Back to Home / #mythtv / 2008 / 03 / Prev Day | Next Day
#mythtv IRC Logs for 2008-03-15

---Logopened Sat Mar 15 00:00:59 2008
00:06-!-superm1 [n=superm1@ubuntu/member/superm1] has joined #mythtv
00:09-!-davilla [] has quit ["Leaving"]
00:14*Captain_Murdoch notices that it appears that programs put into the Deleted recGroup aren't password protected anymore when you deleted them from a password protected group unless you explicitly set a password on Deleted.
00:16-!-jhulst [n=jhulst@unaffiliated/jhulst] has joined #mythtv
00:28-!-ToadP [] has quit []
00:37-!-grokky [] has quit []
00:52-!-joobie [] has quit ["Leaving"]
00:56-!-joobie [] has joined #mythtv
01:39-!-mzuverink [n=mzuverin@unaffiliated/mzuverink] has quit [Read error: 104 (Connection reset by peer)]
01:54-!-hadees [n=hadees@pdpc/supporter/sustaining/hadees] has quit [Remote closed the connection]
02:13-!-czth [n=dbrobins@nat/microsoft/x-55d173cc8bdf9dcd] has joined #mythtv
02:30-!-czth__ [n=dbrobins@nat/microsoft/x-04c63c110db68919] has quit [Read error: 110 (Connection timed out)]
02:33-!-gnome42 [] has quit [Remote closed the connection]
02:49-!-superm1 [n=superm1@ubuntu/member/superm1] has quit [Remote closed the connection]
02:57-!-joobie [] has quit ["This computer has gone to sleep"]
03:06<Captain_Murdoch>the clock from the main menu is showing through on mythflix where the lists are drawing and my crash course on mythui isn't yielding much.
03:18-!-joobie [] has joined #mythtv
03:23<rooaus>Captain_Murdoch: I really like your TV_WATCH_RECORDING action idea, will try and code it soonish, unless you want to ;)
03:23<Captain_Murdoch>rooaus: I'm actually working on that right now while sitting idle on a maintenance window con call for work. :)
03:24<rooaus>Captain_Murdoch: Excellent! :D
03:34<rooaus>Captain_Murdoch: I have been looking (briefly so far) at adding storage groups to mythmusic, is it right to say that the problem is in two parts. (1) Make mythmusic storage group aware. (2) Allow mythmusic to stream content from remote backends using the ringbuffer.
03:37-!-jhulst [n=jhulst@unaffiliated/jhulst] has quit ["Konversation terminated!"]
03:39<rooaus>I have to head out, sorry to ask and run, but I will definitely read scrollback when I get back tonight. May not be in a coherent state (am I ever?) though, going to a concert. :)
03:45<Captain_Murdoch>ok, not sure how long I'll be on this call or playing with this code. 3:45AM here.
03:47-!-superm1 [n=superm1@ubuntu/member/superm1] has joined #mythtv
03:47-!-grokky [] has joined #mythtv
03:59-!-grokky [] has quit []
04:56<gbee>Captain_Murdoch: themer needs to define a background for mythflix, but the default is for anything in the base screen to be visible unless that is done
04:57<gbee>it actually works pretty well in MythCenter and Metallurgy was designed around it (clock is shown on every page)
04:58<Captain_Murdoch>where at? I'm using titivillus and it has the clock in the upper right, but not far enough up to be out of the way of the lists in mythflix.
05:01<gbee>define the background? You'd need to create a copy of netflix-ui.xml in the titvillus folder, then add
05:01<gbee>as the first item in each window
05:01<Captain_Murdoch>no, sorry, I meant where do they put the clock, or I guess you meant that they define backgrounds in mythflix so it's not an issue.
05:03<gbee>well the clock can be moved to a position where it doesn't intefer or a background defined which covers it up - most themes don't have a consistent area of free space that could be used for the clock
05:04<gbee>I will go around sorting out the themes when I'm finished with the port
05:05<Captain_Murdoch>ok, I was mainly curious, haven't messed with mythui much if at all that I can recall.
05:07<Captain_Murdoch>I did make productive use of my maintenance window con call time though and coded up the feature described here: almost makes we want to add it to my menus, but we already have a key bound to popup the rec group chooser so it's not that many extra keys to pick a new one. I can see how people with kids would like to be able to go directly to a group though off th
05:07<gbee> << Space left for the clock to appear on all screens, the background image is the metal, each page defines a background which isn't full screen leaving the items on the edges visible
05:08<Captain_Murdoch>yeah, I remember seeing those now that I look at them again.
05:09<gbee>nice feature that, I think I'd definately use that until we have proper multiuser support
05:09<Captain_Murdoch>one immediate use I thought of is a menu item for "Deleted Recordings"
05:10<Captain_Murdoch>well, use for others, I don't have that feature turned on.
05:10<Captain_Murdoch>lets Theme designers do a little more and users a lot more if they want.
05:12<gbee>the idea behind the clock being in the base is so that if you want it visible on every page, you don't have to keep redefining it in every single window although if you want it to appear in different places, that's also an option
05:12<Captain_Murdoch>yeah, makes sense
05:13<Captain_Murdoch>reusability. :)
05:13<Captain_Murdoch>or inheritance, even better, not sure which word better describes
05:15<gbee>reusability in this case, inheritance has another specific meaning in mythui :)
05:22<gbee>[16560] - inheritance, the windows have textareas/buttonlists but the definition of how they should look comes from a single instance in base.xml
05:30-!-beavis [] has joined #mythtv
05:33-!-joobie [] has quit ["This computer has gone to sleep"]
05:37<gbee>Captain_Murdoch: may be more missing headers as I convert over plugins, mythflix was probably a special case though, it had several headers which weren't used and rather than try to figure out which ones were being used I deleted the lot
05:38<Captain_Murdoch>for that scrolling indicator thing i mentioned, maybe it's just not used on themes I've seen, but i didn't think we supported it. I'm basically talking about something like the sliding indicator you see in the scrollbar alongside a scrollable window like your web browser, etc. something that indicates what position you are in the whole page.
05:38<gbee>ahh, ok
05:39<Captain_Murdoch>yeah, no problem. I'm running ancient installs on some of my machines so I catch things like that. my frontends are newer but my backends and dev machines are rh9
05:39<gbee>yeah, I can probably implement something like that
05:41<Captain_Murdoch>no biggie, it's just something I've thought about for a while, but never dug into the gui enough to even consider coding it. I didn't think it would be that easy to do with the old stuff, but with mythui it might be easier. the image could be small and just scaled up vertically to display how much of the list was visible and the position within the list.
05:47-!-xris [] has quit []
05:51-!-phatmonkey [n=ben@] has joined #mythtv
06:14-!-photon-thug [] has joined #mythtv
06:35-!-beavis [] has quit ["Verlassend"]
06:42-!-photon-thug [] has left #mythtv []
06:52-!-ToadP [] has joined #mythtv
06:54-!-beavis [] has joined #mythtv
07:07<gbee>hmm, interesting problem
07:32-!-beavis [] has quit [Remote closed the connection]
08:37<laga>danielk22: do you think you can take a look at again? this is a showstopper for lots of ubuntu users who need nvidia's legacy driver.. if you need any logs or other information, let me know
08:55-!-joobie [] has joined #mythtv
09:03-!-esperegu [n=jhaarman@] has joined #mythtv
09:13<danielk22>laga: that ticket doesn't say what version of the driver headers this is with. Each ubuntu linux kernel comes with 3 nvidia kernels, and there are dozens of ubuntu linux kernels still in use, and you may even not be up to date with ubuntu on your release... i.e. too many variables remain.
09:19<laga>i finally found our bug report..
09:19<laga>so it happens with version 1.0.7185
09:38<danielk22>laga: you are involved in ubuntu package management?
09:38<danielk22>maybe ubuntu could just fix the glx.h header file to match the driver capabilities?
09:39<danielk22>i.e. it shouldn't report glx 1.4 functions if they aren't implemented...
09:43<laga>yes, i am.. only for the mythtv packages, though.
09:44<laga>danielk22: ideally, nvidia would fix it.. i guess more people will run into this problem than just ubuntu users
09:48<danielk22>heh, nvidia hasn't fixed it in 3 years... glx.h headers have traditionally been thought of as something for os vendor to worry about. Obviously from before the days of graphics cards and non-developers compiling stuff...
09:50<danielk22>if i can do something to detect the broken headers elegantly i'll do it, but if ubuntu is shipping the same headers for drivers with and without glx 1.4 then there is really nothing i can do short of trying to link the library in ./configure and setting a define..
09:51<danielk22>(i understand that ubuntu is just doing what nvidia does.. but the ubuntu packages is in a much better position to fix this than me...)
09:51<laga>danielk22: would fixing the headers actually fix the 'symbol not found' error for packages? excuse my ignorance here, but won't mythtv still try to use glXGetProcAddress unless it was compiled against nvidia-glx-legacy?
09:55<danielk22>mythtv will only use glXGetProcAddress if the headers report that it is available in the library. So fixing the headers would mean that a mythtv compiled with the old headers would only use the the old methods.
09:57<danielk22>The problem is that glXGetProcAddress and it's older cousins are the functions you use to determine all the dynamic linking stuff in GLX..
10:00<danielk22>here's the actual code in MythTV:
10:00<jamesd>danielk22, with your logic we don't need headers at all, the compiler is just suposed to magicly guess the right version to use... if the compiler is that smart why do we need headers
10:01<danielk22>james look at the code, the headers define GLX_VERSION ...
10:02<laga>danielk22: and a mythtv compiled with new headers will fail with nvidia-glx-legacy, correct? normally we just build against the mesa headers
10:02<jamesd>danielk22, ever heard of "garbage in, garbage out"? you have the wrong headers.. you get the wrong code...
10:03<danielk22>laga: only if they actually remove glXGetProcAddressARB or glXGetProcAddressEXT in the new drivers, which I would hope is unlikely since that will break all legacy applications.
10:03<danielk22>jamesd: that's why i'm saying the best thing to do is fix the headers.
10:03<laga>so if we fix the headers it will magically start working even if we build against mesa?
10:04<laga>sorry for being a bit thick here, i'm just getting confused
10:04<danielk22>laga: if you build against a modern mesa it will use glXGetProcAddress()
10:05<laga>so it's gonna break with the legacy drivers, which is what we want to avoid.
10:05<danielk22>ok, so these three functions all perform the same function, to allow you to link to functions that may or may not exist in the driver itself..
10:06<laga>so fixing the headers won't do us any good in our special case here (users just downloading the packages and expecting them to run).
10:06<danielk22>this is why you can't just detect these functions at run-time, because these are the functions you use to detect functions at run-time.
10:06<laga>danielk22: and which function is used is determined at build time.
10:08<danielk22>so.. if you use glXGetProcAddressARB or EXT, if it existed once in the driver it is unlikely to go away, because the driver maintenance people know that you can't detect their absence at runtime very easily. So, using an old one with a new driver is probably safe... only way to *know* is to test..
10:08<laga>and what prevents me from just using glXGetProcAddressARB instead? looking at my glx.h, it still exists in newer versions
10:09<laga>yes, that's bascially what i was thinking
10:09<danielk22>for a package like on ubuntu you can use glXGetProcAddressARB.. just make sure it is still works with the newer drivers... the MythTV construction is to keep you from shooting yourself in the foot when compiling on a system with a non-broken glx.h
10:10<danielk22>if you can test and ARB or EXT works for all supported drivers, use it.
10:10<laga>so there are two options: build against nvidia-glx-legacy-dev or just use glXGetProcAddressARB (assuming it still exists in the headers)
10:10<laga>danielk22: thanks a lot. i'll do just that.
10:12<danielk22>k, if you want to do a patch that adds a configure option --use-glx-proc-addr-arb i'll put it in mainline so you don't need to dpatch.. i think it would make a good packager extension like the --cpu flags..
10:12<laga>oh, that's even better
10:12<laga>i'll make a ticket then
10:13<danielk22>k, thanks
10:14-!-_gunni_ [] has joined #mythtv
10:20-!-xomp [] has joined #mythtv
10:20-!-xomp [] has left #mythtv ["Leaving"]
10:46-!-Dave123 [] has quit ["Leaving"]
10:51-!-ToadP [] has quit [Remote closed the connection]
10:51-!-ToadP [] has joined #mythtv
11:17<gbee>Chutt: mythgallery using mythui -
11:19<Chutt>only thing off is the text size?
11:19<gbee>yeah, need to use a small font for the listbuttons
11:20-!-aevil [] has joined #mythtv
11:20<Chutt>very cool.
11:21<Chutt>you wrote the button grid in mythui, right?
11:22<Chutt>i mean, in the library
11:22<gbee>yeah, it's mythlistbutton with <layout>grid</layout>
11:23<Chutt>so mythvideo can reuse
11:24<Chutt>very nice
11:24<Chutt>thanks for doing this =)
11:24<gbee>and themers can choose to go for a horizontal or vertical list (if they have a good reason to, don't think that really applies to mythgallery though)
11:26<gbee>going to be another day before I'll consider committing mythgallery, still some instability and a couple of things which don't work as they should
11:29<gbee>9 files changed, 430 insertions(+), 1026 deletions(-)
11:33<janneg>sigh, mythfrontend is finally running again
11:36-!-gnome42 [] has joined #mythtv
11:52-!-ead [] has quit ["Ciao!"]
11:52-!-ead [] has joined #mythtv
12:21<Chutt>janneg, does it actually work, though?
12:29<janneg>I haven't tested yet but I assume yes. it was running before the big merge
12:31<janneg>the only major thing not working is the capture card setup, oh and the opengl painter doesn't paint text
12:33<gbee>heh, that's not good :)
12:34<janneg>it paints instead white blocks
12:40-!-asjoyner [] has joined #mythtv
12:45<jams>gbee- are you planning to use the new ui for the settings code?
12:45<GreyFoxx>CDev: So far everything looks fine on my xbox
12:45<gbee>jams: eventually, but it will probably be after the rest of myth is ported
12:45<gbee>that and the OSD
12:45<jams>just checking
12:46<gbee>the settings code is like a tangled ball of string, I think it's pretty likely that instead of porting it to mythui we'd consider starting from scratch
12:47<jams>kind of what i was thinking. I just ran into a dead end while working with the settings code.
12:48<jams>thats fairly far off into the future, so back to finding an alternate route with the current stuff.
12:51<gbee>I'm being extremely polite about the settings code, but only because I don't want to upset anyone
12:55-!-BengalBorn [] has joined #mythtv
12:56-!-BengalBorn [] has left #mythtv []
13:02<janneg>playback works but osd is blank
13:02<danielk22>gbee: you won't upset anyone with the settings code, it's a lot of cruft on top of cruft.
13:03-!-aevil [] has quit ["Ex-Chat"]
13:03<danielk22>it may have started out neat and tidy, I don't know, but it's hasn't gotten much love over the years
13:07-!-Chutt [] has quit [Read error: 110 (Connection timed out)]
13:08<MrGandalf>danielk22: I've finished up my patch for what I call "Now Playing", ie: DVB radio text, but I have a couple small questions if you have a minute
13:09<MrGandalf>or janneg, if he has a minute.. need input on the correct way to do things
13:19<gbee>MythListButton is now 100% mouse friendly
13:23-!-ldkronos [] has joined #mythtv
13:23<janneg>gbee: do you have pending commits for trunk/mythtv? if not I'll do merge from trunk to qt4
13:24-!-dekar1 [] has joined #mythtv
13:25<gbee>janneg: no, nothing important
13:25-!-mjj29 [] has joined #mythtv
13:27-!-esperegu [n=jhaarman@] has quit ["esperanto:"]
13:30<danielk22>mrg: yep?
13:31<MrGandalf>wrt distinguishing the stream
13:32<janneg>wow, just one simple conflict in 50 changesets
13:32<MrGandalf>There is nothing in the payload portion but the text ending in a null
13:32<gbee>janneg: planning on merging back to trunk?
13:32<janneg>not before monday evening
13:32<MrGandalf>and the PES headers give little away. The streamID varies for one of my providers with each song but the other remain constant
13:33<MrGandalf>so I have no real way of knowing for a fact it's a DVB radio stream
13:33<danielk22>mrg: so you are worried that that could conflict with something else stored in the same type of stream..
13:33<danielk22>I think the only thing you can do is make this optional..
13:33<MrGandalf>daniel: well, I've made it optional
13:33<MrGandalf>that's the only thing I can think to do
13:34<danielk22>since there is no standard being followed they haven't bothered to identify their mod..
13:34<janneg>I'll try to fix the setup and osd first
13:34<MrGandalf>besides, before me, avformat didn't support PRIVATE_SECTION streams anyway
13:34<MrGandalf>so it certainly won't conflict with anything current
13:34<danielk22>private sections are scary all around...
13:35<MrGandalf>I filter on stream start, but that's it
13:35<danielk22>you never know what is going to be in there..
13:35<MrGandalf>my second question is wrt patching themes
13:35<danielk22>maybe you could add some sanity checks?
13:36<MrGandalf>daniel: The only thing I'm doing is splitting on 0x0a.. but I suppose something bad may happen if the string doesn't end in 0x00
13:36<MrGandalf>I could put in a few more checks
13:36<MrGandalf>for theme patches, I have one new container
13:37<MrGandalf>and a new graphic
13:37<gbee>janneg: ok, think we should warn users now? I think it needs to be made clear that that trunk from this point is more unstable than usual (mythui + qt4) and that users should probably stick with 0.21 if they aren't comfortable with that
13:37<MrGandalf>should I provide a patch for all themes?
13:37<danielk22>basically always add it to the themes in mythtv/themes, and the default.. allow the rest to do it themselves.
13:38<danielk22>but I am not a theme expert...
13:38<MrGandalf>ok.. but, I found a case where if "program_info" container is mssing, the osd lock never gets unlocked. It's not likely to happen, but I can provide a patch
13:38<danielk22>yeah make sure that nothing bad happens if the "string" doesn't end in a null.
13:39<danielk22>make patches and tickets for bugs, always :)
13:39<MrGandalf>thanks for your help
13:39<danielk22>you might also just want to check that at least 60% of the "characters" are ASCII...
13:40<danielk22>that should work with all European languages at least.
13:40-!-territory [] has joined #mythtv
13:40<MrGandalf>brb, lunch
13:40-!-dekarl [] has quit [Read error: 110 (Connection timed out)]
13:42-!-jarle [] has quit [Read error: 110 (Connection timed out)]
13:43-!-jarle [] has joined #mythtv
13:44<janneg>gbee: I post a announcement today
13:45<gbee>ok, great :)
13:45<gbee>I'd probably have to post a warning about mythui anyway
13:55-!-joobie [] has quit ["This computer has gone to sleep"]
14:29-!-mattwire [] has joined #mythtv
14:31-!-Chutt [] has joined #mythtv
14:32-!-PointyPumper [i=Pintlezz@] has quit [Read error: 104 (Connection reset by peer)]
14:33<janneg>bah, compile time with qt4 seems significant longer
14:34<Chutt>probably need to clean up includes
14:36<janneg>could be, either qt3to4 or CDev replaced specific qt includes with QtCore
14:37<Chutt>which probably pulls in a bunch of stuff
14:37<janneg>yes, all of the base classes
14:40<danielk22>should be easy enough to manually remove all QtCore includes and see what it really needs on recompile... tedious but simple :)
14:41<janneg>it is included in mythcontext.h and other headers so it gets included probably in every file
14:43<Chutt>definitely don't need it there
14:52<janneg>same problem for the other Qt modules. I'll clean it up now. the compile time is just unbearable
14:53-!-ldkronos [] has quit ["ChatZilla 0.9.81 [Firefox]"]
14:54-!-reynaldo [] has joined #mythtv
15:02<Chutt>i just finished chopping 80KB off of a compiled library size (started at 460KB) by removing unnecessary verbose printfs.
15:03<Chutt>and changing functions to static, etc
15:04<janneg>more than 1/6, not bad
15:04<Chutt>and it really wasn't all that many prints
15:05<Chutt>makes me curious about myth =)
15:05<danielk22>hmm, why would they take up so much space??
15:06<Chutt>it adds up
15:07<danielk22>but there must be a multiplier. maybe VERBOSE should call a function instead of everything being in the macro..
15:07-!-xris [] has joined #mythtv
15:07<Chutt>make a compiled out version of VERBOSE
15:08<Chutt>need to change everything to be VERBOSE(("stuff"));, though
15:08-!-xris [] has quit [Client Quit]
15:08<Chutt>or something like that
15:08-!-xris [] has joined #mythtv
15:08<danielk22>There could still be a macro, just all the smarts would be in a function.
15:09<Chutt>that still leaves the strings in the object files
15:09<danielk22>yeah, but those can't be hundreds of KB..
15:09<danielk22>they aren't even translated..
15:09<Chutt>80KB here
15:09<Chutt>in this tiny library
15:10<danielk22>which lib?
15:10<Chutt>work stuff
15:11<Chutt>my openmax implementation
15:13-!-fus10nx [] has quit [Read error: 110 (Connection timed out)]
15:13<danielk22>it's totally possible to use errno's and put the strings somewhere else (and leave them out in release even), but it might be a lot of pain for little gain... waiting until someone needs to port mythtv to a small memory footprint hw might make the most sense...
15:14<Chutt>i was just curious
15:14<danielk22>there are probably other more serious memory hogs that would become apparent, like we had with the themes in 0.20.
15:14<Chutt>well, not for work, i had to do it there =)
15:15<danielk22>put a client is probably paying for it, indirectly or directly.. :)
15:15<Chutt>i did most of the work while waiting on windows media 6 OS compiles
15:15-!-A- [] has joined #mythtv
15:19-!-symptom [] has joined #mythtv
15:20-!-symptom [] has left #mythtv ["Keep your ear to the grindstone"]
15:26-!-phatmonkey [n=ben@] has quit []
15:26-!-|gunni| [] has joined #mythtv
15:27-!-phatmonkey [n=ben@] has joined #mythtv
15:43-!-_gunni_ [] has quit [Read error: 110 (Connection timed out)]
15:58<gbee>kormoc: in #4872, have you tried it with mythvideo and it hasn't worked? The internal player in mythvideo is exactly the same code as it is for recordings so it should just work
15:58<kormoc>Yeah, it's not working
15:59<Chutt>someone needs to measure the cpu overhead of the aspect ratio detect
15:59<kormoc>I played back a wide screen video that in the normal mode would auto zoom, and it didn't
15:59<gbee>hmm, maybe he's added a isVideo() check to exclude non-recording material
16:00<kormoc> if (!videoOutput) return;
16:00<kormoc>would that do it?
16:00<gbee>no, that's just checking that the videooutput object exists (renderer)
16:01<Chutt>sure it was a widescreen video with letterboxing in the video?
16:01<kormoc>I exported it from recordings over, and I know it zoomed under recordings
16:01<gbee>have you turned it on? (Just discounting the obvious :) )
16:01<kormoc>but that said, maybe the export changed something
16:02<kormoc>heh, aye, it's on, at least, I only know of one place to turn it on at :)
16:03<gbee>there is an OSD menu option to enable debugging (or there was in earlier versions) try turning that on
16:03<Chutt>that needs to go before merge, too
16:04<gbee>"Debugg Auto Detect" in the "Adjust Fill" menu
16:04<kormoc>Ahh damn it. I checked another video and it zoomed fine. My export must have screwed it up...
16:08-!-Chutt [] has quit [Remote closed the connection]
16:08-!-aevil [] has joined #mythtv
16:08<kormoc>for some reason ffmpeg didn't preserve the resolution like I thought I asked it to, so it was doing the proper thing
16:09<MrGandalf>has justinh been on recently?
16:09<kormoc>gbee, it also looks like v5 of his patch uses verbrose playback rather then the osd menu
16:09<kormoc>MrGandalf, he's on now
16:10-!-tulbreak [] has joined #mythtv
16:11<gbee>good for Kevin, he's keeping the 0.22 release notes current
16:28-!-Zombie [] has joined #mythtv
16:29-!-phatmonkey [n=ben@] has quit []
16:40-!-superm1 [n=superm1@ubuntu/member/superm1] has quit [Read error: 113 (No route to host)]
16:42<gbee>anyone want to convert mythgallery/singleview.cpp to mythui? :)
16:43<MrGandalf>I could attempt it, but I wouldn't do anywhere near as good a job with it as you. :)
16:44<gbee>it's a jumble of stuff and I'm not looking forward to it :)
16:45<MrGandalf>but you're so good at it.. you'd probably have it knocked out in no time.
16:45*MrGandalf puts on his boots..
16:46*gbee sighs
16:48-!-richards [] has joined #mythtv
16:51-!-ikke_ [] has joined #mythtv
16:53<ikke_>hei, I would have a update for grab_fi_icons_xmltvids for updated icons and xmltvids. Do I send the patch into mailing list or create a ticket? Any advise?
16:59<gbee>ikke_: long term, the best solution would be to talk to the tv_grab_fi author and have them put a channel_ids file with the information on
16:59<gbee>that way we can pull the updated data automatically
17:02<gbee>going forward we really only want to be supporting the icon downloader
17:03<ikke_>ok, i don't know any of that yet, need to read about it
17:08<ikke_>gbee: grabbers nowadays support icon downloads? and timeshift? and mythtv has support for all that?
17:09<ikke_>it seems those files behind your link some have links to icons. need to find grabber developer for finnish icons and talk with him more. thanks
17:14<gbee>ikke_: the uk grabber has support for timeshift, but the Finnish grabber author can probably add support for it as well
17:15<ikke_>the problem with finnish tv grabber is that i don't think anyone really maintains it. the original author told he has no more time on it. that's a long time ago
17:16<ikke_>luckily they take in patches to keep it running
17:17-!-phatmonkey [n=ben@] has joined #mythtv
17:18<ikke_>but i'll try contacting the list. let's see. now to bed...
17:18-!-Nikas [] has joined #mythtv
17:18-!-ikke_ [] has left #mythtv ["g'night"]
17:44-!-tulbreak [] has quit []
17:48<gbee>I'm going to have to start building and working on the same source, keep making silly mistakes when I have to patch over changes to test them
17:49<Anduin>Then you just start checking in things you didn't intend to, you cannot win.
17:50<gbee>yeah, that's why I worked that way until now, I've several incomplete or insufficiently tested patches in my tree
17:55-!-aevil^aw [] has joined #mythtv
18:02<janneg>puh, done with replacing headers
18:03<gbee>has it helped?
18:05<janneg>I'm testing atm but it looks good
18:12-!-aevil [] has quit [Read error: 110 (Connection timed out)]
18:14-!-skamithi [] has joined #mythtv
18:14-!-jmk [] has quit ["Leaving"]
18:23-!-grokky [] has joined #mythtv
18:25<janneg>a complete build without ccache takes now 7:14
18:26<kormoc>what's the base time?
18:27<janneg>I don't know, just started the old build
18:28<kormoc>fair 'nuff
18:35-!-jmk [] has joined #mythtv
18:36<janneg>old build is still in libmythtv
18:36<CDev>janneg: sorry about the headers. On my 8 core box with ccache I didn't notice the slowdown.
18:38<janneg>I was using ccache but it doesn't if you modify a central header like mythcontext.h
18:40<janneg>11:23, over 1/3 faster
18:43-!-degreseven [] has joined #mythtv
18:49-!-TelnetManta [] has quit [Read error: 104 (Connection reset by peer)]
19:02<gbee>any ideas from this backtrace?
19:03-!-jmk [] has quit ["Leaving"]
19:03<gbee>only happens with the QT painter
19:13-!-edit_21 [n=james@unaffiliated/edit21/x-00000001] has left #mythtv []
19:34-!-bunder [n=chris@gentoo/developer/bunder] has joined #mythtv
19:35-!-bunder [n=chris@gentoo/developer/bunder] has left #mythtv []
19:39-!-Chutt [] has joined #mythtv
19:41<janneg>Chutt: cleaning the headers reduce the build time by 1/3
19:47-!-flyback [] has joined #mythtv
19:47-!-flyback [] has left #mythtv ["Ex-Chat"]
19:59*xris curses firewire recording...
20:00<xris>anyone know what the behavior of "delete forever" will do for a program in the deleted group if it was originally placed there via delete+rerecord?
20:01-!-joobie [] has joined #mythtv
20:01<xris>wondering if I should change the mythweb UI if "delete+rerecord" was already requested for a program in the deleted group
20:07-!-degreseven [] has left #mythtv []
20:11-!-richards [] has quit ["ChatZilla 0.9.81 [Firefox]"]
20:19-!-|gunni| [] has quit ["KVIrc 3.2.4 Anomalies"]
20:22<gbee>hmm, somethings really screwed up in the inheritance, when a popup is displayed with the QT painter, the lists in the background change size to match the width of the new list
20:27<gbee>ahh, ok ... they share images when they shouldn't, since the images are scaled differently for each list
20:35<gbee>would QT4 allow us to stop doing full page redraws?
20:36-!-aevil^aw [] has quit [Read error: 110 (Connection timed out)]
20:47-!-elupus [n=elupus@xbmc/staff/elupus] has joined #mythtv
20:48<elupus>i'm trying to figure out how to get the mythbackend to stop recording a program.
20:48<elupus>i'm working on xbmc's mythtv client
20:49<elupus>calling STOP_RECORDING, does stop the recording, however the backend immidiatly starts recording the same show again
20:50<elupus>how would i stop recording the program completly?
20:50<elupus>i've had the same issue doing this from official frontend at some points
20:57<elupus>even when i'm working with the backend tcp protocol directly and not using the frontend? it's more of a how does mythbackend work internally question.
20:58<elupus>this is for the libcmyth implementation to communicate with the backend
20:58-!-phatmonkey [n=ben@] has quit []
20:59-!-PointyPumper [i=Pintlezz@] has joined #mythtv
21:02-!-Nikas [] has quit ["( :: NoNameScript 4.2 :: )"]
21:04-!-mattwire [] has quit ["Leaving"]
21:05-!-kormoc [n=kormoc@unaffiliated/kormoc] has quit []
21:11-!-skd5aner [] has joined #mythtv
21:13<janneg>which qt4 version should we require? I don't think older than 4.3 makes sense. kde 4.0 requries 4.3
21:13<hads>They've shifted qt-copy to 4.4 now haven't they?
21:14<skd5aner>Just curious, I rarely pop my head in IRC, but I've got a problem after I upgraded to current SVN tonight (from a version about a week old)... I can't enter the "Watch Recordings" screen
21:14<skd5aner>is this the appropriate audience to talk to about that?
21:14<janneg>4.4 would be nice since it would allow us to port mythbrowser to qtwebkit and drop the kde dependencies
21:15<janneg>hads: yes, but only for developing 4.1 which is scheduled for july
21:16<hads>Ah yes.
21:17<hads>Looks like 4.4 is scheduled for mid Q2 2008 so I guess it might be a bit optimistic.
21:17<skd5aner>Is anyone else experiencing an issue getting into "Watch Recordings" with SVN 16581?
21:20<gbee>janneg: I think we should go for the most recent version we can possibly get away with, so I guess that means 4.3
21:23<gbee>4.4 would be nice, but I doubt it's in the stable repos of current distro releases
21:23<janneg>i doubt it too given that it is still in beta
21:24<gbee>I think Mandriva have it in their 'Cooker' repo
21:25<skd5aner>gbee and janneg - should I pose my question elsewhere. I apoligize if this is not the proper forum.
21:26<gbee>skd5aner: post to the dev-list, [16581] was Kevin's change to the Watch Recordings screen, he doesn't use IRC and the best way to let him know there is a problem would be through the mailing list
21:28<gbee>skd5aner: wasn't one of my mythui changes since Watch Recordings doesn't use any mythui code
21:30<Anduin>I'm only a few changes away from current and have no issue.
21:30<gbee>those changes would only impact mythappearance, mythflix or mythcontrols (and to a lesser extent the main menus)
21:30<skd5aner>gbee: Thank you, I did post an emails to both dev and users, hopefully Kevin will see it. I appreciate the help
21:30<gbee>I'm one revision behind 16581
21:30<skd5aner>I'll rollback to 16580 and see if the behavior goes away
21:31-!-joobie [] has quit ["Leaving"]
21:31<janneg>so it is 4.3
21:31<gbee>though libmythui is more recent (just didn't install the other libs)
21:31<skd5aner>BTW - if you don't mind me saying, just wanted to tell you folks that I love myth. Read commits daily, and just bought a C++ book and QT book last week
21:32<gbee>well thanks :)
21:32<skd5aner>maybe eventually I'll be a regular in the dev arena, slowly but surely
21:33<skd5aner>you guys do good work, I'm going to go rollback to 16580. I'll report back if I see the behavior return to normal
21:33<gbee>QT painter sure exposes lots of mythui bugs :/
21:34<Chutt>gbee, the qt painter shouldn't be doing full-screen redraws
21:34<Chutt>opengl is, but, should be fine since it's accelerated
21:34<gbee>Chutt: might be confusing the two
21:35<gbee>I knew one of them did and guess I mixed them up
21:36<gbee>wonder how that fits with the differing behaviour I've been seeing
21:37<gbee>FWIW, just updated to [16584] and I'm able to enter the watch recordings screen just fine
21:41<skd5aner>hmmm, thanks for checking. Still updating, let you know in a few if the behavior went away for me at 16580
21:45<janneg>skd5aner: try 16586
21:46<janneg>kevin already replied
21:52<skd5aner>haha, JUST saw that in SVN 4 seconds ago. Thank you Janne (and of course Kevin). I'll update again.
21:57<skd5aner>If you folks don't mind, I did notice one other problem while doing make install
21:58<skd5aner>I believe someone made some changes recently regarding chown on the theme files
21:59<skd5aner>I believe it was 16491
21:59<Chutt>gbee, what's showing up broken with the qt painter?
22:00<skd5aner>now i get these while doing a make install in ubuntu
22:01<skd5aner>chown: missing operand after `../../../../local/share/mythtv/themes//DVR/tv_search.xml' / Try `chown --help' for more information.
22:02<Chutt>hrmph, my hdhr isn't showing up on the network
22:02<skd5aner>... for all theme files
22:02<skd5aner>happens in mythtv and myththemes
22:02-!-elupus [n=elupus@xbmc/staff/elupus] has quit ["Leaving"]
22:04-!-ToadP [] has quit []
22:06<skd5aner>looks like latest SVN fixes my original issue. Thanks!
22:06-!-jmk [] has joined #mythtv
22:09<Chutt>anyone have any idea what a blinking green light on a hdhr means?
22:11-!-skd5aner [] has quit []
22:18<GreyFoxx>static: the wiki has some release notes
22:19<Chutt>damnit, i bet this is dead
22:20<Chutt>i have another 5V/2A dc adapter, but it's the wrong size plug :/
22:22<danielk22>chutt do you have a volt meter?
22:24<danielk22>check it under no load. if it's less than 5V stop there, if it is >= 5, open up the box and see if the voltage is >= 5V under load..
22:25<danielk22>this won't tell you for sure, but it's more likely that the power supply is dead than that the device is taking > 2 amps..
22:25<Chutt>it does the same thing (power supply cycles) with the volt meter
22:26<Chutt>too late now, everything's closed
22:26<danielk22>:) time to dig in the box of transformers.. gerryrig a plug..
22:27<Chutt>wait, inside's +
22:27<Chutt>reverse the leads, it's sitting happy at 5V
22:28<danielk22>inside should be 5 to 6 volts
22:28<Chutt>yeah, looks fine with no load
22:29<danielk22>they often do look fine with no load when broken..
22:29<Chutt>i'll just grab a universal adapter tomorrow
22:29<Chutt>i guess i do have this huge benchtop power supply
22:29<Chutt>i could use that =)
22:29<danielk22>heh, with a dial :)
22:30<Chutt>with knobs and stuff
22:30<danielk22>I had one once that only had wattage control.. so you had to adjust the know depending on the load..
22:31<danielk22>"know" -> "knob"
22:58-!-gnome42 [] has quit [Remote closed the connection]
23:09-!-gnome42 [] has joined #mythtv
23:50-!-skamithi [] has quit ["WeeChat 0.2.3"]
23:55-!-reynaldo [] has quit [Read error: 110 (Connection timed out)]
---Logclosed Sun Mar 16 00:00:43 2008