Back to Home / #mythtv / 2008 / 03 / Prev Day | Next Day
#mythtv IRC Logs for 2008-03-23

---Logopened Sun Mar 23 00:00:26 2008
00:46-!-davilla [] has quit ["Leaving"]
01:50-!-th__ [] has quit [Read error: 113 (No route to host)]
02:39-!-gnome42 [] has quit [Remote closed the connection]
02:54-!-Zaqq [n=Zaqq@] has joined #mythtv
03:36-!-joobie [] has joined #mythtv
03:42-!-xris [] has quit []
03:46-!-xris [] has joined #mythtv
03:56-!-xris [] has quit []
04:35-!-Zaqq_ [n=Zaqq@] has joined #mythtv
04:53-!-Zaqq [n=Zaqq@] has quit [Read error: 110 (Connection timed out)]
05:15-!-Dibblah [] has quit [Remote closed the connection]
05:15-!-Dibblah [] has joined #mythtv
05:20<gbee>rooaus: "cerr << "abc123" << endl;" is ok qith qt4, "QString string = "abc123"; cerr << string << endl;" is not
05:21<gbee>but that said, VERBOSE should always be used instead of cerr/cout
05:21<gbee>if only because VERBOSE messages can be turned off and they are timestamped
05:35<rooaus>gbee: Yeah, meant converting to VERBOSE...
05:35<rooaus>VERBOSE(VB_IMPORTANT, QString("Important message follows: %1").arg(foo)) vs VERBOSE(VB_IMPORTANT, "Important message follows: " << foo)
06:04-!-joobie [] has quit ["This computer has gone to sleep"]
06:06<gbee>rooaus: that's correct
06:12<rooaus>gbee: both you mean?
06:32<rooaus>Gmail is useful... found what I was looking for after some searching, [14990], wasn't a utf8 issue after all. So you think VB_IMPORTANT is best, guess they wouldn't be in there if they weren't important? :)
07:03<gbee>rooaus: use VERBOSE instead of cerr
07:04<gbee>there is even more reason to use VERBOSE with QT4 because of changes to QString
07:06<rooaus>gbee: Thanks, it is appreciated.
07:07<gbee>admittedly VERBOSE(VB_GENERAL, QString("%1").arg(string)); is a little messier than "cerr << string << endl;"
07:08<gbee>but it's more useful in the end
07:12<rooaus>Yeah, just grepped *.cpp for cout|cerr and got 2002 hits, will try and knock a few of them off.
07:16-!-|gunni| [] has joined #mythtv
07:16-!-|gunni| [] has quit [Dead socket]
07:18<janneg>rooaus: don't. that's wasted effort if you do it in trunk and not especially useful if done in qt4 branch
07:18<janneg>except for converting some of the plugins
07:23-!-|gunni| [] has joined #mythtv
07:32<rooaus>janneg: Ok, haven't sorted out a qt4 build environment yet (not a big deal, I hope). Which plugins? From the QT4 branch I assume?
07:47<janneg>yes, from branches/mythplugins-qt4. iirc mythphone had a lot of cerr/cout
08:45-!-grokky [] has quit []
08:51-!-dagar [] has quit [Read error: 110 (Connection timed out)]
08:52-!-dagar [] has joined #mythtv
08:59-!-Agrajag- [] has quit [Read error: 110 (Connection timed out)]
09:43-!-jarzynka [] has joined #mythtv
09:44-!-jarzynka [] has left #mythtv []
09:54<janneg>so, my main backend is again on the qt4 branch
10:22<jams>is there an easy way to tell if a plugin was loaded via the menu vs "mythfrontend pluginname" ?
10:23-!-Zaqq_ [n=Zaqq@] has quit [Read error: 110 (Connection timed out)]
10:39-!-turbo is now known as briand
10:42<janneg>mythweb is partially broken in after the db conversion. broken non ascii chars then mythweb uses the database directly
10:45-!-_gunni_ [] has joined #mythtv
10:57-!-Agrajag- [] has joined #mythtv
11:00-!-wild_oscar [] has joined #mythtv
11:00<wild_oscar>hi there
11:00<wild_oscar>anyone here responsible for the xmltv?
11:02<wild_oscar>I'm having a problem with tv_grab_pt, don't know who I should talk to regarding this issue
11:02-!-|gunni| [] has quit [Read error: 110 (Connection timed out)]
11:20<wild_oscar>there is a dependency issue with the latest version of this file on ubuntu, if someone here knows who's responsible for it please let me know
11:21<laga>wild_oscar: #ubuntu-mythtv
11:22<wild_oscar>cheers, laga
11:35-!-turbo [] has joined #mythtv
11:37-!-briand [] has quit [Read error: 110 (Connection timed out)]
11:42-!-turbo is now known as briand
12:00-!-Dave123 [] has quit ["Leaving"]
12:02-!-ToadP [] has quit [Read error: 104 (Connection reset by peer)]
12:13-!-zion75 [] has joined #mythtv
12:16-!-Dave123 [] has joined #mythtv
12:17<zion75>Hello! Having a problem compiling mythtv from SVN. Using ubuntu 7.10 and is getting this error: ERROR! You must have FreeType installed to compile MythTV.
12:18<laga>zion75: why don't you use the packages?
12:19<zion75>always (two years) been on svn and it has worked just fine
12:19-!-peman [] has joined #mythtv
12:20<peman>morrn morrn
12:29-!-gnome42 [] has joined #mythtv
12:40-!-dekarl [] has joined #mythtv
13:07-!-wild_oscar [] has quit [Remote closed the connection]
13:24-!-dekar1 [] has joined #mythtv
13:41-!-dekarl [] has quit [Read error: 110 (Connection timed out)]
13:55<gbee>The theme (home) is missing a themeinfo.xml file
13:55<gbee>seems to be looking at /home instead of /home/{user}/.mythtv/themes/
13:58<janneg>gbee: that was an error in theme install. in share/mythtv/themes is a directory home
14:12-!-mykeul [] has joined #mythtv
14:23<gbee>janneg: ahh, ok
14:24<gbee>I didn't stop to check
14:35-!-nordenm [] has joined #mythtv
14:54-!-Niklas_E [] has joined #Mythtv
14:54-!-Niklas_E [] has left #Mythtv ["Time makes no sense"]
15:44-!-mykeul [] has quit ["Leaving."]
15:46-!-foxhunt [] has joined #mythtv
15:46-!-foxhunt [] has quit [Client Quit]
16:01<gbee>merge tonight?
16:09-!-gerhard [] has joined #mythtv
16:09<janneg>I don't know
16:10<gbee>want to replace Q3PtrList in mythlistbutton, but I'll wait for a merge first
16:10<janneg>haven't found the conversion in mythweb
16:26<gbee>mythweb/includes/util.php line 353
16:27<gbee>mythweb/classes/Services_JSON.php line 149, line 193
16:27<gbee>possibly ..
16:28<gbee>oh, neither of those seem to be used :/
16:31<gbee>OK, util.php line 253
16:32<gbee>function html_entities() { return htmlentities($str, ENT_COMPAT, 'UTF-8'); }
16:35<gbee>I'm not really helping ... :(
16:35-!-lcase [] has joined #mythtv
16:35-!-lcase [] has left #mythtv []
16:35<janneg>no, there is no conversion. php seems to use utf8 strings. SET NAMES utf8; before the query fixes it
16:37-!-sonium [] has joined #mythtv
16:37-!-lcase [] has joined #mythtv
16:45-!-sonium [] has left #mythtv []
16:48<Anduin>I guess I didn't waste too much time on the plugin conversions... janneg why the explicit charsets for the columns that are not special cased?
16:53<janneg>Anduin: I hope you haven't done anything. I forgot to send my conversion script
16:54<janneg>the explicit charsets are a leftover from debugging the updates
16:55<janneg>I missed a column in conversion and it stayed for mysterious reasons latin1
16:57-!-Nikas [] has joined #mythtv
17:04<janneg>mythweb fixed
17:05<janneg>anything left before the merge
17:12<janneg>gbee: mythfrontend -p segfaults after a db query
17:14<janneg>I thought it gets the db parameters from the discovered backend
17:15<Chutt>i'd prefer to give it another day or so
17:15<Chutt>ie, one day after any big change
17:17<Chutt>don't want something like the utf8 issue to pop up after merging
17:17<Chutt>cuz then certain people will bitch =)
17:19<gbee>janneg: so the db connection is invalid, should be easy to fix
17:21<gbee>Chutt: I think we've given plenty of warnings and given the number of commits I'd guess people are holding back, but another 24 hours isn't going to hurt
17:23<Anduin>There are still some problems with custom record (binding too many values), I'll look at fixing them in a little while (after switching production boxes to qt4)
17:25<gbee>heh, I'm very glad you spotted that Anduin ;)
17:31-!-PointyPumper [i=Pintlezz@] has quit [Read error: 110 (Connection timed out)]
17:33-!-lcase [] has quit []
17:43-!-peman [] has quit [Remote closed the connection]
17:53<gbee>janneg: I'm not sure what's happening with the -p segfault, sorry
17:54-!-feld [] has joined #mythtv
17:56<gbee>damn, is there supposed to be an offset column in recordedmarkup? My table is missing it
17:56<feld>i do use mythtv and all, but has anyone here ever tried to get a pvr-150 MCE to display the composite input via vlc?
17:57<feld>oops my bad, mythtv-users is what i want. ignore that :P
17:57-!-feld [] has left #mythtv []
17:58<Anduin>gbee: Yes, varchar(32) (default NULL)
17:58<gbee>Anduin: thanks
17:59<gbee>think I got ahead of myself and deleted it after it was made obsolete by the recordedseek split
18:03<gbee>Table 'mythconverg.videobookmarks' doesn't exist
18:03-!-JoeBorn [] has joined #mythtv
18:07<gbee>shouldn't that have been removed? Along with the offset column it should have been removed from dbcheck after 0.20 or 0.21 was released
18:09<Anduin>gbee: videobookmarks was removed
18:09-!-foxhunt [] has joined #mythtv
18:10-!-foxhunt [] has quit [Client Quit]
18:10<gbee>Anduin: I can't find a DROP TABLE for it in dbcheck.cpp, it's still getting created on new installs
18:11<Anduin>gbee: Hmm, I don't know when it moved to the main dbcheck.cpp but it is dropped in the myhtvideo dbcheck.cpp
18:12<gbee>Anduin: that would explain why it no longer existed on my install - that will need fixing then because it breaks the utf8() conversion update
18:14-!-czth_ [n=dbrobins@nat/microsoft/x-9c2af3b3227da8ab] has joined #mythtv
18:17<gbee>janneg: I've deleted the changes to videobookmarks
18:17<janneg>so it should be removed from the updates
18:18<janneg>gbee: thanks
18:18<gbee>the DROP TABLE for videobookmarks should be moved from mythvideo/dbcheck but that can wait for now
18:20<Anduin>I'll add a DROP IF EXISTS to libmythtv's
18:21<gbee>Anduin: k
18:31-!-czth__ [n=dbrobins@nat/microsoft/x-09690960a8769dd9] has quit [Read error: 110 (Connection timed out)]
18:32-!-gerhard [] has quit ["ChatZilla 0.9.81 [Firefox]"]
18:43-!-jhulst [n=jhulst@unaffiliated/jhulst] has joined #mythtv
18:57<Anduin>The update seems painfully slow
19:02-!-kiatoa_ [] has joined #mythtv
19:14<gbee>wasn't that slow here, but of course it'll depend on the size of your tables
19:15-!-PointyPumper [n=pintlezz@] has joined #mythtv
19:17<Anduin>My largest table that would be touched is credits at 32MB, 1216 took nearly 18 minutes, program has a large index, but still seems excessive.
19:18<Anduin>scratch that, 1216 took 28 minutes
19:25<gbee>took a minute here
19:28<janneg>it was awfully slow. both updates took 20 seconds each on an empty database
19:31-!-grokky [] has joined #mythtv
19:31<gbee>well it was interrupted twice by the failures I mentioned, so a minute was just a guess and obviously not an accurate one
19:34-!-gardz [] has joined #mythtv
19:44<Anduin>I think doing more than one modify per alter would speed things up, currently it seems to create the temp table, modify, drop old, rename, rinse and repeat for every column.
19:46-!-kiatoa_ [] has quit ["Konversation terminated!"]
19:54<Anduin>1h4m to update
19:54-!-_charly_ [] has quit [Excess Flood]
19:56-!-_charly_ [] has joined #mythtv
20:10<Anduin>Some scheduler bindValues brokenness Scheduler::BuildNewRecordsQueries and UpdateMatches, I have a feeling that unless someone who uses every permutation tests the Qt4 branch, well they will be disappointed.
20:10-!-ahbritto [] has joined #mythtv
20:11<janneg>Anduin: yes, multiple modify statements will be faster. I discovered that in recordseek updates and forgot it unfortunately
20:12<Anduin>janneg: I'll try to work on speeding it up later if no one beats me to it.
20:13<janneg>Anduin: we better fix the scheduler brokenness or bjm gets mad
20:14<Anduin>janneg: Yes, I'll try, any basic power search seems to break it, easy to spot why just hard to find every one of these without hitting them.
20:15<janneg>I'll add a regexp that tests if the bindvalue is in the query
20:16<Anduin>Yeah, I was thinking we could punt on it that way, just output something and collect the broken ones so nothing would break.
20:23-!-Nikas [] has quit ["( :: NoNameScript 4.2 :: )"]
20:34<rooaus>janneg: Is the "ASSERT failure in QWidget" known about in mythphone?
20:38<rooaus>I have compiled all plugins which I don't usually do, will pastebin a bt, just a sec...
20:44-!-jhulst [n=jhulst@unaffiliated/jhulst] has quit [Read error: 110 (Connection timed out)]
20:47<rooaus>janneg: To clarify, I only see the problem if I install mythphone, if I clean out lib/mythtv/plugins it goes away and comes back when mythphone is installed.
20:56-!-reynaldo [] has joined #mythtv
21:06-!-_gunni_ [] has quit ["KVIrc 3.2.4 Anomalies"]
21:07<Chutt>janneg, is qregexp thread safe/reentrant now? (since you made it static in dbcon.cpp)
21:55<Anduin>1215 down to 1m23s from 28m with multiple mods per alter
22:04<Anduin>Yeah, I did change a few DB settings, but they seem to make no difference, I'll finish testing it and check it in.
22:05<Chutt>similar improvements possible for the second update as well?
22:06<Anduin>almost certainly, I don't know why I'm letting my slow test finish before testing that but I am
22:32<Chutt>that's slightly better than over an hour
22:41<Anduin>Apparently it was only 1m7s to the double keybindings context error, just under 2 minutes for all of 1216, still much better. It looks like with a larger log flush size I could have made the old stuff run in about half the time.
22:52-!-jk`gentoo [] has joined #mythtv
23:30-!-jhulst [n=jhulst@unaffiliated/jhulst] has joined #mythtv
23:36-!-clever [] has quit [Read error: 110 (Connection timed out)]
23:36-!-gizmobay [n=gizmobay@VDSL-130-13-22-197.PHNX.QWEST.NET] has joined #mythtv
23:36-!-gizmobay [n=gizmobay@VDSL-130-13-22-197.PHNX.QWEST.NET] has left #mythtv ["Leaving"]
23:40-!-nsaspook [] has quit ["using sirc version 2.211+KSIRC/1.3.12"]
23:45-!-nsaspook [] has joined #mythtv
23:58<Anduin> I'll commit it in a minute, still debating if the the hard-coded mythconverg is worth fixing
---Logclosed Mon Mar 24 00:00:00 2008