Back to Home / #mythtv / 2008 / 09 / Prev Day
#mythtv IRC Logs for 2008-09-24

---Logopened Wed Sep 24 00:00:15 2008
00:12-!-cattelan is now known as cattelan_away
00:14-!-cattelan_away is now known as cattelan
00:16-!-cattelan is now known as cattelan_away
00:17-!-cattelan_away is now known as cattelan
00:23-!-Captain_Murdoch2 [] has quit [Read error: 110 (Connection timed out)]
00:25-!-tsukasa__ [] has joined #mythtv
00:25<tsukasa__>okay, i only have one problem left
00:26<tsukasa__>basically i cant fullscreen the front end
00:26<tsukasa__>the tv is 1366x768, and it works just dandy at any windowed resolution, in both opengl and qt
00:26<tsukasa__>including 1365x767
00:27<tsukasa__>but as soon as its 1366x768 or fullscreened, the graphics fuck up, with the menus being all torn up and distorted
00:27<tsukasa__>so how can i fix this
00:27<tsukasa__>dammit wrong channel again
00:28<tsukasa__>this should be renamed mythtv-devel T_T
00:28-!-tsukasa__ [] has left #mythtv []
00:29-!-Captain_Murdoch [] has joined #mythtv
00:41-!-kardio [] has quit [Read error: 110 (Connection timed out)]
01:03-!-jhulst [n=jhulst@unaffiliated/jhulst] has quit [Read error: 113 (No route to host)]
01:04-!-academy [n=adam@unaffilated/academy] has quit ["Lost terminal"]
01:18-!-turbo [] has joined #mythtv
01:20-!-briand [] has quit [Read error: 110 (Connection timed out)]
01:20-!-gnome42 [] has quit [Remote closed the connection]
01:37-!-turbo [] has quit [Read error: 110 (Connection timed out)]
01:46-!-william5 [i=8eb18897@gateway/web/ajax/] has joined #mythtv
01:48-!-reynaldo [] has quit ["Lost terminal"]
02:01-!-cattelan is now known as cattelan_away
02:01-!-cattelan_away is now known as cattelan
02:08-!-william5 [i=8eb18897@gateway/web/ajax/] has quit [" ajax IRC Client"]
02:09-!-Bentleyx_ [] has quit [Read error: 110 (Connection timed out)]
02:22-!-foxbuntu [] has quit [Remote closed the connection]
02:44<stuarta>interesting set of discussions following on from the dvb s2api announcement
03:10-!-jk1joel [] has quit [Remote closed the connection]
03:11-!-jk1joel [] has joined #mythtv
03:28-!-edannenbe [] has joined #mythtv
03:31-!-AdamMoredo [] has joined #mythtv
03:42-!-AdamMoredo [] has left #mythtv []
03:47-!-briand [] has joined #mythtv
04:00-!-Aval0n [i=aval0n@] has quit [Remote closed the connection]
04:01-!-Aval0n [i=aval0n@] has joined #mythtv
04:04-!-briand [] has quit [Connection timed out]
04:04-!-kormoc [n=kormoc@unaffiliated/kormoc] has quit []
04:04-!-briand [] has joined #mythtv
04:10-!-xris [] has quit ["Leaving"]
04:23-!-nsaspook [] has quit [Remote closed the connection]
04:23-!-nsaspook [] has joined #mythtv
04:25-!-Tapout [] has joined #mythtv
04:33-!-rooaus [] has joined #mythtv
04:34-!-TapoutMMA [] has quit [Read error: 110 (Connection timed out)]
05:02<gbee>stuarta: arguments over the decision to go with S2 instead of multiproto?
05:03<gbee>interesting that one of the most vocal critics throughout has been the guy who wrote multiproto support for VDR - he's upset that it was a waste of time
05:04<stuarta>hah. i've not followed the discussion at all
05:04<stuarta>not had the time
05:06<laga>the guy has a point, though. i doubt the meeting was completely unbiased ;)
05:06<gbee>I've not followed the whole thing through, just checked in on it from time to time
05:07<stuarta>yeah, i had to agree that there was the potential for bias
05:08<gbee>they _do_ have a point, but for the sake of getting some working next gen support in the kernel I think that they should just drop it
05:11<gbee>I can't really debate the technical merits of which API is better, but from my perspective the one which was developed by a group and which is an extension of the old API sounds more promising than the complete re-write by one individual
05:12<stuarta>i agree
05:12<stuarta>also the fact that one of the guys been contributing works for hauppage
05:13<hads>Sounds like it makes sense.
05:16<gbee>stuarta: aye, multiproto - according to a email last week from Manu, multiproto doesn't support any hauppauge devices ...
05:18<gbee>which I guess could be fixed pretty quickly, to be fair
05:20<rooaus>I liked Manu's statement though... "Void is beautiful too."
05:26-!-nsaspook [] has quit [Remote closed the connection]
05:27-!-nsaspook [] has joined #mythtv
05:27-!-nsaspook [] has quit [Remote closed the connection]
05:38<anykey_>let's hope that this quickly comes to an end and we can use dvb-s2 soon :)
05:42<gbee>I think that statement made followed the meeting is conclusive enough, the debate is over and s2api won - the people arguing now probably won't change that
05:43<laga>i think one of these guys (markus?) had his own repo anyways
05:43<laga>so i don't know why wants to contribute to the discussion when he has forked the code already
05:44<anykey_>because. ;)
05:44<gbee>remember that this has been going on s2api vs multiproto for a month - the meeting sealed the fate of multiproto
05:44<stuarta>thought it was a year
05:45<gbee>stuarta: there was unhappiness about multiproto for 18 months, but s2api was only really born in the last couple of months (following an initial proposal about a year ago)
05:45<stuarta>ah, okay
05:45<gbee>lots of people were bitching about multiproto, but there wasn't an alternative on the table until now
05:46<gbee>at least that's the way I've understood things, those subscribed to the list would have a better picture of things
05:48<gbee>if anything, that demonstrates one of the good things about S2API - from work starting to kernel it's going to take 1/3 or less of the time than multiproto
05:49<gbee>simpler changes, more devs - longer term that means more people interested and able to maintain
05:53-!-buttnutt [] has joined #mythtv
06:02<gbee>have there been mpeg stream handler changes which might explain why this leak has been fixed -
06:03<gbee>I don't remember the commit, but it could have slipped past me in one of Erik's patches as I wasn't paying a lot of attention to those
06:04<stuarta>RSAAuthentication yes
06:04<stuarta>PubkeyAuthentication yes
06:04<stuarta>damn windows cut n paster
06:04<gbee>PasswordAuthentication no
06:05<stuarta>i believe there was something in that area from erics patches
06:05<stuarta>i have that already :)
06:07-!-Aval0n [i=aval0n@] has quit ["ircN 8.00 for mIRC (20080413) -"]
06:08<gbee>just checked, user says that it was fixed for them after upgrading -fixes from but there were no commits to libmythtv/mpeg/ which account for it
06:08<stuarta>ooo, heroes s3 starts on bbc on 1st oct
06:09<stuarta>i've long suspected there are memory issues in that area
06:09<stuarta>that only show up when you have a bad signal
06:09<laga>i think someone commented it might be related to EIT handling
06:09<gbee>except maybe this one -
06:09<laga>and janneg committed something related to that. although he claims it couldnt have possibly fixed it :)
06:09<gbee>looks like we have a winner
06:10<stuarta>what commit was that?
06:10<stuarta>much nicer url :)
06:11<gbee>it's the only possible commit, if we assume that the user is right and the problem is definately fixed
06:11<stuarta>close the ticket anyway
06:11<gbee>not that I can see how exactly :/
06:23-!-buttnutt [] has quit [Read error: 110 (Connection timed out)]
06:27-!-dagar [] has quit [Read error: 110 (Connection timed out)]
06:29-!-dagar [] has joined #mythtv
06:37-!-mattwire [] has joined #mythtv
06:42<janneg>gbee: that commit can't have fixed a mem leak. the only change is that the memcpy is done in less cases
06:43<janneg>and if the user hadn't complained about strange chars in EIT before the change it should not make any difference
06:55<gbee>janneg: yeah, it doesn't explain it
07:28-!-academy [n=root@unaffilated/academy] has joined #mythtv
07:29-!-reynaldo [i=reynaldo@] has joined #mythtv
07:29<academy>I have xvid movies split over 2-3 CDs and have chained them so that after mythtv plays the first cd it automatically moves on to the second. I then marked cd2/3 as not browseable, yet they still appear in the video listings. How can I hide them?
07:30<justinh>first of all, go to the correct channel :)
07:46-!-famicom [] has quit [Read error: 110 (Connection timed out)]
07:54-!-moodboom [] has quit [Client Quit]
08:32<gbee>602 open tickets, doesn't seem that long ago when we managed to bring it under 400
08:33<gbee>we need more devs, or at least some people to triage tickets
08:34<stuarta>or some of those time turner things harry potter has
08:34<stuarta>so us devs can actually get some work done
08:37<gbee>aye, that would come in handy
08:38<gbee>can't even blame Erik, only 23 of those tickets are his
08:39<stuarta>oh dear
08:39<stuarta>we have got a lot of work to do...
08:39-!-reynaldo [i=reynaldo@] has quit ["Lost terminal"]
08:42-!-inordkuo [n=inorkuo@] has joined #mythtv
08:43<gbee>problem is that we've fewer people active right now, Chutt, Captain_Murdoch, Gigem, Bruce, Oscar and possibly xris are all inactive at the moment
08:44<gbee>where inactive == not active on MythTV ;)
08:44<stuarta>well i'm only at 5%
08:44<stuarta>if that
08:46<gbee>that leaves just five people able to contribute more than an hour a week - Janneg, Nigel, Danielk, Paul and I
08:47<stuarta>that's pretty much the only people actively doing any commits
08:48<gbee>120 tickets each, conservative average of 30 minutes per ticket, 60 hours total each just to clear the backlog and that ignores the ongoing work we're all doing
08:49-!-onixian [] has joined #mythtv
08:49<gbee>even with my flexible schedule, 60 hours is huge chunk of my spare time ;)
08:49-!-onixian [] has left #mythtv ["Leaving"]
08:50*gbee lights up the Bat signal
08:53*stuarta fires up the windows build on this laptop while waiting for HP engineer
08:54<stuarta>ffs. i'm sitting on a 100mb/s connection to the net and download of perl is doing 40k/s
09:06<stuarta>now to encourage this to build on d: not c:
09:08-!-academy [n=root@unaffilated/academy] has quit ["Lost terminal"]
09:14<justinh>I better learn to code & quick! ;)
09:21-!-Netsplit <-> quits: iamlindoro, Cougar, i_is_cat, _charly_, styelz, jk1joel, Dibblah, otwin
09:22-!-Netsplit over, joins: jk1joel, otwin, i_is_cat, Dibblah, iamlindoro, _charly_, styelz, Cougar
09:23-!-lyricnz [] has joined #mythtv
09:26-!-renato_ [n=renato@] has joined #mythtv
09:33-!-anykey_ [] has quit ["leaving"]
09:34-!-anykey_ [] has joined #mythtv
09:37<janneg>gbee: and only Paul, Nigel and you commit regularly
09:39<janneg>feck, my MBE's root disk just died
09:39<gbee>my condolences
09:40<stuarta>wtf does the windows port need the fortran compiler?!?!?
09:41<stuarta>it's entirely possible thats where tar is for some odd reason
09:42<gbee>forgot Kevin and Stanley in the list of inactive devs, so from ~23 devs 18 months ago we're down to between 3 and 7 active
09:43-!-famicom [] has joined #mythtv
09:43<gbee>13, typo
09:43<justinh>where did everybody go?
09:43<gbee>some have probably moved on, after a while people just find new interests
09:44-!-joe [] has quit ["Terminated with extreme prejudice - dircproxy 1.0.5"]
09:44<gbee>others are just extremely busy and the more time you spend away from a project the harder it is to get back into the work
09:45<GreyFoxx>It would be sweet to be paid to work on Myth all day
09:45<GreyFoxx>At least for a while :)
09:45<GreyFoxx>I'd much rather be doing that than my "real job"
09:45<gbee>there is always a natural turnover, only we've not added any new faces to replace them (although I believe there are some good candidates around)
09:46<gbee>GreyFoxx: aye
09:46<GreyFoxx>I actually got to spend an entire 6 hours messing around in the code on Saturday and loved every minute of it
09:46<GreyFoxx>but that was a rare circumstance :/
09:47<justinh>I love it when stuff works :)
09:47<justinh>just have to keep in mind that when it doesn't work, it's probably not far off working - or I've bitten off more than I can chew
09:48-!-joe [] has joined #mythtv
09:48<rooaus>I wonder if there will be more activity from non core devs when 0.22 is released, ie after Mythui and qt4 port bedded down?
09:49<janneg>the drive isn't even detected anymore
09:50-!-mattwire [] has quit [Remote closed the connection]
09:50<justinh>janneg: ouchy
09:50<gbee>maybe, there is also a seasonal cycle which you see in OOS - during the Summer, when the weather is better and the days longer people don't want to spend their weekends and evenings sitting in front of the computer, so I expect we'll see a small upsurge as we head into winter
09:51<gbee>janneg: that really sucks
09:52<janneg>hard to say but I think it doesn't even spins up
09:52<rooaus>gbee: Did any further discussion happen about the trac triage workflow?
09:53-!-mattwire [] has joined #mythtv
09:54<gbee>rooaus: no, I think we need to pin down Chutt - even if everyone else is in favour he's the only one who can add new people to trac
10:00<rooaus>gbee: np. by the way, I wasn't suggesting myself for the role :). But I do think you could definitely find a couple of suitable people. Just thinking it might help you out with the 600+ tickets.
10:04<gbee>back in a while, have to go out
10:05-!-zabbadapp [] has joined #mythtv
10:06<zabbadapp>easiest way from a frontend to "ping" and see if myth-backend is running? (not the server, but the actual myth-backend)
10:07<rooaus>zabbadapp: try #mythtv-users
10:07<zabbadapp>thanks, sorry
10:08-!-zabbadapp [] has left #mythtv ["Konversation terminated!"]
10:08-!-famicom_ [] has joined #mythtv
10:14-!-famicom [] has quit [Read error: 60 (Operation timed out)]
10:30<stuarta>so much fun watching the win port build....
10:31<laga>are you going to provide binaries?
10:32<stuarta>not atm.
10:32<stuarta>this is more testing for the fun of it
10:32<laga>good for your sanity ;)
10:32<stuarta>i am insufficiently interested in windows to make binaries available
10:33<justinh>just get a domain with some false whois info. no email, no pack-drill
10:33<stuarta>meh, i'd release osx binaries before windows ones
10:41-!-lyricnz [] has quit []
10:46<justinh>join #mythtv-users
10:49-!-dekarl [] has joined #mythtv
10:53-!-dekarl1 [] has quit [Read error: 60 (Operation timed out)]
10:56-!-superm1 is now known as superm1|away
11:09<mattwire>anybody know which theme gbee was testing mythvideo with?
11:09<mattwire>i can't get it to run with any of the themes so far
11:10<gbee>G.A.N.T and MythCenter
11:10<mattwire>ah, you're back.
11:10<justinh>give me a chance & I can update the rest of the core themes
11:10<gbee>mattwire: just in :)
11:11<gbee>I did forget to delete the old video-ui.xml from themes, let me just do that
11:11<mattwire>it works with gant
11:12<mattwire>getting an issue with main menus not displaying any text though
11:13<mattwire>with both qt and gl painters
11:13<laga>i have that in virtualbox
11:14<laga>well, with mythtv itself, not mythvideo
11:14<mattwire>mythcenter and blue display menus but I think they're both using images
11:14<mattwire>yes, mythtv itself
11:14<mattwire>mythvideo seems to be fine in fact
11:16<mattwire>i'm running it on a real system
11:16<gbee>Mythcenter doesn't use images
11:16<mattwire>that's odd then
11:16<gbee>GANT's probably just missing the correct menu font
11:16<gbee>it uses Arial
11:17<justinh>mythcenter & gant use the same fonts AFAIK
11:17<gbee>and QT isn't quite smart enough to find a suitable replacement on most systems
11:17<justinh>oops. trebuchet MS is what mythcenter uses
11:18<gbee>which probably isn't installed either, but QT knows of a similar looking font and uses that instead
11:19<mattwire>strange, i do have arial installed from the msttcorefonts set
11:19-!-iamlindoro_ [n=iamlindo@] has joined #mythtv
11:19<mattwire>can see it in openoffice and stuff
11:19<gbee>ok, well I'm out of ideas ;)
11:19<mattwire>does mythfrontend log when it can't find a font?
11:20<justinh>mattwire: only if a font definition is missing
11:20<justinh>never moans that it can't find a font face
11:21<gbee>it should, just one of those things that I've not got around to adding
11:25<mattwire>nothing obvious in the logs
11:25<mattwire>when it starts up with any of the themes I get five of these messages:
11:25<mattwire> MythUIButton::CopyFrom: Dynamic cast of a child failed
11:25<mattwire>but that's even with mythcenter which works
11:28<gbee>mattwire: hmm, just added that error - wonder why
11:28<gbee>probably a typo
11:32<gbee>fix committed, typo in Erik's patch, missed it when reviewing
11:33-!-gnome42 [] has joined #mythtv
11:41<mattwire>ok. that error has gone now
11:41<mattwire>in mythvideo I can't switch to list view
11:41<mattwire>Get a missing elements error:
11:41<mattwire>Loading from: /usr/share/mythtv/themes/default/video-ui.xml
11:41<mattwire>2008-09-24 16:41:03.609 Theme is missing critical elements.
11:42<gbee>which theme?
11:42<gbee>err, nevermind
11:42<gbee>odd, checking now
11:43<Chutt>make sure that's not an old copy of the theme lying around
11:43<Chutt>ie, it says '/usr/share', and most svn builds wouldn't install there
11:44<mattwire>they do if you use --prefix=/usr
11:44<Chutt>which isn't a good idea, but, whatever you want =)
11:44<mattwire>i know, but it works better that way for me
11:44<mattwire>because I usually build my own packages which install there
11:45<mattwire>and if i do svn builds to /usr/local i end up getting conflicts
11:45<mattwire>but anyway, i did clear the entire /usr/share/mythtv directory out earlier to try and get the fonts to appear
11:46<gbee>aside from forgetting to sync the -wide version of the default theme to match the non-wide, I can't see an obvious cause - it works here
11:50<mattwire>it seems to be getting very confused with paths, missing off subdirectories etc
11:50<mattwire>need to do a bit more investigating to find out exactly what it's doing
11:50<gbee>mythvideo, or the theme stuff?
11:51<gbee>might need switching to filebrowse mode - defaults to database, but the underlying list building hasn't changed
11:52<gbee>Menu > Change View > Enable File Browsing
11:53<gbee>or just do a rescan
11:53<mattwire>filebrowse works ok
11:53-!-superm1|away is now known as superm1
11:55<mattwire>except if i switch it with a usb key connected it opens that instead
11:55<mattwire>with 28 entries and filenames ranging 00001 to 00004
11:55<mattwire>and i can't get out without exiting mythvideo and re-entering
11:56<mattwire>at which point it's no longer in file browse mode
11:56<gbee>fixing the 'no longer in file browse' bug, typo meant the setting wasn't being saved
11:57<gbee>no idea about the USB key - pre-existing bug because I've not changed that side of things at all
11:57<mattwire>the setting in mythvideo settings doesn't make it stick either
11:59<gbee>mattwire: yeah, because the wrong value is being saved as a result of the typo
11:59<gbee>fix committed
12:05<gbee>heh, an up-node is displayed in the root directory in file browse mode
12:06<gbee>up directory node
12:07-!-xris [] has joined #mythtv
12:08<mattwire>think it's working ok now without the usb key
12:08<mattwire>i'll have to look into that
12:08<mattwire>another time..
12:08<mattwire>still can't get into list view though
12:09<mattwire>getting the critical elements error still
12:09<mattwire>any way of telling which elements it's missing?
12:09<Anduin>mattwire: not runnin 18400?
12:09<gbee>mattwire: current trunk?
12:09<mattwire>svn as of about 2 mins ago
12:09<gbee>mattwire: only one element is critical for that screen <buttontree>
12:10<gbee>on the other three it's <buttonlist>
12:12<gbee>can't see where that rogue up directory node is coming from
12:14<gbee>mattwire: pastebin video-ui.xml and I can double check, but it's there in trunk and I know GreyFoxx has been using it without that problem
12:15<gbee>mattwire: it should be in <window name="tree">
12:16<mattwire>it has two lines defined there i think
12:16<mattwire>there's a buttontree definition in there
12:18<gbee>that's all correct
12:20<mattwire>still no joy after cleaning the installed files
12:20<mattwire>i'll build it on my mythbox and see if the problem exists there as well
12:22-!-beavis [] has joined #mythtv
12:24<GreyFoxx>you are seeing an Updirectory icon in the root of the videogallery ?
12:24<GreyFoxx>I've Definitely not seen that
12:25<GreyFoxx>But I'm also doing another update/compile right now so I'll try it again
12:28<laga>GreyFoxx: do you still have that embedded mysql patch somewhere?
12:28<gbee>GreyFoxx: only there in filebrowse mode
12:29<gbee>it's something unrelated to the mythui changes, that child node just shouldn't be added to the root
12:32<gbee>I'll debug later, it's not that serious
12:50-!-Goga777 [] has joined #mythtv
12:53<Anduin>the list view brokenness is not theme related currently, how are you getting there gbee?
12:53<Anduin>(a jump point is the only thing that looks like it will work)
12:57<gbee>Anduin: mythfrontend mythvideo
12:57-!-kormoc [n=kormoc@unaffiliated/kormoc] has joined #mythtv
12:57<gbee>"Video List"
12:58<Anduin>Yeah, cheating :)
12:59<gbee>I know what's wrong - shouldn't take two minutes to fix
12:59<Anduin>I've got several other things to commit tonight, I'll add a fix for this one (had already started on part of it). One of the problems is that the normal menu will jump to the DLG_DEFAULT, the DLG_TREE check will fail, SwitchLayout() doesn't even try currently.
13:04<gbee>Anduin: got a patch for switch layout, not quite sure what the normal menu problem is?
13:04<Anduin>gbee: in runScreen() from the default jump the dialog type is DLG_DEFAULT, bypassing the check to DLG_TREE and resulting in a VideoDialog instead of a VideoTree
13:06<gbee>yeah, so the default screen type check needs to be moved - I'll leave that to you, I might as well commit the switch layout fix first
13:07<gbee>that code was never updated to reflect the need for video tree to be a seperate screen, at the time I still thought it might be possible to combine them into one
13:08<gbee>actually I've since decided it would be possible, but I wasn't going to go back on what I'd already started
13:10<Anduin>It is very close now, having a fuller common base I'll be able to hide it if I can't make it all fit
13:15-!-stoffel_ [] has joined #mythtv
13:16<gbee>it would be possible with some m_type switches in Create() and the currently virtual methods
13:18<gbee>it would eliminate the remaining duplicate code
13:24-!-Goga777 [] has quit [Remote closed the connection]
13:25-!-dageorge [] has quit [Read error: 110 (Connection timed out)]
13:32-!-Goga777 [] has joined #mythtv
13:45<gbee>Chutt: there?
13:51-!-edannenbe [] has quit [Remote closed the connection]
13:53-!-dageorge [] has joined #mythtv
14:04<stuarta>ooo er...
14:04<stuarta>"Theme is missing critical elements"
14:05<stuarta>clearly project grahem doesn't like gbee's latest changes
14:06<stuarta>any way to find out what it wants?
14:07<mattwire>i asked the same questions earlier..
14:07<mattwire>apparently no there isn't a way currently
14:07<Anduin>stuarta: "videos" or button list
14:07<stuarta>ah, that's what you were wondering
14:07<Anduin>he may not be seeing the same thing
14:07<mattwire>but critical elements are buttonlist
14:07<mattwire>or buttontree
14:08<mattwire>for mythvideo
14:08<stuarta>i get it attempting to enter watch videos
14:08*stuarta tries a different theme
14:08<mattwire>it should work with all the built-in themes
14:08<mattwire>or the others if you remove video-ui.xml to force it to use the default
14:10-!-_gunni_ [] has joined #mythtv
14:11<stuarta>should it not pull in a default element if a theme is missing it?
14:11<stuarta>that would allow things to work, albeit look a bit strange
14:15-!-famicom [] has joined #mythtv
14:16*stuarta gets a backtrace from the frontend exit hang
14:23<stuarta>grrr, missing qt4 debug symbols
14:24<stuarta>87.6Mb ??? eeek
14:27<clever>yes the qt4 debug symbols are painfull
14:27<clever>200mb unpacked for my system
14:27<clever>more painfull when you have ~200mb ram
14:28<clever>yes i do have other systems with more ram
14:28*stuarta fires up the frontend again to watch it hang
14:28<clever>but using the core file on the wrong / causes problems
14:28<clever>the binarys dont match up perfectly
14:29<clever>thats one of the nice things from sharing a / with nfsroot
14:29<clever>the core file fits perfectly on every system
14:29<stuarta>that's why i upped my dev machine to 2Gb
14:29<clever>my last crashing problem was a bug within the iobound warning when writing
14:29<clever>which only got triggered often on my 400mhz system
14:31<stuarta>i was hoping they had been fixed with the commits nigel's been doing
14:31<clever>i havent updated much lately
14:31-!-famicom_ [] has quit [Read error: 110 (Connection timed out)]
14:33-!-reynaldo [] has joined #mythtv
14:34-!-ivor [n=user@kde/developer/ivor] has joined #mythtv
14:35<gbee>stuarta: just delete video-ui.xml in the project-grayhem folder
14:36<stuarta>definitely off waiting for threads to exit
14:36<stuarta>gbee: okay
14:37<justinh>stuarta: I'll fix all that stuff when I get round to it - probably won't be too long - course anybody else is welcome to help :)
14:37<Dibblah>I assume kormoc isn't all that active at the moment?
14:37<gbee>it doesn't currently fallback to the default theme, it should and I'll add see about adding that at some point
14:37<Dibblah>Trying to triage #4608...
14:37-!-Goga777 [] has quit [Read error: 110 (Connection timed out)]
14:38<Dibblah>(The issue is that Mythweb updates the schedule, but the backend doesn't stop the in-progress recording, since it's already active)
14:39<Dibblah>The user has expectations set by 'Cancel this schedule' being accessible there.
14:40<Dibblah>It actually MEANS cancel the schedule for recording any of that recording rule - not the instance of the program.
14:42<gbee>Dibblah: being able to cancel an in-progress recording from mythweb would be handy, I generally end up having to delete them instead
14:43<Dibblah>Agreed. Looking at the code now to see how easy it is to genericise the 'do not record this instance' button, which I seem to remember working.
14:44<xris>gbee: can you even do that fro the frontend?
14:44-!-iamlindoro_ [n=iamlindo@] has quit []
14:44<Dibblah>"Do not record this specific showing"
14:44<xris>I tend to just "delete", though that's usually the only reason I'd want to stop a recording, anyway
14:44<gbee>xris: yes
14:44<gbee>in the Watch Recordings menu you can "Stop Recording"
14:46<xris>ah. menu.. that's something I rarely use. heh
14:46<gbee>I occassionally spot a programme I want to record at the last minute which clashes with a lower priority in-progress recording, I might not want to lose half of a recording just because there was something more interesting to record (rare but it happens)
14:46<xris>I'll have to try that next time I have a zero-byte recording.. deleting that when in progress makes the backend segfault.
14:46<xris>not sure if that bug has been fixed or not -- I've been hesitant to upgrade since the qt4 port started.
14:47-!-Vaelys_ [] has quit [Read error: 104 (Connection reset by peer)]
14:47-!-jgarvey [] has joined #mythtv
14:47<gbee>xris: STOP_RECORDING
14:48<xris>you'll have to remind me about that sometime when I'm home
14:48-!-pat__ [n=chatzill@] has joined #mythtv
14:48<xris>I'll probably end up with a bunch of spare time next week. wife is out of town.
14:49-!-iamlindoro_ [n=iamlindo@] has joined #mythtv
14:50<pat__>Hi, I've just set up my myth tv box, and scanned in all the channels (uk/dvb-s/freesat), but now I have hundreds of channels with seemingly random channel numbers. How can I manually reorder them in the guide?
14:50<pat__>Oh soory I guess I should be in -users
14:51*stuarta wanders off to cook some fuud
14:51-!-pat__ is now known as patstew
14:51-!-patstew [n=chatzill@] has left #mythtv []
14:54<gbee>GreyFoxx, Anduin: the flat mode is a per-screen option, should filebrowse be the same? So the user can have some views browse the filesystem and others can use the database without have to constantly flip between them?
14:54-!-ivor [n=user@kde/developer/ivor] has quit [Connection timed out]
14:55<Dibblah>I misremembered. Schedule override doesn't work on in-progress recordings either.
14:56<Anduin>gbee: It was (thus the "[view type] browses files" option), there are disadvantages though, different filter/browse files/etc mean more tree rebuilding during the switch.
14:57<gbee>Anduin: that's why I kept it a global setting, but I just thought I'd check whether that was acceptable
14:59<Anduin>Yeah, I have no problem making it a toggle, I don't think I can get away with eliminating per-view settings anyway.
15:01<Dibblah>Nope. This was a bad ticket to start triaging. I can see the issue from the user's point of view - There is not enough separation between the 'schedule' and the instance of the schedule.
15:01<gbee>might be possible to merge the filebrowsing/database modes at some future point, such that the tree contains both lots of data and we just mask out the entries we don't want
15:03<gbee>at the point where we build the list in loaddata we can check the item against the filter rules before creating a buttonitem
15:03-!-CDev_ [] has joined #mythtv
15:03<gbee>one way of doing it maybe (here I go again)
15:06-!-famicom_ [] has joined #mythtv
15:07-!-jhulst [n=jhulst@unaffiliated/jhulst] has joined #mythtv
15:08<gbee>laga, janneg: mind looking at the T&C for this site and seeing if we can legally use the data -
15:10<gbee>s/thanks/please/ and thank you ;)
15:13<laga>there is no *explicit* rule against it. but i'm not sure if it's allowed.
15:13<laga>you can have one copy of the page for viewing
15:13-!-stoffel_ [] has quit ["leaving"]
15:14<laga>but you may not use it in any way for private or commercial use.
15:14<laga>we could send an email?
15:19-!-CDev [] has quit [Read error: 110 (Connection timed out)]
15:20<gbee>better to ask
15:21<laga>i will compose one tomorrow i guess
15:21<gbee>thanks :)
15:21<Dibblah>Does this sort of thing actually help?
15:21<gbee>we need a legal department ;)
15:21<gbee>Dibblah: sure
15:23<gbee>doesn't necessarily mean that we have extra time to fix an issue, but understanding the problem helps save time wasted on back and forth with a ticket submitter who might never even reply
15:24-!-famicom [] has quit [Read error: 110 (Connection timed out)]
15:30<laga>i think i've done four complete checkouts of mythtv today
15:30<laga>bandwidth is just too cheap
15:33<Dibblah>Appears that the entirety of the patch (modified) has been applied.
15:34<Dibblah>I assume this could be changed into a tracking ticket.
15:34<Dibblah>Oh, dear - It already is :( Sorry for the noise.
15:40*gbee sobs
15:41<gbee>604 open tickets, up two from earlier despite closing tickets
15:42<laga>that's a lot
15:42<laga>and i was complaining about 160 against mythbuntu
15:43<laga>and there are still a lot which needed to be forwarded to trac..
15:43<laga>so, does chutt want other people to triage bugs?
15:45<laga>i've become pretty good at the whole "invalid" and "wontfix" thing.
15:46<gbee>laga: Chutt hasn't said anything on the subject ..
15:46*gbee waves at Chutt
15:47*laga pokes chutt with a broom handle
15:48<gbee>kormoc: want to close as no response?
15:49-!-carunk [] has joined #mythtv
15:54<gbee> requires a protocol change and that means changes to mythweb, perl and python bindings
15:54*gbee requests a pay raise
15:55<kormoc>gbee, bug me tomorrow? I'm so bloody swammped
15:55<gbee>kormoc: sure
15:55<laga>gbee: tell the guy to provide complete patches?
15:56<gbee>laga: nah, despite my whinging it's just quicker to make the changes myself, it's something I'm interested in seeing fixed
15:57<gbee>could wait months for patches, already have - I said what needed doing 7 months ago
15:57<laga>ah. i guess i should read the ticket.
16:12<clever>gbee: about , i had posted that the file played without trouble but i see otherwise now
16:12<clever>the frontend skips like hell, probly from a audio buffer overflow
16:13<clever>but only for 1 short part, probly where the mythtranscode is exploding
16:13<clever>probly simpler to just close the ticket and avoid frame grabbers
16:16<gbee>clever: it's a valid bug which should probably be fixed
16:16<clever>when playing the file back, the audio starts to play real fast
16:16<clever>then the audio starts skiping as the video speeds up
16:17<clever>and after a short while it levels out and plays normaly
16:17<gbee>clever: unlikely to be me fixing it though, so the explanation is wasted ;)
16:17<clever>probly trying to compensate for a lack of audio frames or something
16:18-!-carunk [] has quit [Remote closed the connection]
16:18<clever>i could try to run the file thru plain ffmpeg and maybe forward the ticket to them:P
16:19<gbee>if it offloads the work, sure ;)
16:20<clever>ffmpeg not found!
16:20<laga>you should compile a trunk checkout anyways
16:20<clever>mythtv@d600:/tmp$ /media/mainlv/root/7.10/bin/ffmpeg -i /media/mainlv/mythtv/1028_20071106120000.nuv tribles.avi
16:21<clever>Error while decoding stream #0.0
16:21<clever>Floating point exception
16:21<clever>FFmpeg version SVN-r9260, Copyright (c) 2000-2007 Fabrice Bellard, et al.
16:21<clever>trunk of which program?
16:21<laga>if you want to report a bug with them
16:22<clever>i beleive i do have trunk
16:22<clever>just havent updated recently
16:22<clever>mythtranscode --infine /media/mainlv/mythtv/1028_20071106120000.nuv --outfile tribles.nuv
16:22<clever>Unknown option: --infine
16:22<clever>oops typo
16:23<clever>-rw-r--r-- 1 mythtv mythtv 1023K 2008-09-24 17:20 tribles.avi
16:23<clever>-rw-r--r-- 1 mythtv mythtv 6.6M 2008-09-24 17:23 tribles.nuv
16:23<clever>the crashing seems to be at diff points within the file(or the codecs are just better/worse)
16:23<laga>clever: you do not have to redirect your terminal to this channel
16:23<clever>i'll have to backtrace is more
16:23<clever>and your lucky i didnt cat /dev/tty> channel
16:24<clever>i have tail -F on all my myth log files
16:24<clever>it keeps spewing the entire autoexpire list
16:24<laga>clever: in irssi, you can do /exec -o <your command>
16:24<clever>yes i know
16:24<laga>so you could actually use irssi as your terminal. and annoy everyone in here ;)
16:24<clever>but that would spew all, not a few select lines
16:24<clever>and it tends to drop lines
16:25<clever>it stops reading the fifo when waitpid() returns, not when the fifo is EOF
16:26<clever>and irssi probly doesnt like the line drawing char used to reset to the start of the line
16:27<clever>:(, 4800 baud is painfully slow
16:28<clever>back at the fast xterm now
16:30<clever>ok, cant get a backtrace out of ffmpeg:*
16:31<clever>not a debug build
16:35-!-mattwire [] has quit [Read error: 113 (No route to host)]
16:40<Chutt>gbee, you poked?
16:41<laga>Chutt: do you want people who are not core developers to triage tickets in trac?
16:41<laga>any guide lines?
16:41<Chutt>i dunno
16:41<laga>other than "don't fuck up"?
16:41<Chutt>and "don't be too mean"
16:41<laga>alright, common sense it is then
16:41<laga>i actually signed something that prevents me from being too mean.
16:41<gbee>ok, well it would help if those people had accounts (maybe not full admin) to be able to change ticket properties that average users can't
16:42<laga>dang ubuntu CoC ;)
16:42<Chutt>gbee, i'd have to make a new class of user
16:42<Chutt>i'll look into doing that
16:42<clever>playing with trac perms can get fun:)
16:42<Chutt>for some weird definitions of fun
16:43<gbee>I think we can pick a few people that can be trusted with the task and if it helps us get through the huge backlog any faster/easier then that's a good thing IMHO
16:43<clever>ive messed with my own trac a bit there
16:43<Chutt>gbee, agree
16:44-!-mattwire [] has joined #mythtv
16:44<gbee>Chutt: it's pretty easy Admin -> Permissions -> Put name and a new group name into the "Add Subject to Group" box
16:44<gbee>then give that group permissions as normal
16:45<gbee>just tested it by adding myself to the 'test' group
16:45<Chutt>oh, ok
16:45<Chutt>can it be done through the webui?
16:45<Chutt>i mean, that means any dev can add people :p
16:46<gbee>sure, though they still need to be registered users which AFAIK you need to setup
16:46<gbee>unless we installed one of the registration plugins so users can self-reg
16:47<clever>login still seems to be plain http-auth like on mine, which meant i had to edit the .htdigest
16:47<Chutt>just get people to send me a htdigest line
17:17<Dibblah> (Improved scheduling of consecutive programs with pre-roll/overrecord) - Isn't this patch / ticket obsoleteish with the changes from multirec?
17:17<Dibblah>(Record to multiple files)
17:21<gbee>it's definately a dead ticket, no activity in nearly two years, for digital users it's less of an issue although multirec doesn't handle the situation perfectly - AFAIK, back to back recordings still happening on the same virtual tuner
17:23<gbee>neither Gigem or Bruce are that active right now and I can't think of anyone else who wants to look at the scheduler
17:24<Dibblah>Only DTVRecorder got that ability?
17:27<gbee>yes, it wasn't extended to the other recorders, in fact the way I understand it the ability to record to two different files from the same stream wasn't added at all - multiple instances of dtvrecorder are used instead
17:27<gbee>might be wrong on that point though
17:28<gnome42>gbee: you are correct.
17:29<Dibblah>Yes, I see that now, looking at the code.
17:29<gnome42>I have a patch to improve scheduling with back-to-back shows on the same multiplex.
17:33<gnome42>It needs testing with the shared firewire/STB setup.
17:37-!-superm1 is now known as superm1|away
17:42-!-|gunni| [] has joined #mythtv
17:44-!-jgarvey [] has quit ["Leaving"]
17:47-!-inordkuo [n=inorkuo@] has quit ["Leaving."]
17:52<laga>Chutt: what's your email address?
17:53-!-kormoc [n=kormoc@unaffiliated/kormoc] has left #mythtv []
17:58-!-_gunni_ [] has quit [Read error: 110 (Connection timed out)]
17:59-!-iamlindoro_ [n=iamlindo@] has quit []