n=steve@gentoo/developer/beandog] has joined #mythtv
15:37-!-Anduin [] has joined #mythtv
15:44-!-PointyPumper [i=Pintlezz@] has joined #mythtv
15:50-!-reynaldo [] has joined #mythtv
15:53<danielk22>it compiles... haven't tested it works.
15:54<Chutt>oh wait
15:54<danielk22>well the compile crashed before.. so no :)
15:54<Chutt>it wasn't compiling before =)
15:55<danielk22>da patch,
15:55<danielk22>I don't have time to test it right now.. but if someone else feels like giving it a workout
16:00<Chutt>looks like it should work
16:01<Chutt>i've got to get some real work done today
16:02<danielk22>yeah, i'll test it tonight if no one beats me to it.. but i have tasks to perform right now
16:04-!-Anduin [] has quit ["Leaving."]
16:11-!-CDev [] has joined #mythtv
16:11-!-stoth [] has quit ["Leaving"]
16:13-!-CDev [] has quit [Client Quit]
16:14-!-Netsplit <-> quits: Joemama_, naeo-, cattelan, superm1, jarle
16:14-!-CDev [] has joined #mythtv
16:15-!-Netsplit over, joins: naeo-, Joemama_, jarle, cattelan, superm1
16:16<Chutt>ooh, neat
16:16-!-PointyPumper [i=Pintlezz@] has quit [Read error: 104 (Connection reset by peer)]
16:16<Chutt>QHostInfo::fromName() does not involve the Qt event loop
16:16<Chutt>means we can use it
16:20-!-_gunni_ [] has quit ["KVIrc 3.4.0 Virgo"]
16:22-!-stoffel [] has quit ["leaving"]
16:47-!-reynaldo [] has quit [Read error: 113 (No route to host)]
16:58<teprrr>just out of curiosity, what's the problem with qt's eventloop?
16:59<Chutt>so in qt3, there was one event loop
16:59<Chutt>hostname lookups required the event loop to run - they were always asynchronous
16:59<Chutt>if you wanted to do a hostname lookup in a different thread
16:59<Chutt>you were pretty much screwed
17:00<Chutt>and you could get weird behavior, since the event loop could start processing other random stuff
17:01<mattwire>Chutt: libmythdb should be free of qt3 headers right? So the fact that it's trying to use them is most likely a problem with my system?
17:02<Chutt>what's the error?
17:03<mattwire>oh it's getting loads of errors because it seems to have reverted to using includes from /usr/include/qt3
17:03-!-iamlindoro___ [n=iamlindo@] has quit []
17:03<mattwire>thought I'd just check that was wrong before poking further :)
17:05<mattwire>seems to be compiling ok if I explicitly run qmake-qt4 in the libmythdb dir
17:05<Chutt>delete all the makefiles in the tree
17:05<Chutt>and re-run configure
17:06<mattwire>make distclean doesn't get rid of them all?
17:06<Chutt>not the top level one
17:06<Chutt>you can try..
17:16<gbee>the Nova-S cards have been relisted if anyone is interested
17:17-!-foxbuntu [] has quit ["Leaving"]
17:25-!-_gunni_ [] has joined #mythtv
17:26-!-Netsplit <-> quits: jwhite, jamesd, Hannibal-, _charly_, jgarvey, sutula, rooaus, asjoyner_
17:26-!-Netsplit <-> quits: anykey_, j-rod
17:27-!-Netsplit over, joins: jgarvey, rooaus, jamesd, _charly_, anykey_, asjoyner_, sutula, Hannibal-, jwhite, j-rod
17:28-!-andreax [] has quit ["Leaving."]
17:33-!-Anduin [] has joined #mythtv
17:40-!-beandog [n=steve@gentoo/developer/beandog] has quit ["Leaving"]
17:42-!-slfalcon [] has joined #mythtv
18:00-!-Cardoe [n=Cardoe@gentoo/developer/Cardoe] has quit ["Leaving"]
18:02-!-Anduin [] has quit ["leaving"]
18:02-!-Anduin [] has joined #mythtv
18:02-!-Anduin [] has quit [Client Quit]
18:06-!-DerDracle [] has joined #mythtv
18:07<DerDracle>Is PTS an objective value associated with seconds? Or is it arbitrary?
18:15-!-Anduin [] has joined #mythtv
18:17-!-PointyPumper [i=Pintlezz@] has joined #mythtv
18:23-!-npmccallum_ [] has joined #mythtv
18:24-!-ricosecada [n=ricoseca@] has joined #mythtv
18:24<ricosecada>Why does mythtv need a mysql database?
18:25<Chutt>because it does.
18:25<mattwire>so it can store pictures of elephants
18:27<ricosecada>ooh, you are so fucking funny.. damned lame, thats what it is.. tv viewer that needs a database..
18:27-!-ricosecada [n=ricoseca@] has quit [Client Quit]
18:27<danielk22>matt: shhh, don't let out the secret
18:28<iamlindoro>lame is thinking it's just a TV viewer
18:28-!-jgarvey [] has quit ["Leaving"]
18:30-!-_gunni_ [] has quit [Remote closed the connection]
18:35-!-ricosecada [n=ricoseca@] has joined #mythtv
18:36<ricosecada>What a piece of crap software.. I have been using linux for more than ten years, and for the first time I needed to record some old video recordings. Took a look at mythtv.. damned.. database installation.. crap setup.. this really sucks big time!
18:36-!-ricosecada [n=ricoseca@] has quit [Client Quit]
18:37<gbee>so you've been using it since you were 3 years old?
18:39<gbee>no, I must resist troll baiting, that was out of order - sorry
18:40<iamlindoro>Don't worry, he doesn't stay long enough to see it :)
18:41<gbee>heh, I've got join/part messages hidden
18:44-!-joobie [] has joined #mythtv
18:45-!-jmk [n=jmk@] has quit ["Leaving"]
19:05-!-grokky [] has joined #mythtv
19:08-!-grokky [] has quit [Client Quit]
19:15-!-XChatMav [] has joined #mythtv
19:16-!-jhulst [n=jhulst@unaffiliated/jhulst] has joined #mythtv
19:32-!-npmccallum_ [] has quit []
19:32-!-MavT [] has quit [Read error: 113 (No route to host)]
19:33-!-noisymime [n=josh@nat/ibm/x-51b9c3253510dda1] has joined #mythtv
19:45-!-mattwire [] has quit ["Leaving"]
19:45-!-npmccallum [] has joined #mythtv
19:46-!-famicom [] has quit [Read error: 110 (Connection timed out)]
19:46-!-npmccallum [] has quit [Client Quit]
19:48-!-npmccallum [] has joined #mythtv
20:08<Chutt>libmythupnp looks tricky to de-qt3support-ify
20:23-!-stoth [] has joined #mythtv
20:35-!-jhulst [n=jhulst@unaffiliated/jhulst] has quit [Remote closed the connection]
20:35-!-xris [n=xris@] has quit []
20:41-!-npmccallum_ [] has joined #mythtv
20:45-!-npmccallum__ [] has joined #mythtv
20:53<CDev>Chutt: which parts?
20:54-!-noisymime [n=josh@nat/ibm/x-51b9c3253510dda1] has quit [Read error: 104 (Connection reset by peer)]
20:55-!-noisymime [n=josh@nat/ibm/x-0930334ca19beffc] has joined #mythtv
20:55-!-slfalcon [] has left #mythtv []
20:56-!-npmccallum [] has quit [Read error: 110 (Connection timed out)]
20:59-!-npmccallum_ [] has quit [Read error: 110 (Connection timed out)]
21:03-!-jhulst [n=jhulst@unaffiliated/jhulst] has joined #mythtv
21:04-!-jhulst [n=jhulst@unaffiliated/jhulst] has quit [Remote closed the connection]
21:06-!-greenmoss [] has joined #mythtv
21:06-!-greenmoss [] has left #mythtv []
21:08-!-jhulst [n=jhulst@unaffiliated/jhulst] has joined #mythtv
21:09<Chutt>CDev, mainly the QPtrList stuff
21:09<Chutt>no more autodelete
21:10<Chutt>CDev, probably the easiest way would be to not make them lists of pointers, i think
21:19-!-xris [] has joined #mythtv
21:35-!-jpabq_hs [] has joined #mythtv
21:38<clever>QList<void *>
21:38<clever>QList<QObject *>
21:38<Chutt>doesn't help with the autodelete stuff
21:38<clever>(make all classes qobjects)
21:39<clever>qobjects emit a destroyed signal when deleted
21:39<Chutt>that doesn't exactly work with threads
21:40<Chutt>myth has a _few_ threads :p
21:40<clever>i saw how the custom event loop murdered the signals from QUdpSocket
21:40<clever>i 'fixed' it by polling hasDataGram() in the mythtv event loop
21:40<clever>lets assume it doesnt sleep much:P
21:40<Chutt>and there's a lot of issues with qobjects and threads
21:40<clever>which it doesnt seem to do
21:41<clever>cpu usage doesnt go down 1 bit when i pause
21:41<Chutt>paused is still displaying video
21:41<clever>and mythmusic seems to use MORE cpu then video:P
21:41<clever>yes but paused doesnt have to DECODE
21:41<danielk22>I'm getting all itchy at the mention of using a QObject in the backend.. much less signals :-o
21:41<clever>its just a solid frame
21:41<clever>which takes 0% cpu on mplayer
21:41<Chutt>the OSD
21:42<Chutt>decoding's minimal cpu, really
21:42<Chutt>at least for anything non-HD
21:42<clever>the osd doesnt doesnt change that much
21:42<clever>its fairly static
21:42<Chutt>blended in with every frame
21:42<danielk22>heh, these days anything non-h.264
21:42<clever>blend once
21:42<clever>wait for a change
21:42<Chutt>yeah, it does
21:43<Chutt>depends on the output method
21:44<danielk22>mythmusic uses the OSD now?
21:44<Chutt>no, i meant the cpu useage during paused
21:44<Chutt>usage, too
21:45<clever>yeah the mythmusic ui is alot diff
21:45<clever>backspace in the OSD hides any OSD popups
21:45<clever>like escape
21:45<clever>but it cant end playback
21:45<Chutt>the visualizers are all pretty cpu intensive
21:45<clever>it doesnt hide the volume bar in mythmusic
21:45<clever>which still times out
21:45<clever>Chutt: but i had no viz on:P
21:46<clever>and putting the black viz fullscreen made the controls lag even worse
21:46<clever>drawing black in fullscreen is hard on the cpu:P
21:46<clever>xterm does it all month long on practicaly 3mhz
22:03<Chutt>i really don't get why they made a Q3TextStream
22:03<Chutt>the changes are minor in QTextStream
22:18<clever>but if the way a function reacts changes
22:18<clever>it may not cause compile errors
22:18<clever>so you may not notice the bugs until a week later
22:19<Chutt>it hasn't really changed for most usage
22:19-!-jhulst [n=jhulst@unaffiliated/jhulst] has quit [Remote closed the connection]
22:19<Chutt>and they renamed functions that did
22:19<Chutt>found one difference
22:20<Chutt>in QTextStream
22:20<Chutt>it doesn't do charset stuff on strings
22:20<clever>which could be major depending on how you use it
22:21<Chutt>libmythupnp was using it to ensure everything was utf8
22:21<Chutt>so, yeah, major
22:26-!-stoth [] has quit [Read error: 110 (Connection timed out)]
22:30-!-stoth [] has joined #mythtv
22:32<CDev>Chutt: When I first started the qt4 port and looked at libmythupnp, I was thinking of removing all of the QPtrList and other QT containers and replacing them all with STL containers.
22:32<CDev>The more I looked at it, the more I wanted to refactor the entire library, then I got distracted with my house and work. Oh well.
22:37-!-jhulst [n=jhulst@unaffiliated/jhulst] has joined #mythtv
22:37<CDev>If you want to put off libmythupnp for a while, I should be freeing up by the middle of August. I can take care of reworking the library not to need qt3support.
22:38<Chutt>i'll probably give it a go
22:38<Chutt>we'll see, though
22:38<Chutt>it looks like QTextStream does the proper thing when operating on a QByteArray
22:38<Chutt>(ie, it supports the codec stuff)
22:39<CDev>that helps.
22:39<Chutt>and i _think_ most of the parts that matter in there are on QCStrings
22:39<Chutt>which become QByteArrays in qt4
22:39<Chutt>and, the autodelete stuff isn't so bad
22:39<CDev>Sounds about right.
22:40<CDev>A lot of the classes are refcounted, so it shouldn't be too back taking care of destroying the rest.
22:40<Chutt>all the socket code transfers, since i copied in QSocketDevice
22:42<CDev>It helps not having to re-write all the communication code... however, I have learned a lot since I wrote it originally... it's not the most effient or scaleable code.
22:42<Chutt>if you're intending on more drastic reworkings
22:42<Chutt>then i'll leave it alone. =)
22:43<Chutt>though i'll probably go through to make sure there's no unsafe ascii()/latin1()/utf8() usage
22:43<CDev>I am, but I can't commit to making them for at least a few months...
22:43<Chutt>no, no rush
22:43<Chutt>i just had some extra time this past weekend and a 3 day work week, so figured i should get some of the old tasks out of the way =)
22:44<CDev>I still have big plans for upnp support, embedded setup html pages, and live streaming... just no time.
22:44<Chutt>there's nothing really _wrong_ with using qt3support
22:44<CDev>Enjoy it while you can :)
22:44<CDev>I didn't like using it, but I figured it was the first step to the port.
22:45<clever>ive often learned things on programs i made the first time thru
22:45<Chutt>well, 2 of the libraries are free of it, so =)
22:45<clever>so when i remade it from scratch it looked alot better
22:46<CDev>It's a good start. Plus, thanks for removing the circular dependancies... long over due.
22:46-!-jamesd__ [] has joined #mythtv
22:46-!-jamesd [] has quit [Read error: 110 (Connection timed out)]
22:46<CDev>got to go... have a good night.
23:01-!-jhulst [n=jhulst@unaffiliated/jhulst] has quit [Remote closed the connection]
23:06-!-jhulst [n=jhulst@unaffiliated/jhulst] has joined #mythtv
23:16<GreyFoxx>You would likely have to map the keys
23:24-!-nvc [] has joined #mythtv
23:25<nvc>hi, i'm having an issue with FreeType on ubuntu 8.04, mythTV won't compile without it and i've installed the only 2 'freetypes' i've found on the package manager
23:26<nvc>anyone have any ideas?
23:28-!-jhulst [n=jhulst@unaffiliated/jhulst] has quit [Remote closed the connection]
23:31-!-nvc [] has quit ["leaving"]
23:31-!-nvc [] has joined #mythtv
23:33<Anduin>nvc: build issues still belong in #mythtv-users
23:33<nvc>oh, sorry, didn't realise there was a seperate chan
23:34-!-npmccallum__ [] has quit []
23:41<Chutt>i still can't figure out how best to get the mediamonitor working again
23:47<Chutt>it doesn't really belong in libmythui
23:47<Chutt>but the handler/events it sends to the mainwindow need it there or lower
23:52<Chutt>actually, it _can't_ go into mythui since it depends on the old ui code
23:52<Chutt>such a mess :/
23:58-!-nvc [] has quit ["Lost terminal"]
---Logclosed Wed Jul 02 00:00:31 2008