Back to Home / #mythtv / 2008 / 02 / Prev Day | Next Day
#mythtv IRC Logs for 2008-02-19

---Logopened Tue Feb 19 00:00:12 2008
00:01-!-JoeBorn [] has joined #mythtv
02:02-!-sphing [] has joined #mythtv
02:05-!-sphing [] has left #mythtv ["Leaving"]
02:10-!-joe_ [] has joined #mythtv
02:10<joe_>hello, what exactly is mythtv
02:13-!-joe_ [] has quit [Client Quit]
02:18-!-gnome42 [] has quit [Remote closed the connection]
02:28-!-onyxsoft_ [] has joined #mythtv
02:29-!-onyxsoft [n=david@] has quit [Read error: 113 (No route to host)]
02:34-!-eskil [] has quit ["Ex-Chat"]
02:50-!-mykeul [n=mykeul@] has joined #mythtv
03:09-!-xris [] has quit []
03:18-!-Puhi [] has joined #mythtv
03:41-!-czth__ [n=dbrobins@nat/microsoft/x-6540f79868dc39fe] has joined #mythtv
03:58-!-czth_ [n=dbrobins@nat/microsoft/x-5d994f23f62f0bc0] has quit [Read error: 110 (Connection timed out)]
04:45-!-MaverickTech [] has joined #mythtv
05:02-!-MavT [] has quit [Read error: 113 (No route to host)]
05:16-!-kormoc [n=kormoc@unaffiliated/kormoc] has quit []
05:19-!-ma9mwah|c2d [] has joined #mythtv
05:31-!-ma9mwah|tired [] has quit [Read error: 110 (Connection timed out)]
06:10-!-Balachmar [] has joined #mythtv
06:10<Balachmar>Hi guys, someone at work (who has a bit more C++ experience) helped me out. And now I hjave found out that in order to let my code work I need to add md5.hh and to the project so that
06:11<Balachmar>These are compiled. Because at this moment there is no md5.o, this would probably be the cause of the runtime error
06:11<Balachmar>But how do I add files to the libmythmusic?
06:23-!-BleedAway [] has joined #mythtv
06:41-!-MrGandalf [] has quit ["Leaving"]
07:18-!-TelnetManta [] has quit ["Ex-Chat"]
07:39-!-lsobral [n=sobral@] has joined #mythtv
07:42-!-TelnetManta [n=benwilli@] has joined #mythtv
07:47-!-MavT [] has joined #mythtv
07:57<Balachmar>ok, now I know for sure, that all I need to learn now is know how I should add libraries to the mythmusic project so that make does the right thing
07:59<Balachmar>Where do I specify the libs that are needed to compile mythmusic? I need to add the openssl library
08:03-!-Dibblah [] has quit [Read error: 113 (No route to host)]
08:04-!-Dibblah [] has joined #mythtv
08:05-!-MaverickTech [] has quit [Read error: 113 (No route to host)]
08:12<gbee>superm1: Could you help this user supply the right output from apport?
08:32-!-renatofilho^ [n=renato@] has joined #mythtv
08:32<gbee>danielk22: F7 is bound to two conflicting contexts in tv_play.h, Signal Monitor and ITV - seems logical to move the signal monitor to another key, any preference?
08:36-!-dekarl [] has joined #mythtv
08:52-!-dekar1 [] has quit [Read error: 110 (Connection timed out)]
09:01-!-otwin_ [n=otwin@] has joined #mythtv
09:08-!-otwin__ [n=otwin@] has joined #mythtv
09:08-!-otwin_ [n=otwin@] has quit [Read error: 104 (Connection reset by peer)]
09:12-!-otwin [n=otwin@] has quit [Read error: 110 (Connection timed out)]
09:12-!-otwin__ is now known as otwin
09:15-!-doschdn [] has joined #mythtv
09:16-!-doschdn [] has left #mythtv []
09:23-!-Cardoe [n=Cardoe@gentoo/developer/Cardoe] has joined #mythtv
09:32-!-R[OnUaJcu [n=otwin@] has joined #mythtv
09:33<gbee>would it stretch the definition of feature too far to include ?
09:34<gbee>the attached patch is bogus btw
09:35-!-otwin [n=otwin@] has quit [Read error: 110 (Connection timed out)]
09:35-!-R[OnUaJcu is now known as otwin
09:57<superm1>gbee, i added a comment. if the guy doesn't respond with what I said in a timely fashion, nuke the bug with a comment indicating to submit apport reports directly to us in the future (so we can link them)
09:57<gbee>superm1: thanks
10:02-!-Hannibal- [] has joined #mythtv
10:10-!-jwhite [] has quit [Remote closed the connection]
10:12-!-jwhite [] has joined #mythtv
10:17-!-canZzZz is now known as canatella
10:20-!-JoeBorn [] has quit [Read error: 110 (Connection timed out)]
10:26-!-`\c^^]vOq [n=otwin@] has joined #mythtv
10:33-!-otwin [n=otwin@] has quit [Read error: 110 (Connection timed out)]
10:33-!-`\c^^]vOq is now known as otwin
10:50-!-Balachmar [] has quit ["Ex-Chat"]
11:03-!-bilbo75 [] has joined #mythtv
11:06-!-djc_ [n=djc@] has quit [Read error: 110 (Connection timed out)]
11:06-!-gnome42 [] has joined #mythtv
11:07-!-djc_ [n=djc@] has joined #mythtv
11:10-!-bilbo75 [] has left #mythtv []
11:19-!-MrGandalf [] has joined #mythtv
11:29-!-eskil [] has joined #mythtv
11:42-!-splAt1 is now known as splat1
11:48-!-BathoryQuorthon [] has joined #mythtv
11:50-!-eskil [] has quit [Read error: 110 (Connection timed out)]
11:58-!-mykeul [n=mykeul@] has left #mythtv []
12:07<gbee>danielk22: MHEG is always on, it's not like Teletext where the teletext buttons are only supposed to work when you bring up the menu - so long as MHEG is enabled in the global settings keys in the ITV Menu context will always override those in the playback context
12:08<gbee>and I forgot you were in here, so I've already said the same to the list
12:09<gbee>the "ITV Menu" context shouldn't really exist, it's a misunderstanding of how action contexts are supposed to work - those keys should be in the "TV Playback" context
12:10-!-siXy [i=siXy@] has joined #mythtv
12:10-!-xris [] has joined #mythtv
12:22<danielk22>gbee, ok move em to TV Playback then, to avoid confusion
12:22<danielk22>I thought we made MHEG hidden by default in MythTV..
12:26<gbee>it's off by default, but if it's enabled then the binding for the signal manager clashes
12:27<gbee>since most UK users will want it enabled, then it makes sense to move the signal monitor to something which won't clash with any other default bindings
12:29<xris>can someone else verity the length of recorded.basename in the current db schema?
12:29<xris>I got an email report from a user that 128 (the last official value I know of) is too short for some people
12:30<gbee>`basename` varchar(255) NOT NULL,
12:31-!-systemloc [] has joined #mythtv
12:32-!-systemloc [] has left #mythtv []
12:41-!-Balachmar [] has joined #mythtv
12:44<Balachmar>Hi, in what file do I have to add the libraries that I use for the mythmusic plugin?
12:46<Chutt>xris, 255 here as well.
12:47<xris>ok, cool. looks like I or someone else fixed.
12:48<xris>I vaguely recalled fixing it, but wasn't sure.
12:52<gnome42>Aha! A nice PIP fix ... creating ticket.
12:56-!-geu [] has joined #mythtv
12:57-!-geu [] has left #mythtv ["Verlassend"]
12:57<gbee>do we want to add to the list of bugs which need fixing for 0.21 by assigning a few 'Unknown' bug tickets, or do we call a halt and just work through the bugs we've already chosen?
13:05<danielk22>gbee: I use the "unknown" tag to mean "unknown, like not likely to be fixed by 0.22" so I'd be more inclined to move "0.22" bugs up to "0.21".. but there are enough "0.21" bugs that I don't think we'll be able to fix them all by release time, and every bug fix carries the risk of introducing new bugs...
13:06<danielk22>of course, all new bugs are tagged as unknown and need to be classified..
13:06<gbee>well all bugs are assigned to unknown by default, meaning that unless someone reviews them and assigns them
13:07<danielk22>yeah, no harm in reading through "unknown" and looking pushing the more serious ones up to 0.21
13:07<gbee>heh, yeah - I think we'd be ok assigning bugs to 0.21 even if we don't intend to fix them for release, fixes will continue to go to the branch for future point releases
13:11-!-renatofilho^ [n=renato@] has quit [Remote closed the connection]
13:11<gbee>can't hurt to give some of the segfault tickets a second look before 0.21
13:13-!-renatofilho^_ [n=renato@] has joined #mythtv
13:17<xris>gbee: so my mythweather setup is completely screwed up.
13:17<xris>can't add sources because I don't have screens, and can't add screens because I don't have a source for certain data points (it lists a bunch of stuff like icon, etc)
13:18-!-lcase [] has joined #mythtv
13:19<danielk22>do all the themes need some certain minimums to support mythweather? I got it working here a few weeks ago, but then it broke again on one of my svn up's
13:19<gbee>xris: truncating the tables didn't work? It should scan for sources when you enter setup, if it's not finding them then something is pretty wrong
13:20-!-canatella is now known as canZzZz
13:20<gbee>danielk22: it uses the base theme - it's probably broken because I moved the script location out of necessity but mythwether doesn't handle that too well at the moment
13:20<xris>truncate didn't help
13:20<xris>it may find the sources, but I can't go in to configure sources because it tells me I don't have any screens set up
13:21<gbee>xris: are the folders in scripts readable by the mythfrontend user?
13:21<xris>yeah, I tested as root
13:21<xris>and executable, and no perl compile errors when I run them
13:21<xris>I have sources listed in the db now
13:21<xris>just can't add them via the UI
13:22<gbee>ack, somethings really screwed up then, thinking we might just have to disable mythweather for 0.21 unless someone wants to give me a hand with a page long list of fixes
13:22<gbee>xris: that's weird, works fine here
13:22<xris>gbee: truncate all 3 tables, then go into frontend config and try to configure a source, and it works?
13:23<xris>if I do that, it tells me I need to configure a screen first...
13:23<xris>which I can do, but when I try to "finish" the screens config, I get an error complaining about data points, or something like that.
13:23<gbee>xris: you need to go into the "Screen Setup" page first?
13:25<gbee>xris: setup goes something like this, select a screen from the inactive list, select it in the active list and press enter - brings up a menu, select location, then finish
13:25<xris>ok, I may be doing it wrong, then.
13:25<xris>if I select it from the active list, it deactivates it.
13:29<gbee>xris: can you pastebin a dump of your weather tables?
13:29<gbee>I'll take a look after I've eaten
13:30<gbee>mythweather is rapidly becoming a thorn in my side, it needs a lot of work
13:30<xris>only weathersourcesettings has data
13:30<xris>and it's here:
13:38<Balachmar>janneg are you there?
13:44<Balachmar>Hi guys, sorry to bore your again with my md5 stuff. Janneg actually mailed something that looks ok. But I can't get that code to work
13:45<Balachmar>Also I had it working, but I was using a external library, and then read that mythtv doesn't like that very much :)
13:45<Balachmar>So I am now trying to reuse the md from avutils
13:45-!-feiner_ [] has quit [Remote closed the connection]
13:48-!-feiner [] has joined #mythtv
13:55-!-davilla [n=davilla@] has joined #mythtv
13:57<Chutt>gah, i really need to look at why we take so long to start playing these days
13:57<gbee>xris: ok, I'm lost - you should be able to add screens as described above
13:58<xris>gbee: I'll try again when I get home tonight
13:59<gbee>think I'll have to dedicate this weekend to working on mythweather
14:00<jams>xris- after the "active screen" is highlighted are you hitting the menu key?
14:00<jams>that little step is the one that eluded me for some time.
14:01<xris>jams: no, I was "
14:01<xris>"selecting" like gbee said.
14:01<xris>and that explains why it was deactivating instead of showing the menu
14:01<xris>is that info in the help text?
14:02<xris>maybe that's all that's missing -- instructions. :)
14:04<gbee>in current trunk "select" should work as well as "menu"
14:04<Balachmar>OK, I have rewritten to code to use the avutil md5. and it compiles, but now I need to specify somewhere that it should be linked, in which file I need to put that?
14:05-!-_gunni_ [] has joined #mythtv
14:06<jams>gbee- your right select works, but menu doesn't.
14:15-!-davilla [n=davilla@] has quit ["Leaving"]
14:18-!-ma9mwah|tired [] has joined #mythtv
14:31<Anduin>Balachmar: You may find better help for that sort of thing someplace like #qt, there is no shortage of examples, even inside the file you've already changed (
14:31-!-ma9mwah|c2d [] has quit [Read error: 110 (Connection timed out)]
14:32-!-catinpan is now known as catinpan_away
14:32-!-catinpan_away is now known as catinpan
14:32<Balachmar>Anduin: Well, if it was a library I would have known, but it is a header and that I don't know
14:32<janneg>Balachmar: the code I emailed was buggy, sorry. which libs are linked is in the *.pro file
14:32<Balachmar>I need to link to this: #include "../../../mythtv/libs/libavutil/md5.h"
14:33<janneg>add a "LIBS += -lmythavutil-$$LIBVERSION"
14:33<Balachmar>@janneg: the code works now, I have fixed it, I only need to get the linking right
14:33<Chutt>you don't need to add anything to libs.
14:33<Chutt>and that include path is horribly wrong.
14:34<Balachmar>@Chutt, what path does it need to be then? Cause this was proposed to me on mailinglist
14:34<Anduin>Balachmar: plugins can only compile against the installed headers, you can't make assumptions about what relative path will work.
14:34<Chutt>anything but that?
14:35<Chutt>look where the main myth install puts md5.h
14:35<Chutt>and include it from there.
14:35<janneg>Balachmar: it should be something like <mythtv/libavutil/md5.h>, but since we don't install the libavutil headers
14:35<Chutt>janneg, sure we do
14:35<janneg>it wonm't work atm
14:36<Balachmar>ok I'm confused...
14:36<Balachmar>I have this on my pc : /usr/local/lib/
14:36<Balachmar>isn't that the lib I need?
14:36<Chutt>are you confusing libraries and headers?
14:37<Chutt>don't just tell the guy what to do :p
14:37<janneg>Chutt: indeed, but proably not all since I ignored new headers in the ffmpeg sync
14:37<Chutt>he'll nevr stop asking basic questions
14:37<Balachmar>@Chutt maybe I am
14:37-!-ahbritto [] has quit [Client Quit]
14:37<Anduin>Why I suggested #qt
14:38<Balachmar>Hey, I'm just learning, I do some experience just not with bigger projects
14:38<Balachmar>@Chutt, but aren't libraries collections of headers and code?
14:39<Anduin>Balachmar: sure, but other channels will be better (and more willing) to answer basic C++/Qt development questions.
14:39<Balachmar>Anduin, I understand but I must admit stuff like plugins can only compile against the installed headers make it mythtv specific
14:39<Chutt>that's not myth specific.
14:40<Balachmar>Because I already had a working version using openssl's md5
14:40<Balachmar>I am rewriting it now, since I have read that mythtv doesn't really want many dependencies
14:41<Chutt>yes, we wouldn't add an extra dependency for something that's already in the source tree.
14:41<Balachmar>I can understand that :)
14:43-!-ma9mwah|tired [] has quit []
14:44-!-mykeul [] has joined #mythtv
14:44<Balachmar>Although probably someone from qt would tell me to use qt's md5 which is in qt4 :)
14:47<stuarta>evening all
14:52-!-briand [] has quit [Read error: 110 (Connection timed out)]
14:54-!-skamithi [] has joined #mythtv
14:55-!-jblack [] has joined #mythtv
14:55-!-briand [] has joined #mythtv
14:56-!-t0ny-p40 [n=t0ny-p40@] has quit [Remote closed the connection]
14:56<jblack>Hi. I've been getting a lot of segfaults lately, so I got myth running under gdb, and have managed to track it down to opengl rendering.
14:57<danielk22>um, opengl rendering has a known segfault, that's why it's disabled by default..
14:57*stuarta didn't know that
14:57<Chutt>danielk22, he probably means the painter
14:58<danielk22>oh, the painter -- nevermind
14:58<jblack>Ok. I'm getting a segfault in mythpainter_ogl.cpp, on line 59; realParent->makeCurrent()
14:58<jblack>That's up on the fifth frame. When I change the painter from Opengl to Qt, I still get segfaults along the opengl codepath.
14:59-!-t0ny-p40 [n=t0ny-p40@] has joined #mythtv
14:59<laga_>jblack: did you restart the frontend afterwards?
14:59<jblack>I did.
14:59<jblack>When the painter is qt, I get the segfault in ~MythMainWindow (frame 7), with glXMakeCurrentReadSGI at frame 1
15:00<stuarta>jblack: stick the entire backtrace into a pastebin
15:00<jblack>With qt as the painter?
15:01<stuarta>which ever one you are currently trying to describe
15:01<jblack>That'll be with qt as the painter
15:02<stuarta>you need to do 'thread apply all bt full'
15:02<jblack>Ok. I'll generate a new segfault
15:03<Chutt>you should probably just check to make sure your video card/opengl drivers are installed properly.
15:05<jblack>a quick check shows that gl should be fine... no broken packages listed, glchess works fine. I'll do a more careful check after I get a good backgrace
15:07<danielk22>It looks like there is a problem with the python install:
15:07<danielk22>make[2]: Entering directory `/myth/mythtv-0.21-fixes-living/mythtv/bindings/python'
15:07<danielk22>-o python -L/usr/share/qt3/lib -L/usr/X11R6/lib -lqt-mt -lXext -lX11 -lm -lpthread
15:07<danielk22>python install --root=""
15:07<danielk22>running install
15:07<danielk22>running build
15:07<danielk22>running build_py
15:07<danielk22>creating build
15:07<danielk22>error: could not create 'build': Permission denied
15:07<danielk22>it wants to write to the mythtv sources directory during install.. but if you use NFS with root squash, root can't write to the directory...
15:08<danielk22>who's in charge of the python binding code?
15:09<hads>danielk22: There's a problem with an empty INSTALL_ROOT Anduin was going to take a look at it.
15:12<Anduin>Yeah, going to check a fix in later, and move the other fixes checked in to trunk over tonight.
15:19-!-skamithi [] has quit ["WeeChat 0.2.3"]
15:24<hads>Although, that doesn't look like the same issue, it will still need to create the build dir.
15:27<jblack>I'm having a lot of trouble getting the segfault with the qt painter, so I went back to the other segfault, with the opengl painter. That gave me:
15:27-!-gerhard [] has joined #mythtv
15:27<hads>I guess we could add a target to 'python build' when running make rather than doing everything at make install time.
15:31-!-absalom [n=simon@] has joined #mythtv
15:32<absalom>Hello, after much hastle I think Ive managed to set up the db and back end, but when I run the front end it says something like "Could not connect to back-edn, are you sure its on?" The tv-show I want to watch started two minutes ago, Please help me! :(((
15:33*stuarta points helpfully at the topic
15:33<absalom>roger that =)
15:33<Balachmar>@ Anduin and Chutt: I have just learned to check libraries for the required method and I can't find that method in
15:34<Balachmar>Although that is very it should be I guess, so maybe janneg was right? And it won't work atm?
15:36<absalom>awww, no one responding there :(
15:37<Balachmar>And since there are only headers in: /usr/local/include/mythtv/ffmpeg
15:37<Balachmar>I guess there is nothing to link to...
15:42-!-kormoc [n=kormoc@unaffiliated/kormoc] has joined #mythtv
15:44-!-gerhard [] has quit [Remote closed the connection]
15:46<jarle>if I have xmltv data for a "channel", is there some way I can use the same xmltv data for "channel+1"? does tmoffset in the channel table have anything to do with this?
15:47-!-JoeBorn [] has joined #mythtv
15:47<jarle>I can not find any info about tmoffset in
15:47<stuarta>we've never got around to writing that
15:48<jarle>stuarta: was the last comment referring to my question?
15:48*stuarta nods
15:49<jarle>stuarta: so tmoffset has got nothing to do with xmltv data?
15:50-!-BathoryQuorthon [] has left #mythtv ["Great Hall Awaits a Fallen Brother"]
15:50<stuarta>sorry i meant there is no way currently of feeding data across to the +1 data
15:51-!-mick_home [n=clamwin@clamwin/admin/mickhome] has quit [Connection timed out]
15:52<jarle>stuarta: but the tmoffset is meant to be used for this functionality?
15:53<stuarta>we do have a concept of timezone offset for program data
15:54<stuarta>can't remember exactly where it's stored.
15:54<jarle>stuarta: yeah I know about that...
15:55-!-jblack [] has left #mythtv []
15:56<jarle>stuarta: what I am looking for is being able to use for two channels, having one of the channels add +1 hour to all the data...
15:56<stuarta>that's the bit we haven't written
15:56<jarle> is just an xmltv example I made up..
15:57<jarle>stuarta: and I was wondering if there has been added fields in the db for such functionality, or is both code and db-fields missing?
15:58<stuarta>it depends if you can work out if that DB field is used.
15:59<jarle>I was thinking that adding a offset field in the channel table to work out how much has to be added or subtracted on a per channel basis...
16:00<stuarta>it's part of a larger problem of feeding data across to other channels
16:01<stuarta>such as data only available via EIT that can be fed across to other datasources
16:01-!-jgarvey [] has joined #mythtv
16:03<hads>I think the uk xmltv grabber might have implemented something like that recently.
16:05<jarle>hads: how? (I'm using the uk xmltv grabber, among others)
16:05<hads>jarle: No idea, I think I saw it on the xmltv-devel list.
16:06<jarle>In my opinion it would be best if you could chosse between xmltv or EIT on a channel per channel basis...
16:09<stuarta>well you can
16:09<stuarta>it's just not obvious
16:10-!-lcase [] has quit [Read error: 113 (No route to host)]
16:10<stuarta>allow eit on the datasource, and untick use EIT on every channel you don't want EIT on
16:10-!-lcase [] has joined #mythtv
16:10-!-Balachmar [] has quit ["Ex-Chat"]
16:12<jarle>yeahm I've seen that option in the db, guess I should automate this so any channel that had an xmltvid will automatically have the use EIT unticked...
16:12-!-absalom [n=simon@] has quit [Remote closed the connection]
16:12<stuarta>that would be welcomed i'm sure
16:12-!-BathoryQuorthon [] has joined #mythtv
16:13<jarle>wouldn't this be a smart thing for myth to do internaly? in mythtv-setup for example?
16:14-!-catinpan [] has quit [Read error: 104 (Connection reset by peer)]
16:14<jarle>having both EIT and xmltv for the same channel is bound to give you scheduling problems when EIT changes your EPG...
16:15<stuarta><thinking out loud> perhaps a patch to the eit code that removes channels with an xmltvid from the list
16:16<jarle>stuarta: sounds like a good way of "attacking" the problem...
16:17<stuarta>there will be a DB select in there somewhere that pulls the chanids from a table
16:17<gbee>jarle: update to the latest xmltv 0.5.51 and the +1 channels are all handled for the uk_rt grabber
16:17<stuarta>however the mixing eit & rt data with a small patch is a separate problem
16:18<stuarta>aren't there still some channels that don't have data from the xml feeds?
16:18<stuarta>(and i suspect there always will be some)
16:19<gbee> << Seems familiar, wasn't this fixed in trunk?
16:19<jarle>stuarta: yeah I have seen that, so just altering the select to leave out channels that had a xmltvid !=0 would probably fix it..
16:19<stuarta>jarle: yeah, xmltvid != '' or xmltvid != 0
16:19*stuarta checks with some mysql-fu
16:20<gbee>stuarta: a handful, but all the Freeview channels are covered with the exception of radio/shopping etc
16:20<stuarta>the radio ones where what i was thinking about
16:20-!-catinpan [] has joined #mythtv
16:20-!-_gunni_ [] has quit ["KVIrc 3.2.4 Anomalies"]
16:21<stuarta>jarle: looks lie those 2 are equivalent
16:21<gbee>+1 and differing schedules for Freeview hours are now handled by xmltv by reusing the existing data and either timeshifting or excluding certain time frames (dunno if I can take credit for suggesting that change)
16:22<jarle>gbee: this xmltv update just gives you a separate xmltvid for the channels then?
16:22<gbee>there were rumours that radio channels would soon be available from RT, but I've not heard anything more about that
16:22<stuarta>i'm up for a putting in a bit of code that doesn't do eit updates if there is an xmltvid
16:22<stuarta>anyone see a problem with that idea?
16:23<gbee>jarle: yeah, also uses a channel_ids file which is stored on so you get updates automatically
16:23<gbee>new ids can be found here:
16:23<jarle>stuarta: would be good to be able to turn on EIT for all my sources without of fear of xmltv data being overwritten, I'd say go for it!
16:24<stuarta>was kinda hoping you'd write it ;-P
16:24<janneg>clearing/reentering a xmltvid is more work than switching an bool and it breaks my setup
16:24<stuarta>i knew someone would say that
16:24<stuarta>what's the scenario it breaks?
16:25<janneg>using eit and xmltv on a single channel
16:25<stuarta>well, yeah, that's what it's trying to avoid
16:25<stuarta>is there a good reason for doing so?
16:26<janneg>with a modified mythfilldb updating only long term data
16:26<stuarta>ah, the 0.01% case :)
16:27<janneg>eit has unfortunately data for only 3 days
16:27<stuarta>see mixing them like that here is pointless, since the details in the data vary wildly between the 2 sources
16:27<jarle>stuarta: I figured you already had your dev-environment setup, I haven't started looking at the myth code yet...
16:28<janneg>I could switch to xmltv only for that channels but I don't want to rely on xmltv
16:28<jarle>stuarta: but hey, one patch commit has to be my first :)
16:28<stuarta>so if the box in mythtv-setup on the datasource that says "use eit data" became a combo box with an added setting "use eit data when no xml data is available" would work
16:29<stuarta>jarle: we all have to start somewhere. i just picked eit :)
16:29<stuarta>mainly cause i can't be arsed with grabbers when there is reasonable EIT data
16:30<janneg>stuarta: mythtv-setup makes that distinction already. you can't enter a xmltvid if eit is enabled for that channel
16:30<jarle>I'm still thinking that buggy EIT code is the main reason for my backend crashing on a regular basis :(
16:30<stuarta>okay. wasn't aware of that.
16:31<stuarta>however wouldn't that new concept be simpler?
16:31<jarle>janneg: then you have peep like me that just enter xmltv manually into the channel table...
16:32<stuarta>heh. i've an sql script to do it for me from the last time i bothered
16:32<stuarta>and also isn't mfdb capable of setting up the xmltvid's for you?
16:32<janneg>it has be removed when we implement proper mixing of eit and xmltv
16:32<stuarta>this sounds like step 1 to me
16:32<janneg>stuarta: only for new channels
16:33*stuarta sighs at the 143 mythtv messages to go in the inbox
16:33<janneg>jarle: when you edit the channel table you can change useeit
16:35<jarle>janneg: Yeah, I know... but a more automatic approach would have been better....
16:35<stuarta>which is what we've just worked out
16:36<janneg>stuarta: it makes it more complicated. if we add this we could have used eit.mythtv as special xmltvid instead of the bool useeit
16:36*stuarta thought it was *much* simpler
16:37<janneg>jarle: it is automatic if you use mythtv-setup
16:37-!-lcase [] has quit []
16:37<jarle>janneg: oki...
16:37<janneg>how are two conditions simpler than one
16:38<stuarta>simpler from the setup point of view
16:38<stuarta>add the combo box, remove the individual useeit flags
16:38<stuarta>(from the setup, not the db)
16:39*stuarta suspects a bit more thought is required
16:40-!-MrGandalf [] has quit ["home"]
16:42-!-TelnetManta [n=benwilli@] has quit ["Ex-Chat"]
16:42<janneg>stuarta: you're speaking of the video source screen?
16:42<stuarta>i guess i'm aiming for a more usable default behaviour
16:42*stuarta fires up setup to check
16:44<stuarta>"perform EIT Scan" sounds like what i was talking about
16:44*stuarta checks channel config page as well
16:45<stuarta>the 2nd part of what i was thinking about is in the channel detail
16:46<stuarta>so i'm suggesting 2 options at the data source level when a grabber is configured
16:47<stuarta>1. what is currently called "Perform EIT scan" which does what it does now.
16:48<stuarta>2. something like "use EIT data for channels grabber doesn't provide data for"
16:48*stuarta tries to think of a nice short name for that
16:48<stuarta>for new installs the default should be 2.
16:49<stuarta>that make sense?
16:50<gbee>v4l api changed recently? Something that might explain this error: "VIDIOCGCHAN: Invalid argument"?
16:55-!-XChatMav [] has joined #mythtv
17:03-!-MavT [] has quit [Read error: 148 (No route to host)]
17:03-!-mykeul [] has left #mythtv []
17:06<janneg>stuarta: leaving the channel details as they are now?
17:06<stuarta>yeah, no need to change them
17:06<stuarta>more thinking in terms of new user to mythv
17:07<stuarta>standard setup, scan for channels use grab. expect to see decent program data
17:07<stuarta>since the useeit is initially populated by the scanner
17:09<janneg>gbee: which driver/card?
17:10<gbee>janneg: looking at this ticket,
17:11<janneg>stuarta: iirc mythfilldb/xmltv setup won't reuse the scanned channels. user has to enter xlmtvids manually and disables useeit
17:11<stuarta>do they setup channels with mfdb?
17:11-!-ahbritto [] has joined #mythtv
17:12<gbee>allowing mfdb to insert channels is pointless for everything except SchedulesDirect - no tuning information normally
17:12<stuarta>i like that answer :)
17:13<jams>gbee you wanted something
17:14<gbee>doesn't stop mfdb trying if you enter channels in the xmltv config but the xmltvid isn't already assigned to a channel in the database - chicken and egg, but something I'd like to fix for 0.22
17:14<gbee>jams: did I? hmm, forgotten
17:17<stuarta>gbee: that theory we just came up with is for those that enable EIT data and use a grabber. it should default to using EIT only when there isn't an xmltvid for the channel
17:18<stuarta>to stop the obliteration of xml data via eit
17:18<gbee>stuarta: sounds good
17:18<jams>gbee- ok..guess my nick highlight went off for no reason
17:19<danielk22>janne, when you added the pref_max_cpus setting did you do any DB update for existing profiles somewhere? VideoDisplayProfile::SaveDB() uses "UPDATE .." statements for existing entries, so if there is no pref_max_cpus setting, it gets ignored...
17:19<stuarta>now have i got any time to whip up a patch
17:19-!-reynaldo [] has joined #mythtv
17:19*stuarta ponders
17:19<gbee>not feeling motivated to do any work atm, just trawling through open tickets with the tv on in the background
17:20<stuarta>i got up at 5am. dunno why i'm not in bed
17:20<janneg>danielk22: I know. I didn't add since I decided it was easier to handle the missing value in the code
17:21<danielk22>? but you can never save the any pref_max_cpus value in an existing profile..
17:21<danielk22>the any -> any
17:22<danielk22>so it will always be "1"
17:22-!-gerhard_ [] has joined #mythtv
17:23<janneg>yeah, I know it's buggy since I tested unfortunately only a new profile. I'm fixing it
17:24<danielk22>thx.. are you adding smarts to SaveDB? or a dbcheck.cpp db update? something else?
17:25<janneg>I started to write the db update and will remove the code handling the missing value
17:27<jarle>hmm.. how do I return only records where xmltvid is set? "SELECT * FROM `channel` WHERE `xmltvid` !=0" does not seem to do the trick?
17:27<janneg>!= ""
17:27<stuarta>!= ''
17:30<jarle>using phpMyadmin for simplicity, this does give me also records without xmltvid set:
17:30<jarle>SELECT *
17:30<jarle>FROM `channel`
17:30<jarle>WHERE `xmltvid` != CONVERT( _utf8 '""'
17:30<jarle>USING latin1 )
17:30<jarle>COLLATE latin1_swedish_ci
17:30<jarle>AND `visible` =1
17:30<jarle>LIMIT 0 , 30
17:37<jarle>would this be correct mysql syntax for turning off EIT scanning for the channels where I have xmltv: "SELECT * FROM `channel` WHERE `xmltvid` !=" " SET 'useonairguide'=0;" ?
17:38<stuarta>update channel set useonairguide=0 where xmltvid != '';
17:38<stuarta>that's pure mysql
17:40<jarle>stuarta: thnx.
17:42*stuarta goes to bed
17:43-!-carvajal [n=carvajal@] has quit [Connection timed out]
17:44-!-gerhard_ [] has quit ["ChatZilla 0.9.80 [Firefox]"]
17:45-!-skamithi [] has joined #mythtv
17:50-!-buk___ [] has joined #mythtv
17:55-!-buk___ [] has quit ["DSOrganize IRC"]
18:00-!-Cardoe [n=Cardoe@gentoo/developer/Cardoe] has quit ["Leaving"]
18:00-!-mattwire [] has joined #mythtv
18:29-!-BathoryQuorthon [] has left #mythtv ["Great Hall Awaits a Fallen Brother"]
18:30-!-rinaldi1 [] has joined #mythtv
18:35<janneg>danielk22: fixed
18:36-!-sc00p [] has quit [Connection reset by peer]
18:39<rinaldi1>would this be the place to report any bugs or unsolveds?
18:40<janneg>no, use
18:40<rinaldi1>k thanks
18:41<gbee>rinaldi1: just be sure that it is an unfixed bug (run trunk) and follow the guide here:
18:42<danielk22>thx, janne
18:42<gbee>triaging tickets is an endless job
18:42<danielk22>i think my fix for #4679 may actually fix the cpu issue too..
18:44<danielk22>hmm, that new db backup functionality doesn't work for me..
18:45<danielk22>mysqldump --defaults-extra-file='/tmp/mythtv_db_backup_conf_xxhc02' --host='localhost' --user='mythtv' --add-drop-table --add-locks --allow-keywords --complete-insert --extended-insert --lock-tables --no-create-db --quick 'mythconverg' > '/video/testing/mythconverg-1211-20080219184335.sql'
18:45<danielk22>Could not open required defaults file: /tmp/mythtv_db_backup_conf_xxhc02
18:45<rinaldi1>gbee: thanks already been there :)
18:46<gbee>sphery: ^^
18:46<danielk22>the mythtv user on the dev box can of course access /tmp/ ...
18:48<danielk22>should we maybe disable this for 0.21 ? mysqldump might not even be a valid executable, or might refer to the wrong mysql server instance, etc. etc.
18:50<Chutt>or could be any random program.
18:50<danielk22>esp, if you refer to the --host='localhost' even if we added a --port='blah', it gets ignored when the host is 'localhost' in preference to using a unix socket, so if you have multiple mysql servers running this can refer to the wrong server (which is why you normally replace localhost with, but this may not work on some distro's..)
18:51<janneg>danielk22: no error before the failing mysqldump?
18:52<danielk22>janne, no error before failing mysqldump.. but it didn't report the error, I reran that on the command line without the stderr redirect to get an error message, it may be the wrong message..
18:52<janneg>Chutt: we aren't checking if wget is really wget in the datadirect client
18:52-!-rinaldi1 [] has left #mythtv []
18:52<Chutt>well, yeah :p
18:52-!-MrGandalf [] has joined #mythtv
18:52<danielk22>we should probably run some distro defined script for the backup.. and put a sample in contrib...
18:53<danielk22>and we shouldn't be using wget at all :)
18:53<janneg>danielk22: the temprary file was deleted in dbutil.ccp
18:54<danielk22>ic, so I should patch to remove the stderr redirect to get the real error...
18:56<Chutt>helps to svn up the /var/www dir
18:56<janneg>for debugging you could rerun the command line without --defaults-extra-file='/tmp/mythtv_db_backup_conf_xxhc02' and add -p
18:58<danielk22>mysqldump: Got error: 1045: Access denied for user 'mythtv'@'localhost' (using password: NO) when trying to connect
18:58<janneg>Chutt: how is the documentation on created? we should switch it to create it from the release branch
18:58<danielk22>did I mention 'localhost' was a bad idea for a hostname?
18:59<Chutt>it's magic
18:59<Chutt>i can switch it, though
19:00-!-skamithi [] has quit ["WeeChat 0.2.3"]
19:01<janneg>danielk22: edit ~/.mythtv/mysql.txt
19:01-!-gbee [] has quit [Remote closed the connection]
19:01<danielk22>when I add --password='secret extracted from mysql.txt' I get a little further.. but mythweather tables are choking things.
19:03<janneg>Chutt: please do. having the trunk documentation there is confusing
19:04<janneg>danielk22: the secret is in the file specified by --defaults-extra-file
19:10-!-rooaus [] has quit [Read error: 113 (No route to host)]
19:11<danielk22>k, dropped all the weather tables and it ran ok.. the weather tables were all messed up anyway.
19:14<danielk22>chutt: it makes more sense to have the magic use the release tree for docs. then we can update trunk with docs for the next release as we go, and still update the release docs with additional info
19:15<Anduin>danielk22: 16170 should fix your python install problem (works here with a ro build dir)
19:16-!-jthomas_work [] has joined #mythtv
19:16<jthomas_work>hi, do you use any radio module for mythtv ?
19:16<jthomas_work>i mean internet radio
19:17<jthomas_work>ah maybe going to mythtv-users :D sorry
19:17-!-jthomas_work [] has left #mythtv []
19:18-!-siXy [i=siXy@] has quit ["bye!"]
19:20-!-davilla [] has joined #mythtv
19:23-!-jhulst [n=jhulst@unaffiliated/jhulst] has joined #mythtv
19:24-!-gbee [] has joined #mythtv
19:24<danielk22>thx anduin, i'll give it a try
19:24-!-czth_ [n=dbrobins@nat/microsoft/x-097ddcff262e0bd3] has joined #mythtv
19:28<Captain_Murdoch>danielk22: you might be the best person to answer this. DiSEqCDevSwitch::SwitchTypeTable is declared as [9] in diseqc.h and [7] in diseqc.cpp. my gcc 4.1.0 box is failing to compile because of the difference. I don't know anything about that code, do you know which one should change?
19:30<Captain_Murdoch>I assume the [9] in the .h file should be [7].
19:30-!-gbee [] has quit [Read error: 113 (No route to host)]
19:32<Chutt>next time the docs update script runs, it should pull from release-0-21-fixes
19:32<Chutt>i believe that's once an hour
19:33<Chutt>there, just kicked it manually
19:37-!-sc00p [] has joined #mythtv
19:46<Captain_Murdoch>hmm, must have been a stuck file from SVN on this one machine, my other boxes have [7] in their .h files.
19:47-!-czth__ [n=dbrobins@nat/microsoft/x-6540f79868dc39fe] has quit [Read error: 110 (Connection timed out)]
19:48<danielk22>anduin: it seems to have installed without incident. :)
19:48<Anduin>danielk22: Good, thanks for the confirmation.
20:00-!-reynaldo [] has quit [Read error: 113 (No route to host)]
20:01<Captain_Murdoch>I think my "svn switch" failed somewhere along the line on that box.
20:02<Captain_Murdoch>building -fixes on the via 733.
20:06<danielk22>anduin: I just got the "error: could not create 'build': Permission denied" on a different machine. I think maybe I was wrong, I'll try a distclean and try again.
20:07<Anduin>danielk22: Ok, it does need to create the build directory during a make, but make install shouldn't need to.
20:08<Anduin>and yeah, distclean, the make file generation changed (old install depended on all so even the new fix would have failed)
20:11-!-jgarvey [] has quit ["Leaving"]
20:12-!-mick_laptop [n=mweiss@clamwin/admin/mickhome] has joined #mythtv
20:14-!-Idle [] has left #mythtv []
20:25-!-carvajal [n=carvajal@] has joined #mythtv
20:30-!-xris [] has quit []
20:35-!-mattwire [] has quit ["Leaving"]
20:38-!-onyxsoft_ [] has quit ["Ex-Chat"]
20:42<Chutt>is there an upstream for libdvdnav?
20:43<Anduin>didn't we borrow the one from xine?
20:43-!-beoba` [n=fsoh@unaffiliated/beoba] has joined #mythtv
20:44<Chutt>been lots of little fixes to it, that's all
20:44<Chutt>we should resync
20:44<Anduin>The news at is from 2004
20:44<superm1>if at all possible wouldn't it make more sense to link to an external library?
20:44<Chutt>not when you have local modifications
20:45<Chutt>and upstream is dead
20:45<beoba`>hi, ive noticed that remotely controlling mythtv over the network doesnt work in cases when mplayer is used to play a video file. is there a workaround for this?
20:45<GreyFoxx>be : wrong channel
20:45<superm1>well the last upstream release was in 06
20:45<superm1>so yeah i suppose it might be dead
20:45<beoba`>oh woops right
20:45-!-beoba` [n=fsoh@unaffiliated/beoba] has left #mythtv ["Buy a better life from the comfort of your sofa."]
20:46<janneg>Chutt: libdvdnav is afaik dead
20:46<Chutt>superm1, also, it's sometimes easier to just pull in small libraries like that, than deal with different versions installed on different users' systems
20:46<superm1>yeah i can see that
20:57-!-xris [] has joined #mythtv
21:05-!-onyxsoft [] has joined #mythtv
21:08-!-TelnetManta [] has joined #mythtv
21:17-!-mick_laptop [n=mweiss@clamwin/admin/mickhome] has quit [Read error: 104 (Connection reset by peer)]
21:46-!-Tokayla [] has joined #mythtv
21:46-!-Tokayla [] has left #mythtv []
22:05-!-mick_laptop [n=mweiss@clamwin/admin/mickhome] has joined #mythtv
22:17-!-JoeBorn [] has quit [Read error: 110 (Connection timed out)]
22:27-!-ahbritto [] has quit [Client Quit]
22:44-!-onyxsoft [] has quit ["Ex-Chat"]
22:44-!-reynaldo [] has joined #mythtv
22:46-!-mick_laptop [n=mweiss@clamwin/admin/mickhome] has quit [Read error: 104 (Connection reset by peer)]
22:55-!-JoeBorn [] has joined #mythtv
23:06-!-feiner_ [] has joined #mythtv
23:06-!-feiner_ [] has quit [Read error: 104 (Connection reset by peer)]
23:06-!-jhulst [n=jhulst@unaffiliated/jhulst] has quit [Read error: 110 (Connection timed out)]
23:10*xris finally figured out mythweather
23:10<xris>what a mess
23:11<Anduin>xris: Do the inputs have a priority?
23:11<xris>Anduin: I think one of them is higher than the other
23:11<xris>should that matter if both are available, though?
23:12<xris>er, or something like that
23:12<Chutt>i thought the 'deinterlace on pause' bug was fixed?
23:14-!-mick_laptop [n=mweiss@clamwin/admin/mickhome] has joined #mythtv
23:14<mick_laptop>hi everyone, does anyone have a vector version of the logo?
23:14<mick_laptop>i only have a vector of the text
23:14<Anduin>xris: I don't remember exactly what weight it is given, but yes it can make a difference.
23:15-!-onyxsoft [] has joined #mythtv
23:15<xris>mick_laptop: as far as I know, just some high res png files in the source
23:16<mick_laptop>ah thanks
23:16<xris>I thought that one of those was made from an svg file, but if it was, the source has been lost.
23:46-!-kormoc [n=kormoc@unaffiliated/kormoc] has quit []
23:49-!-jamesd [] has joined #mythtv
23:53-!-jamesd [] has quit [Remote closed the connection]
23:54-!-jhulst [n=jhulst@unaffiliated/jhulst] has joined #mythtv
---Logclosed Wed Feb 20 00:00:49 2008