:: #mythtv

Daily chat history

Current users (94):

aloril, Anssi, anykey_, Beirdo, brfransen, brtb, CaCtus491, cattelan, cesman, Chutt, clever, coling, Cougar, damaltor, danielk22, Dave123, davide, dekarl, dlblog, ElmerFudd, foobum, foxbuntu, ghoti, gigem, gregL, GreyFoxx, hank, highzeth, J-e-f-f-A, j-rod|afk, jams, jarle, jcarlos, joe_, joki, jpabq, jpabq-, jpabq_, jstenback, justinh, jwhite, jya, k-man, kenni, knightr, kormoc, kurre2, kwmonroe, laga, lolcat`, mag0o, Malard, mike|2, mrand, MythBuild, MythLogBot, mzanetti, OldEnK, peitolm, pheld, ponyofdeath, poptix, purserj, Rasperin, rhpot1991, rsiebert_, skd5aner, Slasher`, sphery, sraue, stuarta, stuartm, superm1, sutula, tgm4883, ThisNewGuy, tomimo, tris, Unhelpful, vallor, wagnerrp, wahrhaft, warped, wookey_, wseltzer, xavierh, xavierh_, XChatMav, XDS2010_, xris, ybot_, zCougar, zombor, _charly_
Friday, March 2nd, 2012, 00:42 UTC
[00:42:33] jpabq_ (jpabq_!~jpabq@mythtv/developer/jpabq) has quit (Quit: jpabq_)
[01:01:18] stichnot (stichnot!chatzilla@mythtv/developer/stichnot) has quit (Ping timeout: 252 seconds)
[01:03:36] noahric (noahric!~noahric@ has quit (Quit: noahric)
[01:35:45] andreax (andreax! has quit (Read error: Connection reset by peer)
[01:58:46] danielk22: Heh, I just tried using the OpenGL and OpenGL-lite video renders on my fairly modern i5 based T510 laptop.. The OpenGL renderer one can't handle 720p, and neither can the OpenGL-lite if deinterlacing is configured (on neither 720p nor 1080i material.)
[01:59:22] danielk22: The graphics chip is a GT218 [NVS 3100M].
[02:02:34] sphery: Yeah, I can use OpenGL video renderer at extremely high CPU usage on my Athlon X2 250 3.0GHz system in performance mode (not freq scaled), but can't use timestretch without significant frame dropping. I can go all the way to 2.0 timestretch with VDPAU or Xv renderers.
[02:04:22] danielk22: sphery: I've been looking a bit at the renderers since I'm going to be the one maintaining them. The only ones that really look viable right now are VDPAU and XVideo.. I like the OpenGL one in theory, but the performance just isn't there.
[02:04:59] davide (davide! has quit (Remote host closed the connection)
[02:05:12] sphery: yeah, in the "drop Xv" plan, there was an assumption that doing so would allow focusing on OpenGL renderer to improve efficiency of it
[02:05:24] davide (davide! has joined #mythtv
[02:05:59] sphery: but, IME, it would take a /lot/ of improvement to even be usable for most
[02:06:10] sphery: (not even considering the underpowered systems)
[02:06:19] danielk22: sphery: That still might be a viable option. I haven't looked at it enough yet, but I thought we were closer since there was already talk of dropping XVideo and going with OpenGL.
[02:08:15] sphery: FWIW, I'm extremely happy with my VDPAU rendering, but until I got a VDPAU-capable card, I was using Xv because I refuse to live without timestretch :)
[02:08:47] skd5aner (skd5aner! has quit (Quit: Leaving)
[02:09:34] sphery: (and, BTW, my GeForce 7800GTX on the same Athlon X2 250@3GHz couldn't do OpenGL renderer at 1x, let alone timestretch. It's only since I got my GT220 that 1x OpenGL is possible with that CPU.
[02:13:54] danielk22: sphery: VDPAU is the gold standard right now, no question about it.
[02:27:24] noahric (noahric!~noahric@ has joined #mythtv
[02:34:27] kormoc (kormoc!~kormoc@mythtv/developer/kormoc) has quit (Quit: kormoc)
[03:05:59] jya: is libmp3lame required? does it add any advantages?
[03:06:16] Captain_Murdoch: mp3 encoding for mythtranscode and software encoders
[03:06:31] Captain_Murdoch: not sure if we can rip CDs to mp3 or not.
[03:08:07] Captain_Murdoch: libav* still uses libmp3lame for encoding mp3.
[03:09:33] jya: any particular reasons it's not enabled by default?
[03:09:59] Captain_Murdoch: not sure. I thought we were using it for software encoding and mythtranscode by default.
[03:10:47] Captain_Murdoch: ah, that's why, NuppelVideoRecorder uses libmp3lame directly, so we don't need to enable it for libav*
[03:11:18] Captain_Murdoch: to use mp3 with HTTP Live Streaming, you do need to enable it since the HLS code uses libav* for encoding, not NVR.
[03:12:06] jya: ok… on all my system I tried , after running configure, it shows libmp3lame : no
[03:12:08] Captain_Murdoch: I had a patch years ago to convert NVR to use libav* for encoding mp3, but never put it in.
[03:12:29] jya: so I can add --enable-libmp3lame and everything will be dandy then?
[03:12:56] ** jya still struggling to build a universal mythfrontend on the mac.. but getting there **
[03:13:02] Captain_Murdoch: I think you might need --enable-nonfree as well
[03:13:42] jya: when you say NuppelVideoRecorder uses libmp3lame directly, you mean it incorporate the soruce code of libmp3lame ?
[03:14:56] Captain_Murdoch: no, we link to the lib and NVR uses lame* routines to encode audio instead of libav* routines.
[03:15:18] Captain_Murdoch: libmythtv gets linked with -lmp3lame
[03:17:00] jya: i see… ok so if I understand correctly, the way I should read "mp3lame yes" means libmp3lame is compiled for libav*
[03:17:13] jya: not directly related to libmyth*
[03:17:21] Captain_Murdoch: but libav* won't get linked with -lmp3lame unless you --enable-libmp3lame and --enable-nonfree, so no HLS support. we should just enable it by default now, but for some reason I was thinking that there was one distro that disabled linking with libmp3lame. we could enable by default and make them --disable-libmp3lame.
[03:17:35] Captain_Murdoch: right, --enable-libmp3lame is just for linking libav* with libmp3lame
[03:17:42] jya: ok… thanks
[03:17:55] ** dblain is finally making progress on the WSDL support... 99.9% done. One more thing and a lot of testing and it should be done. **
[03:17:55] Captain_Murdoch: which as far as I know is only used with HTTP Live Streaming.
[03:18:03] Captain_Murdoch: dblain, nice.
[03:22:19] zombor_ (zombor_!~zombor_@kohana/developer/zombor) has joined #mythtv
[03:22:56] k-man: dblain, what is WSDL? Web Services Description Language?
[03:23:57] dblain: yes. It will allow any tool that understands wsdl to generate a proxy class to access the Service API. I'm testing it with C# and WCF service references.
[03:28:17] zombor_ (zombor_!~zombor_@kohana/developer/zombor) has quit (Remote host closed the connection)
[03:30:57] zombor_ (zombor_!~zombor_@kohana/developer/zombor) has joined #mythtv
[03:32:11] zombor_ (zombor_!~zombor_@kohana/developer/zombor) has quit (Remote host closed the connection)
[03:33:50] stichnot (stichnot!~chatzilla@mythtv/developer/stichnot) has joined #mythtv
[04:08:14] k-man: dblain, what is the practical upshot of that?
[04:10:12] jya_ (jya_!~jyavenard@ has joined #mythtv
[04:10:12] jya_ (jya_!~jyavenard@ has quit (Changing host)
[04:10:13] jya_ (jya_!~jyavenard@mythtv/developer/jya) has joined #mythtv
[04:12:54] clever: k-man: the client can create a class and then simply call methods on it, and the RPC is done automaticaly
[04:13:33] clever: WSDL defines the methods on the class, so it can automaticaly create a compatible class
[04:18:18] kormoc (kormoc!~kormoc@mythtv/developer/kormoc) has joined #mythtv
[04:19:41] jya (jya!~jyavenard@mythtv/developer/jya) has quit (Quit: jya)
[04:19:42] jya_ is now known as jya
[04:23:29] k-man: oh i see
[05:20:46] saintdev (saintdev!~saint@unaffiliated/saintdev) has joined #mythtv
[05:35:04] saintdev (saintdev!~saint@unaffiliated/saintdev) has quit (Remote host closed the connection)
[05:41:48] j-rod is now known as j-rod|afk
[06:13:06] OldEnK (OldEnK! has joined #mythtv
[06:17:24] cattelan is now known as cattelan_away
[06:19:28] cattelan_away is now known as cattelan_away_aw
[06:20:25] cattelan_away_aw is now known as cattelan_away
[06:20:33] cattelan_away is now known as cattelan_away_aw
[06:21:19] cattelan_away_aw is now known as cattelan_away
[06:22:01] cattelan_away is now known as cattelan_away_aw
[06:23:43] cattelan_away_aw is now known as cattelan_away
[06:24:59] noahric (noahric!~noahric@ has quit (Quit: noahric)
[06:26:35] OldEnK (OldEnK! has left #mythtv ("Leaving")
[06:29:26] saintdev (saintdev! has joined #mythtv
[06:29:26] saintdev (saintdev! has quit (Changing host)
[06:29:26] saintdev (saintdev!~saint@unaffiliated/saintdev) has joined #mythtv
[07:37:35] kormoc (kormoc!~kormoc@mythtv/developer/kormoc) has quit (Quit: kormoc)
[08:08:48] warped (warped!~piotro@ has joined #mythtv
[08:39:39] cecil (cecil! has joined #mythtv
[08:40:40] cesman (cesman!~cesman@pdpc/supporter/professional/cesman) has quit (Ping timeout: 276 seconds)
[09:37:01] stuartm: Captain_Murdoch: we can rip CDs to MP3 and Flac
[09:37:33] stuartm: oh, and ogg
[09:38:18] stuartm: obviously support for other formats would be possible if we leveraged libav*, but we don't do that yet
[09:39:30] warped (warped!~piotro@ has quit (Quit: warped)
[09:46:06] stuartm: danielk22, sphery: I'm a little suprised at the opengl performance, back when we very first got the opengl video rendering I remember using it instead of Xv and not having any trouble with it and that was on an old AMD Turion 1.6Ghz (single core) with NVidia 7200 (iirc)
[09:46:39] stuartm: so are we saying that performance has regressed? Or did I never try it with HD?
[09:50:58] stuartm: the hardware can handle it, at least in theory, so it's really just a question of implementation and drivers
[10:03:04] jya: stuartm: I remember early when markk added the OpenGL deinterlacers. The Yadiv 2X would take 100% CPU on my 3GHz Pentium 4.. completely unusable..
[10:09:28] stuartm: yeah, it might be the advanced deinterlacing stuff which killed the performance
[10:40:50] jya: I see when on my mac the screen saver kicks in and some fancy OpenGL stuff comes on screen, the CPU usually goes over 200%, and that's with a quite beefy GPU (ATI 6970)
[10:47:59] highzeth (highzeth! has quit (Ping timeout: 252 seconds)
[10:48:38] highzeth (highzeth! has joined #mythtv
[10:58:31] seeker (seeker!~seeker@unaffiliated/seeker) has joined #mythtv
[11:05:02] mike|3 (mike|3! has quit (Remote host closed the connection)
[11:05:50] mike|2 (mike|2! has joined #mythtv
[11:07:59] xavierh: I notice the list of theme to choose from, depends on what theme I am using. using Terra I have 12 themes, using Mythcenter, I only have 3 to choose ???
[11:08:43] xavierh: using a month old version from trunk
[11:08:46] stuarta: wonder if that's a 4:3 vs 16:9 thing
[11:08:59] stuarta: mythcentre is 4L:3 iirc
[11:09:37] xavierh: using mythcenter wide, and I can choose mythcenter mythcenter-wide and Terra
[11:11:45] xavierh: I there any tools to check the theme, mythcenter 4:3 is broken, I cannot go in the appearance menu
[11:12:43] xavierh: for mythcenter 4:3 brokenness, is my fault
[11:17:49] MythBuild: build #443 of master-osx-snow-leopard is complete: Failure [failed compile_2] Build details are at . . . d/builds/443 blamelist: Jean-Yves Avenard < >
[11:26:22] jya: damn
[11:33:21] stuarta: mysqlclient lib is in .osx-packager/build/lib/mysql
[11:33:59] stuarta: must another -L used in other parts
[11:34:02] stuarta: that's missing
[11:35:39] jya: yeah.. I hadn't noticed, because in my new script I have the -L$prefix/lib/mysql provided as LFLAGS
[11:35:50] jya: have to use regex then...
[11:36:00] ** jya Reading qmake regex doc **
[11:38:04] aloril (aloril! has quit (Ping timeout: 245 seconds)
[11:39:13] jya: what they describe there just doesn't work:
[11:48:47] jya: anyone expert enough in Qt on how to remove from a variable any such pattern /-arch [a-z0–0_]*/
[11:49:09] jya: the VAR ~= s/regex// doesn't work for me, and neither replace
[11:49:42] jya: so say VAR= "-Iblah -Lpath -arch i386 -arch x86_64" become "-Iblah -Lpath"
[11:50:02] aloril (aloril! has joined #mythtv
[11:50:16] stuarta: is there some arch based selector you can use?
[11:50:24] peitolm: jya, i think your regex is broken
[11:50:28] peitolm: (as an aside
[11:50:49] peitolm: you probably want a space on the end and to replace the * with a +
[11:51:22] jya: stuarta: the issue is that my mysql lib has been compiled in universal mode. so if it links my 32 bits only .o with a library, using -arch i386 -arch x86_64, it will error as it can't find the relevant 64 bits entries
[11:51:55] jya: peitolm: I actually have ( |$)
[11:51:59] jya: after
[11:52:10] jya: but it still doesn't work
[11:52:56] jya: peitolm: if I do: MYSQLIBS ~= s/arch/foo/
[11:53:01] jya: MYSQLIBS become
[11:53:19] jya: -F/Library/Frameworks -foo i386 -arch x86_64
[11:53:27] jya: you see, it has only replaced one occurence of it
[11:53:36] knightr (knightr!~knightr@mythtv/developer/knightr) has quit (Read error: Connection reset by peer)
[11:53:50] knightr (knightr!~knightr@mythtv/developer/knightr) has joined #mythtv
[11:53:52] stuarta: ahah!
[11:54:10] stuarta: s/arch/foo should only do 1 instance, s/arch/foo/g should do all
[11:54:30] peitolm: :)
[11:54:33] stuarta: in standard sed type regexp
[11:54:38] peitolm: and pcre
[11:56:12] jya: of course...
[11:56:30] stuarta: don't you just love regexp's :)
[11:56:43] jya: yet..
[11:57:21] jya: MYSQLIBS ~= s/arch x86_64//g
[11:57:39] jya: -F/Library/Frameworks -arch i386 -arch x86_64
[11:57:52] jya: as soon as I have a space, it doesn't work
[11:58:31] stuarta: try replacing the space with a dot '.'
[11:59:14] jya: nope :(
[11:59:18] jya: gosh, this is annoying
[12:01:58] jya: well, that will do
[12:01:59] jya: MYSQLIBS ~= s/-arch|i386|x86_64|ppc64|ppc7400//g
[12:02:15] jya: I can't have any different values anyway
[12:09:01] sphery: xavierh: the reason you see fewer themes in MythCenter is because some themes default to showing all themes--including remote themes--whereas others (MythCenter among them) show only locally-installed themes. On those themes, you need to hit MENU then select the option to show remote themes.
[12:09:31] mrand (mrand!~mrand@ubuntu/member/mrand) has quit (Ping timeout: 246 seconds)
[12:12:45] peitolm: that won't work surely
[12:13:27] peitolm: try /-arch\ \w+/
[12:13:48] peitolm: well it might work, but be overly greedy
[12:14:11] jya: peitolm: I had try that… if it see a "\ ", I get a warning unescaped \ are deprecated error
[12:15:04] jya: peitolm: but I'm fairly confident in the limited subset of cases I will encouter, my work around is sufficient
[12:15:22] jya: it seems that the ~= operator is designed to work on one word only at best
[12:16:19] jya: I know it will always be -arch blah
[12:16:33] jya: where blah is i386, x86_64, ppc64 or ppc7400
[12:17:02] jya: the day Apply supports a new processor , I'll hopefully remember to adjust it
[12:17:23] peitolm: but that's not what your regex is doing :)
[12:17:40] peitolm: well, it kind of is
[12:17:47] peitolm: but if it's working, *shrug*
[12:17:48] jya: it removes all -arch and all i386|x86_64|
[12:17:51] peitolm: untill it breaks
[12:18:07] jya: as it will always be one of the case I mentioned abouve
[12:18:14] jya: not pretty, but will do
[12:18:23] peitolm: i wonder
[12:18:36] peitolm: \s is a whitespace char isn't it?
[12:18:45] jya: hes
[12:18:46] jya: yes
[12:19:00] jya: I tried -arch\si386 too
[12:19:02] peitolm: /-arch\s\w+/ not work?
[12:19:22] jya: no, as I said: it gives me the error that unescaped \ are deprecated
[12:20:34] peitolm: /-arch[^a-zA-Z0–9_][a-zA-Z0–9_]/
[12:21:31] jya: I'm starting to be fairly certain that the scope for the string replacement is related to words only
[12:21:46] jya: so you can remove something like -Lpath
[12:21:59] jya: not "-prefix name"
[12:22:30] peitolm: :(
[12:22:43] peitolm: not really a regext then :)
[12:23:36] davide (davide! has quit (Remote host closed the connection)
[12:23:56] davide (davide! has joined #mythtv
[12:27:25] xavierh: sphery: what is the purpose of this? seems confusing for the user point of view.
[12:32:27] seeker (seeker!~seeker@unaffiliated/seeker) has quit (Quit: Colloquy for iPhone -
[12:33:22] jya: peitolm: Just tried your last regex, it does no change at all.. that confirms my suspicions really
[12:37:19] MythBuild: build #28 of 0.24-osx-snow-leopard is complete: Failure [failed compile] Build details are at . . . rd/builds/28 blamelist: Stuart Morgan < >
[12:38:19] MythBuild: build #297 of 0.24-linux-64bit is complete: Failure [failed compile core] Build details are at . . . t/builds/297 blamelist: Stuart Morgan < >
[12:41:34] MythBuild: build #122 of 0.24-freebsd-64bit is complete: Failure [failed compile core] Build details are at . . . t/builds/122 blamelist: Stuart Morgan < >
[12:44:10] stuartm: ah crap
[12:45:41] gregL (gregL! has quit (Read error: Connection reset by peer)
[12:46:49] MythBuild: build #275 of 0.24-linux-32bit is complete: Failure [failed compile core] Build details are at . . . t/builds/275 blamelist: Stuart Morgan < >
[12:47:07] mrand (mrand!~mrand@ubuntu/member/mrand) has joined #mythtv
[12:57:35] MythBuild: build #130 of 0.24-linux-ppc is complete: Failure [failed compile core] Build details are at . . . c/builds/130 blamelist: Stuart Morgan < >
[12:58:48] jya: damn… this Qt is sooooo buggy
[12:59:22] jya: in the .pro qmake file
[12:59:23] jya: I have:
[12:59:34] jya: QMAKE_FLAGS=-arch i386
[12:59:47] jya: I do QMAKE_LFLAGS += TMP
[12:59:56] jya: where TMP = -arch x86_6
[13:00:14] jya: resulting QMAKE_FLAGS becomes: -arch i386 x86_64
[13:00:27] jya: as it thinks there's already a -arch, it doesn't add it again
[13:01:32] peitolm: try it in quotes?
[13:01:56] jya: peitolm: there's no quote to speak of.. this is all the process of qmake
[13:02:53] gregL (gregL! has joined #mythtv
[13:05:01] davide (davide! has quit (Remote host closed the connection)
[13:05:25] davide (davide! has joined #mythtv
[13:05:52] jya: according to the doc += always add at the end of the list, *= only adds if it's not present.
[13:05:59] jya: looks like they got the doc wrong
[13:06:11] peitolm: i mean QMAKE_LGLAFS += TMP where TMP = "-arch x86_64"
[13:06:46] jya: I've never seen " " being used in qmakefiles
[13:06:53] jya: but that's worth a shot
[13:08:55] MythBuild: build #298 of 0.24-linux-64bit is complete: Success [build successful] Build details are at . . . t/builds/298
[13:16:25] MythBuild: build #276 of 0.24-linux-32bit is complete: Success [build successful] Build details are at . . . t/builds/276
[13:17:49] MythBuild: build #123 of 0.24-freebsd-64bit is complete: Success [build successful] Build details are at . . . t/builds/123
[13:29:26] jya: peitolm: you're right… it needs to be in quote to behave properly
[13:29:53] jya: the question being how to add quote to start with.. the arguments come from copying LDFLAGS which doesn't use quote
[13:30:11] stuarta: "$LDFLAGS"
[13:30:18] peitolm: "${var}"
[13:30:22] Jordack (Jordack! has joined #mythtv
[13:30:25] peitolm: maybe?
[13:30:36] ** stuarta likes peitolm's version better **
[13:30:50] peitolm: although mine might be a bash-sim, i find it far easier to read, and less likely to break
[13:31:10] peitolm: it also handles fun like ${foo}${bar}
[13:31:24] peitolm: where as $foo$bar could be missed
[13:31:50] jya: ohhhh.. that seems to work
[13:32:18] jya: echo "QMAKE_LFLAGS+=\"$LDFLAGS\"" >> $TMPMAK
[13:32:46] peitolm: ${} shouldn't need escaping if it works
[13:32:58] jya: I wonder if it's not the same thing with my regex
[13:33:11] peitolm: prossibly
[13:34:08] jya: if I did something like MYSQLFLAGS = "$$system(mysql_config --lib"
[13:34:13] jya: let's try that
[13:37:57] stuarta: stupid buildbot, it failed the git step, and now it's re-cloning the whole repo
[13:40:29] jya: peitolm: that works !!
[13:40:42] jya: MYSQLIBS = "$$system(mysql_config --libs)"
[13:40:42] jya: MYSQLIBS ~= s/-arch +[a-z0–9_]*[ $]//g
[13:41:05] jya: instead of MYSQLIBS = $$system(mysql_config --libs)
[13:41:52] peitolm: your regex still looks wrong to me, but if it works..
[13:42:07] jya: I can\t use \s
[13:42:13] jya: so " +" will do
[13:42:24] jya: the * should be +
[13:42:47] peitolm: but why anchor it?
[13:43:10] jya: interestingly, what you find in the generated Makefile is: -F/Library/Frameworks -L/Library/Frameworks -L/Users/jyavenard/Work/mythtv/.osx-packager/build/x86_64/lib -Wl,-syslibroot,/Applications/Custom/Development/ atforms/MacOSX.platform/Developer/SDKs/MacOSX10.6.sdk\ -mmacosx-version-min=10.5\ -L/Users/jyavenard/Work/mythtv/.osx-packager/build/lib\ -F/Library/Frameworks\ \
[13:43:10] jya: -L/Users/jyavenard/Work/mythtv/.osx-packager/build/lib/mysql\ -lmysqlclient\ -lz\ -lm -framework QtSql -framework QtCore -framework QtGui
[13:43:15] peitolm: the last [ $] is irrelevant
[13:43:17] jya: see all te \ for the space
[13:43:20] jya: you're right
[13:43:32] jya: it looks more fancy though
[13:43:39] stuarta: :)
[13:43:53] peitolm: is that seriously the generated make file?
[13:44:08] jya: i reckon, regex should always written so the person reading them need to think or google
[13:44:08] peitolm: where are the '\ ' coming from?
[13:44:26] jya: because I puse the "$$system… " instead
[13:44:31] ** peitolm bops jya with the bat of maintainability **
[13:44:56] stuarta: meh
[13:45:00] jya: as the string that is returned by system contains space, it escapes them all
[13:45:09] peitolm: o.k.
[13:45:17] jya: that's my guess anyway
[13:45:27] peitolm: if it works, it works
[13:45:36] stuarta: i was about to say that
[13:46:10] jya: how long have we spent on that regex thing ?
[13:46:19] jya: over an hour!
[13:46:38] stuarta: you love it really...
[13:46:43] jya: sure..
[13:47:02] jya: I compile a universal mythfrontend bundle now !
[13:47:06] peitolm: time spent finding a sucessful solution is rarely wasted if everone learns from it
[13:47:31] peitolm: woo
[13:47:33] stuarta: so far we've learnt. Qt can't make a make framework lib release for toffee
[13:47:47] peitolm: are your fixes going to be fed back into osx-packager?
[13:47:57] jya: stuarta: as I haven't managed to compile ffmpeg in universal mode , what I do is loop the build for all the architecture, loop through all the generated files and combine all the files together
[13:48:23] stuarta: peitolm: post 0.25 we will switch osx to using the framework based Qt, which is where this is all going
[13:48:25] jya: peitolm: no…. As I will be the only one to appreciate the changes, I keep them for myself
[13:48:41] peitolm: lol @jya
[13:49:01] peitolm: 0.25 is almost upon us isn't it?
[13:49:09] jya: I have already committed the changes to the code in mythtv so the new builder works
[13:49:45] jya: quite a few bug that were never noticed earlier because no one tried to compile a 64 bits version, or using dynamic libs rather than static ones
[13:49:56] peitolm: I guess my question is, if I fired up a osx builder now, would it work?
[13:50:01] jya: all those changes need to be backported to 0.24 too
[13:50:08] jya: right now
[13:50:09] jya: no
[13:50:19] jya: I broke the mythzmserver in the plugins
[13:50:24] peitolm: :)
[13:50:34] jya: which is what the regex is about
[13:50:40] stuarta: damn mysql lib paths...
[13:50:57] jya: problem is I've built mysql in universal mode now
[13:51:13] jya: so I can link it against 64 bits and 32 bits version of the frontend
[13:51:37] jya: but that mysql_config always try to link both architecture
[13:51:51] ** peitolm idly wonders how slim a network based usb recorder could be **
[13:52:13] jya: did you see that dual core ARM9 USB stick/
[13:52:17] jya: look pretty awesome
[13:52:24] jya: just a USB connector
[13:52:26] jya: HDMI out
[13:52:33] jya: that's it.. with BT connectivity
[13:53:23] jya: . . . preorder.ars
[13:53:43] ** jya need to look into this Mali chip **
[13:53:57] jya: and wifi!
[13:54:06] jya: $199.. ouch
[13:54:14] jya: lots of R&D I'm guessing to cover
[13:56:56] peitolm: ouch, that's 8x the price of the cheap r_pi
[13:58:56] peitolm: and just under 6x the cost of the Pi with ethernet
[13:59:43] stuartm: better specs though
[14:00:07] stuartm: wifi, better cpu, more ram and smaller
[14:00:17] peitolm: as long as it can drive a USB capture card and spit it out of ethernet, it will do for a recording node
[14:00:22] jya: it's great.. now you can have a full size PC, that you will use just to power the USB frontend
[14:02:39] jya: ohhh… actually, the fix for actually fix the compilation of mythzoneminder on my linux PC
[14:02:50] jya: I had never been able to get it to compile on that box..
[14:02:56] jya: it was just those damn quotes!
[14:03:04] stuartm: given the immediate success of the R pi, I'd expect to see the market flooded with similar designs
[14:03:22] peitolm: designs yes
[14:03:28] peitolm: but at that price point,.. dunno
[14:03:44] stuartm: right, the prices will have to be lower to sell
[14:03:47] peitolm: they are helped out by the fact one of the leaders of the project is reasonably high up in broadcom
[14:04:21] MythBuild: build #30 of 0.24-osx-snow-leopard is complete: Success [build successful] Build details are at . . . rd/builds/30
[14:04:58] jya: stuarta: what did you change to make it compile
[14:05:31] stuartm: in bulk though the component prices aren't very high, and Rpi was just talking about batches of 10,000 – bigger well funded companies could expect to turn out the product for less, it's all a question of whether they'd be happy with the small margins
[14:06:28] peitolm: *nods* no idea what the next batches will be size-wise, we now have some reasonably large companies in the mix
[14:08:49] stuartm: even if it turns out that my Raspberry Pi isn't as fun as I hope it will be, at just £30 (postage + VAT) I'm not going to suffer any buyer's remorse, unlike my first netbook which ultimately was a waste of money
[14:09:42] peitolm: ditto
[14:10:21] peitolm: a friend of mine bought a hauppauge net box to use as a front end, then sony fixed/enabled book marks on the ps3 dnla client so it's redundant
[14:10:37] peitolm: at least a pi might be able to do cutting/trimming
[14:10:51] skd5aner (skd5aner! has joined #mythtv
[14:18:26] MythBuild: build #3241 of master-linux-64bit is complete: Failure [failed compile plugins] Build details are at . . . /builds/3241 blamelist: Jean-Yves Avenard < >
[14:20:39] MythBuild: build #131 of 0.24-linux-ppc is complete: Success [build successful] Build details are at . . . c/builds/131
[14:21:10] MythBuild: build #2024 of master-freebsd-64bit is complete: Failure [failed compile plugins] Build details are at . . . /builds/2024 blamelist: Jean-Yves Avenard < >
[14:21:18] jya: rhaaaa
[14:21:58] stuarta: jya: that's 0.24 which stuartm broken
[14:21:59] jya: so here it can't be in qupte !
[14:22:05] stuarta: he fixed it
[14:22:18] jya: master-linux-64bits
[14:22:35] jya: that looks definitely related to the change I made
[14:22:36] MythBuild: build #2986 of master-linux-32bit is complete: Failure [failed compile plugins] Build details are at . . . /builds/2986 blamelist: Jean-Yves Avenard < >
[14:22:46] jya: "unrecognized option '-Bsymbolic-functions -rdynamic -L/usr/lib/mysql -lmysqlclient'"
[14:22:48] stuarta: certainly is, by the look of it
[14:22:55] jya: it doesn't like being in quote there
[14:23:03] jya: weird, cause it works all fine for me
[14:23:17] jya: I compiled on linux 64 bits, and mac 32/64 before committing
[14:23:52] jya: and on my PC mysql_client --libs returns "-Wl,-Bsymbolic-functions -rdynamic -L/usr/lib/mysql -lmysqlclient"
[14:26:56] stuartm: wagnerrp: are we going to switch to the tmdb v3 API after 0.25 is out?
[14:29:18] MythBuild: build #1756 of master-debian-stable-64bit is complete: Failure [failed compile plugins] Build details are at . . . /builds/1756 blamelist: Jean-Yves Avenard < >
[14:38:38] stuarta: jya, the '\ ' need to be turned back into ' '
[14:38:46] jya: I know!
[14:38:51] jya: but how do you do that ?
[14:38:52] stuarta: :)
[14:39:02] jya: I tried eval, quote etc..
[14:39:06] stuarta: s/\ / /
[14:39:09] stuarta: s/\ / /g
[14:39:44] jya: I end up with LIBS = quote(-L/blah\
[14:40:12] jya: I have an easier , less elegant solution
[14:40:18] jya: if mac -> use with quotes
[14:40:22] jya: if not, use without
[14:40:27] jya: like it was before for linux
[14:41:27] dekarl-too (dekarl-too!51c8c614@gateway/web/freenode/ip. has joined #mythtv
[14:42:15] zombor (zombor!~zombor_@kohana/developer/zombor) has joined #mythtv
[14:42:23] dekarl-too: stuartm: <- the icon search for channels without space in their name seems to be broken
[14:43:10] jya: stuarta: replacing the \ do not work, and I dont know how to escape it
[14:44:59] j-rod|afk is now known as j-rod
[14:49:24] Captain_Murdoch: xavierh, I see all themes using MythCenter here. there is no code that I know of that makes the # of themes available dependent on the theme currently in use
[14:50:19] Captain_Murdoch: xavierh, there are currently no (released) tools to check theme brokenness. :| I started on one a while back but didn't get too far.
[14:50:37] Captain_Murdoch: it sounds like you're only seeing the local themes, not the downloadable ones.
[14:56:48] Captain_Murdoch: sphery, do you know of any code that makes the theme chooser act like that? I can't recall any code that allows a theme to specify whether to show downloadable themes or not.
[14:58:37] dekarl-too (dekarl-too!51c8c614@gateway/web/freenode/ip. has left #mythtv ()
[14:59:31] seeker (seeker!~seeker@unaffiliated/seeker) has joined #mythtv
[15:01:01] jya: if any Qt guru wants to answer my questions:
[15:01:01] jya: . . . 6#post215226
[15:10:04] seeker_ (seeker_!~seeker@unaffiliated/seeker) has joined #mythtv
[15:13:29] seeker (seeker!~seeker@unaffiliated/seeker) has quit (Ping timeout: 245 seconds)
[15:13:31] seeker_ is now known as seeker
[15:15:11] jya: stuarta: this does the trick… LIBS += $$system(echo $$MYSQLIBS) however, I'm uncertain on how portable that is
[15:16:18] stuarta: hmmm, most likely to fail is windows
[15:17:17] jya: having said that, I doubt the $$system(mysql_config --libs) would work!
[15:18:39] jya: ah, there's $$sprintf($$MYSQLIBS)
[15:18:56] jya: but that would fail if any of the path or libs contain %1 to %9
[15:19:03] jya: most likely, very unlikely
[15:22:22] cattelan_away is now known as cattelan
[15:24:23] stuarta: ah it won't matter as the windows buildbot has zoneminder disabled by default
[15:33:23] wagnerrp: stuartm: no, waiting until the tmdbv3 API is actually released
[15:33:29] MythBuild: build #2003 of master-linux-ppc is complete: Failure [failed compile plugins] Build details are at . . . /builds/2003 blamelist: Jean-Yves Avenard < >
[15:35:23] wagnerrp: stuartm: the grabber is more or less functional right now, but there are a number of things missing or broken in the API still
[15:35:48] wagnerrp: for one, the search mechanism does not page at all like it's supposed to
[15:36:36] ** jya waive proudly the flag on who broke the most builds at once **
[15:38:46] ** stuarta awards jya a badge **
[15:38:54] stuarta: we've all done it
[15:40:21] jya: are all the binaries generated when compiling myth found in $prefix/bin ?
[15:41:30] seeker (seeker!~seeker@unaffiliated/seeker) has quit (Quit: Colloquy for iPhone -
[15:41:50] stuarta: yes i believe so
[15:42:07] stuarta: but the plugins live in a different dir
[15:42:14] stoffel (stoffel! has joined #mythtv
[15:42:34] jya: they seem to be in lib/mythtv/filers or lib/mythtv/plugins
[15:42:41] MythBuild: build #3242 of master-linux-64bit is complete: Success [build successful] Build details are at . . . /builds/3242
[15:43:00] stuarta: \o/
[15:43:08] jya: yeahh !
[15:46:25] MythBuild: build #2025 of master-freebsd-64bit is complete: Success [build successful] Build details are at . . . /builds/2025
[15:47:34] sphery: Captain_Murdoch: No. I've never looked at the code--only observed the behavior.
[15:48:34] MythBuild: build #2987 of master-linux-32bit is complete: Success [build successful] Build details are at . . . /builds/2987
[15:51:54] MythBuild: build #2004 of master-linux-ppc is complete: Success [build successful] Build details are at . . . /builds/2004
[15:57:06] MythBuild: build #1757 of master-debian-stable-64bit is complete: Success [build successful] Build details are at . . . /builds/1757
[15:59:27] stuarta: jya: Unknown replace function: sptrinf <- osx build
[15:59:33] seeker (seeker!~seeker@unaffiliated/seeker) has joined #mythtv
[15:59:39] jya: it does?
[15:59:49] jya: you're running Qt 4.7 right?
[16:00:00] stuarta: 4.6
[16:00:04] jya: ah
[16:00:08] jya: I run 4.7.4
[16:00:24] jya: I'll look into tomorrow, I've had way too litlle sleep in the past few days
[16:00:32] stuarta: heh
[16:00:42] jya: hopefully, someone will have answer my questions on the qt forum
[16:01:47] jya: ah damn
[16:01:51] jya: it's a spelling mistake
[16:02:38] stuarta: that's what i though
[16:02:42] jya: weird, it compiles on my mac
[16:02:43] stuarta: sprintf is what you wanted?
[16:02:46] jya: yes
[16:03:01] ** stuarta declare jya's mac weird **
[16:03:50] jya: way too late
[16:05:13] peitolm: :)
[16:05:28] peitolm: does this mean i need to get the lion slave up and ready?
[16:06:05] stuarta: we can certainly start hooking it up to a test master
[16:06:19] jpabq_ (jpabq_!~jpabq@mythtv/developer/jpabq) has joined #mythtv
[16:06:43] peitolm: cool
[16:07:51] stuartm: wagnerrp: ok thanks, I want to get the certification statetype up and running for 0.26 but that depends on supporting the locale stuff for tmdb
[16:08:22] wagnerrp: right now, i basically do nothing with the locale in the grabber
[16:08:31] wagnerrp: its something i still need to add
[16:08:41] wagnerrp: plus i need to add proper local fallthrough
[16:09:01] wagnerrp: as it is, the tmdbv3 API does NOT fallthrough to english values, it just leaves them empty
[16:09:22] stuartm: dekarl: odd, I don't remember a change which would have broken that
[16:10:36] stichnot (stichnot!~chatzilla@mythtv/developer/stichnot) has quit (Ping timeout: 260 seconds)
[16:10:55] wagnerrp: stuartm: plus, for proper certification, were going to have to start passing locale
[16:10:59] stuartm: wagnerrp: ok, I might modify the existing scripts to support it in the v2 API then,
[16:11:01] wagnerrp: right now, we just do language
[16:11:41] rsiebert_ (rsiebert_! has joined #mythtv
[16:11:51] wagnerrp: stuartm: the v2 API is considered dead by travis (tmdb dev) and no longer receiving any updates
[16:11:57] stuartm: wagnerrp: I know, I did already add it for tmdb, but iamlindoro wanted it reverted until all tv/movie grabbers were updated to accept it (even if they ignored it)
[16:12:12] wagnerrp: chances are very good the v3 will be out by the time 0.26 is released
[16:12:33] stuartm: wagnerrp: v2 supports locale, it's just we have never used it
[16:12:49] wagnerrp: if you want something to play around with now, it wouldn't be difficult to make the necessary additions to the v3 grabber
[16:15:01] rsiebert (rsiebert! has quit (Ping timeout: 276 seconds)
[16:17:02] xavierh: Captain_Murdoch: I guess it is not that easy to check brokenness of a theme
[16:23:40] MythBuild: build #448 of master-osx-snow-leopard is complete: Success [build successful] Build details are at . . . d/builds/448
[16:23:47] stuarta: \o/
[16:25:41] seeker: wagnerrp: stuartm: there is a patch around to update all sources to accept the locale arguments (even if they don't do anything with it)
[16:25:57] seeker: I attached it to the ticket I made about it some
[16:26:03] seeker: Time before feature freeze
[16:26:18] warped (warped!~piotro@ has joined #mythtv
[16:26:40] seeker: #10240
[16:26:48] wagnerrp: what was the final decision on how to handle that?
[16:26:56] wagnerrp: combined with language? separate argument?
[16:27:08] seeker: wagnerrp: Separate
[16:29:04] seeker: My reasoning was that although tmdb uses en-GB format, other Apis may not, so better to keep them seperate (as they are already separate in the DB) and combine them in the script if needed rather than screwing around with string parsing
[16:30:46] seeker (seeker!~seeker@unaffiliated/seeker) has quit (Read error: Connection reset by peer)
[16:31:56] wagnerrp: tmdb does not use a combined format
[16:32:13] wagnerrp: it only takes the 2 character ISO 639 format
[16:32:26] wagnerrp: its up to the script to handle locale filtering
[16:32:57] wagnerrp: assuming that isnt instead handled by mythtv itself
[16:40:47] seeker (seeker!~seeker@unaffiliated/seeker) has joined #mythtv
[16:41:55] stuartm: wagnerrp: tmdb v2 does take a combined format
[16:42:08] Seeker`: eugh, connection died
[16:42:19] stuartm: lang-country, ISO3166 and ISO639
[16:42:54] stuartm:
[16:43:02] wagnerrp: ah, so it does
[16:43:16] wagnerrp: odd that all their use examples just show the 639 language code
[16:43:57] stuartm: wagnerrp: early version of v2 didn't support it, I guess their examples weren't updated
[16:44:52] stuartm: it took a bit of lobbying by myself and others before things like localized certification and titles were added
[16:45:10] wagnerrp: well v3 is explicitly 639 only (for searches)
[16:45:33] wagnerrp: however many of the data points provide 3166 region codes for client-side filtering
[16:46:16] Seeker`: wagnerrp: if the script has to do filtering you probably want the . seperate too
[16:46:39] Seeker`: or youll just have to pull them apart again
[16:47:11] wagnerrp: production countries, filming locations, alternate titles, release information (date and certification)
[16:47:31] wagnerrp: the '.'?
[16:47:39] Seeker`: them
[16:47:49] Seeker`: on an iphone :P
[16:47:55] wagnerrp: ah
[16:54:53] stuartm: alternate titles and release info are the ones we're interested in, although I'm not sure how we sort the correct alternate title from the list when it encourages you to add unofficial titles as well
[16:55:41] stuartm: and there is no mechanism for adding the official localised title
[16:55:54] wagnerrp: stuartm: from that list, production companies and filming locations would obviously all be passed
[16:56:16] wagnerrp: release information isnt harming anything by passing everything, since that gets populated into a table, and the correct one is chosen at time of playback
[16:56:38] wagnerrp: which leaves alternate titles, would could similarly be put into a table, and filtered for display
[16:56:53] wagnerrp: allowing you to change locales, or remove the filter, if so desired
[17:03:40] jmusits (jmusits! has joined #mythtv
[17:03:41] seeker (seeker!~seeker@unaffiliated/seeker) has quit (Read error: Connection reset by peer)
[17:04:12] jmusits (jmusits! has quit (Client Quit)
[17:04:14] stichnot (stichnot!~chatzilla@ has joined #mythtv
[17:04:14] stichnot (stichnot!~chatzilla@ has quit (Changing host)
[17:04:14] stichnot (stichnot!~chatzilla@mythtv/developer/stichnot) has joined #mythtv
[17:17:39] stoffel (stoffel! has quit (Ping timeout: 245 seconds)
[17:24:16] stoffel (stoffel! has joined #mythtv
[17:27:27] jya: stuarta: I think I'm done ...
[17:27:47] ** stuarta prescribes a weekend of sleep **
[17:28:16] jya: what am I going to do next now ? :(
[17:28:44] stuarta: beer?
[17:29:36] jya: hum… under a blanket it's going to be hard
[17:29:38] mrand: mythtv is done?
[17:29:50] mrand: :-)
[17:30:00] jya: mrand: how did you guess?
[17:30:22] stuarta: jya: did you commit it somewhere?
[17:30:30] jya: no...
[17:30:46] jya: I hate committing stuff right after I'm done… cause I always screw up
[17:31:10] stuarta: well have a good sleep then
[17:31:31] jya: I need to backport the changes on fixes/0,24
[17:31:48] jya: so it can compile in 64 bits too
[17:32:30] jya: actually, mythmusic didn't compile because it's missing vorbis and flac… oh well
[17:32:37] stuarta: cool, right, i'm off home, nn
[17:32:43] jya: time for bed.. the hardest is done..
[17:33:00] jya: I can copy the diff somewhere for you
[17:33:02] jya: hold a sec
[17:33:51] jya: actually, I'd appreciate if you could test the compilation with the updated osx-bundler
[17:33:59] jya: but with the old
[17:34:05] jya: it should be 100% compatible
[17:34:10] jya: but I may have missed things
[17:36:20] jya:
[17:36:37] jya: do ./ -man
[17:37:03] jya: I've put a short doc on how to call it, according to the type of Qt SDK you've installed (Qt SDK or libraries DMG)
[17:38:27] jya: I need to resurect my powerPC powerbook and see how it does
[17:47:20] stoffel (stoffel! has quit (Ping timeout: 260 seconds)
[17:53:02] wagnerrp: stuartm: looks like the pages has been silently fixed
[17:53:07] wagnerrp: *pager
[18:04:20] joe_ (joe_!~jmk@ has quit (Ping timeout: 252 seconds)
[18:04:40] davide (davide! has quit (Remote host closed the connection)
[18:05:02] davide (davide! has joined #mythtv
[18:26:23] stoffel (stoffel! has joined #mythtv
[18:32:40] joe_ (joe_!~bob@ has joined #mythtv
[18:38:53] kormoc (kormoc!~kormoc@mythtv/developer/kormoc) has joined #mythtv
[18:45:34] warped (warped!~piotro@ has quit (Quit: warped)
[18:52:55] noahric (noahric!~noahric@ has joined #mythtv
[19:29:51] stuartm: jpabq: Junction is proving itself to be a real pain, if you start displaying accented characters, especially upper case e.g. À then the string is cropped, and it really doesn't seem to matter what the height of the textarea is
[19:41:58] jpabq_: stuartm: that would imply that even FontMetrics does not know the "true" height of that character.
[19:42:33] stuartm: :/
[19:47:47] jpabq_: stuartm:
[19:48:32] jpabq_: Pure hack, but it would interesting to see if it helps.
[19:52:19] jpabq_: stuartm: the problem with doing a hack like that, is it breaks vertical scrolling. We could just yank vertical scrolling out of the tool box, and instead use a character height that is twice what QFontMetrics says it needs.
[19:54:07] jpabq_: Or, I guess we could change the code such that only if vertical scrolling is not being used, pad the ascent/descent with a bit extra. That is probably the cleanest "fix".
[19:56:51] warped (warped!~piotro@ has joined #mythtv
[20:04:13] stuartm: the hack fails spectacularly with no text being rendered at all
[20:04:36] stuartm: and lots of 'Invalid Canvas' errors
[20:06:13] stuartm: doesn't Junction have a negative ascent? so lineheight + ascent would cancel each other out
[20:13:38] stoffel (stoffel! has quit (Remote host closed the connection)
[20:27:50] jpabq: stuartm: I have an idea, but it will have to wait until Ivey home, to work it out
[20:28:09] jpabq: Get home
[20:31:29] cecil is now known as cesman
[20:31:47] cesman (cesman! has quit (Changing host)
[20:31:47] cesman (cesman!~cesman@pdpc/supporter/professional/cesman) has joined #mythtv
[20:52:02] jpabq_: stuartm: you are correct. That should have been a subtraction instead of an addition.
[20:52:20] jpabq_: Unfortunately, thinking more about it, my idea will not work.
[21:11:38] Beirdo: stuartm: for #9446, I'll take another look, but last I looked it was one of those "fix it for this client, break it for this other one" issue
[21:11:40] stichnot (stichnot!~chatzilla@mythtv/developer/stichnot) has quit (Ping timeout: 272 seconds)
[21:11:58] Beirdo: can't hurt to take another look though
[21:12:32] Beirdo: in the middle of un-borking mythtranscode (fifo mode) right now
[21:12:53] Beirdo: then we'll have to retest the other modes to make sure that they still work right
[21:13:37] Beirdo: in fifo mode, audio and video got consumed differently after the HLS patches, totally messing up sync :)
[21:13:53] Beirdo: I have it reworked so it SHOULD work for all of the cases, but still testing
[21:15:11] Beirdo: this should close a few nuvexport-related reports, and possibly others too
[21:16:57] stuartm: Beirdo: if there's no way of fixing it for the Sony client without breaking others, or if it's one of those times when we're following the spec and the client is at fault then probably best to close as Won't Fix?
[21:18:32] stichnot (stichnot!~chatzilla@mythtv/developer/stichnot) has joined #mythtv
[21:18:32] Beirdo: yeah, but not sure that we are following spec :)
[21:18:44] Beirdo: I'll try to take a look at it all tonight
[21:19:08] wagnerrp: Beirdo: do you know anything about git subtree?
[21:19:13] Beirdo: thanks for reminding me. Doing the occasional re-triage sweep is useful
[21:19:16] stuartm: well following the spec is always the best approach, if that ends up breaking other non-compliant clients that's not our fault
[21:19:19] Beirdo: wagnerrp: haven't used it
[21:19:40] Beirdo: well, the problem is, we don't HAVE the spec, it's $500 to just evaluate it
[21:19:55] Beirdo: and $10k a year for DLNA membership that allows you to implement it
[21:19:59] stuartm: there are no leaked copies floating around?
[21:20:06] Beirdo: I'm sure there are
[21:20:35] Beirdo: but that puts us on legal thin ice should we actually use it
[21:20:38] stuartm: fwiw, I thought it had been ruled illegal to protect a protocol in that way?
[21:21:06] Beirdo: I hadn't heard. Maybe it is in the civilized world of Europe :)
[21:21:33] Beirdo: definitely worth a revisit though, it's been a while since I looked it
[21:22:13] Beirdo: doing a re-triage at this point is a good thing. Thanks for going through the mess again.
[21:45:06] xavierh_ (xavierh_! has joined #mythtv
[21:45:51] OldEnK (OldEnK! has joined #mythtv
[21:47:24] Jordack (Jordack! has quit ()
[21:59:33] sphery: wagnerrp: I wonder if the "parental level doesn't work" tickets may be due to users enabling file browse mode? (Some in the -users thread said it works fine for them, others said it didn't, so...)
[22:00:12] joki- (joki-! has joined #mythtv
[22:00:18] joki (joki! has quit (Ping timeout: 260 seconds)
[22:00:18] joki- is now known as joki
[22:00:26] wagnerrp: sphery: i wouldnt be surprised
[22:00:28] sphery: (i.e. file browse mode without enable metadata in file browse mode)
[22:01:56] Beirdo: OK, can we get some concerted effort to knock down the new warnings?
[22:02:27] Beirdo: I see 2 unused variables in core build... and a constructor initialization issue
[22:02:58] Beirdo: and another 3 unused in plugins (2 vars, one func)
[22:03:27] Beirdo: I'd be happy to take care of them, but if they are in your code, please feel free to do it :)
[22:04:33] Beirdo: be good to have them gone before release :)
[22:15:08] johntramp (johntramp! has joined #mythtv
[22:15:38] johntramp (johntramp! has left #mythtv ("WeeChat 0.3.2")
[22:23:49] EagleIJoe (EagleIJoe! has joined #mythtv
[22:29:13] mrand (mrand!~mrand@ubuntu/member/mrand) has quit (Ping timeout: 245 seconds)
[22:31:58] mrand (mrand!~mrand@ubuntu/member/mrand) has joined #mythtv
[22:42:50] stuartm: stichnot: thanks, missed that :)
[22:43:28] stichnot: oh, I thought that was one of the constructor warnings you wrote about...
[22:48:40] warped (warped!~piotro@ has quit (Quit: warped)
[22:50:18] Beirdo: it was ;)
[22:52:24] j-rod is now known as j-rod|afk
[22:57:49] stichnot: oops, I got stuartm and Beirdo confused... how'd that happen?
[22:58:36] Beirdo: no clue :)
[22:59:37] jpabq_: stuartm: I just fixed a typo in MythUIText that could result in an incorrect canvas size. Can you test and make sure that was not your whole problem?
[23:00:29] zombor (zombor!~zombor_@kohana/developer/zombor) has quit (Remote host closed the connection)
[23:01:42] stuartm: jpabq_: sure
[23:01:56] stuartm: jpabq_: is that with, or without the earlier patch?
[23:02:06] jpabq_: Without
[23:07:06] stuartm: that's fixed it
[23:09:34] Beirdo: YAY
[23:10:06] jpabq_: stuartm: I am VERY glad. The idea of putting a hack in was not appealing.
[23:10:07] Beirdo: this fixed mythtranscode for fifo mode (well, I haven't tried with cutlists yet, that will be tonight)
[23:10:29] stuartm: jpabq_: should it be ascent*2 or ascent+decent?
[23:10:31] Beirdo: OK, I'll work with Captain_Murdoch to make sure I didn't break the other modes in the process :)
[23:13:30] jpabq_: stuartm: The viewport (drawRect) is in a fixed position. The canvas can move around (behind the viewport) and that is how scrolling is achieved. Because of that mechanism, the canvas actually needs move "down", but then still needs room to show the area — so it has to grow by 2*ascent, there.
[23:15:26] stuartm: ah, ok – at 11pm that only half makes sense to me, but it sounds convincing ;)
[23:16:00] jpabq_: stuartm: actually, instead of doing a setY that probably should be a moveY, then it would not need to be *2
[23:18:40] warped (warped!~piotro@ has joined #mythtv
[23:19:29] zombor (zombor!~zombor_@kohana/developer/zombor) has joined #mythtv
[23:20:21] stuartm: I'm surprised that my scrolling technique is still alive, I didn't have any clue how to do scrolling text 'properly' but idea of moving the text bitmap relative to the draw rect come to me and kinda worked, I just never spent any time on ironing out the bugs
[23:20:31] stuartm: s/come/came/
[23:23:55] jpabq_: stuartm: Yup. It is not logically that hard, but it is easy to get confused by the three "areas" interacting with each other.
[23:26:57] EagleIJoe (EagleIJoe! has quit (Quit: EagleIJoe)
[23:38:05] noahric (noahric!~noahric@ has quit (Quit: noahric)
[23:51:35] stuartm: aye, it takes a bit of mental juggling to keep it straight ;)
[23:56:23] stuartm: Beirdo: one of the things I'm talking about with user experience can be nicely illustrated by the recent livetv work, that's not stuff dealing with bling or the UI but lower level code, it hasn't changed what the user sees but it has improved the behaviour and removed one significant source of frustration
[23:59:57] wookey_ (wookey_! has joined #mythtv

IRC Logs collected by BeirdoBot.
Please use the above link to report any bugs.