Back to Home / #mythtv / 2003 / 07 / Prev Day | Next Day
#mythtv IRC Logs for 2003-07-29

00:02<thor>Chutt, only knows to grab the disc name
00:02<thor>is there a dvd equivalent to cddb ?
00:02<thor>should we start one =) ?
00:05<just1nux>does imdb do anything like that?
00:07<just1nux>just movies right, not dvds
00:07<thor>libdvdread has a unique MD5 checksum routine for calculating a unique DVD id
00:07<thor>I'd be happy to query it against the web automatically, but I'm not aware of anywhere to query
00:08<just1nux>i dont see anyplace to query either. you'd think somebody woulda started one.
00:09-!-mecraw [] has joined #mythtv
00:11<paulproteus>Amazon API?
00:11<paulproteus> ?
00:16-!-hadees [] has joined #mythtv
00:17-!-michal [] has quit [Read error: 110 (Connection timed out)]
00:17<Captain_Murdoch>Chutt: that 2-line fix to DoRemove() didn't fix the problem. I'm wondering it's a timing issue maybe caused by my slow machine. It worked OK on a 1/2 hour show but not on a 1-hour show.
00:17<Captain_Murdoch>Also: got any ideas on how to stop the auto-expire from deleting a show that's currently being watched?
00:18<Captain_Murdoch>no way to tell if a frontend is currently watching a show or not right now right?
00:21<Timon>Chutt: Thor helped me figure it out. . . Had to edit and put lcdmenuitem.[cpp|h] in there
00:22<Captain_Murdoch>Could have a flag set whenever a show is being watched and unset when done. have auto-expire not delete if the flag is set, but the regular manual delete popup could delete whether the flag was set or not.
00:27-!-tmk [] has joined #mythtv
00:31<bline>dvdplay is quite easy to use it seems, it uses a callback to tell you states and has nice methods to navigate/read the dvd.
00:32<thor>bline, mythdvd could definitely use a native player ... especially if it could do DVD menu stuff .......
00:33<bline>well this isn't a player, it gives you the means to read and navigate though, which is half the battle :)
00:33<bline>I was just browsing the code
00:33<bline>it does seem pretty clean though
00:34<Chutt>captain_murdoch, could you get a backtrace of the hang again?
00:34<Chutt>with that fix in place..
00:34<bline>what plays data from a VOD file, is it it's own codec?
00:34<bline>err VOB
00:34<Captain_Murdoch>it doesn't hang anymore, it just closes the dialog and doesn't delete the file.
00:34<thor>bline, menu stuff is based on the notion of a DVD virtual machine
00:35-!-mechou [] has joined #mythtv
00:35<thor>simple states to show segments, toggle some registers, etc.
00:35<Captain_Murdoch>I don't think it would react to up/down keys though, I might have had to hit ESC to get out of the playbackbox. I'll try to get a bt when I can. not sure if I have anything I can delete.
00:35<bline>this seems to give you that atleast it looks like it does, but I see nothing for rendering
00:36<thor>could also steal a codebase from xine/ogle
00:36<Captain_Murdoch>found a show I can try deleting. I'll give it a shot under gdb.
00:36<mechou>anyone know what errno 145 is is mysql? It seems I lost all my data in the myth db....
00:36<Chutt>try commenting out the unlink/delete from db calls
00:37<Chutt>see if that changes anything
00:38<bbeattie>does anyone here use the qt3 libraries and development packages from debian unstable?
00:39<Chutt>you'll at least be able to delete things multiple times =)
00:39<Chutt>they should get removed from the ui, then re-appear in a bit
00:39<bline>ogle's code is not pretty
00:41<Captain_Murdoch>this is weird. there are VolumeControl::VolumeControl () lines in my backtrace and I just started mythfrontend and went straight to the playbackbox and hit the delete key then CTRL-C-ed.
00:44* blinegoes to bed
00:45-!-bline [] has quit ["Client exiting"]
00:45<mechou>OK, need some help re mysql here. Last night while I was sleeping mythtv log ate up /var partition, and now mysql complains "can't open program.MYI", but it exists.
00:45<mechou>any chance I can still recover the db?
00:46<Chutt>you can try mysqlcheck
00:47<mechou>mysqlcheck -r mythconverg?
00:47<Chutt>thor, you could always try to reverse engineer microsoft's dvd lookup stuff =)
00:48<thor>do they do that ?
00:48<thor>network call ?
00:48<Chutt>believe so
00:48<Chutt>i don't know any details
00:48<Chutt>and i'd say it's low priority =)
00:48<thor>(the data may be on the disc somewhere ... I know very little about DVDs =)
00:49<Chutt>it's not always on disc
00:49<thor>packet sniffer ..... but I'd have to install Windows somewhere ...
00:49<thor>ah, laptop has XP ..... does it work on XP
00:50<Chutt>i think it'd almost be easier to just include the imdb stuff
00:50<Chutt>from mythvideo
00:50<Chutt>allow people to input a title
00:50<thor>I'm inclined to start
00:50* paulproteuswould love to see the MS protocol reverse-engineered and abused :)
00:50<thor>seems crazy it doesn't exist already
00:50<mechou>chutt, "mysqlcheck --use-frm" reports two tables as crashed (size discrepancy). Show I just go ahead and do "mysqlcheck -r mythconverg"?
00:50<Chutt>there's a couple
00:51<Chutt>mechou, i have absolutely no idea
00:51<mechou>well, I did that anyway, now for the real test...
00:51-!-hadees [] has quit [Remote closed the connection]
00:55<mechou>seems to have worked. Thanks for the suggestion, chutt.
00:58<bbeattie>anyone: I just upgraded frmo qt2 to qt3 on debian, when trying to compile some programs such as mythepg, I get thousands of errors like this: ../../libs/libmyth/ undefined reference to `QComboBox::QComboBox(bool, QWidget *, char const *)'
00:58<Chutt>you need to recompile mythtv.
00:58<Chutt>and i highly doubt you upgraded from qt2
00:58<bbeattie>I am.. I've even done distclean
00:58<paulproteus>Probably didn't remove qt2.
00:59<bbeattie>I did an apt-get dist-upgrade to debian unstable from debian woody
01:00<mechou>you did double check QTDIR variable?
01:01<mecraw>how do I find out what is using /dev/dsp? mythbackend says it can't open it
01:01<paulproteus>Hmm, fuser. I'll look at that.
01:05<bbeattie>I had been using libqt 2.3.1-22 , I removed the entire qt2 library and anything related, I'm doing a distclean and make now.
01:06<mechou>don't forget ldconfig afterwards....
01:06<bbeattie>did that too, :)
01:06<mechou>well, just reminding ya...
01:08<bbeattie>thank you..
01:08<bbeattie>.. you arn't running debian by any chance are you?
01:08<bbeattie>I have libqt3 files under /usr/share/qt3/lib and /usr/lib,
01:09<mechou>sounds ok.
01:14<mechou>but I don't have qt3 files anywhere else besides /usr/lib/qt3, aside from /usr/bin and my docs dir...
01:14<bbeattie>what is your QTDIR set to?
01:15<Captain_Murdoch>Chutt: I don't even see it sending a delete request across the network.
01:17<Captain_Murdoch>PlaybackBox::showDeletePopup() sets noUpdate = true but never unsets, so when doDelete then doRemove are called, noUpdate is true so it returns rigth away or am I missing something.
01:29<Chutt>captain_murdoch, ok, in doDelete()
01:29<Chutt>move the noUpdate = false; to the beginning of that function
01:31<Chutt>that won't work.
01:31<Captain_Murdoch>I had it this way:
01:31<Captain_Murdoch>void PlaybackBox::doDelete()
01:31<Captain_Murdoch> popup->hide();
01:31<Captain_Murdoch> doRemove(delitem);
01:31<Captain_Murdoch> noUpdate = false;
01:31<Captain_Murdoch>the noupdate=false is unnecessary in that case really since doremove sets it false.
01:31<Chutt>it should be set before the doRemove, though
01:32<Captain_Murdoch>doremove sets it true right away though then false again.
01:32<Chutt>since doRemove should bail if it's set
01:32<Captain_Murdoch>so move the noupdate=false above doremove but after hide?
01:32<Chutt>might as well try that
01:33<Chutt>i think it may still hang, though =)
01:33<Chutt>or at least bring it back
01:33<Chutt>ah well
01:33<Captain_Murdoch>ok. I'll test and checkin to cvs to get it back working.
01:33<Chutt>actually, lemme do something
01:33<Chutt>i'll include that change
01:34<Chutt>and another additional check
01:34-!-mechou [] has quit [Remote closed the connection]
01:39<Chutt>checked in
01:40<Chutt>should fix it, though i didn't test it at all =)
01:43<Captain_Murdoch>cvs updateing now.
01:46<thor>ah, progress
01:47<Chutt>as in, you got progress working right? =)
01:47<thor>counting disc sectors
01:47<-- tmkhas quit ()
01:48<thor>just working out a little time calculation
01:50<bbeattie>Chutt: do you know of anyone who runs debian unstable?
01:50<Chutt>i do.
01:51<bbeattie>could you post the result of a dpkg -l|grep libqt please?
01:52<Chutt>all i have are the 3.1.1 debs that are in unstable.
01:54<bbeattie>.. there are actually 2 choices for which packages you are using, such as libqt3-dev vs libqt3c102-dev and libqt3-mt vs libqt3102-mt, etc
01:55<Chutt>you have to use the -mt packages
01:55<Chutt>it says that in the howto.
01:55<bbeattie>I am,
01:56<bbeattie>do you use libqt3-mt or libqt3c102?-mt
01:57<Chutt>there's no such package libqt3-mt
01:57<bbeattie>rc libqt3-mt 3.0.3-20020329 Qt GUI Library (Threaded runtime version).
01:57<bbeattie> - There was at one point, so I was wondering if you were using it, but I'll take that as a no
01:58<Chutt>there's no such package in unstable as that.
01:58<Chutt>if you have that choice, you're not running unstable.
02:09<bbeattie>Chutt: on your debian unstable, do you use gcc 3.2.3 or gcc 3.3.x?
02:15-!-billytwowilly [] has joined #mythtv
02:16<thor>you know what would be righteously kickass ?
02:18<Timon>whats that?
02:19<billytwowilly>naked ladies to change the channels instead of a remote?
02:19<billytwowilly>was I close?
02:19<thor>if we had a native dvd player, then you could have a preview window on the title ripping screen ... selecting angle, audio channel, subtitles, etc.
02:20<thor>or the nekid women
02:20<Timon>They have to have big boobies though :-)
02:20<billytwowilly>any sort of dvd player would be nice.
02:21<billytwowilly>heh, you could tweak the right nipple left for channel plus, right for channel minus and tweak the other nipple like that for volume plus or minus;)
02:21<billytwowilly>it's the perfect system;)
02:21<billytwowilly>thor: I recommend you commence work on this feature;)
02:21<thor>hard to get the right driver compiled for lircd though
02:22<billytwowilly>Just threaten the "compiler" with the "b1tch slap"
02:22<billytwowilly>works fine after that;)
02:24<billytwowilly>I'm trying to glean the timing of the next release of mythtv from the mailing list archives and it isn't working...
02:29-!-billytwowilly [] has quit ["Client exiting"]
02:29<Timon>thor, how do I get the title of the currently selected item in the mythmusic tree?
02:30<thor>hang on
02:31<Timon>k, thanks
02:32<thor>currently naviagted to, or currently playing (current or active)
02:32<Timon>Currently navigated to
02:33<thor>Managed Tree list emits nodeEntered(int, IntVector*)
02:33<thor>the int (if it's greater than 0), is a metadata index
02:34<Timon>k, thanks
02:45<thor>also, Chutt put in a new method in earlier today to just get the GenericTree node that is currently navigated to
02:45<Timon>hmm, ok
02:45<Chutt>it'd be nice to be able to store data in there as well
02:46<Chutt>so you didn't have to keep a second copy outside the tree
02:46<bbeattie>bah, found the problem, gcc was linked to 3.2 but g++ was linked to 2.95.. dang little bugger is an anoying problem on my part.
02:46<Chutt>and look things up all the time
02:46<thor>then it wouldn't be Generic =)
02:46<thor>that's work wonders
02:47<Timon>where is the managed tree mumbo jumbo? I can't find the .h for it
02:48<Chutt>thor, the mythgame stuff i did today builds the tree on the fly, it's nice and speedy =)
02:48<Timon>thor, the managed tree stuff is for playbackbox, I'm working with databasebox right now
02:49<thor>Chutt, cool
02:49<thor>Timon, oh ... sorry ...
02:49<thor>not paying attention
02:49<Timon>np :-)
02:50<thor>just ask the ListView
02:50<Timon>K, I was looking in ListViewItem
02:50-!-bbeattie [] has quit ["Client Exiting"]
02:50<Timon>lemme look at listivew
02:55-!-Edgan [] has quit ["Client exiting"]
02:57<thor>progress in
02:58<thor>now I can sleep, and tomorrow QProcess("transcode") ... yah
03:03<Timon>bah, sleep
03:17<Chutt>who the fuck signed mythtv-users up to the gmane archives
03:22<Chutt>some stupid mail->news gateway
03:25<Chutt>thor, progress code seems to work well
03:28<Chutt>eww, nasty
03:28<Chutt>select: Bad file descriptor
03:29<Chutt>mtd went dead
03:44-!-choenig [] has joined #mythtv
03:58-!-hfb [] has quit ["Client exiting"]
04:11-!-orangebits [] has joined #mythtv
04:22-!-Markos [] has quit ["Client exiting"]
04:58-!-Markos [] has joined #mythtv
06:46-!-choenig [] has quit [Remote closed the connection]
06:51-!-michal [] has joined #mythtv
07:10-!-choenig [] has joined #mythtv
07:27-!-poptix [] has quit [Read error: 60 (Operation timed out)]
07:45-!-lunk_ [~lunk@] has joined #mythtv
07:45-!-lunk [] has quit [Read error: 104 (Connection reset by peer)]
08:14-!-jkolb [] has joined #mythtv
08:51-!-moegreen_ [] has joined #mythtv
08:51-!-jkolb [] has quit [Read error: 104 (Connection reset by peer)]
09:07-!-moegreen [] has quit [Read error: 110 (Connection timed out)]
09:15-!-poptix [] has joined #mythtv
09:32-!-bline [] has joined #mythtv
09:38-!-jkolb [] has joined #mythtv
09:41-!-moegreen_ [] has quit [Read error: 60 (Operation timed out)]
10:01-!-mecraw [] has quit [Read error: 54 (Connection reset by peer)]
10:09<thor><Chutt> select: Bad file descriptor
10:09<thor>how did that happen?
10:09<Chutt>thor, when the mtd died
10:09<Chutt>i've fixed that, though
10:10<Chutt>going to send you a patch in a bit
10:10<Chutt>since i changed some other stuff
10:11<thor>great, thanks
10:11<thor>what killed the mtd ?
10:12<Chutt>i have no idea
10:12<Chutt>can't reproduce that
10:13<thor>hmmm ... ok ... I'll figure it out
10:14<Chutt>may be a good idea to make the default option be -n
10:15<thor>for development purposes, at least
10:16-!-Edgan [] has joined #mythtv
10:16-!-bline [] has quit ["Client exiting"]
10:24-!-lunk_ is now known as lunk
10:36-!-sfr [] has joined #mythtv
10:39<thor>Chutt, any of your changes touching the mtd ?
10:41-!-bline [] has joined #mythtv
10:41-!-schultmc [] has joined #mythtv
10:42<Chutt>i changed how the scan thread ran
10:42<thor>ah, I bet you keep the same thread
10:42<thor>and recycle it
10:43<thor>ok, nothing on job threads though ... thats where I am at the moment
10:44<thor>there shoudl really be some way to set a signal to say, I'll wait here, you tell me if the media changes
10:44<thor>from the kernel
10:47<sfr>any comments on the mythmusic patch i sent? no news isn't always good news.
10:47<thor>sfr, no time ... let me a have a quick look now
10:48<sfr>didn't mean to push anyone, i'm just courious what others think.
10:49-!-hfb [] has joined #mythtv
10:51<thor>sfr, rather then interleaving the SQL stuff, I'd break out to a different function() once you've figured out you're in netjukebox "mode"
10:52<sfr>thor, with function() you mean the SQL-related ones of Metadata, AllMusic and so on? or complete new classes
10:54<thor>sfr, yeah, the SQL-related ones
10:55<thor>just cleaner to say, if (netjukebox) goSomewhereElse() else keep going
10:55<bline>you guys never sleep
10:56<thor>lots of FIXME's .... =)
10:57<sfr>thor, well it's work in progress or not even that. :-)
10:57<thor>but nothing I'm idealogicaly opposed to.
10:57<thor>if you could isolate your code to a few separate functions, that would be nice
10:57<KikoV>is mailing list failing?
10:58<sfr>thor, but wouldn't i still have some if then... to check wether to call the netjuke function or mythmusic's one?
10:58<thor>would also make it easier to #ifdef around it in case we want to do a ./configure --netjukebox-support
10:58<thor>sfr, yup
10:58<thor>but only once, at the top
10:59<thor>rather than checking down through function blocks
11:00<Chutt>kikov: <>: Sorry, no mailbox here by that name.
11:00<sfr>thor ok, i didn't really know how to gon, so i started with the one with the least effort. But if separate functions are preferred, i'll go that route.
11:00<thor>also, instead of passing the flag around all over the place, you may want to have allmusic and playlists have an internal flag
11:00<KikoV>Chutt: it's pretty strange...
11:00<Chutt>you just bounced off.
11:00<Chutt>fix your mail server.
11:01<KikoV>Chutt: but I receive mails... :/
11:01<Chutt>your server is telling the lists that you don't exist
11:01<Chutt>it's _not_ a problem on my end
11:01<KikoV>Chutt: Ok... Thx a lot...
11:01<KikoV>Chutt: I have been using this email account for months... :P
11:01<sfr>thor, hm i pass one in 'new AllMusic' no?
11:02<Chutt>that really doesn't matter.
11:02<thor>sfr, you seem to be passing a string called musicmode around a lot. I'd just have a bool or an int in one place that knows what the mode is
11:03<thor>sfr, but I'm just scrolling through the code, so don't take anything I say as gospel
11:03<sfr>thor, ok, i thought for the future when XY different schemas need support :-)
11:04<Chutt>sfr, i certainly hope not
11:04<Chutt>that's really stupid.
11:04<sfr>Chutt, what about 1 ?
11:04<Chutt>i'm feeling that it's totally unnecessary as well
11:04<sfr>err, i mean 2.
11:04<Chutt>why not add support to netjuke for the mythmusic stuff?
11:05<thor>sfr, or .. what is it about netjuke that myth doesn't do ?
11:05<sfr>Chutt, what do you mean?
11:05<jkolb>Chutt: Have you always been so blunt when you disagree with someone, or is this a result of leading a successful project where you recieve lots of bad ideas?
11:05<Chutt>sfr, why dirty up mythmusic for a feature that only a couple people are going to use?
11:06-!-tmk [] has joined #mythtv
11:06<tmk>hey chutt
11:06<Chutt>jkolb, pretty much always =)
11:06<tmk>guess what someone just commited to the -devel cvs
11:06<tmk>fb driver
11:06<sfr>netjuke is just a web application, which allows streaming songs to remote players, download of songs,
11:06<Chutt>who wrote it?
11:06<tmk>i haven't tried it
11:06<tmk>matt yourst
11:06<thor>sfr, moves audio over http ?
11:07<tmk>but he says "it works great"
11:07<sfr>thor streaming, yes. over http i don't know for sure.
11:07<tmk>i gotta run to work
11:08-!-tmk [] has quit [Client Quit]
11:08<thor>sfr, you can always keep a private code tree, but I bet Chutt (and a lot of users) would be very happy to have some Myth-based way to do the same thing
11:08<Chutt>i just don't want the mythmusic code all messed up with different db schemas
11:09<thor>I understand
11:09<sfr>thor the streaming? probably, i just don't have the means to do it.
11:09<thor>sfr, ah
11:09<thor>well ... I'd never written a thread before ....
11:09<thor>now I'm droppin 'em with impunity
11:11<sfr>i wrote one in perl (or better used the skeleton from the perl cookbook) and extended it. but have zero knowledge of C/C++/Qt.
11:11<sfr>one = single-threaded server, tsts, hitting return too fast.
11:13-!-orangebits [] has quit ["Client exiting"]
11:14<sfr>and mythmusic isn't really meant as a 'music server'. e.g. listening music isn't possible while watching muted TV.
11:15<Chutt>it could be.
11:15<sfr>yeah, someone just needs to write the code... i just don't see me being capable of it.
11:16<sfr>but wouldn't the backend need to be extended for that? not mythmusic itself.
11:17<Chutt>not really
11:17<Chutt>the music decoding code should move elsewhere
11:17<Chutt>into libmyth or whatnot
11:19<KikoV>Chutt: could you give me the hostname who bounce my emails?
11:19<Chutt>which hostname?
11:19<Chutt>the one that's sending?
11:19<sfr>Chutt which books to learn C/C++ can you recommend? ;-)
11:19<KikoV>Chutt: or the IP...
11:20<KikoV><Chutt> kikov: <>: Sorry, no mailbox here by that name. -> where ?
11:20<Chutt>sfr, i can't, sorry.. didn't learn from a book, so..
11:21<KikoV>Chutt: thx!!! :] solved!
11:21<Snow-Man>Stupid motherfuckers.
11:21<Chutt>was what accepted the mail from colo
11:25<thor>sfr, don't know about completely introductory texts (and I think you're already beyond that). Anything by Meyers (Effective C++, more Effective C++, etc.)
11:26<mdz_>is there any way to disable blank frame marking?
11:26<Chutt>isn't it one of the config options?
11:26<bline>sfr: it's not really that hard to learn if you know a language already
11:26<sfr>thor, thanks will look into it.
11:28<sfr>bline right, i was actually surprised how much i (believe i) understood.
11:28-!-dopez [] has joined #mythtv
11:30-!-moegreen_ [] has joined #mythtv
11:33-!-Drikus [] has joined #mythtv
11:36-!-StarHeart [] has joined #mythtv
11:38-!-StarHeart [] has quit [Client Quit]
11:43<Captain_Murdoch>mdz: do you mean disable it during software encoding?
11:45<mdz_>Captain_Murdoch: yes
11:45<mdz_>I believe I turned off the option, but it is still recording the blank frames in the db
11:45<Captain_Murdoch>the setting doesn't affect the record-time detection. is it affecting performance?
11:45<mdz_>Captain_Murdoch: I'm trying to eliminate factors
11:46<mdz_>Captain_Murdoch: which database connection does it use?
11:46<mdz_>(in 0.10)
11:46<Chutt>why don't you just update to cvs?
11:47<Captain_Murdoch>oh. no way to turn it off in setup. easy to comment out though. just coment out a few lines around line 673 in NuppelVideoRecorder.cpp that declare and setup the CommDetect object.
11:48<Captain_Murdoch>I think it uses the main one and it locks/unlocks around uses.
11:49<Captain_Murdoch>if the commDetect variable is null then it doesn't try to detect or write to the db.
11:52-!-StarHeart [] has joined #mythtv
12:00<Chutt>thor, sent you that patch (finally)
12:03<thor>Chutt, thanks ... why not just commit ?
12:03<Chutt>wanted to make sure it was ok with you first
12:04<thor>ah ...
12:04<thor>I'd just change the wait(1) to wait(keepGoing())
12:04<thor>so that quit will work
12:04<thor>I'll deal with it
12:05<Chutt>ok =)
12:22<rkulagow_>chutt: please sync the docs. bbl (anya calls)
12:22<Chutt>i will, thanks =)
12:23<thor>Chutt, I don't think we want that have_disc = false before possibly sleeping for 6 seconds
12:24<thor>good god, I was calling run() directly .... doh!
12:25<Chutt>it's just on startup, right?
12:37-!-mchou [] has joined #mythtv
12:39<mchou>I have a disappearing mouse cursor problem in myth tv ui. I can't see the mouse cursor over combo boxes and the like in mythtv setup (not the setup binary)
12:39<mchou>anyone know how I ought to go about fixing this?
12:40<thor>arrow up/down to get to the box. arrow left/right to change values.
12:40<mchou>yeah, the highlight is tough to see, and s/t I end up changing the wrong option....
12:41<mchou>ist the missing cursor now the "expected" behavior?
12:42<Chutt>it's never had a cursor.
12:42<mchou>on one machine I get a cursor in the setup ui, on another I don't
12:43<lunk>mchou: use the keyboard?
12:43<mchou>I thought the cursor would show up on spinner controls....??
12:44<Chutt>it's not supposed to.
12:44<Chutt>use the keyboard.
12:45<mchou>yeah, I'm using the keyboard. hmm, so you guys are saying that my other machine with cursor has "unexpected" behavior...
12:45<lunk>i'm saying i don't know what a mouse is
12:51-!-rkulagow___ [] has joined #mythtv
12:59<-- mchou( has left #mythtv
13:09-!-Markos [] has quit ["Client exiting"]
13:17-!-choenig [] has quit [Remote closed the connection]
13:33-!-Markos [] has joined #mythtv
13:34<Markos>has anyone had any weird errors like VIDEOSYNC0: error
13:35<Markos>in their mythbackend.log
13:35<Markos>last night myth failed to record a program and filled up the log with that
13:35<Markos>to the tune of about two gigabytes worth
13:37<Chutt>that's why you don't log stuff like that.
13:37<Chutt>basically, your tuner card decided to die
13:38<Chutt>well, the driver, really, but..
13:38<Markos>hmm, that is strange i've left this computer on for weeks and the card has always worked
13:39<Markos>mythbackend was running for only 5 days or so when that happend
13:39<Markos>dmesg showed no bttv errors or anything like that, looked fine
13:39<Chutt>that's an error from bttv
13:40<Markos>it showed VIDEOSYNC0: and VIDEOSYNC1:
13:40<Markos>over and over
13:41<merf>Chutt: did anyone ever figure out about that lockup-on-delete bug?
13:41<Chutt>merf, that's fixed in cvs.
13:41<Markos>is this a mythtv problem or a bttv problem, its never happened with tvtime or xawtv
13:41<Chutt>markos, bttv
13:42<merf>and while i think of it.. any reason why myth would record for waay longer than it's supposed to (one hourlong episode of a show ended up being 6gb)?
13:42<Chutt>i've said that twice.
13:42<merf>Chutt: when's it gonna find its way into the debs? :D
13:42<Markos>well i'd be inclined to think that its myth since its the only software that has done this
13:42<Chutt>uh, next release?
13:43<merf>that makes sense
13:43<Chutt>markos, why the hell are you arguing with me?
13:43<merf>you said on the news that the .15 XMLTV bug made its way into the debs, but .15 wouldn't work with them (at my end anyways)
13:43<merf>for what it's worth
13:43<Chutt>have you even looked at the code in question?
13:43<Markos>i'm not arguing, i'm just expressing my opinion
13:43<Markos>unless that isnt allowed in this channel
13:43<Chutt>it's not a matter of opinion.
13:44<Chutt>the _only_ way that error shows up is if the bttv driver fails.
13:44<merf>and when i say bug, i mean bug fix. cause i'm an idiot.
13:44<Chutt>it's not possible for it to be a mythtv problem.
13:44<Markos>i find it odd that it would only happen with mythtv and it not be a mythtv problem, is all
13:44<Chutt>i find it odd that you have absolutely no idea what you're talking about, yet you're still talking
13:44<merf>Markos: have you tried using any other tv-related apps since it's happened?
13:45<merf>cause maybe your driver just never took a crap all over itself before
13:45<Markos>yeah they work fine
13:45<Markos>Chutt: i was just asking questions no need to be prissy
13:45<Chutt>do you _want_ to be removed from the channel?
13:45<Chutt>i answered your question.
13:45<Chutt>you ignored my answer.
13:46<Chutt>why the fuck do i even bother.
13:46<merf>cause you're a damn nice guy.
13:46<Markos>wow, kicked for discussing a possible bug
13:46<Chutt>it's not a bug, dumbass.
13:46<Chutt>the bttv driver _failed_
13:46<merf>you guys need some booze
13:47<Chutt>mythtv asked it to queue a buffer to record from, it said 'no, i can't do that'
13:47<thor>I'll drink to that
13:47<Chutt>mythtv asked it again to queue a buffer to record from
13:47<Chutt>it failed again
13:47<Chutt>rinse, repeat.
13:47<Markos>why would this behaviour only happen with mythtv, and why would the other tv apps still work?
13:47<jkolb>You forgot wash.
13:48<Chutt>markos, because it's an intermittant bug?
13:48<merf>maybe your tuner hates you and you should sell your computer and buy a poster
13:48<sfr>Markos, do you leave 'other tv' apps running for days without stopping them?
13:48<Markos>well thats a bit better then calling people dumbasses and shit
13:48<Chutt>if you don't want to be called a dumbass, don't act like one
13:48<Chutt>it's really simple.
13:49<Markos>i just asking for more clarifacation and you jump all over me
13:49<Markos>i dont think i've done anything to warrent that kind of response
13:49<Chutt>you didn't ask for clarification
13:49<merf>you weren't obliged to get an answer, either
13:49<Chutt>you contradicted me
13:49<merf>but you did
13:49<merf>so quit whining
13:50<Markos>if you didn't want to help me you didn't have to reply
13:50<Markos>another case of elitest coders it seems
13:51<Chutt>no, another case of dumbass user.
13:51<merf>i'm not a coder
13:51<merf>hell, mythtv's source scares the piss out of me
13:52<Markos>the developers scare me, this is like the freebsd mailing lists or something
13:52<Markos>ask a question and get flammed to hell
13:52<Chutt>you didn't get flamed until you contradicted me.
13:52<Chutt>more than once, in fact
13:52<Markos>and rather then tell me why i was wrong, you go out of your way to insult me
13:52<Chutt>i told you why you were wrong
13:52<Markos>such a childish way to act
13:52<merf>he did tell you why you were wrong
13:53<Chutt>you didn't listen the first two times.
13:53<merf>like four times
13:53<Chutt>then i got pissed off
13:53<Chutt>i've got better things to do, later.
13:54<Markos>yeah and after asking a couple of follow up questions he bites my head off
13:54<Chutt>[13:43] <Markos> well i'd be inclined to think that its myth since its the only software that has done this
13:55<Chutt>yeah, that's a follow up question
13:55<Chutt>oh, wait, where's the question in there?
13:55<Markos>yes thats what i thought
13:55<Markos>and if you just explained why i was wrong, i would have said
13:55<Markos>ah i see, thank you
13:56<Markos>not you are an idoit because you dont understand the source
13:56<merf>which he did
13:56<merf>and you didn't
13:56<Markos>not before calling me a dumbass
13:56<merf>but he explained it
13:56<merf>dumbass or not
13:57<Chutt>the very first thing i said explained it.
13:57<merf>Chutt: is that myth-recording-till-the-next-day thing i described above a mythtv bug or something to do with xmltv giving the wrong times?
13:57<Chutt>merf, i haven't seen it do that, so..
13:58<Chutt>it's easy to check to see if xmltv was wrong, to rule that out
13:58<Markos>yeah and i thought you were mistaken, and then you just get all defensive and surly and shit
13:58<merf>when i get home then, i'll take a looksee
13:59<Chutt>see, that's the issue. if i bother to answer about a bug, i'm not mistaken.
13:59<Chutt>especially something as simple as this.
14:00<Markos>and rather then tell me i'm wrong you feel the need to insult me and then tell me why i'm wrong in an elitest way
14:00<Markos>really not a way developers should be actiong
14:00<Chutt>i told you you were wrong _several_ times.
14:00<Chutt>i can act however i damn well please
14:00<Snow-Man>Exactly the way developers should be acting.
14:00<Chutt>i don't give a damn if you use my software or not
14:01<Snow-Man>This is why open source works, developers of open source software don't have to give a shit about users. :)
14:01<Markos>its childish, rather then explain something just flame people because they ask a question
14:01<merf>he did explain it
14:01<merf>and now you've been whining for 15 minutes
14:01<Chutt>i fucking _explained_ it to you before anything
14:01<merf>instead of taking the valuable answer and working on the bug yourself
14:01<merf>you could probably have it fixed by now
14:01<Snow-Man>Oh, just fucking ban him already.
14:01<Chutt>i'd have to identify with nickserv
14:01<Chutt>and that's a pain
14:02* Snow-Manshrugs.
14:02<Chutt>'sides, i'm just waiting for this slowass box to recompile stuff
14:02<Markos>now i get a ban, wow, this is like nazi germany in here
14:02<sfr>Markos, you lost.
14:03<Snow-Man>Amusing distraction for a few seconds, eh? :)
14:03<Chutt>have you gotten banned?
14:03<Chutt>you're still talking in here
14:03<Markos>i feel that its only a matter of time at this rate ;>
14:03* thorcan't believe someone said nazi ... isn't that a usenet rule
14:03<Chutt>thor, yeah
14:03<Chutt>it is =)
14:03<sfr>well this is like usenet in realtime
14:03<Chutt>thor, i can't get mtd to die again
14:04<merf>Chutt: well, according to mythconverg the times are alright. could it have something to do with the show being played twice on the same channel (once at 2pm, once at midnight)?
14:04<thor>Chutt, that's a good thing, right?
14:04<Chutt>merf, shouldn't be
14:04<merf>i know nothing about how myth decides when to stop recording
14:04<Chutt>thor, yeah
14:04<Chutt>thor, 13 minutes to rip a 2 hour movie
14:04<Chutt>not bad :p
14:04<thor>try it agian with updatedb on =)
14:05<Chutt>actually, i need to figure out why the entire computer slows down so much when just doing that
14:05<Chutt>even reniced down low
14:05<merf>dma is off ;D
14:05<Chutt>naw, dma's on
14:05<bline>what filesystem?
14:05<Chutt>it's using 10% cpu
14:05<thor>drive contention, no ?
14:05<merf>maybe a gremlin stole your read buffer
14:05<bline>maybe try reiserfs
14:05<Chutt>works fine with recording video
14:05<Chutt>should be fine to just copy off of a dvd :p
14:05<merf>speaking of resierfs.. is it normal to have reiser use 8gb for its journal (presumably) on a 120gb disk?
14:06<thor>nice load test for the mtd though
14:06<bline>I've seen a massive speedup on servers we put reiserfs on
14:06<Chutt>bline, it shouldn't really matter for huge files
14:06<bline>Chutt: I thought it was updatedb that was slowing things down
14:06<Chutt>no, copying off of a dvd
14:07<Chutt>thor, how often does the UI update itself, btw?
14:07<Chutt>X is taking 10% cpu =)
14:07<thor>every second, I think
14:07<bline>is the dvd on the same channel as the harddrive?
14:07<merf>can you append two videos with mencoder?
14:07<Chutt>bline, nope
14:07<thor>progress bitmaps have alpha maybe ?
14:07<Chutt>yeah, maybe
14:07<merf>i tried with avidemux and it messed up a couple hundred frames where it joined them
14:08-!-vektor [~bbiggs@nat-pc.cs.Dal.Ca] has joined #mythtv
14:08<Chutt>vektor, yo
14:09<Chutt>not much
14:09<bline>someone was looking for you yesterday
14:10<thor>that was me
14:10<thor>too late
14:10<thor>I figured it out
14:10<merf>vektor: you go to dal?
14:11<vektor>do you?
14:12<thor>dal as in Dalhousie ?
14:12<vektor>thor: yeah
14:12<thor>as in Nova Scotia ?
14:12<vektor>yes, that's where i am
14:13<thor>got a cousin in the law school there
14:14<merf>no, i don't go to dal. i may the year after next though.
14:14<merf>my mom's a prof there, however
14:14<vektor>that's cool.
14:15<thor>merf, which department ?
14:15<merf>she teaches history for the costume studies department
14:16<merf>and my dad was the stage director for the cohn five years ago or so
14:16-!-bbeattie [] has joined #mythtv
14:16<thor>cohn ?
14:16<merf>incoming dumb question: how the hell do i let root display windows on a user's X session?
14:16<merf>thor: the rebecca cohn auditorium.. it's in the dal arts center
14:17<thor>xhost +
14:18<merf>and all this time i'd been doing xhost +localhost
14:19<thor>note, xhost can be very dangerous
14:20<merf>i know ;)
14:22-!-Timon [] has quit [Remote closed the connection]
14:23-!-bbeattie [] has quit []
14:23-!-dopez [] has quit []
14:23-!-hfb [] has quit []
14:23-!-sfr [] has quit []
14:23-!-Edgan [] has quit []
14:23-!-Justin_ [] has quit []
14:23-!-bigguy [] has quit []
14:23-!-jfm_ [] has quit []
14:23-!-tdb30_ [] has quit []
14:23-!-fulbert [] has quit []
14:23-!-bbeattie [] has joined #mythtv
14:23-!-dopez [] has joined #mythtv
14:23-!-hfb [] has joined #mythtv
14:23-!-sfr [] has joined #mythtv
14:23-!-Edgan [] has joined #mythtv
14:23-!-Justin_ [] has joined #mythtv
14:23-!-bigguy [] has joined #mythtv
14:23-!-jfm_ [] has joined #mythtv
14:23-!-tdb30_ [] has joined #mythtv
14:23-!-fulbert [] has joined #mythtv
14:25<-- Markos( has left #mythtv ("Client exiting")
14:25<Chutt>guess i'll add a backend protocol version
14:25<merf>i wish i could get alsa working
14:27-!-choenig [] has joined #mythtv
14:28<Chutt>choenig, did you ever send an updated mythmusic patch to the list?
14:29<choenig>Chutt: nope, having a test tomorrow I had now time yet
14:35<bbeattie>Chutt: I posted a message, it may have been lost in the net split?
14:36<Chutt>i didn't see it
14:37<bbeattie>basically I still havn't been able to get channel change working. A simple program I wrote to change it does, and I put the changes into myth where I could see that they were needed, but I must have missed something.
14:38<bbeattie>The ioctl I set is ioctl(chanfd, VIDIOCSCHAN, &ch [ = struct video_channel {4,"chan --",1,VIDEO_VC_TUNER,VIDEO_TYPE_TV,VIDEO_MODE_ATSC}];), then I do a freq set ioctl, that's it.. where all may I be missing ioctl sets?
14:38<bbeattie>does the videosource have anything that could be needed to be done, or just tv_rec, channel, and the hdtvrecorder?
14:38<Chutt>channel is the only place that sets the channel
14:38<Chutt>it may not be doing the VIDEO_MODE_ATSC, though
14:38<Chutt>unless you changed that
14:39<bbeattie>I changed that, I set the mode for that,
14:39<bbeattie>I shouldn't need to change the freq ioctl at all I'd think,
14:40<bbeattie>in the structure I posted, do you see anything besides the VIDEO_MODE_ATSC that could be different and needed in myth?
14:40<Chutt>i don't think so
14:40<Chutt>but i'm not entirely remembering that
14:40<bbeattie>(I also set the videomode to V4L2_STD_ATSC_8_VSB, maybe that's wrong?
14:41<bbeattie>Myth actually fails to start until I set the ioctl in my program, then it just shows whatever channel I set it to last. :-)
14:42<bbeattie>myth makes a lot more calls then I need ;)
14:43<bbeattie>such as the setformat,
14:45<bbeattie>.. what does that do? something for myth only?
14:45<Chutt>which call?
14:45<Chutt>at the top of mpegrecorder?
14:45<bbeattie>No, in channel
14:45<Chutt>that goes through and gets all the inputs on the cards
14:45<Chutt>and sets them to NTSC mode or whatnot
14:45<bbeattie>If the video4linux wasn't selected in the setup for the card, that is bypassed, correct?
14:45-!-moegreen_ [] has quit []
14:45-!-ahbritto [] has quit []
14:45-!-vektor [] has quit []
14:45<Chutt>no, not unless you made a new type and a new channel class for it
14:47<bbeattie>I made a new cardtype, but that was mostly it (and the hdtvrecorder class
14:47<Chutt>you may want to do your own channel class
14:50-!-mecraw [~mecraw@] has joined #mythtv
14:54-!-moegreen_ [] has joined #mythtv
14:54-!-vektor [~bbiggs@nat-pc.cs.Dal.Ca] has joined #mythtv
14:54-!-moegreen_ [] has quit [Read error: 60 (Operation timed out)]
14:54-!-moegreen_ [] has joined #mythtv
14:55-!-ahbritto [] has joined #mythtv
15:02-!-michal [] has quit [Read error: 113 (No route to host)]
15:11<jkolb>Does Chris Petersen (New Mythweb guy) show up here?
15:15<Chutt>i've never seen him to
15:16<Chutt>unless he's been quiet and used a non-obvious nick =)
15:17<jkolb>He replaced the MythTV logo with a png, and to get the transparency to work in IE, he does ... odd things with it. When I tried to port over the preview stuff, the transparency hacking he does makes all my images disappear.
15:19<jkolb>It's frustrating to see them download, display, then get replaced by a black 1x1 image.
15:19<Chutt>yeah, i can imagine
15:20<jkolb>So I fixed it by commenting out where he replaces the image src with the 1x1 gif, but then the transparency doesn't work on the logo.
15:21<Chutt>different image type for the logo, then?
15:22<jkolb>I was thinking gif.
15:22<Chutt>would be fine with me
15:25<jkolb>I'd make one, but I don't have an image editor.
15:27<jkolb> wouldn't be needed any longer, either.
15:27-!-ahbritto [] has quit []
15:28<Chutt>you'd have to email it to chris
15:28<Chutt>want stuff to go through him for his code
15:28<jkolb>Will do.
15:28<Chutt>cool, thanks =)
15:30<jkolb>Um, do you have an address other than, or do you think that would be the best address to use?
15:30<Chutt>use that, i think
15:33-!-ahbritto [] has joined #mythtv
15:33-!-ahbritto [] has quit []
15:33-!-ahbritto [] has joined #mythtv
15:39-!-ahbritto [] has quit []
15:51-!-ahbritto [] has joined #mythtv
15:55<jkolb>Well, suck. I just destroyed my recorded table because starttime is a timestamp, and I wasn't careful.
15:56<Chutt>there's a perl script in contrib/ that reconstructs that
15:56<Chutt>i think
15:56<Chutt>and i will accept a patch that changes the column types there
15:56<mdz_>jkolb: but your database is a few kb smaller :-)
15:56<jkolb>I didn't have anything I can't get again, so I just nuked the table.
15:56<Chutt>just no one's ever submitted one
15:56<jkolb>And my drive has ~100G more free space
15:58<sfr>100G more? my drive only has ~110G. tsts
15:59<jkolb>I got a 200G for my myth box.
15:59<Chutt>i'm going to have to buy a larger drive
15:59<jkolb>I was actually down to ~10G at one point.
15:59<sfr>and you watched all recordings?
16:00<jkolb>Nah. I record Law & Order anytime, anywhere, so it's damn near impossible to watch everything I record.
16:00-!-michal [~michal@] has joined #mythtv
16:01<sfr>jkolb but it's a good feeling knowing you could, right?
16:01<jkolb>If I stayed up all night and came into work all bleary-eyed every day, I *suppose* I could.
16:01<jkolb>I don't think they'd like me here too much if I did that, though.
16:03<thor>tee hee
16:14<jkolb>Is the starttime in the transcoding tables actually used as a timestamp?
16:15<jkolb>It's just a copy from the recorded table, right?
16:17<jkolb>Ok, then there's a patch available at
16:18<Chutt>does it work? :p
16:18<jkolb>I really should have worked nazi into the title somehow.
16:25<jkolb>It seems that Chris didn't know myth had an IRC channel.
16:32<jkolb>Damn. That patch is incomplete.
16:39<jkolb>I'm assuming that the lastplayed column in musicmetadata is really a timestamp, and not just a datetime in disguise?
16:42<jkolb>Right, but you actually use the auto-updating feature of a timestamp column, and don't just use it to store a datetime, right?
16:42<thor>uhm ... let me check
16:46<thor>playbackbox tells the metadata what time it is whenever a track is played, the track stores that in memory
16:46<thor>when you quit mythmusic, that change is written to the database
16:46<thor>lastplay is explicitly set
16:46<jkolb>So you set it explicitly? It should be a datetime, and not a timestamp?
16:47<thor>as you may exit mythmusic several days after the value was set
16:47<thor>since it only touches the db on exit
16:51<thor>excellent, nazis
16:51<jkolb>Yeah, it's not very clever, but I had to.
16:58-!-mechou [] has joined #mythtv
17:08<-- mechou( has left #mythtv
17:13-!-sfr [] has quit ["Client exiting"]
17:16<jkolb>Chutt: Acceptable? That's all the timestamps in the db.
17:27<Chutt>yeah, probably
17:27<Chutt>i'm wondering if the insert code in mythtv is compatible
17:27<Chutt>as don't they have different formats?
17:27<jkolb>Do they?
17:27* jkolbsighs
17:28<jkolb>Well, crap.
17:28<Chutt>timestamp is yyyyMMddhhmmss
17:28<Chutt>datetime is yyyy-MM-ddThh:mm:ss
17:28<jkolb>This is going to suck if I actually have to do *work*.
17:28<Chutt>i dunno
17:28<Chutt>the server may be smart enough to translate
17:28<Chutt>just needs testing
17:29<jkolb>YYYY-MM-DD HH:MM:SS for datetime
17:29<jkolb>So the T is the difference. For 'Timestamp,' maybe?
17:29<Chutt>no, the T's part of the iso date specification
17:30<jkolb>Heh. I read what you said backwards.
17:32<jkolb>Datetime, Date, and Timestamp can all accept the same formats on an insert/update, it seems. Or are you thinking about selects?
17:32<Chutt>i dunno
17:32<Chutt>was going for both
17:32<Chutt>i think qt may deal with them both internally
17:32<mdz_>I think it is smart enough to do everything right on writes
17:32<Chutt>there may be code in there that assumes the timestamp format for selects
17:32<mdz_>on select, everything should get converted into a qdatetime anyway
17:33<Chutt>mdz, i'm not sure if anything's just dumping it to a qstring, though
17:33<mdz_>except for stuff done in the query itself, which should be compatible
17:33<jkolb>So it's only mythweb that may suffer?
17:33<mdz_>no idea about mythweb
17:36<jkolb>Well, I ran that on my db, and mythweb2 can still display the program listings.
17:37<jkolb>And it's giving sane times for scheduled recordings.
17:37<Chutt>need to look at all the selects in mythtv, though
17:37<Chutt>to make sure they're converting directly to qdatetimes and not going to a qstring
17:53-!-Drikus [] has quit ["toedeledoki"]
18:09-!-tmk [] has joined #mythtv
18:16-!-choenig [] has quit [Remote closed the connection]
18:24<mecraw>would it be possible to include the diffs in the cvs commit emails?
18:27-!-jkolb [] has quit [Read error: 110 (Connection timed out)]
18:34<Chutt>mecraw, i've thought about doing that, but it's just as easy to go through viewcvs for em
18:43<mecraw>i just find it a little easier to gradually get into the code to see the day to day changes without bringing up a web page
18:48-!-jkolb [] has joined #mythtv
18:49<jkolb>Well, that was fun. The power company measured my meter a week ago and came up with a number 1600 KWh higher than it reads today.
19:00-!-mecraw [] has quit ["Trillian ("]
19:15-!-michal [] has quit [No route to host]
19:21<Chutt>moegreen, do you see the issue with mythweather (from the most recent post to -dev)?
19:22<Chutt>the debugging output explains it =)
19:24<just1nux>chutt: is there any support for this chipset? Conextant Fusion 878a Chip
19:24-!-tmk [] has quit ["Client exiting"]
19:25<Chutt>just1nux, probably the bttv driver
19:26<just1nux>hrm... but this card is supposed to be an mpeg-2/1 encoder as well?
19:26<Chutt>which card?
19:28<just1nux>kworld computer mpeg TV Station card. picked it up cheap just to see if it would work at all.
19:28<thor>mpeg is in software, very misleading marketing speak on the box
19:29<just1nux>how is 500 mhz pc supposed to do software mpeg? according to box specs.
19:29<Chutt>very low resolution
19:30<just1nux>i guess it would have to be.
19:37-!-bbeattie [] has quit ["Client Exiting"]
19:43<just1nux>they shouldnt be allowed to put mpeg in the name of a product unless the card has a hardware encoder on it. Its a nice low profile card though, came with a neat little remote, fm-tuner, only 40 bucks.
19:45-!-dopez [] has quit [Read error: 104 (Connection reset by peer)]
19:47<moegreen_>Chutt: I still can't reproduce it. Though he has finally given me what I asked for in the original email I sent him
19:49<moegreen_>and his original backtrace showed him breaking in a different place
19:49<Chutt>moegreen, see what it's doing, though?
19:50-!-schultmc [] has quit ["Client exiting"]
19:50<Chutt>so it's sitting in that loop, waiting for a response, or a timeout
19:50<Chutt>the urlTimer finally timesout, so in weatherTimeout(), it calls GetWeatherData()
19:51<Chutt>and restarts the entire process
19:51<Chutt>it stacks things up
19:51<Chutt>and resets stopProcessing in the process
19:52<Chutt>so stuff eventually just gets stuck in a loop
19:53<Chutt>there probably needs to be another way to restart the query
19:53<Chutt>instead of calling GetWeatherData() again
19:54<Chutt>also, i think you're leaking the INETComms objects you're newing there =)
19:54<moegreen_>if it's trying to get the weather map - which his email shows, it's using the QNetworkOperation with the urlTimer stopped
19:55<Chutt>are you looking at the bottom one, though?
19:55<moegreen_>delete weatherMapLink2; <- unless I added that last night (and the others)
19:56<Chutt>right, url timer's stopped
19:56<Chutt>but since it's in GetWeatherData for the second or third time
19:56<Chutt>it's waiting for the the urlTimer to shoot and set stopProcessing to true or whatnot
19:57<Chutt>i _think_ it can be fixed by just removing the GetWeatherData() from the timeout function
19:57<Chutt>to not make it recursive
19:59<moegreen_>but how should it restart then? I guess I could just set one timer for the GetWeatherData call from UpdateData()
19:59<Chutt>i dunno
20:00<Chutt>i'd probably make the thing that calls GetWeatherData() originally try to call it again if it failed
20:03<Chutt>that all make sense?
20:04<moegreen_>yeah, I think this should fix it :)
20:05-!-hfb [] has quit ["Client exiting"]
20:07<moegreen_>i can't think clearly today ... it's been a real problem
20:09-!-bline [] has quit ["Client exiting"]
20:09<moegreen_>and to think - it wasn't his strace, the backtrace that made it clear what was going on. think :)
20:10<moegreen_>I like the message sent in earlier: I do NOT allow http traffic to leave my network directly
20:11<moegreen_>heh, someone else requested that earlier, but since I don't have a proxy server ... it'd be hard for me to be sure a solution actually works
20:11<Chutt>doing proxy stuff without a server to test on is a pain
20:12<moegreen_> that INETComms is in libmyth - someone could send in a patch to do the work there :)
20:15<jkolb>Chutt: I have a squid proxy set up if you'd like to test with it.
20:15<jkolb>Not that it will be fast or anything.
20:15<Chutt>there isn't an easy way to do proxy stuff in Qt
20:16<Chutt>no 'setProxyHost' or anything =0
20:16<Chutt>err, =)
20:16<Chutt>if it were easy, i'd do it, but since it's not, i'll leave it to someone that wants that functionality
20:16<jkolb>Ok. I hadn't read moe's statement before saying that. I thought you were looking for one.
20:21<moegreen_>There might be something new in 3.2, but I haven't looked
20:29<moegreen_>Chutt: just committed a fix, let me know what you think :)
20:30-!-StarHeart [] has quit ["Client exiting"]
20:30<Chutt>i'll check it out
20:32<Chutt>that looks right
20:40<Chutt>bugs left: joe caputo's long startup times
20:41<Chutt>well, major ones
20:41<Chutt>that i can think of from the mailing list
20:41<thor>long startup time on what?
20:41<Chutt>he says it takes 10 seconds or so to start livetv mode
20:41<thor>ah ... odd
20:42<Chutt>i'm thinking something's off in the prebuffering code
20:42<thor>naw, he's just got a really slow computer
20:42<Chutt>my nehemiah should be slower, and that's fine
20:42<Chutt>i dunno what's up
20:42<thor>irblaster thing?
20:43<Chutt>doubt it
20:47<-- jkolbhas quit ()
20:56-!-mikegrb [] has quit [Read error: 60 (Operation timed out)]
21:00--> Thexder-( has joined #mythtv
21:00<Thexder->Anyone have issues compiling lirc and have a compile error with TOPDIR: parameter null or not set ??
21:16-!-Timon [] has joined #mythtv
21:26-!-mikegrb [] has joined #mythtv
21:32-!-kslater [] has joined #mythtv
21:33<kslater>I need some visual help from anyone with a PVR-250.
21:35<-- Thexder-has quit (Read error: 110 (Connection timed out))
22:02-!-bline [] has joined #mythtv
22:19-!-bbeattie [] has joined #mythtv
22:59-!-kslater [] has quit ["Client Exiting"]
23:16-!-rkulagow___ [] has quit [Read error: 110 (Connection timed out)]
23:16-!-rkulagow___ [] has joined #mythtv
23:17-!-rkulagow_ [] has quit [Read error: 110 (Connection timed out)]
23:17-!-rkulagow_ [] has joined #mythtv
23:19-!-tmk [] has joined #mythtv
23:21* tmkgets a pen
23:22* tmklooks at chutt expectantly
23:22<Chutt>so the fb driver isn't quite stable =)
23:23<bigguy>fb driver?
23:23<tmk>oh ya?
23:23* bigguyperks up
23:24<Chutt>first of all, i couldn't get it to unload
23:24* tmkdidn't write it
23:25<Chutt>yeah, i knwo
23:25<tmk>where's the patch?
23:25<Chutt>and some stuff worked on it
23:25<tmk>did it output video tho?
23:25<tmk>he said he updated it today again
23:26<Chutt>hm, so he did
23:27<Chutt>what else changed in there?
23:27<tmk>uhmm.. i think he changed the iomap() type
23:28<tmk>and merged in the iomap patch from the main tree
23:28<tmk>I didn't do anythin'
23:29<tmk>i checked and v4l2 doesn't even address a codec-decoder device
23:29<tmk>so i'm on my own
23:29<Chutt>basically, all i really need is a way to get the current playing position
23:29<tmk>oh that's eays
23:29<tmk>by frame ok?
23:29<Chutt>that's best
23:29<tmk>well, it will be done
23:29<Chutt>and it'd be nice to give it some data and tell it to play a max of X frames
23:30<tmk>it'd be best to send it a frame at a time
23:30<Chutt>if that's possible?
23:30<tmk>and just stop sending it data
23:30<Chutt>just thinking about the edit-mode stuff
23:30<Chutt>where i need frame-exact seeking
23:30<tmk>i don't think it can rewind past its buffer
23:30<Chutt>no no
23:31<tmk>there is a FF mode
23:31<Chutt>i'll be sending it data starting at the previous keyframe
23:31<Chutt>to where it should go
23:31<Chutt>it's just the < 15 frames to decode after that
23:31<tmk>hmm what about a 'video-meta' device
23:31<tmk>that'd spit out info about the current cap
23:32<tmk>ie "just decoded frame 1232'
23:32<Chutt>that'd be fine
23:32<Chutt>or just an ioctl i can query every now and then
23:32<tmk>that's easier
23:32<tmk>i think it may send an irq after each decoded frame
23:32<tmk>haven't checked that
23:33<Chutt>you _can't_ just send it frames of data and stop
23:33<Chutt>the whole coded order vs display order thing
23:33<Chutt>an mpeg file is played as IBBPBBPBBPBB, right
23:34<Chutt>it's in the file as IPBBPBBPBBPBB
23:34<Chutt>out of order
23:35<Chutt>so frame-exact positioning can't really be done by just sending it the right number of frames
23:35<tmk>it does have a FF mode
23:36<tmk>i wonder if you can pause it
23:36<tmk>then run fwAPI call 01
23:36<tmk>0x1 rather
23:37<tmk>it has a 0-based frame to start from
23:37<Chutt>i can potentially get around that by decoding in software
23:37<tmk>so pause, then say 'start playback from frame 1000'
23:37<Chutt>and sending it the image
23:37<tmk>if you stop playback, you can tell it to keep the image on screen
23:37<tmk>not jsut do black
23:38<Chutt>hey, another thought
23:38<Chutt>is there any way to send raw audio data through the card?
23:38<Chutt>like, to play other stuff..
23:38<tmk>i hvaen't seen anything
23:39<Chutt>guess i can buy a couple y adapters or something
23:39<tmk>presumably you could send mpg
23:39<tmk>it's possible that you could send data directly
23:39<tmk>and that it's just not documented
23:39<Chutt>and hook the decoder audio outs to the regular audio outs
23:39<tmk>ie set sub-stream type to 2 or something
23:40<tmk>so all you really want is a 'current frame' ioctl
23:40-!-lmetzger1 [] has joined #mythtv
23:41<Chutt>and osd stuff to blend crap onto the video
23:41<tmk>Chutt: see fwAPI 0x15
23:41<Chutt>i imagine that's going to be something like 'here's a buffer, draw to it, tell me when it's updated'
23:41<tmk>does that do you for frame info?
23:41<tmk>do you want it to block until updated
23:41<tmk>or notify you later somehow
23:42<Chutt>blocking's fine
23:42<Chutt>so would be just returning the current info
23:42<Chutt>either way, 25/30 times a second, so it wouldn't block for long
23:43<tmk>looks like it CAN do rewind
23:43<tmk>but you have to be smart about which data you're sending it
23:43<Chutt>i don't really need to do rewind
23:43<Chutt>i keep track of all GOP starts
23:43<Chutt>so i can just seek to those and start sending data from there
23:44<tmk>any 'nice-to-haves'?
23:44<Chutt>not really, no
23:44<Chutt>i can't imagine any of these apis being _terribly_ complicated
23:44<tmk>cool.. really sounds like no problem
23:45<tmk>nah, they're pretty basic
23:45<Chutt>d'ya ever see the 'bottom half of frame corrupted with old video data' bug?
23:46<Chutt>never happens on my -350, get it fairly often on my -250-2
23:46<tmk>no can't say that I have
23:46<tmk>i have a 250-rev1
23:46<Chutt>it's pretty noticeable =)
23:46<tmk>and a 350
23:46<tmk>do you have latest firmware?
23:46<Chutt>latest from their website
23:46<tmk>i think i'm using the hauppaute 1.6 version
23:46<Chutt>not the latest beta that was posted
23:46<Chutt>same firmware on both cards, though
23:47<Chutt>no biggy
23:47<tmk>is it bottom half of frame
23:47-!-evil_alex [~Alex@] has joined #mythtv
23:47<tmk>or second field
23:47<Chutt>bottom half
23:47<tmk>that's an encoder bug
23:47<tmk>does it happen mid-stream
23:47<Chutt>just was wondering if you've seen it is all
23:47<tmk>or when starting after stoping
23:47<Chutt>not blaming you =)
23:48<tmk>i cant think of anything i'm doing that would cause that unfortunately
23:48<Chutt>i'm pretty sure it's a firmware bug
23:48<tmk>wierd that it's only on the -250-2
23:48<tmk>why not ask the list if anyone there sees it
23:48<Chutt>it was discussed on the ivtv list recently :p
23:49<tmk>anthony toste or whomever is the guy who runs forums
23:49<Chutt>and on the mythtv list
23:49<tmk>he knows ALL about these cards
23:49<Chutt>and i believe i saw it mentioned on those forums
23:49<Chutt>oh yeah, he emailed me off list
23:49<Chutt>i never replied to him :(
23:49<tmk>he's kind of crazy
23:49<tmk>deaf, REALLY bad grammar
23:49<Chutt>he was all like 'i live near you!!' in broken english
23:50<evil_alex>what nationality is he ?
23:50<fulbert>hour or so ago, full system locks. now the backend process freaks out after 10-15 mintues of running and dumps lots of this stuff:
23:50<fulbert>DB Error (position map insert):
23:50<fulbert>Query was:
23:50<fulbert>INSERT INTO recordedmarkup (chanid, starttime, mark, type, offset) values ( '1053', '20030729223000', 3195, 6, "1139519526");
23:50<Chutt>fulbert, mysql shat itself
23:50<Chutt>use mysqlcheck
23:50<Chutt>i think
23:50<Chutt>to repair things
23:50<tmk>Chutt: dunno, didn't ask
23:50<fulbert>I added a third 250 to the system and now it likes to lock a lot
23:51<Chutt>fulbert, blame tmk
23:51<tmk>get latest cvs
23:51<fulbert>as of yesterday I do.
23:51<tmk>did you get the tarball
23:51<fulbert>it was fine for hours, then it stated to lock a lot.
23:51<tmk>or from CVS
23:51<fulbert>no cvs from server.
23:51<tmk>hows the heat in the case
23:51<tmk>also check fro shared irq's
23:51<fulbert>I disabled more onboard devices I am not using to free irq, now all cards on their own.
23:52<fulbert>I only have one usb, ata, video and eth plus cards now.
23:52<fulbert>it's the same system that has been running two cards fine for a month now.
23:52<tmk>what cards are they
23:52<Chutt>tmk, i'm surprised at how much heat the -350 puts out
23:52<tmk>rev1 or rev2
23:52<Chutt>it _really_ needs a fan
23:52<fulbert>heat is good. all cards rev 2
23:52<tmk>what motherboard
23:53<fulbert>I have crazy fans in the case and on cpu.
23:53<Chutt>i almost burned myself the other day swapping the -250 and -350, and i hadn't even been capturing with the -350
23:53<fulbert>it's an intel 865 mATX board.
23:53<fulbert>I want to say its a d865glxc
23:53<fulbert>it's the only 865 based intel board with onboard gigE.
23:53<Chutt>tmk just wanted to blame it on via
23:53<fulbert>heh ;-)
23:53<tmk>yeah, guilty as charged
23:54<Chutt>fulbert, try the latest cvs tarball
23:54<Chutt>not actual cvs, since it may be old
23:54<Chutt>other than that..
23:54<fulbert>oh ok
23:54<fulbert>I assumed the cvs server was more current.
23:54<fulbert>I will try the tarball in a few.
23:54<fulbert>so should I give up and start a new DB with all those errors now?
23:54<Chutt>sourceforge anon cvs is a day or so old
23:55<Chutt>try mysqlcheck
23:55<Chutt>see if it can repair it
23:55* fulbertis off to tinker