Back to Home / #mythtv / 2008 / 01 / Prev Day | Next Day
#mythtv IRC Logs for 2008-01-05

---Logopened Sat Jan 05 00:00:37 2008
00:01<xris>roothorick: cool. sounds a bit like a -users thing, though (and definitely a good thing to put into the wiki).
00:02|-|rn114__ [] has quit [Read error: 113 (No route to host)]
00:11<roothorick>xris: channel changing has been widely reported as not working with the 4250HDC; my box is a 'black sheep' if you will
00:11<xris>roothorick: not a development topic, though.
00:11<roothorick>well... it is and isn't
00:13<roothorick>it is in the sense that a box with unusual behavior compared to other boxes of the same model has surfaced and researching the cause of the deviance could lead us to a method that works on all 4250HDCs
00:21|-|davilla [] has quit ["Leaving"]
01:22|-|reynaldo [] has joined #mythtv
01:42|-|jhatch [] has quit [Read error: 104 (Connection reset by peer)]
01:57|-|xris [] has quit ["Leaving."]
01:57|-|xris [] has joined #mythtv
02:34|-|rtsai [n=rtsai@] has joined #mythtv
02:50|-|roothorick [] has quit ["leaving"]
02:53|-|gnome42 [] has quit [Remote closed the connection]
03:36|-|MavT [] has joined #mythtv
03:54|-|MaverickTech [] has quit [Read error: 110 (Connection timed out)]
04:04|-|xris [] has quit ["Leaving."]
04:17|-|jhulst [n=jhulst@unaffiliated/jhulst] has quit [Read error: 113 (No route to host)]
04:23|-|gandalfcome [] has joined #mythtv
04:35<stuarta>anyone else got 2 actions bound to W by default?
04:35<stuarta>it started many moons ago when daniel split out the zoom mode handling and fill mode
05:05|-|gandalfcome [] has quit []
05:06|-|ascii___ [] has quit [Remote closed the connection]
05:47|-|grokky [] has joined #mythtv
05:57<gbee>janneg: until you've come up with a better solution, do you think there is any value in niceing the mythbackend preview generation process?
06:04|-|gandalfcome [] has joined #mythtv
06:08|-|mattwire [] has joined #mythtv
06:16|-|gandalfcome [] has quit []
06:20|-|mattwire [] has quit ["Leaving"]
06:22|-|grokky [] has quit []
06:30|-|mattwire [] has joined #mythtv
06:31|-|rn114__ [] has joined #mythtv
06:57|-|beavis [] has joined #mythtv
07:11[~]stuarta valgrinds the static EITCache to see what happens
07:11|-|superm1 [n=superm1@ubuntu/member/superm1] has quit [Read error: 104 (Connection reset by peer)]
07:15<rooaus>Is there a simple way in Qt to tell if a directory is a child (or the same dir)?
07:16<rooaus>I guess getting canonical path and seeing if the parent dir is the prefix of the child should do it?
07:18<rooaus>stuarta: Is it possible to valgrind LiveTV? I think I have a leak that pushed the frontend into 700MB of swap in a dedicated box with 1GB ram.
07:21|-|superm1 [n=superm1@ubuntu/member/superm1] has joined #mythtv
07:34<stuarta>not with 1g of ram
07:34<stuarta>i've 2g and that make doing the backend possible without swapping
07:36<rooaus>stuarta: ok. I use LiveTV quiet a bit and suspect it related to that. Is there anything else I could do to try and narrow it down?
07:36<stuarta>well livetv is basically 1x recording, 1x playback of recording
07:37<stuarta>so do each bit separately
07:37<Dibblah>rooaus: Is this current svn?
07:37<stuarta>there is something that grows every time a recording happens according to my graphs
07:38<rooaus>Dibblah: Yeah, as of a few days ago.
07:38<Dibblah>My bet would be the preview generator... Again.
07:38<stuarta>or still
07:38<stuarta>right that's enough of a valgrind.
07:41<rooaus>Anyone know the *right* way to do the dir check? I am trying to add a separate ripdir setting to mythmusic and only want to update the DB with new tracks if the rip dir is accessible from the music dir.
07:47|-|johnp__ [] has joined #mythtv
07:49<stuarta>morning johnp__
07:51<johnp__>stuarta:hey stuart
07:51<stuarta>fixup 8.5 commited
07:51<johnp__>so I saw
07:51<johnp__>just checking it out now.
07:52<stuarta>fyi there's an extra space floating around in your patches, it's after a }
07:52<stuarta>you'll see it if you look at the diff in trac
08:05<johnp__>I think it looks ok, although the lose (or rather not gain) of the episode numbers is probably a bit of an issue. (Think I'll just keep that as a private patch).
08:06<stuarta>which bit?
08:07<stuarta>where i don't remove them from the description?
08:11<stuarta>i'm undecided atm.
08:12<stuarta>i kinda like seeing them there
08:12<stuarta>but it's not consistent with the RT data
08:14<johnp__>could you give an example (or are you not bothered :) )
08:14<stuarta>examples are good
08:27<johnp__>need to go
08:27|-|johnp__ [] has left #mythtv []
08:28|-|AndyBurnsUK [] has joined #mythtv
08:33|-|johnp__ [] has joined #mythtv
08:40<stuarta>johnp__: i'll be around all weekend
08:45<johnp__>I think we could probably go with what we've got at the moment. (bits remaining are, "New." stuff, BBC4 removal and of course the episode number removal or not )
08:45<stuarta>i was thinking the CBBC regexps look a bit agressive
08:45<johnp__>oh, how come ?
08:46<stuarta>or not quite right
08:46<stuarta>are you trying to grep "CBBC." & "CBBC " ?
08:46<johnp__>ah well I can't vouch for that :)
08:46<johnp__>trying to
08:47<stuarta>k. then the regexp which is currently ^CBBC\\s*[\\.]
08:47<johnp__>maybe we don't need it ?
08:48<stuarta>should probably be "^CBBC[\\s*|\\.]"
08:48<stuarta>so the or is done on the whitespace and the dot
08:48<stuarta>or is the dot meant to be optional?
08:49<johnp__>dot should be optional
08:49|-|rn114__ [] has quit [Read error: 104 (Connection reset by peer)]
08:49|-|rn114__ [] has joined #mythtv
08:50<stuarta>with or without spaces?
08:50<stuarta>can't say i've ever seen "CBBC ."
08:50<johnp__>no it's probably overkill.
08:51<stuarta>same with CBeebies
08:51<stuarta>i'll fix that as well
08:51<johnp__>of course
08:51|-|AndyBurnsUK [] has quit ["Leaving."]
08:56|-|rn114__ [] has quit [Read error: 104 (Connection reset by peer)]
09:00|-|robthebob [] has joined #mythtv
09:05[~]stuarta goes to make a bit of lunch
09:48|-|brfransen [] has quit [Read error: 110 (Connection timed out)]
09:58<janneg>gbee: I doubt it but worth a try
09:59|-|cattelan [] has joined #mythtv
10:00<janneg>I haven't look closely at the bug and I'm not sure that the hang is caused by too much backend processes
10:13<stuarta>the backtrace on 4112 is perplexing
10:20<gbee>stuarta: qstring related, something to do with the query (insert?) being empty
10:21<stuarta>don't see how it could be
10:21<gbee>the sql error in the log and if you look at the backtrace, there is a "where" at the end which shows the crashing thread
10:22[~]stuarta looks at the log
10:22<gbee>stuarta: well not being familiar with the code, I can't say either but is it possibly iterating past the end of a list of events to INSERT? pure guesswork
10:23<gbee>the actual segault seems to have something to do with the destruction of the QString
10:24<gbee>the original ticket author suggests memory corruption, which is another possibility?
10:26<stuarta>i've been looking a locking/mutexes in the EITCache
10:27<stuarta>one thing i can't find is anywhere in the MSQLQuery code where there is locking to prevent multiple accesses through the DB
10:28<stuarta>another pair of eyes on that would be good
10:35<gbee>mysql handles a lot of that locking itself, or do you mean to the db object?
10:38<stuarta>the db object.
10:38<stuarta>what happens if we have multiple db objects?
10:39<stuarta>ah see, there's a lock in MDBManager
10:44|-|robthebob [] has quit [Read error: 113 (No route to host)]
11:01|-|dekar1 [] has joined #mythtv
11:05|-|mattwire_ [] has joined #mythtv
11:05|-|mattwire [] has quit [Read error: 104 (Connection reset by peer)]
11:05|-|djc_ [n=djc@] has joined #mythtv
11:06<djc_>!seen GreyFoxx
11:08|-|Dibblah [] has quit ["Kopete 0.12.4 :"]
11:11<gbee>are we now going to need to check every change we make for Windows compatability?
11:12<stuarta>i think once the compat.h stuff becomes fleshed out it'll not matter
11:14|-|aburns [] has joined #mythtv
11:15|-|dekarl [] has quit [Read error: 110 (Connection timed out)]
11:16<Daviey>gbee: Is windows now an offically supported platform?
11:18<stuarta>just a few people are hacking on it
11:18<gbee>Daviey: that's the question I guess I'm asking, a lot of work has gone into letting it build on Windows, but whether it's officially supported is another matter
11:18<Daviey>gbee: I suggest commiting known working code for linux, and it it causes a bug on windows - then a windows dev can submit a patch :)
11:19<gbee>I don't think anyone would say that OSX is officially supported, generally if there are problems with building on OSX then Nigel deals with it but linux devs are not expected to fix these problems
11:19<Daviey>It's not as if many people will use mythtv on windows (i know that quote wil lcome back to bite me)
11:19<gbee>Daviey: yeah
11:20<stuarta>famous last words they are...
11:20<gbee>Daviey: maybe I'm just prejudiced, but I hope almost no-one will want to use mythtv on Windows
11:20[~]Daviey would like to see some hardcoded references to the OS, just so it can't work on windows :)
11:21<aburns>I rarely even use mythTVplayer, so am not very likely to want a full win32 frontend.
11:23<gbee>think we maybe need OSX and Windows components in Trac
11:24<Daviey>or just ship a virtualbox/vmware image of linux+mythtv
11:25<stuarta>don't need to with all the myth distros out there
11:29|-|cattelan [] has quit [Read error: 110 (Connection timed out)]
11:47|-|cattelan [] has joined #mythtv
11:50<gbee>seems Asus are due to launch a second gen Eee at the same price as the existing one, think I'll probably wait for that
11:50<laga>it's not like i can buy the first one in germany anyways..
11:51<stuarta>gbee: even tho they are planning to install windows by default not linux?
11:52<gbee>stuarta: hadn't heard that ... pretty stupid decision if you ask me, but that will be the MS "marketing" dollars at work :(
11:53<gbee>hard to see how they could justify selling a Windows version at the same price as the linux one, so I'll wait and see
11:53<stuarta>the existing one comes with a windows install disk
11:53<stuarta>i guess they are planning to reverse the sitation
11:55<gbee>it does? hmm, thought it just came with instructions for installing Windows .... not sure I like the idea of paying for a Windows license I'll never use
11:55<stuarta>i'm pretty sure it does.
11:56<stuarta>guy at work has one
11:57<johnp__>stuarta: different subject, I think we could get rid of m_ukSubtitled from eitfixup.cpp
11:58<stuarta>i want to rework that whole area
11:58<stuarta>it needs to be one big globbing regexp which can pull out all the possible permutations
11:59<stuarta>of S,SL,AD
11:59<stuarta>and i've also seen HD as well
12:00<stuarta>btw. you seen what happens to Non-Stop Hits on Q???
12:00<stuarta>title: non subtitle: stop hits...
12:02<stuarta>back to the earlier thing. there's an outstanding ticket i have
12:02<stuarta>which is to allow things like sign language to modify the priorities of recording
12:03<stuarta>fixing the S,SL,AD handling is part of that.
12:03<stuarta>so you can ignore that area :)
12:03<stuarta>unless bored
12:04[~]stuarta puts on the regexp fu hat
12:07|-|cattelan [] has quit [Read error: 110 (Connection timed out)]
12:08<stuarta>possibly :)
12:08[~]stuarta makes a cuppa
12:11<gbee>stuarta: the HD, AD and I think SL stuff would be handled through additional descriptors once Janne has done that work, so reading it from the description isn't necessary
12:11<Daviey>stuarta: i haven't opened the cd yet - but thought it just included window drivers - not an installation disk
12:11<stuarta>i don't think we have the descriptors in the data stream at this point
12:12[~]stuarta is no expert on the eePC
12:12<stuarta>if you have 1, you have more than me
12:12<Daviey>i have 2 :)
12:12<Daviey>and gave one to the parents for xmas
12:13<stuarta>gbee: i'd actually prefer to get it from descriptors if possible
12:13<stuarta>people with a clue generally do that bit
12:13|-|_rik_ [] has joined #mythtv
12:14<gbee>stuarta: they are there - component_type: 32 (0x20) == Content&Component: (= DVB subtitles (for the hard hearing) with no monitor aspect ratio critical)
12:14<gbee>Content&Component: (= video, 16:9 aspect ratio without pan vectors, 25 Hz)
12:15<stuarta>but in the EIT data or the streams themselves?
12:15<gbee>EIT data
12:15<gbee>not looked at the streams themselves
12:15[~]stuarta does some dvbsnoops
12:16<gbee>there is also data on the audio tracks available, stereo/mono/surround, pretty sure AD gets described too, but it's all in the big table in the specs
12:16|-|cattelan [] has joined #mythtv
12:17|-|_rik_ [] has left #mythtv []
12:18|-|jamesd_ [] has joined #mythtv
12:18<stuarta>ah, see it's in the PMT for the currently active program
12:18<stuarta> MPEG-DescriptorTag: 10 (0x0a) [= ISO_639_language_descriptor]
12:18<stuarta> descriptor_length: 4 (0x04)
12:18<stuarta> ISO639_language_code: eng
12:18<stuarta> Audio_type: 3 (0x03) [= visual impaired commentary]
12:19|-|gnome42 [] has joined #mythtv
12:19<Daviey>Is anybody actively working on the recording of multiple streams concurrently on the same multiplex?
12:19<stuarta>you been living under a rock?
12:19<gbee>yeah, there are descriptors for AD audio as well - page 36 of ETSI EN 300 468
12:19<stuarta>don't mean they are broadcast
12:20[~]stuarta digs out his specs
12:20<gbee>Daviey: already done, just waiting to be merged into trunk
12:20<Daviey>ahh, i saw the code in svn - but assumed it wasn't mature enough
12:21<gbee>stuarta: I know, but at least we can add support for those places which do broadcast the info, if it turns out that the UK isn't one of those then we can still parse it out of the programme description
12:21|-|reynaldo [] has quit [Read error: 113 (No route to host)]
12:21<gbee>Daviey: will probably be in trunk this time next week
12:21<stuarta>gbee: the UK will probably implement it first
12:21<Daviey>gbee: awesome!
12:22<gbee>stuarta: that's what I was thinking, in fact I kinda guessed that the Freeview consortium were responsible for that addition since I'm not aware of any other country which offers AD audio streams
12:23<gbee>well by Freeview I mean that group of interested parties here in the UK
12:24<gbee>not literally the Freeview group since it probably didn't even exist at the time
12:24<justinh>does that mean we can have the option to not record subs & AD then? :)
12:25<stuarta>justinh: probably in the longer term
12:25<gbee>justinh: yeah, or to record HD showings in preference to SD
12:25<stuarta>or record the version without the little man waving his hands around
12:25<stuarta>or with if you prefer
12:25<justinh>yay for that
12:26<gbee>all that was in the back of my mind when I expanded the audio/video/subtitle properties
12:26<gbee>going to prepare some food, back later
12:28<stuarta>okay it is being broadcast already in the UK
12:30<justinh>did I ever finish doing all the icons for that or not? can't remember
12:33<Daviey>justinh: did you make a tarball with more chan' icons than the current one linked from parker1?
12:34<justinh>no but I make a script to download all the ones with xmltvids available from lyngsat. it's in /contrib - or was
12:34<justinh>*but I did make a...
12:34<gbee>justinh: yeah, all the icons got done
12:36<gbee>Daviey: that's been superceeded by the stuff in trunk (the more people use it, the better it gets, so yes I'm pushing everyone to try it)
12:36<gbee>there is a script in contrib or a buggy integrated client in mythtv-setup
12:38<Daviey>oww, didn't realise that - will check it out
12:38<stuarta>s/buggy/work in progress/ :)
12:38<Daviey>who looks after the domain
12:38<stuarta>da boss
12:44|-|jamesd_ [] has quit [Remote closed the connection]
12:47<stuarta>well now i know that we go to summer time at 2008-03-30 01:00:00 (UTC)
12:50<gnome42>oh, the MythEvent extradata patch already got committed ... no need for ticket then.
12:50<stuarta>heh, i was planning to then noticed it had been done :)
12:52<janneg>23:44 < janneg> gnome42, stuarta: I'll apply the patch in a couple of minutes
12:52<gnome42>yeah, cool.
12:52<gnome42>janneg: oh, missed that ... right before my system meltdown probably :)
12:55<janneg>no problem
12:55<stuarta>janneg: btw. currently running with the static EITCache
12:58<janneg>bah, the software encoder bitrate bug is nasty. it looks like a problem in libavcodec. I'm moving to something else
13:01<janneg>stuarta: I think we need to change the pruning. iirc the conplete cache is written back to the db on every transponder change
13:02<stuarta>we do
13:02<janneg>which is bad if the cache is shared
13:02<stuarta>every time it changes channel it writes to the DB and drops the current cache
13:02<stuarta>which immediately gets reloaded
13:03<stuarta>don't mind the writing
13:05<stuarta>i'll continue working on the caching. it's something i want to sort out
13:05<janneg>the writing is not that bad since it writes only modified entries
13:06<stuarta>it's a nice checkpoint mechanism
13:06<janneg>ok, I'll implement the EIT component parsing for gbee
13:07<stuarta>i was just having a quick look at the raw data
13:13<gbee>janneg: I started writing a patch but didn't finish it after you said you would do it, it might not apply to trunk, but I can dig it out and it might save you some time
13:15<janneg>gbee: there was not much work in the basic eit code. I would guess that it still applies
13:16<gbee>just trying to find it, have over 200 files in my patches directory
13:16<janneg>I wasn't aware that you had already started. it might have been finished by now if I hadn't inferred
13:16<stuarta>you tried using quilt?
13:17<janneg>I'ld guess quilt created the patches dir
13:18<stuarta>ho hum...
13:20<janneg>gbee: grep -rn "\+\+\+.*eit" patches
13:20<gbee>janneg: can't find the patch, grepping only finds one which changes the hdtv,stereo and subtitled variables from bools to ints
13:21<gbee>stuarta: not using quilt, most of these patches were applied a long time ago I just never deleted them
13:21<janneg>no problem, changes are simple enough
13:21<gbee>I need to spring clean
13:28<stuarta>heh, i've 88 patches in my quilt patch directory
13:29<stuarta>only 14 of which are still in the quilt
13:29<stuarta>most suffering from bitrot
13:29<johnp__>stuarta:now at eitfixup level 10 , dunno if you want it though.
13:30<gbee>superm1: any chance you can get xmltv 0.5.49 (or maybe a higher version) backported to the feisty backports/update repo? When Mythtv 0.20.2 was backported it increased the required version of xmltv and there are fixes in 0.5.47 and 0.5.50 which were designed to reduce the stress on source websites
13:30<stuarta>hmmm, looks like i had a bash at the audio descriptors last jan :)
13:30<gbee>I've been promising myself that I would look at quilt for nearly a year now
13:31<stuarta>johnp__: pastebin it and i'll have a look
13:32[~]stuarta goes to heat up some food
13:32<gbee>superm1: mythtv 0.20.2 used with feisty's xmltv 0.5.44 is causing problems for uses and where we get the data from
13:39|-|KnyghtMare [] has joined #mythtv
13:40|-|KnyghtMare [] has left #mythtv []
13:43<gbee>another one for the CSI treatment, Law and Order:SVU (etc)
13:46<stuarta>the CSI's work :)
13:48[~]stuarta gets concerned by the sound of his dinner exploding
13:49<stuarta>johnp__: that looks okay.
13:49<stuarta>thou ukBB3 & 4 can be combined
13:50<superm1>gbee, yeah i can try to
13:50<stuarta>"BBC \\w+ on BBC \\w+\\."
13:51<gbee>stuarta: not for the Law and Order's they don't ;)
13:52<gbee>superm1: thanks
13:53<stuarta>hmmm yeah, no different to superman however
13:53|-|Chutt [] has joined #mythtv
13:53<stuarta>smallville i mean actually
13:54<stuarta>there are quite a few like that.
13:54<gbee>I'm sure there are plenty more examples, that was just one I noticed for the couple of days I was using EIT
13:56[~]stuarta nods
13:59|-|Toxicity999 [] has joined #mythtv
14:00|-|Toxicity999 [] has quit [Client Quit]
14:00|-|Toxicity999 [n=bryan@unaffiliated/Toxicity999] has joined #mythtv
14:02<superm1>gbee, 0.5.49 is more feasible than 0.5.50, since it would likely be a backport coming from gutsy (otherwise gutsy needs the backport from hardy first)
14:09<gbee>superm1: no problem, 0.5.49 would fix the biggest problem
14:09<gbee>and thanks for trying, even if you can't make it happen
14:09<superm1>i should be able to, only problem is i cant stay on top of people this week to make sure it happens as i'll be taking vacation for a week
14:09<superm1>i'll at least get the test builds and bugs filed for it
14:10<gbee>superm1: ok
14:54|-|reynaldo [] has joined #mythtv
14:57<superm1>gbee: here is the bug if you'd like to track it this week
14:57[~]GreyFoxx envisions a future of "commit xyz wont compile on ABC platform" becoming more common now
15:12|-|rhp [] has left #mythtv ["Leaving"]
15:16<jams>at least this last one was simple to fix. I honestly didn't give other platforms any thought.
15:18<GreyFoxx>I was gonna make the change anyway after I found the other method when I was doing it for the upnp reload stuff yesterday :)
15:18|-|xris [] has joined #mythtv
15:21|-|jamesd_ [] has joined #mythtv
15:36|-|_rik_ [] has joined #mythtv
15:47|-|_rik_ [] has quit ["leaving"]
16:05|-|jhulst [n=jhulst@unaffiliated/jhulst] has joined #mythtv
16:10<superm1>gbee, there is a slight issue that one of the dependencies of xmltv needs to be upgraded too, and by doing so many reverse dependencies of that dependency may break (there is a compatibility change according to the Changelog)
16:11|-|davilla [] has joined #mythtv
16:16|-|aburns [] has left #mythtv []
16:21<xris>superm1: lame
16:21<superm1>so the most likely solution will be to encourage folks to upgrade to gutsy
16:44<davilla>the current svn db version is 1205, anyone know which version when from 1202 to 1203. These are not being listed in the commits.
16:45<stuarta>something that touches dbcheck.cpp
16:47<GreyFoxx>Daviey: there was a commit which included a MySQL 5 specific change
16:47<GreyFoxx>if you are using anything older it would fail
16:48|-|johnp__ [] has quit [Read error: 104 (Connection reset by peer)]
16:52<davilla>I'm running 15221 but need to jump to 15279 to pickup OSX 10.5 IR remote fixes. 15279 fails to build on
16:53<stuarta>i just build the latest but on osx 10.4
16:54<janneg>davilla: is quite handy
16:59<davilla>Thanks, I'll bookmark that page for future reference. Looks like 15279 would be currentDatabaseVersion 1204 which burns me anyway, so manual patch time.
17:07|-|grokky [] has joined #mythtv
17:11<davilla>For the OSX MythFrontEnd fans, for running under 10.5 FrontRow, see
17:13|-|jhulst [n=jhulst@unaffiliated/jhulst] has quit [Read error: 110 (Connection timed out)]
17:28|-|reynaldo [] has quit ["Lost terminal"]
17:31|-|mattwire_ [] has quit [Read error: 110 (Connection timed out)]
17:35|-|davilla [] has quit ["This computer has gone to sleep"]
17:37|-|mattwire [] has joined #mythtv
18:26|-|feiner [] has quit [Remote closed the connection]
18:30|-|feiner [] has joined #mythtv
18:33|-|beavis [] has quit ["Verlassend"]
18:34|-|mpontill1 [] has joined #mythtv
18:37|-|feiner [] has quit [Remote closed the connection]
18:42|-|feiner [] has joined #mythtv
18:43|-|tulbreak [] has joined #mythtv
18:44|-|mattwire [] has quit ["Leaving"]
18:48|-|tulbreak [] has quit [Remote closed the connection]
18:49|-|tulbreak [] has joined #mythtv
18:52<lucas2>anyone around who tried a win32 compile?
19:08|-|aburns [] has joined #mythtv
19:16<justinh>here we go...
19:21|-|tulbreak [] has quit [Remote closed the connection]
19:26<GreyFoxx>luncas: Try #mythtv-users
20:06|-|Toxicity999 [n=bryan@unaffiliated/Toxicity999] has quit [Remote closed the connection]
20:07|-|Toxicity999 [n=bryan@unaffiliated/Toxicity999] has joined #mythtv
20:10<GreyFoxx>any expert on the housekeeper code around ?
20:16|-|javatexan [] has joined #mythtv
20:16<javatexan>howdy all!!!
20:17<javatexan>well I have a new glich....LOL...the line out of my mobo assumes the speakers it is plugged into will be amplified, so sound to my TV is really quiet. Anyone know of a cheap fix?
20:17<GreyFoxx>You might wanna try #mythtv-users :)
20:18<javatexan>sorry greyfoxx....wrong bookmark!!!
20:18<GreyFoxx>no worries :)
20:18<javatexan>how you been?
20:19<GreyFoxx>always busy :)
20:19<javatexan>I guess that is do you fit into the dev of mythtv?
20:20<GreyFoxx>just one of the crowd. work on what bugs me or interests me. Currently upnp stuff
20:21<GreyFoxx>trying to get some stuff ready before 0.21
20:21<javatexan>yeah i bet, that interface usually looks a lot like i18n eh? LOL
20:22<javatexan>well if I had a mentor, I could try to help; but I am way too green to help at this point
20:23<javatexan>so what are you trying to set up for UPNP....
20:24<GreyFoxx>just working on support for the not so universal upnp clients
20:24<GreyFoxx>xbox360 support tweaks, support for WMP11
20:25<GreyFoxx>and an arms length of other stuff
20:25<javatexan>LOL...I know the feeling. That's why I had to take a couple years off open source dev....My arms got too heavy to lift. LOL
20:29[~]GreyFoxx wonders where his kid put the DVD she wants to watch
20:31|-|aburns [] has left #mythtv []
20:32[~]javatexan understands...his kids do the same thing
20:33|-|Toxicity999 [n=bryan@unaffiliated/Toxicity999] has quit [Remote closed the connection]
20:34|-|Toxicity999 [n=bryan@unaffiliated/Toxicity999] has joined #mythtv
20:41|-|Toxicity999 [n=bryan@unaffiliated/Toxicity999] has quit [Remote closed the connection]
20:41|-|Toxicity999 [n=bryan@unaffiliated/Toxicity999] has joined #mythtv
20:42|-|Toxicity999 [n=bryan@unaffiliated/Toxicity999] has quit [Remote closed the connection]
20:42|-|Toxicity999 [n=bryan@unaffiliated/Toxicity999] has joined #mythtv
20:44|-|Toxicity999 [n=bryan@unaffiliated/Toxicity999] has quit [Remote closed the connection]
20:44|-|Toxicity999 [n=bryan@unaffiliated/Toxicity999] has joined #mythtv
20:44|-|Toxicity999 [n=bryan@unaffiliated/Toxicity999] has quit [Remote closed the connection]
20:44|-|Toxicity999 [n=bryan@unaffiliated/Toxicity999] has joined #mythtv
20:46|-|feiner [] has quit [Read error: 110 (Connection timed out)]
21:02|-|nemik [] has joined #mythtv
21:02<nemik>hello. is anyone here working on mythstream? i wrote a parser i'd like to check in of possible
21:03|-|Disconnect [] has joined #mythtv
21:06<laga>nemik: mythstream is not maintained in mythtv's version control system. you'll have to contact the author
21:06|-|Baudrate [] has joined #mythtv
21:07|-|Baudrate [] has left #mythtv ["Ex-Chat"]
21:07<nemik>laga: ok cool. thanks
21:10|-|feiner [] has joined #mythtv
21:22|-|Toxicity999 [n=bryan@unaffiliated/Toxicity999] has quit [Remote closed the connection]
21:23|-|Toxicity999 [n=bryan@unaffiliated/Toxicity999] has joined #mythtv
21:36|-|jhulst [n=jhulst@unaffiliated/jhulst] has joined #mythtv
21:48|-|davilla [] has joined #mythtv
22:09|-|knowledgejunkie [n=knowledg@unaffiliated/knowledgejunkie] has joined #mythtv
22:10<knowledgejunkie>xris: about?
22:10|-|javatexan [] has left #mythtv ["Leaving"]
22:10<knowledgejunkie>xris: I've compiled a list of the outstanding channels/icons coverd by the uk_rt XMLTV grabber that do not return an icon with the channel_icons script
22:11<knowledgejunkie>xris: it's in CSV format as XMLTV,URL - can I pastebin it for you?
22:11<xris>knowledgejunkie: returns no icon?
22:11<xris>put it in a ticket
22:11<knowledgejunkie>xris: ok
22:11<xris>crap.. need to fix that grabber script so it has all of the new lyngsat URLs
22:17<xris>knowledgejunkie: actually... let me rerun this lyngsat scan and then have you research again
22:17<knowledgejunkie>xris: ok - almost pressed the submit button then
22:20<xris>scan is running now.. I'll let you know when it's done (probably 30 mins)
22:20<knowledgejunkie>xris: q - Am I correct in thinking that the script/mythtv-setup still retrieves icon URLs against XMLTV IDs?
22:21|-|otwin [n=otwin@] has quit [Read error: 110 (Connection timed out)]
22:21<xris>and dvb id numbers
22:21<xris>or atsc id numbers
22:22<knowledgejunkie>xris: OK - so how does the update script pull in new links between XMLTV IDs and URLs? Is it parsing the channel_ids files from the XMLTV grabbers?
22:24<xris>not sure what you mean. local mythtv sends all of the info it has to, which has a db based on user submissions for similar matches
22:27<knowledgejunkie>xris: explaining in another way i) I maintain the uk_rt XMLTV grabber and channel list ii) I create the XMLTV IDs and find suitable icons for those channels iii) I semi-regularly run a scan against the services DB to identify which XMLTV IDs have no icon associated with them iv) I want to bulk populate the services DB so that UK/Eire users can get icons based on any XMLTV IDs they submit
22:28<xris>won't happen
22:28<xris>the current system is based on user submissions for what they think the correct icon is
22:29<xris>once we get 10 or so submissions from different users, we mark that as the valid icon
22:29<knowledgejunkie>OK - so the icon that the XMLTV project decides to use is not good enough?
22:29<xris>that'll probably change in the future as I rethink things (like maybe get something hosted at with standard info about all xmltv-known channels)
22:29|-|unkn0wn2u [] has joined #mythtv
22:30<xris>xmltv project isn't a cohesive group (I got that one from the current maintainer of the project as a whole)
22:30<knowledgejunkie>they're the icons that everyone using XMLTV will have anyway who uses the UK grabber with MFDB for their listings - it seems sensbile (only to me it seems) to have that data already available for users
22:31<xris>yeah. but I don't actually have a way to do that
22:31<xris>the system only lists icons that people have submitted through the script
22:32<xris>and I'm not about to rewrite the server code to allow it when I'm going to completely redesign it for this other stuff that Robert (xmltv maintainer) and I are talking about
22:33|-|fenriz_ [] has joined #mythtv
22:33|-|fenriz_ [] has quit [Client Quit]
22:39<knowledgejunkie>No worries. I'm surprised the current system doesn't allow simple DB updates, and I'll wait for the redesigend system. Is the 'other stuff' SchedulesDirect related? I've not seen anything mentioned on xmltv-dev
22:42<xris>it could, but it's a lot of joined tables in SQLite, and I'm lazy (and a bit buzzed) at the moment
22:42<xris>anyway, scan is done now
22:43<knowledgejunkie>ok - lemme rerun my script
22:44|-|feiner_ [] has joined #mythtv
22:45|-|feiner [] has quit [Read error: 104 (Connection reset by peer)]
22:57<knowledgejunkie>xris: OK - the same failures as before - all bad lookups return with "ERROR: Unknown xmltvid"
22:57<xris>knowledgejunkie: I think that means that things haven't been submitted yet
22:58<xris>database doesn't know about xmltvid fields until they've been submitted
22:58<knowledgejunkie>xris: is there are submit url that can accept both xmltvid and icon url?
22:59<knowledgejunkie>xris: or does it require the icon ID from the DB?
22:59<xris>at the moment, you have to run the channel icon script in contrib
23:00<xris>it pulls a bunch of stuff from the client db as well as the services db
23:01<knowledgejunkie>xris: last year (March I think) we went through this rigamarole for the then-current XMLTV IDs - what did we do then?
23:03<xris>no clue
23:03<xris>too long ago
23:03<xris>not sure I even had set up back then
23:03<knowledgejunkie>I'm sure you did, because I was running the same script (I think)
23:04<xris>wow, yeah, commit logs go back to feb
23:04<knowledgejunkie>if I want to submit them via the script, do I need the channels configured in the client DB?
23:04<xris>the way it works now, yeah
23:05<xris>esp. because I'd like to see more than just xmltvid for things... because it would eventually be nice to have a dvb->xmltvid magic linkup.
23:06<knowledgejunkie>xris: that's in the works on xmltv-dev
23:08<xris>funny.. I just mentioned it to robert the other day and he said he didn't think anything like that was being talked about
23:09<knowledgejunkie>xris: see - we were discussing it on the list last October
23:10<xris>not quite the same thing, but close
23:10<xris>lineup would also be handy, though
23:11<xris>basically duplicate some of what SD is buying from TMS...
23:11<xris>esp. since things like local broadcast lineups are hard to come by
23:11<knowledgejunkie>in the rest of the world, things don't seem to be so complex
23:11<xris>broadcast is.
23:12<knowledgejunkie>by broadcast do you mean OTA, or all sources?
23:12<xris>since it's about geography... you might get a channel, but 3 blocks away might be behind a hill and not get reception worth talking about
23:14|-|feiner_ changed nick to feiner
23:14<xris>TMS info about which channels you can and can't receive by postal code is relatively inaccurate.. either way over-estimates, or misses a bunch of channels that come in just fine
23:16<knowledgejunkie>Here in the UK we don't have the same affiliate-based system like the US.
23:16<xris>my first stop for this was to come up with all possible ways to represent a channel (or station as TMS calls them)... xmltvid, tms_id (which SD can't auto-fill, but users could probably fill in if it shows up in the feed data), name, broadcast frequency info, callsign.
23:16<xris>knowledgejunkie: sure you do... just not OTA.
23:17<xris>but yeah, you don't have the same issues with being able to get BBC2, but not BBC1 because the broadcast tower is in a different direction.
23:18<xris>anyway, once we have a way to reference channel information, then we could start thinking about a way to maintain lineups by region/provider.
23:18<knowledgejunkie>with digital/cable/satellite/analog you essentially get the same content with only very minor regional variations
23:20<xris>unless you have sky or another sat provider
23:21<knowledgejunkie>in the UK there is only Sky currently
23:24<xris>no c-band?
23:24<knowledgejunkie>and when FreeSat starts up, they'll show the same content on the same channels as Sky
23:24<xris>or people pulling data from satellites for other countries? asian tv, etc?
23:25<xris>anyway, still need to account for non-UK users.
23:26<knowledgejunkie>I guess if you've got a big enough motorised dish
23:26<knowledgejunkie>in terms of commercially available platforms joe user has in the UK, it's only Sky at the moment, with FreeSat to follow this year
23:27<xris>makes it easy for the UK folks, I guess.
23:28<xris>though you still have to keep track of id numbers for dvb-s, dvb-c, dvb-t, analog tv and xmltvid.. and get them all connected to channel name/network/icon/etc info
23:28<knowledgejunkie>if you want cable TV, you have VirginMedia
23:29<knowledgejunkie>not really very competitive here :)
23:29<knowledgejunkie>xris: that's what the lineup thing should do
23:29<xris>no.. lineup is "which channels are available via which source".. I'm still talking about how to find a specific channel, regardless of source
23:31<xris>as in, you can know that "scifi uk" is xmltvid, 11:34234:12312 on dvb-s (serviceid:freqid:whatever) and "channel 232 on sky"
23:32<knowledgejunkie>that's also discussed in the lineup proposal - you have detailed <channel> elements with that information, and can then associate those channels with service providers to give you a specific lineup
23:33<knowledgejunkie>there's a thread called "Start planning a lineup xml-format" which discusses this
23:33<xris>yeah. and I'm talking about the database to keep track of all of that info, along with channel logo, network logo, related channels/networks, etc.
23:38<xris>basically, I'm talking about the data that your lineup stuff lists....
23:39<knowledgejunkie>is the DB to site alongside XMLTV as a spearate entity, or be part of it?
23:39<xris>not sure.
23:39<knowledgejunkie>a sort of global TV tuning DB?
23:39<xris>I'm hoping it can be a joint venture between xmltv, schedules direct and the media wiki folks (composing an email to them right now)
23:40<xris>sort of an open source version of, but way better.
23:41<knowledgejunkie>have you seen
23:45<xris>but that's similar to what I was thinking... but unfortunately a pure-wiki type setup doesn't allow proper searching by specific field names...
23:46<xris>guess I should look into getting in touch with them, too
23:48<knowledgejunkie>seems like there's a lot of info on the site
23:49<xris>but no email address contact info
23:49<xris>but I'll dig around
23:50<knowledgejunkie>cool - I better get some sleep
23:50<knowledgejunkie>i posted a ticket - #4435 with the xmltvid/icon info - use it if you can, and close it if you can't - at least the info is there now
23:56<xris>cool, thx
23:57<jhulst>What do most people do for developing? Use a virtual machine? Are there ways of simulating tuner inputs?
23:57<xris>I dev on my live box
23:57<xris>other people set up a dev box with its own tuner
23:57<xris>virtualization doesn't work so well because of the specialized tv hardware
23:58<GreyFoxx>I use both vms for some stuff and live setup for other
23:58<GreyFoxx>using a vm for the upnp stuff right now
23:59<jams>jhulst- virtual tuners are also possible.
23:59<jhulst>I'm looking into multi-rec with the HD Homerun, I'll probably need a live machine for that
23:59<GreyFoxx>jhulst: I don't see why that wouldn't work with avm
---Logclosed Sun Jan 06 00:00:04 2008