Back to Home / #mythtv / 2003 / 03 / Prev Day | Next Day
#mythtv IRC Logs for 2003-03-28

00:04<moegreen>PeteCool: I probably will, but i'm sure someone will find a reason to complain - because it does look pretty bad on a monitor
00:04<Chutt>moegreen, qt's already got fairly generalized xml reading classes
00:04<Chutt>i'm just envisioning a lot of code shared between the different screens to read in layout files
00:06<moegreen>Chutt: right ... i haven't coded in the layout data yet, just the colors. And it check to see if there is a file in the theme's directory to replace the default png files
00:06<Chutt>been playin zelda
00:06<Chutt>no time for code =)
00:27<-- DShadow( has left #mythtv
00:34<moegreen>(there will be a quiz in the morning)
00:35<nyquiljer>moegreen: that is really cool
00:39<moegreen>and it's *Official* :)
00:43<Chutt>why do people submit patches with lots of useless indenting/whitespace changes?
00:43<Chutt>like, totally unrelated sections to what they've modified, aside from adding a space or two
00:47<inman>maybe they are trying to clean up the code?
00:48<Chutt>maybe you don't know what you're talking about?
00:48<inman>i know that the code doesn't use consistent ws.
00:49<Chutt>then cleaning it up would be making it consistent
00:49<Chutt>not introducing yet another style
00:49<inman>your comment didn't clearly indicate that they were introducing new style.
00:52<foom>well in any case cleaning-up-spacing patches shouldn't be submitted with any changes embedded
00:52<inman>i agree with that. :-)
00:54<Ndiin>Perhaps simply ask people to use diff -uBb, that would eliminate such problems in the future.
00:55<Chutt>that makes things even harder to read
00:55<Chutt>i don't apply patches without reading them over, first
00:55-!-Soopaman [] has joined #mythtv
00:55<Chutt>and i usually apply them by hand
00:55<Ndiin>Bb is just newline and whitespace changes ignored
00:56<Chutt>ndiin, and it makes very messy diffs
00:56<Chutt>if, for instance, you need to change indenting because of a new control clause
00:56<inman>Ndiin: sometimes indentation changes
00:57-!-nziarek [] has joined #MythTV
00:57-!-hfb [] has joined #mythtv
01:00-!-NonToxic is now known as zZzToxic
01:01<Chutt>ndiin, i was going to apply your mythmusic changes, but now it's going to wait until i have more time :p
01:03<Ndiin>Thats fine with me.. Most of the extra changes in it were actually because of changes in thor's playlists, I basically just adapted Joe's patch to work with the changes
01:04<Chutt>figured it'd be easier to use your updated patch, rather than redo what you did in updating it already
01:05<Ndiin>I actually took out quite a few silly changes (moving open brackets to new lines, etc) that were in the straight diff
01:05<Ndiin>but that became irksome, to say the least
01:12<inman>anyone notice problems with pgup/pgdn jumping?
01:12<Chutt>which screen?
01:12<inman>i'm seeing really weird jumping if i just jump all over the place.
01:13<inman>i'm waitching a 3hr recording.
01:13<inman>initially, it works fine. jumps in 5min increments.
01:13<inman>but if i pound on the keys and really get crazy, it eventually gets really confused.
01:14<inman>starts jumping in increments of just a couple seconds, gets confused as to where in the stream it is.
01:14<Ndiin>So not letting it seek fully first?
01:14<inman>at one point, jumping backwards was actually incrementing my position in the stream.
01:14<inman>Ndiin: i guess.
01:15<inman>i'm just going by what the osd says.
01:16<inman>it does seem to work for stuff earlier in the stream
01:16<inman>eg. i'm 30mins into the stream and it works fine in those first 30 minutes.
01:17<Chutt>should be fairly easy to debug
01:18<Chutt>but i need to sleep, so =)
01:18<inman>:-) g'nite. i'll take a look.
01:20<inman>it's as though the second keypress which occurs while myth hasn't yet reached 5mins in the future,
01:20<inman>simply seeks to the "end" of the stream that we've buffered.
01:21<Chutt>it shouldn't actually be seeking again until its done seeking from the first keypress
01:21<inman>if i hold down pgdn, the osd updates at 1sec intervals.
01:21<Chutt>but the UI might still be updating
01:21<inman>Chutt: right.
01:21<Chutt>i think the OSD stuff gets updated regardless of if that keypress is going to actually generate a seek or not
01:21<foom>oh yeah, i noticed the editor doesn't work that way
01:21<foom>if you hold down a key it keeps going long after you released it
01:22<Chutt>yeah, it doesn't discard keypresses like it should
01:22<Chutt>ah well, g'nite
01:27-!-Zytheus [] has joined #mythtv
01:27<Zytheus>anyone alive?
01:28<Zytheus>just did a debian install (pretty sweet, btw)...trying to run for first time and i'm seeing it didn't install any themes
01:28<inman>are you using CVS?
01:28<Zytheus>no osd theme...wondering where i could get it
01:28<Zytheus>deb sources
01:28<inman>did you see any errors during `make install`?
01:28<inman>eg. from the themes directory?
01:28<Zytheus>didn't do make install
01:28<Zytheus>used apt-get
01:29<Zytheus>maybe i should email the debian guy...
01:29<inman>huh. well, i've never used binaries and i'm not running debian.
01:29<inman>what's in /usr/local/share/mythtv?
01:29<Zytheus>matt zimmemarman
01:29<inman>did you do setup?
01:30<Zytheus>dir does not exist
01:30<Zytheus>"apt-get install mythtv" ...debian's standard package installer
01:30<inman>yeah, i get it.
01:30<inman>can you search your filesystem?
01:30<Zytheus>search for what?
01:31<inman>eg. search for mainmenu.xml
01:31<foom>dpkg -L <packagename> shows the files it installed
01:31<inman>the themes directory is a peer of that file.
01:31<Zytheus>if i do "dpkg -l \*myth\*" i can see they're all installed
01:31<foom>i installed the unstable debs and they worked
01:31<foom>dunno about woody ones
01:32<Zytheus>i'm on woody
01:32<Zytheus>hold on...i found it
01:32<Zytheus>... /usr/share/mythtv/mainmenu.xml
01:32<Zytheus>needed to run updatedb :)
01:32<inman>is there a themes directory alongside?
01:33<inman>in /usr/share/mythtv
01:33<Zytheus>cool..found my themes
01:33<Zytheus>...but myth is looking in the wrong place for them
01:33<Zytheus>i'll just make a symlink
01:33<inman>update the settings.
01:34<foom>btw it's dpkg -L (capital L) that shows the files the package installed
01:35<foom>anyways, good luck, i'm going to sleep.
01:35-!-foom [] has quit ["sleep"]
01:36-!-Soopaman [] has quit [Read error: 104 (Connection reset by peer)]
01:37<inman>you all set, Zytheus?
01:38<Zytheus>i'm rerunning setup
01:40<Zytheus>running /etc/cron.daily/mythtv-backend
01:40<Zytheus>....because it said to after mythtv-setup
01:40<Zytheus>doh--database error
01:42<Zytheus>is there a forum somewhere? ...where i can search posts/questions/answers before bugging you guys?
01:42<Zytheus>obviously, i'll read docs first
01:42<inman>bugging people here is faster.
01:42<Zytheus>heh...okay if you don't mind
01:43<inman>well, /i/ don't, no.
01:43<Zytheus>but i try to rtfm/stfw my share
01:43* inmannods.
01:43<Zytheus>is the update supposed to have occasional db errors?
01:44<Zytheus>getting duplicate entry errors
01:44<inman>check your *.xmltv file in ~mythuser/.mythtv
01:44<inman>make sure there aren't duplicate channels defined.
01:44<inman>comment them out if so.
01:44<inman>else post the error message here.
01:45<Zytheus>heh..i don't have that file
01:45<inman>you must have an xmltv file somewhere.
01:46<inman>the name matches that of your channel-lineup
01:46<Zytheus>.mythtv is dir not file...i tried vi on it
01:47<Zytheus><-- moron
01:47* inmanwinks.
01:47<Zytheus>looking for dups
01:47<Zytheus>no dups
01:47<Zytheus>okay, update done
01:50<Zytheus>hmmm...getting a host of errors now when running "mythtv"
01:51<inman>run mythbackend, then run mythfrontend, and RTFM.
01:51<Zytheus>first, says "could not open network socket"
01:51<Zytheus>it says that before it says "connecting to backend server"
01:51<Zytheus>backend is running
01:51-!-Timon [] has quit ["My damn controlling terminal disappeared!"]
01:53<Zytheus>you're right...i was using the wrong exe
01:53<Zytheus>all appears to be well
01:54<Zytheus>thanks for fielding my newb crap
01:54<inman>np ;-)
02:05<-- nyquiljer(coyote@ has left #mythtv
02:23<inman>bug? after editing a recording, i can't fast-forward
02:23<inman>jumping ahead increments time in the osd, but not in the stream.
02:23-!-nziarek [] has quit [Read error: 110 (Connection timed out)]
02:36<Zytheus>sorry, i'm not quite that advanced yet...having installed half an hour ago and all
02:39<inman>just logging it here in case someone reads it later. don't have time to check it out right now.
02:54-!-RagnaRoC [] has joined #MythTv
02:58-!-inman [] has quit ["bbiab"]
03:18-!-Received a CTCP TIME from RagnaRoC (to #MythTv)
03:18<-- RagnaRoC( has left #MythTv (""You run, never stop, got to win, got to run till you drop"")
04:22-!-choenig [] has joined #mythtv
07:26-!-Ndiin [] has quit [Read error: 110 (Connection timed out)]
07:28-!-Zytheus [] has quit ["My damn controlling terminal disappeared!"]
08:13-!-Ndiin [] has joined #mythtv
08:14-!-nziarek [] has joined #MythTV
08:44-!-RagnaRoC [] has joined #MythTv
08:45<RagnaRoC>Vektor: You awake/alive?
08:46<RagnaRoC>join #livid
08:46<RagnaRoC>Morning to you
08:47<RagnaRoC>Wanted to tell you stereo is working MTV on with my speakers in Prologic mode and the ground outside shakes :)
08:47<vektor>Ok, I've been working on improving that patch, just so you know.
08:47<vektor>There are some 'issues'.
08:47<RagnaRoC>I have been noticing what appears to be frame rate problems that seem to have started up on occasion
08:48<vektor>In tvtime? We should discuss this on #livid.
08:48<RagnaRoC>define issues...I have thought I saw/heard some occasinal problems but not 100 %
08:48<RagnaRoC>okI am in there allready
09:03-!-nziarek [] has quit [Read error: 110 (Connection timed out)]
09:09-!-hfb [] has quit [Read error: 110 (Connection timed out)]
10:10-!-inman [] has joined #mythtv
10:14-!-paperclip [] has joined #mythtv
10:26-!-hfb [] has joined #mythtv
11:01-!-RagnaRoC [] has quit [Read error: 113 (No route to host)]
11:04--> |Uli|( has joined #mythtv
11:42-!-zZzToxic [] has quit [Read error: 110 (Connection timed out)]
11:44-!-chinande [] has joined #mythtv
11:58<inman>i'm getting segfaults in the frontend just from viewing recordings briefly and deleting them. no time to debug atm.
12:00<Chutt>honestly, saying it segfaults without providing a backtrace is useless
12:13-!-nziarek [] has joined #MythTV
12:13<inman>i know, i will debug it in a bit.
12:14<inman>i was looking more for a "me too" than a "i think i know why".
12:21-!-Chang [] has joined #mythtv
12:26<Chang>I'm seeing a segfault using the debian binaries
12:26<Chang>When I'm watching an in progress recording and the recording stops frontend segfaults
12:26<Chang>Anybody know if this is present in the 0.8 compiled from source
12:27<rkulagow>chutt: i just got a segfault when i tried to "watch recordings". here's a bt: (this is current cvs; make clean distcleaned)
12:27<rkulagow>urpmi libcdda0 libcdda0-devel libvorbis0 libvorbis0-devel
12:27<rkulagow>duh, wrong paste. sorry.
12:28<|Uli|>i have problems running mythtv with suse 8.1
12:28<rkulagow>(gdb) run
12:28<rkulagow>Starting program: /usr/local/bin/mythfrontend
12:28<rkulagow>[New Thread 16384 (LWP 29357)]
12:28<rkulagow>connecting to backend server:
12:28<rkulagow>Program received signal SIGSEGV, Segmentation fault.
12:28<rkulagow>[Switching to Thread 16384 (LWP 29357)]
12:28<rkulagow>0x4067691c in QString::QString(QString const&) () from /usr/lib/qt3/lib/
12:28<rkulagow>Current language: auto; currently c
12:28<rkulagow>(gdb) bt
12:28<rkulagow>#0 0x4067691c in QString::QString(QString const&) () from /usr/lib/qt3/lib/
12:28<rkulagow>#1 0x0808279e in PlaybackBox::getPixmap(ProgramInfo*) (this=0xbfffe3d0, pginfo=0x0) at playbackbox.cpp:1973
12:28<rkulagow>#2 0x080798d8 in PlaybackBox::updateVideo(QPainter*) (this=0xbfffe940, p=0xbfffe3d0) at playbackbox.cpp:640
12:28<rkulagow>#3 0x08077e97 in PlaybackBox::paintEvent(QPaintEvent*) (this=0xbfffe940, e=0xbfffe6e0)
12:28<rkulagow> at playbackbox.cpp:489
12:28<rkulagow>#4 0x40425a67 in QWidget::event(QEvent*) () from /usr/lib/qt3/lib/
12:28<|Uli|>when i run setup, it says: QSqlDatabase warning: QMYSQL3 driver not loaded
12:29<Chutt>rkulagow, i'll fix that in a sec
12:29<rkulagow>chutt: thanks.
12:29<rkulagow>Uli: did you read the HOWTO?
12:29<Chutt>uli, no, you didn't read the howto
12:30<Chutt>the howto says you need qt's mysql driver installed
12:30<Chutt>you do not have qt's mysql driver installed
12:30<|Uli|>i have installed qt3-mysql
12:31<Chutt>do you have QTDIR set?
12:31<moegreen>Chutt: I'm working on the playbackbox.cpp right now, i can do the fix for rkulagow
12:31<rkulagow>uli: what sitro?
12:31<Chutt>moegreen, sure
12:31<|Uli|>and there is: /usr/lib/qt-3.1.1/plugins/sqldrivers/
12:32<|Uli|>and /usr/lib/
12:33<|Uli|>i found out that i need QTDIR, libmysqlclient and libqsqlmysql
12:33<|Uli|>and now i don't know what else to test and check
12:34<Chang>Uli: What version of qt-devel did you compile against?
12:34<Chutt>ah, people have reported this before on the list
12:34<Chutt>the rpms are broken, apparently =)
12:35<Chang>Chutt: I can vouch for that. I wasted 4 days on those RPMS
12:36<Chutt>whatever packaging system suse uses nowadays
12:36<Chutt>their qt stuff is broken
12:36<Chutt>use a better distro
12:37<|Uli|>oh, i thought you meant mythtv-rpms
12:55-!-sc00p [~oldendic@] has joined #mythtv
13:03-!-nziarek [] has quit [Read error: 60 (Operation timed out)]
13:34<|Uli|>i have solved my problem:
13:34<|Uli|>i had to do: export LD_LIBRARY_PATH=$QTDIR/lib:$LD_LIBRARY_PATH
13:35<|Uli|>and in contrast to QTDIR=/usr/lib/qt3 as in the instructions, i had to do export QTDIR=/usr/lib/qt-3.1.1
13:42<inman>add /usr/lib/qt-3.1.1/lib to /etc/ and run ldconfig (as root); then you can omit the LD_LIBRARY_PATH step
13:57<-- Chang( has left #mythtv
13:57-!-aleitner [] has joined #mythtv
13:57<-- aleitner( has left #mythtv
14:02-!-aleitner [] has joined #mythtv
14:02<aleitner>hi anybody using mythtv on debian unstable?
14:03<aleitner>seems like the woody and sarge debs are not instablable on unstable
14:04<Chutt>there are debs for unstable.
14:13<mdz_>aleitner: at the top of the page: These packages are intended for Debian 3.0 (woody). Packages for Debian unstable are now separate.
14:13<mdz_>(with a hyperlink)
14:15<aleitner>mdz_: thanks alot. i got it now. sorry for my stupid question
14:32<aleitner>hm, sorry if this is again a stupid question, but when installing mythtv-database it fails because it cannot connect to the database
14:32<aleitner>looks like it doesnt know the password (where from should it know)
14:33<aleitner>(it didnt ask me for one)
14:33<aleitner>is there something i am doing wrong? (pretty sure i am :)
14:51<aleitner>oh, got it, needed to do a -force reconfigure
14:57<aleitner>What does the following mean:
14:57<aleitner>Session management error: Authentication Rejected, reason : None of the authentication protocols specified are supported and
14:57<aleitner>host-based authentication failed
14:57<mdz_>it means that you are not authorized to connect to the X server
14:57<aleitner>oh i see
14:58<aleitner>its not mythtv related (;
14:58<aleitner>is this something i should be worried about?
14:59<aleitner>xclock starts fine (with the same warning) and i did a xhost +
15:00<aleitner>'/etc/init.d/mythtv-backend' start complains about:
15:00<aleitner>Starting MythTV server: mythbackendQSettings: error creating /root/.qt
15:00<aleitner>anybody knows how i can correct this or what it means?
15:03<gimpy>main.o(.text+0x7a): In function `main':
15:03<gimpy>: undefined reference to `QApplication::QApplication[in-charge](int&, char**)'
15:03<gimpy>main.o(.text+0xa3): In function `main':
15:03<gimpy>i compiled qt with -qt-sql-mysql..
15:03<gimpy>and went to compile mythtv and i get that
15:05<Chutt>you're compiling mythtv with a version of gcc that's different than what qt was compiled with.
15:05<gimpy>i compiled qt yesterday..
15:05<Chutt>you've got an older version lying around, then
15:05<aleitner>anybody knows what /root/.qt is for?
15:06<Chutt>aleitner, qt settings files?
15:06<gimpy>ya i have another gcc .. i have two of them installed..
15:06<Chutt>of qt
15:06<Chutt>not gcc.
15:06<gimpy>and two qt's.. but its using /usr/lib/qt which is a symlink of my current qt
15:06<aleitner>Chutt: do you have any idea, why mythtv might fail to create them?
15:07<Chutt>it doesn't have write permission?
15:07<Chutt>gimpy, it's still trying to link against a version of qt compiled with an older gcc
15:07<aleitner>user mythtv in dir /root/ i hope not!
15:08<gimpy>[gimpy:/usr/lib]$ echo $QTDIR
15:08<gimpy>you are right
15:08<aleitner>Chutt: do you know if /etc/init.d/myth-backend is supposed to create this file?
15:08<gimpy>i.. put it in my /etc/profile to.. hmm
15:08<Chutt>aleitner, no, it's not
15:08<gimpy>oh well ill manually export it again and try, thanks
15:08<Chutt>why would it?
15:09<aleitner>Chutt: i dont know. i am totally new to mythtv
15:09<aleitner>Chutt: all i do is start the backend server via the init.d script
15:09<aleitner>but it fails
15:09<gimpy>/usr/bin/ld: cannot find -lqt-mt thats the error im getting now =/
15:09<moegreen>gimpy: you don't have the qt threads compiled in
15:10<gimpy>so i should recompile qt?
15:10<Chutt>you should use a real distro.
15:10<Chutt>aleitner, i wasn't talking to you
15:11<aleitner>oh, sorry (;
15:12<moegreen>gimpy: if you're going to recompile, be sure to use the -thread when doing you ./configure, also be sure to use the -qt-sql-<driver> flag
15:12<gimpy>ya im doing that now
15:13<moegreen>Well get it right this time :)
15:13<gimpy>hehe.. ya this took all night to compile oh well imma go get a haircut anyway
15:14<gimpy>and there was something in /etc/profile.d/ that kept setting qt i suppose
15:14<inman>please don't let myth come between you and your personal grooming.
15:37<mdz_>aleitner: that message is completely harmless
15:37<mdz_>aleitner: you'll notice that the backend starts up fine in spite of it
15:38<mdz_>assuming everything else is in order
15:39<mdz_>if Zytheus was using the debs and had no themes, he was probably running the wrong binary
15:41<Chutt>all the coords for text and things are pixmap relative
15:41<Chutt>in the osd
15:41<Chutt>i should make em all absolute
15:53-!-kain [~kain@] has joined #mythtv
16:00<inman>mdz: it wouldn't be the first time, for Zytheus.
16:04-!-sc00p [] has quit ["Client exiting"]
16:42<moegreen>Chutt: are you around?
16:44<moegreen>ok, i'm getting an X Error (X Error: BadWindow (invalid Window parameter) 3). I can reproduce it, to do so, I go into Watch Recordings, pick a recording, watch it for any given amount of time, then hit escape a couple of times (once to exit the recording, once to exit the watch rec screen, and maybe one or two more times.
16:44<Chutt>trying to draw on a window that's no longer there, perhaps?
16:44<moegreen>I thought it might be the playbackbox trying to start/display the small playback preview again so I added a Key_Escape handler to ensure that the timer and player is stopped
16:44<moegreen>but that didn't do it
16:45<moegreen>it shows my 'exitWin()...killing app.' statement i added just before the emit call...then a 'Changing from None to None', then the X Error
16:48<moegreen>I have to hit escape pretty quickly, but it's happened to me once or twice on accident
17:08<poptix>i need to find out what i did to my setup
17:08<poptix>i suddenly can't do any capture on my bt878 card
17:08<poptix>but overlay works fine
17:10<inman>kernel changes?
17:10<poptix>just rebooted
17:11<inman>are your modules loaded correctly?
17:11<poptix>i think it's because it's sharing the IRQ with 2 network cards and my emu10k1
17:11<inman>does xawtv work?
17:11<poptix>xawtv works fine in overlay mode, but not in grabdisplay mode
17:11<poptix>tvtime doesn't work either, because it uses the capture mode
17:11<inman>btw, who are you getting to port the ivtv driver to myth?
17:11* inmanhas a pvr-350.
17:11<poptix>inman: trying to get a buddy on efnet to do it
17:12<inman>there are several products that could benefit from the same interface to v4l2.
17:12<poptix>dvb stuff
17:12<inman>what's the difference between overlay mode and grabdisplay mode?
17:12<inman>technically, i mean.
17:13<poptix>inman: in overlay mode it just makes a box on the screen that's a certain color
17:13<Chutt>overlay mode sends the data to the video card for display
17:13<poptix>then the tv card sends the data to the video card, or the video card driver, and it gets put on top of that special colored box
17:13<Chutt>grabdisplay hits the system ram, first
17:13<inman>overlay is pretty easy to understand just by watching it
17:13<Chutt>poptix, not everything does it with colorkey like that
17:14<poptix>that's why when you try to take a screenshot of your DVD playing, you usually get a pink box
17:14<poptix>Chutt: hmm, i was fairly sure everything did
17:14<inman>i'm probing at what could be different to cause one to work while the other does not.
17:14<poptix>Chutt: does it just give position info?
17:14<Chutt>think so, yeah
17:14<inman>i hate overlay
17:15<poptix>i dropped in a drive with win98 on it, and dscaler is working
17:15<poptix>so i know the card isn't broken
17:15<inman>if overlay works, i'd say the card ain't broken, but what do i know.
17:15<poptix>but i also made windows disable the use of IRQ 3 =p
17:16<inman>does it work okay if you don't load your emu10k drivers?
17:16<poptix>inman: haven't tried yet
17:16<mdz_>surely php must have a strftime equivalent
17:16<poptix>i've got usb, emu10k1, bt878, and 2 netork cards (eepro100) on irq 5 in windows
17:16<poptix>along with ACPI stuff
17:17<inman>i'm suggesting that you unload emu10k1 modules
17:17<poptix>i miss the days of full irq control in the bios
17:17<inman>then reload your bttv stuff and see if you can get capture again
17:17<inman>nonsense, it's better now.
17:18<poptix>pfft =p
17:18<poptix>it groups everything onto one IRQ
17:18<mdz_>poptix: the days? I bought a motherboard a few months ago with a bios that lets me assign IRQs manually
17:18<mdz_>optionally, of course
17:18<Chutt>turn off acpi
17:18<poptix>mdz_: well, this is a cheap motherboard
17:19<Chutt>and it shouldn't group everything on one irq anymore
17:19<inman>Chutt: i doubt he changed that between reboots...?
17:19<poptix>everything else works great with this board, it just bunches up stuff on the same IRQ
17:19<inman>poptix: it's not a problem if it works.
17:19<Chutt>inman, i've got a broken motherboard that randomly turns it on and off inbetween booting =)
17:19<poptix>i don't reboot this system often
17:19<poptix>who knows what the bios did
17:19<poptix>i think it randomly reassaigns everything everytime i reboot
17:20<mdz_>I have ACPI on on my mythbox and it assigns the IRQs reasonably
17:20<inman>my board has 64bit pci slots that only run at 32bit
17:20* inmansmirks.
17:20<poptix>inman: i've got a 64 bit compaq NC3131 (quad port ethernet) card in a 32 bit slot, so there =p
17:20<inman>poptix: are you running a kernel from your distro or a custom-compiled?
17:20* inmannods.
17:21<mdz_>Chutt: do you think it would help if there were separate links in the howto for stable and unstable?
17:21<mdz_>probably not
17:21<mdz_>that last guy found both woody and sarge but not unstable, somehow
17:22<poptix>mdz_: aren't you the one maintaining the myth debian packages?
17:22<poptix>those precompiled packages need to be staticly linked
17:23<poptix>it would fix a few things =p
17:23<inman>i don't envy the sap who has to download all those staticly linked binaries
17:23<Chutt>i should just make one big huge binary package with all the libs it needs
17:24<Chutt>oh, wait, nm
17:24<poptix>Chutt: =p
17:24<inman>i rebuilt this machine and went from qt-3.0.5 to qt-3.1.2 and i still don't have a selection bar in `manage recordings`
17:26<mdz_>poptix: yes
17:26<mdz_>poptix: and no, they don't need to be statically linked
17:27<mdz_>they work fine as long as the user selects the right branch
17:27<Chutt>that database password thing gets a lot of people
17:27<Chutt>maybe the question should be made higher priority?
17:29<poptix>mdz_: i just don't trust people running debian, who can't compile their own packages
17:29<poptix>(or read instructions)
17:30<poptix>it's like an oxymoron.
17:30<mdz_>poptix: I'm not going to degrade the quality of the packaging to compensate for people who can't read :-)
17:30-!-keyhack [] has joined #mythtv
17:31<mdz_>Chutt: that one makes no sense to me
17:31* keyhackis upset how his 1.1 mhz cannot handle MythTV
17:31<poptix>mdz_: in eggdrop we've got the 'die' command in like 3 different places within the default conf
17:31<Chutt>maybe they're entering their root password or something?
17:31<mdz_>Chutt: that question is asked at priority medium
17:31<poptix>die "you still didn't go through the whole conf..."
17:31<Chutt>keyhack, that's nice.
17:31<mdz_>which is the default
17:32<keyhack>Chutt: thanks for the sarcasm
17:32<mdz_>so the only people who should be missing it are those who explicitly raise their priority above medium
17:32<Ndiin>keyhack: How much do you notice that extra 0.1 MHz? heh.
17:32<mdz_>in which case they also won't see the message from mysql telling them to set a password
17:33<mdz_>so apparently there are all these people running around who are compelled to set a password for mysql, but aren't comfortable enough with the system to see priority: medium debconf prompts
17:33<mdz_>or else there is a bug somewhere
17:34<mdz_>I could raise it to priority high
17:34<mdz_>but then presumably people would get confused because they have never set a mysql root password, and don't even know that there is one
17:34<mdz_>and yet it is asking them for it
17:35<mdz_>at priority high, mysql-server will leave the root password blank and never even say anything about it
17:35<Chutt>mdz, maybe that's what they're doing, and are inserting their root password in that box
17:35<Chutt>and the mysql password is still blank
17:36<mdz_>could be
17:36<keyhack>I did that
17:36<keyhack>I caught the mistake later
17:37<mdz_>maybe I'll add a sentence to the template saying that unless they have explicitly changed it, leave it blank
17:44-!-NonToxic [] has joined #mythtv
17:48-!-keyhack [] has quit ["Client Exiting"]
17:49<aleitner>mdz_: thanks (sorry was away)
17:51<aleitner>no, i have the backend running, do a 'su mythtv' and then run 'mythtv', it tells me it is connecting to the backend server and then nothing happens (no window pops up, no nothing)
17:52<Ndiin>"su" leaves your environment variables untouched, btw.
17:53<aleitner>Ndiin: how else should i log in? 'su -'? then for some reason user mythtv cannot connect to the x server anymore...
17:55<Ndiin>That depends on your setup.. for me, 'su -' doesn't mess with the DISPLAY variable, in fact, the only case it should would be if your login scripts change it, I believe.
17:55<Chutt>if nothing happens at all, then it's most likely attempting to open the sound card and blocking on it
17:56<aleitner>btw, what should i run 'mythtv' or 'mythfrontend'? (Docs say mythfrontend i believe)
17:57<aleitner>ahh, will try and reset $DISPLAY after a 'su -' login
17:57-!-Captain_Murdoch [] has joined #mythtv
18:10<aleitner>btw, is there a mythvideo deb as well?
18:29<aleitner>erm, anybody knows what debian packages are required to compile mythvideo?
18:34<moegreen>aleitner: i don't think there are packages for mythvideo
18:39<aleitner>moegreen: thanks. i am trying to compile it on my own now
18:39<aleitner>but i have a hard time finding the package that provides:
18:40<aleitner> /usr/share/qt/mkspecs/linux-g++/qmake.conf
18:40-!-TheAsp [] has joined #mythtv
18:41<TheAsp>hmm, i have some programs i've never recorded before marked as being repeats
18:44<aleitner>also, how can i find out what keys have what effect in mythtv (to find a good layout for my remote)
18:45-!-r1ckt3r [] has joined #mythtv
18:45<TheAsp>oh, odd, must have been preempted becuse of the war...
18:46<Captain_Murdoch>aleitner: keys.txt
18:51<aleitner>Captain_Murdoch: thanks alot!
18:55<aleitner>anybody knows where the qmake.conf files comes from. i have installed all build-depends i found in the mythtv source package but its still nowhere in sight
18:56<TheAsp>hey, how well does that mpeg4 patch work?
18:57<bigguy>TheAsp: I doubt anyone's really tested it yet
18:57<TheAsp>mdz aparently tried to apply it . . . :P
18:58-!-kain [] has quit ["i hate computers"]
19:00<Chutt>mdz was talking about an earlier patch to mythweb
19:03<TheAsp>i guess i cant read
19:04<TheAsp>how about i apply it and get back to you
19:08<PeteCool>mpeg4 patch? which?
19:08<Chutt>just allows you to turn on hq mode, basically
19:08<Chutt>ie, the non-realtime stuff =)
19:10<PeteCool>Chutt: do you think vorbis could be an interesting audio encoder for myth?
19:11<aleitner>when i try to import channels via 'mythfilldatabase --manual' (de/at listings) it downloads the xmltv listing but then tells me it is ignoring every channel because it is an unkown channel
19:12<bigguy>A man's occupation is to stick his cockulation in a woman's ventilation to increase the population of the younger generation. I got this information from the board of education if you want a demonstration -- LIE DOWN!
19:12<bigguy>soory mischan
19:13<aleitner>should mythvideo be compiled with gcc 3.2 or earlier?
19:13<Chutt>petecool, not unless the vorbis stuff's changed recently, really
19:13<choenig>aleitner: maybe you should have a look at there is something about using german xmltv stuff
19:14<TheAsp>hey, my screen is stretched in myth's menu's
19:14<Chutt>petecool, with vorbis, you can't tell it to encode audio and give you the encoded stuff back immediately, you have to wait for it (up to several seconds later)
19:14<aleitner>choenig: great, thanks!
19:14<TheAsp>next/prev are off the screen
19:14<Chutt>ie, not really good for realtime encoding =)
19:14<Chutt>theasp, with that patch?
19:15-!-hfb [] has quit ["Huh! Wasn't me!"]
19:15<Chutt>figured, the additional options make the screen too big
19:17<PeteCool>Chutt: hmm, that's bad.... what about flac? (Not asking you to do it or anything... I think when I get the schoolwork done by Sunday, it'll make something interesting to try, but I don't want to try something useless)
19:17<TheAsp>seems to work well in livet
19:19<PeteCool>Chutt: oh wait it has to be possible to put it in a nuv container...
19:21<TheAsp>is hq on by default?
19:21<TheAsp>(without the patch)
19:21<mdz>TheAsp: I didn't try to apply any patch
19:21<TheAsp>yeah, i saw, i cant read
19:22<mdz>the vhq mode makes encoding a whole bunch slower
19:22<mdz>and looks very marginally better in my experience
19:22<TheAsp>not much speed diff here
19:22<TheAsp>without 4mv
19:22<TheAsp>does look better though
19:24* TheAspwaits for the comercial skip code to give up in live tv...
19:25-!-aleitner [] has quit ["My damn controlling terminal disappeared!"]
19:27<TheAsp>it's funny how much cpu top uses
19:28<Captain_Murdoch>theasp, I've got it detecting the ratings logo they show in the upper left hand corner now also. need to modify it a little to make it detect white-on-black since I see Fox uses that some where others use a black-on-white rating 'logo'
19:28<PeteCool>TheAsp: yeah when any useful refresh timeout, it makes myth drop frames
19:29<TheAsp>heh, neat idea
19:30<Captain_Murdoch>it uses that to make sure it doesn't false detect a commercial.
19:30<TheAsp>i spose there is no way to detect transparent logos from the bottom right . . .
19:31<PeteCool>TheAsp: there's a virtualdub plugin to remove them, so they are indeed detectable
19:32<PeteCool>I don't know how efficient it is though
19:33<moegreen>Captain_Murdoch: I've seen that tv rating logo at the beginning of a south park commercial
19:35<Captain_Murdoch>Liam is working on code to do that, but I need to try to figure out how to do it offline. his patch works while you're watching a show. I'm trying to do stuff offline so you have instantaneous skips when you are playing back.
19:36<Captain_Murdoch>moegreen: I'm almost thinking there could/should be a bunch of options that can be turned on/off that are used to turn on/off the various detection methods. no single one is giong to be perfect for everyone/everywhere and putting them all together is probably bad as well.
19:37<Captain_Murdoch>the worst-case scenario right now is that it wouldn't flag that commercial since it has the rating logo in it.
19:37<moegreen>Captain_Murdoch: right, I just wanted to let you know. :)
19:38<Captain_Murdoch>thanks. I've debated making a small program that people could use to grab out a section of a .nuv file and send so I could test things like that.
19:45<TheAsp>a maximum comercial break time would be nice too
19:45<TheAsp>if there isntr one already
19:47<PeteCool>Captain_Murdoch: what would really be useful would be per-channel detection mode(s) to use
19:47<Captain_Murdoch>well, it detects individual commercials right now, so unless the actual show has blank frames every 15,20,30,45,60,90,or 120 seconds then it shouldn't go off into never-never land.
19:47<TheAsp>i have a channel that comes in in light and dark diagonal stripes, which can defeat the blank screen detector i think
19:48<TheAsp>i had an 800 second skip the other night
19:48<Captain_Murdoch>PeteCool: that's something else I have typed into my TODO file. :) I was thinking something like that, as well as a flag to say "don't bother checking this show/channel for commercials"
19:50<TheAsp>oldish build though
19:50<Captain_Murdoch>right now it detects the individual commercials then appends them all together to get the commercial break. I'll make a note to put in a sanity check for the commercial break length. are you using the auto commercial flagging mode or just hitting 'Z'.
19:50<Captain_Murdoch>ok, so you're just hitting Z.
19:51<Captain_Murdoch>current CVS can (if you have auto flagging turned on) run a thread after a recording that will go through the video looking for blank frames and will generate a commercial list and put it in the database.
19:51<Captain_Murdoch>then when you hit 'Z', it's an instantaneous skip. You can also use 'Q' to skip backwards.
19:51<TheAsp>yeah, didnt it used to do that at playback>?
19:51<TheAsp>recording, not playback
19:52<TheAsp>up until a week or so ago
19:52<Captain_Murdoch>it does it after the recording finishes.
19:52<Captain_Murdoch>it flags blank frames during recording, but then it analyzes after it finishes, looking for the rating logo and eventually other stuff.
19:53<TheAsp>i have io concerns with that new thread
19:53<Captain_Murdoch>it runs as a nice-d process so it uses very little cpu time, but can take a while because of that.
19:53<TheAsp>yeah, i saw that :)
19:54-!-choenig [] has quit [Remote closed the connection]
19:54<Captain_Murdoch>the thread doesn't even show up in the top of 'top' when I run it here since it's sleeping every frame in addition to being niced.
19:54<TheAsp>i upped it to 10 trying to play games . . .
19:55<Captain_Murdoch>I'm running 352x480 on an Athlon 700, so I'm keeping an eye on cpu utilization. I've scheduled back-to-back-to-back recordings to make sure it doesn't mess up the 2nd & 3rd recording. During the 3rd recording, mythbackend was recording and there were 2 threads doing the commercial searching.
19:55<Captain_Murdoch>upped what to 10?
19:55<TheAsp>the niceness of that thread
19:57<TheAsp>i'm using 512x384 on a 1.1ghz
19:57<Captain_Murdoch>I think current cvs code just does the nice. my current code sleeps every frame as well so it won't consume all cpu even if there's nothing else running. I played around with trying to intelligently throttle cpu usage, but decided to just have it run slow & steady & predictable.
19:57<Captain_Murdoch>19 is the nicest you can get, so you told it to be less nice and use more cpu?
19:58<TheAsp>it was running at 4...
19:58<Captain_Murdoch>it does a "nice(19)" how could it be at 4?
19:58<Captain_Murdoch>tv_rec.cpp has "nice(19);"
20:00<Captain_Murdoch>the newer code is even nicer because it sleeps. I can be recording @ 352x480 and flagging 2 shows for commercials at the same time and still have 20% cpu free on my Athlon 700.
20:00<TheAsp>maybe another case of me not being able to read? )
20:00<TheAsp>(possibly read off the priority)
20:01<Captain_Murdoch>well, newer will be better. :) I hope to have that committed in the next few days.
20:04<TheAsp>damn, that res aint bad
20:05<Chutt>there, new osd code is shaping up
20:05<Chutt>draws images and text again, at least =)
20:05<TheAsp>is the second screen of the recording profiles necessary? the one with the pulldown with the profile name in it
20:06<Chutt>it will be once you can rename stuff
20:06<-- r1ckt3r( has left #mythtv
20:07<TheAsp>i still have profiles from when you could :P
20:07<TheAsp>or add them...
20:09<Captain_Murdoch>anybody seen setup segfault in QWidget::focusProxy?
20:11<Captain_Murdoch>I'm trying to setup a 2nd backend on my dev machine and getting a segfault running setup. calls go SetupMenu -> ThemedMenu -> parseSettings -> parseButtonDefinition -> parseText -> QWidget::setFont -> QWidget::hasFocus -> QWidget::focusProxy. sound like a problem with a theme I have or something?
20:11<Captain_Murdoch>mythfrontend runs fine on this box.
20:13<TheAsp>hmm, frontend just seg'd on me
20:13<TheAsp>going into the watch recordings screen
20:13<Chutt>you know what to do
20:13-!-NonToxic is now known as ddrToxic
20:14<Captain_Murdoch>TheAsp: I've had it do that quite a few times on me lately since the new playback screen went in. I haven't tracked it down yet through. haven't put the code on my main box yet and won't. :)
20:14<TheAsp>i do
20:14<TheAsp>i just dont want to, as it's recording
20:14<Chutt>theasp, have you updated today?
20:14<TheAsp>few hours ago
20:15<Chutt>rkulagow reported a bug a big ago
20:15<Chutt>moegreen said he'd check in a fix, dunno if he's gotten to that yet
20:15<TheAsp>nothing new since i built
20:15<TheAsp>rebuilding with debug
20:23<PeteCool>any way to know how much free cpu time is left, without using top? top uses too much cpu
20:24<TheAsp>i use the gnome cpu usage panel applet
20:25<TheAsp>there is a file in /proc that shows it to if i remember right
20:30<TheAsp>still building
20:38<TheAsp>ahh, on the epg
20:38<Chutt>still have the frontend and backend to go =)
20:41<TheAsp>btw, those new screens look sweet when they work
20:41<Chutt>bugs should be getting ironed out as people send in backtraces
20:42<TheAsp>show deletion planned for the next release?
20:42<TheAsp>to free up space
20:42<Chutt>automatic, you mean?
20:43<Chutt>i dunno
20:43<Chutt>captain_murdoch, that guy's filldatabase problem is that the source id needs to be the sourceid field from the database, not something like 'Television'
20:44<Captain_Murdoch>that's another Chris. :)
20:44<Chutt>oh, oops
20:44<Captain_Murdoch>same initials C.P though. :)
20:44<Chutt>sorry, i've taken to just skimming the mailing list posts =)
20:44<Captain_Murdoch>ditto. falling behind on users, only read that every few days. glad it took a lot of traffic off dev
20:45<Chutt>yay, new osd code actually scales coordinates properly now
20:45<Chutt>unlike the old stuff, which scaled some and not others =)
20:46<Captain_Murdoch>I got setup to run by ssh-ing to another machine then back to my own and tunning it through ssh. weird... have my workstation setup as a slave backend and scheduled 2 recordings for 9pm EST both pointed at the same nfs-mounted storage directory.
20:46<Chutt>make sure you're running recent cvs code
20:47<Chutt>if you're scheduling timeslot/channe/anytime recordings that commonly don't have subtitles/descriptions
20:48<Captain_Murdoch>oh yeah, forgot about that. oh well, if it works it works, if not then that's ok. just testing anyway. don't want to upgrade my main machine until I know it won't crap out on me when I got into playback. hate to have to drag the keyboard out.
20:48<Captain_Murdoch>most/all of my stuff has that. these are just 2 things I picked out of the blue to test dual recording.
20:49<Chutt>that's why it took me so long to track that down :(
20:49<Chutt>all of the stuff i record has subtitles and descriptions
20:50* TheAspgrumbles something about never doing this again
20:51<Chutt>theasp, i've switched 5 or 6 times between release + debug today
20:52<Captain_Murdoch>got it so I can record SciFi now, for another worst-case commercial detection channel. :)
20:52<Captain_Murdoch>compiling isn't that bad, try it on a machine with only 128M of ram or inside vmware even. :)
20:52<Chutt>the stuff did _not_ do well with the children of dune stuff
20:53<Captain_Murdoch>yeah, I was hoping to get it hooked up before then but my IR transmitter didn't want to work with my DishNetwork Sat receiver. I can receive codes fine and can transmit to my VCR fine but the sat receiver doesn't work too well and the lircd sample conf even has a note about the touchiness of it.
20:54<TheAsp>FUCK. it worked
20:57<Chutt>just paste it
20:57<Captain_Murdoch>TheAsp: try watching something then exiting the playback screen. that's when it's showing up most for me.
20:58<TheAsp>#0 0x40609edd in QString::QString(QString const&) ()
20:58<TheAsp> from /usr/lib/
20:58<TheAsp>#1 0x080820fb in PlaybackBox::getPixmap(ProgramInfo*) (this=0xbfffeae0,
20:58<TheAsp> pginfo=0xbfffec30) at playbackbox.cpp:1973
20:58<TheAsp>#2 0x080790eb in PlaybackBox::updateVideo(QPainter*) (this=0xbfffeae0,
20:58<TheAsp> p=0xbfffe540) at playbackbox.cpp:640
20:58<TheAsp>#3 0x08077567 in PlaybackBox::paintEvent(QPaintEvent*) (this=0xbfffeae0,
20:58<TheAsp> e=0xbfffe850) at playbackbox.cpp:489
20:58<Chutt>theasp, same thing as earlier, moegreen knows 'bout that one
20:59<TheAsp>i have a messed up png...
21:00<TheAsp>for a show
21:04<Captain_Murdoch>well, 2 recordings fired off, if only I had remembered to put the mythtv user into the right group to be able to write to the storage directory. :)
21:08<TheAsp>oh, the priority is 19, nice is still 4
21:08<Captain_Murdoch>hmm, should be something like priority 34, nice 19
21:08<TheAsp>23604 mythtv 19 4 67724 46m 7272 R 59.6 6.1 2:39.72 mythbackend
21:09<TheAsp>23616 mythtv 10 -15 67724 46m 7272 S 30.3 6.1 3:19.25 mythbackend
21:27-!-Timon [] has joined #mythtv
21:30<Timon>mdz, chutt: either of you here?
22:10-!-SarahEmm [] has joined #mythtv
22:19-!-ddrToxic is now known as NonToxic
22:20<SarahEmm>i'm getting a segfault running 'setup'... it's a duron, so i don't think it's a -march problem.. any ideas?
22:20<SarahEmm>hihi nontoxic
22:20<Chutt>compile for debugging like it describes in the howto, send in a backtrace
22:20<NonToxic>heya sarah
22:24<SarahEmm>duh.. i should have known that..
22:34<NonToxic>bsaunder: no
22:34<NonToxic>er, whoops.
22:34<NonToxic>wrong win
22:45<-- |Uli|has quit (Remote closed the connection)
22:46* inmanbeats up a PHP designer.
22:46<inman>the reference system in PHP is so fscking stupid.
22:56-!-kain [~kain@] has joined #mythtv
22:57<kain>i don't suppose anyone might know why i can only seem to read 8 bit 8kHz mono audio from /dev/dsp? it's driving me quite insane
22:58<inman>is /dev/dsp a capture device?
22:58<kain>no its my sound card
22:59<vektor>kain: You wrote a recorder app?
22:59<vektor>kain: You're using the OSS API ?
22:59<vektor>kain: What's the soundcard?
23:00<kain>using alsa with the snd-cmipci module
23:00<kain>im just using aplay as per the test in the FAQ, i got the mixer settings right to mythtv but the sound quality is stinking
23:01<inman>i've found that aplay doesn't work as well as, eg. play (sox)
23:01<vektor>And does the driver support better than 8bit 8khz? :)
23:01<vektor>sox probably uses the OSS API.
23:02<inman>i'm using alsa, though, and aplay chops on simple wav samples, while play doesn't.
23:02<vektor>anyway, if that actually works better than aplay in this situation i'd be surprised
23:02<inman>yeah it surprised me, too. :-P
23:02<vektor>inman: have you investigated/reported the bug?
23:02<inman>vektor: nope.
23:03<vektor>inman: or considered how aplay caches/looked at your HD performance?
23:03<inman>to hear chutt tell it, i'm running a frankenstein system here -- not to be trusted.
23:03<vektor>you may want to do that.
23:03<inman>vektor: i'm talking about 10k samples
23:03<vektor>well whatever, anyway.
23:03<inman>actually closer to 5k
23:03* inmannods.
23:03<vektor>that could be confounding your test.
23:04<inman>kain: define "stinking" sound quality
23:04<inman>kain: hisses/pops/static what?
23:05<kain>well it's 8k mono
23:05<vektor>the file is?
23:05<kain>no, that's all i can make anything read from /dev/dsp at
23:05<inman>did you load the driver with, eg. rate=8000 ?
23:06<kain>i can't seem to find any documentation that tells me i can do that
23:06<inman>try it.
23:06<kain>i shall
23:13<SarahEmm>it segfaults in 640x400
23:13<PeteCool>the cmipci driver is fine, I'm using it here
23:13<SarahEmm>i changed to 800x600 and it's fine
23:13<kain>PeteCool: are you passing it any special options or anything?
23:17<PeteCool>kain: nope
23:17<PeteCool>kain: it just magically works :)
23:17<PeteCool>kain: maybe the test is screwed
23:17<PeteCool>kain: make sure to set the line volume to 100%
23:25<PeteCool>Chutt: If I have a backend recording a show, can I watch it as it records, from multiple frontends? Or would it hit disk latency issues?
23:25<Chutt>should be able to
23:29-!-TheAsp [] has quit [Read error: 54 (Connection reset by peer)]
23:32<PeteCool>is this a bttv chip: CX23881/883 audio/video decoder
23:33<bigguy>PeteCool: that lindows guy promised a driver
23:33<bigguy>PeteCool: but I don't know if it ever really was delivered
23:34<bigguy>I think he sent mdz something off list tho
23:34<Chutt>it was
23:34<bigguy>anyway I'm off to read some then get some sleep
23:34<Chutt>the bttv guy has it
23:34<Chutt>and is working on an actual working driver for that chip
23:35* bigguyis [away -={ read -> sleep }=- ]
23:37<PeteCool>Chutt: playback is touchy for me with current CVS... If I ff/rw a little it slows, then hangs... only since the Nuppel* changes this afternoon
23:38<Chutt>those only touched the preview pixmap generation
23:38<Chutt>nothing else
23:40<inman>when playback ends, isn't it supposed to ask me what i want to do with the recording?
23:40<Chutt>if you have that setting on
23:40<inman>ah, cool
23:41<inman>when i delete something from `manage programs`, it sometimes prints a "O MB" in the proper column but a line above the topmost program listed.
23:44<Chutt>i'll let moegreen know
23:44<inman>i'm looking at it, but nothing's jumping out at me.
23:48-!-NonToxic is now known as Toxic-Byebye-OPN
23:49<mdz>bigguy: the stuff he sent me was up on the web somewhere anyway by that point
23:58<inman>myth's UI works so much better without Xinerama, these days. the only "rogue" window is the MythTV window itself.
23:59<inman>the hanging dialog problem i was having mysteriously resolved itself with one of the patches in the last couple weeks.
23:59<inman>nice. :-)