:: #mythtv

Daily chat history

Current users (71):

aloril, amessina, andreaz, Anssi, arescorpio, caelor, Captain_Murdoch2, Chutt, clever, coling, Cougar, CrypticSquared, dblain, dekarl, dreamcat4, eee-blt, ElmerFudd, fetzerch, FrankD_Work, Gibby, gregL, GreyFoxx, J-e-f-f-A, jams_, jarle, jarryd, jheizer_, joki, jpharvey, jwhite, jya, kc, knightr, kormoc, kurre2, kwmonroe, MythBuild, MythLogBot, nephyrin, nyloc, peper03, poptix, purserj, rhpot1991, robink, rsiebert_, ryan_turner|MTW, Seeker`, seld, Sharky112065, sheedy-away, skd5aner, sl1ce, sphery, sraue, stuartm, sunkan, superm1, taylorr, tgm4883, toddc, toeb, tonsofpcs, tris, unforgiven512, wagnerrp, wseltzer, XDS2010_, xris, zentec, _charly_
Thursday, July 24th, 2014, 00:41 UTC
[00:41:41] wagnerrp: jya, dekarl: the "accepts" parameter allows a grabber to specify that it can use inetrefs from another grabber
[00:42:06] wagnerrp: so could say it accepts inetrefs prepended by, and automatically upgrade them when called
[00:51:17] sheedy-away is now known as sheedy
[01:08:05] fetzerch (fetzerch!~quassel@unaffiliated/fetzerch) has quit (Read error: Connection reset by peer)
[01:08:07] joki (joki! has quit (Ping timeout: 256 seconds)
[01:08:37] joki (joki! has joined #mythtv
[01:13:39] fetzerch (fetzerch!~quassel@unaffiliated/fetzerch) has joined #mythtv
[01:25:53] J-e-f-f-A (J-e-f-f-A!~J-e-f-f-A@unaffiliated/j-e-f-f-a) has quit (Quit: Oh No!!!! ;-))
[01:46:50] J-e-f-f-A (J-e-f-f-A!~J-e-f-f-A@unaffiliated/j-e-f-f-a) has joined #mythtv
[02:57:55] sheedy is now known as sheedy-away
[02:58:06] mad_enz (mad_enz! has joined #mythtv
[03:01:12] peper03 (peper03!~peper03@mythtv/developer/peper03) has quit (Ping timeout: 260 seconds)
[03:01:48] peper03 (peper03!~peper03@mythtv/developer/peper03) has joined #mythtv
[03:18:18] MythBuild: build #1952 of master-ubuntu-12_04-lts-64bit is complete: Failure [4failed compile core] Build details are at . . . /builds/1952 blamelist: David Blain < >
[03:19:16] MythBuild: build #5160 of master-freebsd-64bit is complete: Failure [4failed compile core] Build details are at . . . /builds/5160 blamelist: David Blain < >
[03:20:23] cappicard (cappicard! has joined #mythtv
[03:27:23] MythBuild: build #1054 of master-fedora-32bit is complete: Failure [4failed compile core] Build details are at . . . /builds/1054 blamelist: David Blain < >
[03:27:48] MythBuild: build #581 of master-f20–64bit is complete: Failure [4failed compile core] Build details are at . . . t/builds/581 blamelist: David Blain < >
[03:29:46] MythBuild: build #2259 of master-debian-wheezy-64bit is complete: Failure [4failed compile core] Build details are at . . . /builds/2259 blamelist: David Blain < >
[03:30:47] MythBuild: build #1149 of master-f19–64bit is complete: Failure [4failed compile core] Build details are at . . . /builds/1149 blamelist: David Blain < >
[03:32:27] MythBuild: build #455 of master-f20-qt5–64bit is complete: Failure [4failed compile core] Build details are at . . . t/builds/455 blamelist: David Blain < >
[03:35:14] MythBuild: build #1911 of master-ubuntu-current-64bit is complete: Failure [4failed compile core] Build details are at . . . /builds/1911 blamelist: David Blain < >
[03:39:26] dblain: :(
[03:58:22] fetzerch (fetzerch!~quassel@unaffiliated/fetzerch) has quit (Ping timeout: 260 seconds)
[03:59:34] fetzerch (fetzerch!~quassel@unaffiliated/fetzerch) has joined #mythtv
[04:06:45] MythBuild: build #1953 of master-ubuntu-12_04-lts-64bit is complete: Success [3build successful] Build details are at . . . /builds/1953
[04:07:37] MythBuild: build #2260 of master-debian-wheezy-64bit is complete: Success [3build successful] Build details are at . . . /builds/2260
[04:10:26] MythBuild: build #1912 of master-ubuntu-current-64bit is complete: Success [3build successful] Build details are at . . . /builds/1912
[04:13:16] MythBuild: build #5161 of master-freebsd-64bit is complete: Success [3build successful] Build details are at . . . /builds/5161
[04:19:00] MythBuild: build #582 of master-f20–64bit is complete: Success [3build successful] Build details are at . . . t/builds/582
[04:19:33] MythBuild: build #1055 of master-fedora-32bit is complete: Success [3build successful] Build details are at . . . /builds/1055
[04:20:49] MythBuild: build #1150 of master-f19–64bit is complete: Success [3build successful] Build details are at . . . /builds/1150
[04:22:10] skd5aner (skd5aner! has quit (Ping timeout: 260 seconds)
[04:23:43] MythBuild: build #456 of master-f20-qt5–64bit is complete: Success [3build successful] Build details are at . . . t/builds/456
[04:26:22] mad_enz (mad_enz! has quit (Quit: Leaving)
[04:31:32] cappicard (cappicard! has quit (Remote host closed the connection)
[04:37:08] arescorpio (arescorpio! has joined #mythtv
[04:53:09] arescorpio (arescorpio! has quit (Excess Flood)
[05:18:27] skd5aner (skd5aner! has joined #mythtv
[05:56:33] SteveGoodey (SteveGoodey! has joined #mythtv
[05:56:54] SteveGoodey (SteveGoodey! has quit (Remote host closed the connection)
[05:59:55] SteveGoodey (SteveGoodey! has joined #mythtv
[06:03:32] SteveGoodey (SteveGoodey! has quit (Client Quit)
[06:13:02] toeb (toeb! has joined #mythtv
[06:25:01] Tobbe5178 (Tobbe5178! has joined #mythtv
[06:25:55] terravires (terravires! has joined #mythtv
[06:26:12] terravires (terravires! has left #mythtv ()
[06:26:59] CrypticSquared (CrypticSquared!~CrypticSq@unaffiliated/crypticsquared) has quit (Quit: Leaving...)
[06:37:44] CrypticSquared (CrypticSquared!~CrypticSq@unaffiliated/crypticsquared) has joined #mythtv
[07:03:56] stichnot (stichnot!stichnot@mythtv/developer/stichnot) has joined #mythtv
[07:10:27] superm1 (superm1!uid4318@ubuntu/member/superm1) has quit (Quit: Connection closed for inactivity)
[07:19:48] dreamcat4: can someone please look at this bug? --->
[07:21:06] dreamcat4: it is mythweb. i have included necessary steps to reproduce the issue.
[07:49:20] stuarta: lemme see if i've modified my nginx config since i committed the sample
[07:50:06] andreaz (andreaz! has joined #mythtv
[07:50:12] stuarta: damn where did i put the sample config
[07:52:10] Chutt_ (Chutt_!~ijr@2605:a000:1208:c08c:c457:ef1d:60fd:2177) has joined #mythtv
[07:53:56] Chutt (Chutt!~ijr@2605:a000:1208:c08c:c457:ef1d:60fd:2177) has quit (Ping timeout: 240 seconds)
[07:54:45] stuarta: dreamcat4: do you have selinux enabled?
[07:55:48] dreamcat4: stuarta: i'm on FreeBSD. it doesn't seem to be anything related to nginx (apache users can get this same bug)
[07:56:05] stuarta: weird
[07:56:10] stuarta: using nginx?
[07:56:32] dreamcat4: stuarta: the real issue is in the very last comments, about the database schema, "case 3"
[07:57:06] dreamcat4: stuarta: i use nginx because i find it is much easier to configure php on nginx than apache
[07:57:33] stuarta: i use it because it's more efficient. i wrote the original sample nginx config
[07:58:15] dreamcat4: stuarta: ah right man, thank you!
[07:58:17] stuarta: have you upgraded recently?
[07:58:34] dreamcat4: stuarta: it is a band new (fresh) install. 1st timer.
[07:58:40] stuarta: k
[07:59:19] dreamcat4: i can pastie up all my steps, including what i did for creating the MYSQL database
[07:59:33] dreamcat4: it'll take a couple of minutes
[07:59:47] stuarta: i'm reading all the available data atm
[08:01:20] stuarta: does your phpinfo show the mysql modules loaded and working
[08:01:49] stuarta: i ask because it's failing on a DB query
[08:03:34] JohnBergqvist (JohnBergqvist!~JohnBergq@ has joined #mythtv
[08:04:06] ** stuarta fetches cuppa **
[08:04:47] dreamcat4: stuarta: that could be "no"
[08:06:30] dreamcat4: maybe mythweb's php code could check that required modules are indeed loaded, before trying to use them ?
[08:07:00] dreamcat4: and print / echo out an appropriate error message to tell the user?
[08:07:03] stuarta: i would be an idea
[08:07:12] stuarta: patches welcome :)
[08:07:22] dreamcat4: it would have saved me many hours of utter grief
[08:07:40] Merlin83b (Merlin83b! has joined #mythtv
[08:08:53] dreamcat4: stuarta: i don't know any php, but maybe google can show me (if it is simple enough)
[08:09:24] dreamcat4: we shall see. first can you tell me which php module(s) are needed, it is just only for the mysql ?
[08:13:38] dreamcat4: stuarta: i check my phpinfo and it shows "mysql support enabled"
[08:14:41] dreamcat4: . . . fo-html-L206
[08:15:49] dreamcat4: so if that is indeed the bug, then i don't understand.
[08:17:05] stuarta: maybe not
[08:19:12] dreamcat4: i github comment it to kormoc because he added those lines for the db update (git blame)
[08:19:36] stuarta: what it requires is a judicious sprinkling of Data::Dumper to identify what is wrong where
[08:20:34] stuarta: dreamcat4: can you enable the web developer tools in your browser and see if any of the component parts 404?
[08:20:41] dreamcat4: . . . 8b08cf50fac8
[08:21:06] dreamcat4: stuarta: i would love to turn on extra debugging (if I can understand how to do that!)
[08:22:07] stuarta: can you query your db? what is the result of "select * from settings where value='WebDBSchemaVer';
[08:22:42] dreamcat4: stuarta: can you bear with me? i don't run sql queries very often
[08:22:56] stuarta: np, i have plenty of work to be doing in the meantime
[08:23:32] dreamcat4: ERROR 1046 (3D000): No database selected
[08:23:47] stuarta: use mythconverg;
[08:24:47] dreamcat4: stuarta: result –
[08:25:16] stuarta: good, so mythweb and your db agree on the db schema version
[08:26:09] dreamcat4: stuarta: it used to be '3', before i hacked it to '4'
[08:26:23] stuarta: the db or mythweb?
[08:26:30] dreamcat4: so you can assume the bug occurs when the schema version = 3, not 4
[08:27:05] dreamcat4: stuarta: when you hack mythweb and comment out the first 3 lines of the "case 3" block...
[08:27:32] dreamcat4: then the 4th line beneathe it will increment the db's version from 3-->4
[08:27:58] stuarta: it should never enter that block unless the schema was 3
[08:28:33] dreamcat4: stuarta: yes, well like i just told you, the schema was 3 until i messed around with it (hacking about)
[08:31:21] stuarta: select * from settings where value in ('recommend_enabled', 'recommend_server', 'recommend_key');
[08:32:15] dreamcat4: stuarta: gist updated with db query result –
[08:32:48] stuarta: same as mine
[08:33:11] dreamcat4: stuarta: you may be able to reproduce this issue on your own installation of mythweb
[08:33:36] dreamcat4: by hacking the file "includes/db_update.php"
[08:33:57] dreamcat4: and manually setting the variable "$db_vers = 3;"
[08:36:51] dreamcat4: hmm, the mythbackend service was not started
[08:37:13] dreamcat4: i try again while backend is running
[08:38:48] stuarta: might help ;-)
[08:39:13] dreamcat4: stuarta: OK, it worked! so that tells a lot
[08:39:52] dreamcat4: if a patch to error out, it should check if mythbackend service is running and quit
[08:44:23] stuarta: i thought that was already in there. maybe not
[08:44:52] stuarta: with mythweb, it'll need to try to talk to the backend, with a quick timeout, and then present a nice error message
[08:45:15] stuarta: but i though it already did. maybe not a nice message bit it did present something
[08:47:14] natanojl (natanojl!~jonatan@mythtv/developer/natanojl) has joined #mythtv
[08:47:53] dreamcat4: stuarta: i saw such an error message like "backend not running" or something BUT
[08:48:46] dreamcat4: the message comes too late in execution. you only see it if have already hacked the db_ver from 3 -> 4
[08:50:00] dreamcat4: i think we need it to perform the test earlier, before the problem code is run, and print out a worning as a more basic PHP exception
[08:50:30] dreamcat4: since the skins / template haven't been loaded yet
[08:51:10] stuartm: dig -x
[08:51:29] stuarta: or better yet, make sure the skins / template are up, and then make a nice output
[08:51:35] ** stuarta hands stuartm a shovel **
[08:52:09] dekarl1 (dekarl1! has joined #mythtv
[08:52:29] stuartm: resolves to 'localhost' here ... found that interesting
[08:53:00] stuartm: was in my denyhosts block log
[08:53:09] andreaz (andreaz! has quit (Read error: Connection reset by peer)
[08:53:40] dekarl (dekarl! has quit (Ping timeout: 250 seconds)
[08:53:54] stuarta: stuartm: same here
[08:54:05] JohnBergqvist (JohnBergqvist!~JohnBergq@ has quit (Quit: Leaving)
[08:54:07] dreamcat4: stuarta: (for myself, a user perspective) – just throw a basic PHP exception early and be done with it.
[08:54:37] stuarta: interesting, so that's an attempt at DoS based upon triggering active firewalls
[08:54:48] stuarta: naive at best
[08:55:10] natanojl (natanojl!~jonatan@mythtv/developer/natanojl) has quit (Ping timeout: 264 seconds)
[08:57:10] stuarta: so now after re-installing f20 on my mac mini, i discover that they no longer support UEFI booting on 32 bit platforms
[08:57:13] stuarta: ffs
[08:57:41] stuarta: might be time to try mint
[08:57:58] stuarta: or just stick with ubuntu
[09:22:17] dekarl1 is now known as dekarl
[09:50:32] Sharky112065 (Sharky112065! has quit (Read error: Connection reset by peer)
[09:55:15] Sharky112065 (Sharky112065! has joined #mythtv
[09:55:57] natanojl (natanojl!~jonatan@mythtv/developer/natanojl) has joined #mythtv
[10:30:54] ** stuarta downloads mint to try later **
[10:34:02] natanojl (natanojl!~jonatan@mythtv/developer/natanojl) has quit (Ping timeout: 245 seconds)
[11:02:53] joki (joki! has quit (Ping timeout: 240 seconds)
[11:08:32] mad_enz (mad_enz! has joined #mythtv
[11:09:41] joki (joki! has joined #mythtv
[11:17:10] dreamcat4: hi again. what is the effect of not having 'lame' pkg dependancy installed on the backend ?
[11:19:11] dreamcat4: on freebsd 9 repo
[11:26:55] stuarta: it's in ports
[11:27:05] stuarta: cd /usr/ports/audio/lame
[11:29:34] dreamcat4: stuarta: the problem is that LAME is restricted, so can't be distributed as a pkg
[11:30:03] dreamcat4: stuarta: which in turn means mythtv isn't installable as a binary pkg either
[11:30:38] stuarta: yep, if it'll build without it, you lose transcoding and ability to record from framegrabber cards (that is no loss tho)
[11:30:38] dreamcat4: as mythtv pkg is built with lame as default option (default dependancy)
[11:31:11] dreamcat4: stuarta: i guessed that. it isn't really important enough to warrant being excluded from pkgs
[11:31:41] stuarta: we have a ticket open somewhere to make it an optional dependency
[11:31:47] stuarta: it's not optional atm iirc
[11:32:29] dreamcat4: stuarta: do you mean to make it optional in freebsd port Makefile, or in mythtv srcs itself ?
[11:32:42] stuarta: src
[11:32:54] stuarta: if you don't have it installed, the configure error's out
[11:33:33] dreamcat4: right ok, so that's a problem for FreeBSD users who want an easy install (from binary pkgs)
[11:36:04] ** stuarta heads out for lunch. biab **
[11:38:09] dreamcat4: i'm going to have to re-compile myth right now anyway, so will double-check whether or not compiling with lame is really madatory
[11:54:06] dreamcat4: yes, it is. still in master branch: . . . #L6020-L6022
[11:55:13] dreamcat4: couldn't find any ticket for removing it though
[12:21:04] mad_enz (mad_enz! has quit (Quit: Leaving)
[12:27:33] jya (jya!~jyavenard@mythtv/developer/jya) has quit (Ping timeout: 256 seconds)
[12:30:45] jya (jya!~jyavenard@mythtv/developer/jya) has joined #mythtv
[12:34:46] stuarta: dreamcat4: #12019
[12:34:46] ** MythLogBot **
[12:47:18] dreamcat4: thanks stuarta
[12:54:35] stuarta: fyi adding the irc log to the ticket doesn't actually help in this case
[12:57:19] stuarta: the mint installer is pretty at least
[12:57:30] stuarta: very slick and no nonsense
[12:57:40] stuarta: i can see why people are using it
[13:00:34] jpharvey (jpharvey! has quit (Ping timeout: 264 seconds)
[13:01:02] dreamcat4 (dreamcat4!~dreamcat4@ has quit (Ping timeout: 245 seconds)
[13:01:12] dreamcat4 (dreamcat4!~dreamcat4@ has joined #mythtv
[13:03:28] stuarta: damn it. i just wiped the fedora install and forgot to backup the ansible playback i'd started to install stuff
[13:03:31] stuarta: arse
[13:06:00] stuarta: *playbook
[13:13:31] jpharvey (jpharvey! has joined #mythtv
[13:57:36] superm1 (superm1!uid4318@ubuntu/member/superm1) has joined #mythtv
[14:06:54] Merlin83b (Merlin83b! has quit (Remote host closed the connection)
[14:07:18] Merlin83b (Merlin83b! has joined #mythtv
[14:11:57] wagnerrp: i don't understand #12019
[14:11:57] ** MythLogBot **
[14:12:14] wagnerrp: if you're building mythtv, what does it matter that you have to build lame as well
[14:15:47] wagnerrp: or is it more that it means mythtv cannot be built and distributed as a binary package due to the lame dependency?
[14:15:48] stuartm: he's building packages for distribution
[14:16:20] wagnerrp: lame is marked restricted, but ffmpeg isnt?
[14:16:37] stuartm: the present available bsd packages are based on an older version
[14:16:56] stuartm: wagnerrp: guess ffmpeg can at least be built with the problematic bits disabled ...
[14:18:50] wagnerrp: can we be built with the problematic ffmpeg bits disabled?
[14:19:35] wagnerrp: support for framegrabbers means needing an mp3 encoder, which seems like it would be just as restricted as lame
[14:24:32] wagnerrp: ah, reading the first message on that ticket, framegrabbers were specifically mentioned
[14:24:42] stuarta: wagnerrp: the main reason i opened #12019 is because you have to explicitly go find it and build it, before mythtv will build, while ffmpeg is included
[14:24:42] ** MythLogBot **
[14:24:57] stuarta: it's yet another lib
[14:25:06] stuarta: which now has limited use
[14:33:57] stichnot (stichnot!stichnot@mythtv/developer/stichnot) has quit (Ping timeout: 272 seconds)
[14:34:28] natanojl (natanojl! has joined #mythtv
[14:55:25] stuarta (stuarta!~stuarta@mythtv/developer/stuarta) has quit (Quit: restarting)
[14:56:27] natanojl (natanojl! has quit (Ping timeout: 245 seconds)
[14:58:35] andreaz (andreaz! has joined #mythtv
[14:58:43] stuarta (stuarta! has joined #mythtv
[14:58:43] stuarta (stuarta! has quit (Changing host)
[14:58:43] stuarta (stuarta!~stuarta@mythtv/developer/stuarta) has joined #mythtv
[15:09:39] natanojl (natanojl!~jonatan@mythtv/developer/natanojl) has joined #mythtv
[15:23:56] stichnot (stichnot!~stichnot@ has joined #mythtv
[15:23:56] stichnot (stichnot!~stichnot@mythtv/developer/stichnot) has joined #mythtv
[15:23:56] stichnot (stichnot!~stichnot@ has quit (Changing host)
[15:28:33] stuartm: wagnerrp: we don't have to use mp3 any more, could just as easily use ogg or better still MP2 (since that's the broadcast standard)
[15:31:06] Jordack (Jordack! has joined #mythtv
[15:32:35] stuartm: although it seems ffmpeg doesn't support MP2 natively, you need libtwolame, which is a shame
[15:33:41] stuartm: oh, no I mis-read that, it does support encoding natively
[15:34:11] stuartm: but you may optionally use twolame instead
[15:36:29] rsiebert (rsiebert! has joined #mythtv
[15:38:17] Chutt (Chutt! has joined #mythtv
[15:38:20] Chutt_ (Chutt_!~ijr@2605:a000:1208:c08c:c457:ef1d:60fd:2177) has quit (Ping timeout: 240 seconds)
[15:49:49] dreamcat4: stuartm: twolame would work for us on FreeBSD, because twolame is not license-restricted
[15:50:54] dreamcat4: twolame does pull in 4 more pkgs deps: libogg flac libvorbis libsndfile, but at least they will install fine / ok
[15:53:08] dreamcat4: we still don't wish for it to continue being some mendatory dependacy though
[15:56:25] stuarta: dreamcat4: are you working on packaging mythtv for freebsd?
[15:56:28] stuartm: well we'd not need twolame anyway since ffmpeg does support native encoding, and native encoding is the best option for a core feature like recording as it keeps the dependency count down
[15:56:40] ** stuarta agrees with stuartm **
[15:57:20] stuartm: mp2 is much more widely supported than the alternative of oga, seems like the ideal choice
[15:58:59] wagnerrp: "widely supported" is only a concern if we're going to move away from nuv as well
[15:59:09] stuartm: encode video in mpeg2 or H.264, audio in mp2 and you end up with stuff that's identical to all digital broadcasts, so it also reduces the number of codec combinations that we need to support perfectly in playback
[15:59:25] stuartm: wagnerrp: I'd move to mpeg-ts
[15:59:31] stuarta: wagnerrp: we should move away from nuv to a more common format
[15:59:46] stuartm: or ps
[16:00:42] stuarta: heck even mkv
[16:01:09] stuartm: as standard at least, maybe with the option to use mkv, but since hardly anyone still uses framegrabbers it hardly seems worth the effort of supporting multiple codecs/containers
[16:02:26] stuartm: stuarta: mkv wouldn't be a great choice for a default since it limits the devices you can play the video on – all hardware/software supports mpeg-ts/mpeg2/mp2 including upnp etc
[16:02:40] stuarta: ah yes, that would be a much better idea
[16:05:16] stuartm: I'm sure there would be a call for H.264, although mpeg2 still beats mjpeg comfortably and the resources required to encode H.264 are so much higher and device support isn't nearly so good
[16:06:48] stuartm: again, we're talking 1% of the user base, so there's no need to spend much time on adding bells and whistles
[16:07:41] dreamcat4: stuarta: we already have a FreeBSD port, but it is broken as a pkg (binary pkgs get auto-build from ports)
[16:07:41] stichnot (stichnot!~stichnot@mythtv/developer/stichnot) has joined #mythtv
[16:10:36] dreamcat4: stuarta: what i'm currently working on is a How-To for FreeBSD, to install MythTV + MyThweb with nginx/php-fpm
[16:11:48] dreamcat4: the FreeBSD port maintainer expressed interest in removing lame dep just recently in email to me
[16:12:16] stuarta: i'd love to see lame gone. i think it's an unneccessary pain in the arse
[16:47:30] dekarl1 (dekarl1! has joined #mythtv
[16:48:10] dekarl (dekarl! has quit (Ping timeout: 250 seconds)
[16:48:45] stuartm: the whole nuv thing is a bit of an embarrassment now
[16:53:07] Merlin83b (Merlin83b! has quit (Quit: Leaving)
[16:55:41] dekarl1: can't we use the avfwriter and stick aac+ say mpeg-2 or whatever the GPU can create (H.264?) into a Matroska container?
[17:03:43] stuartm: read what I was saying above
[17:05:05] stuartm: mkv is the worst possible container you could choose – fine if you only ever want to play those files in MythTV or on the desktop, but useless for tablets, upnp, streaming etc
[17:06:43] stuartm: and offering the option to use h.264/mkv would be fine as an optional extra to a default of mpeg2/mpeg-ps(ts), but it wouldn't be a great use of our limited developer resources
[17:14:27] SteveGoodey (SteveGoodey! has joined #mythtv
[17:19:12] dekarl1: stuartm, doh. but then, lets fix mythtranscode, so it doesn't convert to MPEG-PS?
[17:19:23] dekarl1 is now known as dekarl
[17:19:42] dekarl: I don't like that we lose all language information for the audio tracks when transcoding
[17:19:50] dekarl: s/transcoding/cutting/
[17:25:22] andreaz (andreaz! has quit (Read error: Connection reset by peer)
[17:27:36] dekarl (dekarl! has quit (Ping timeout: 250 seconds)
[17:59:32] natanojl (natanojl!~jonatan@mythtv/developer/natanojl) has quit (Ping timeout: 260 seconds)
[18:09:44] dblain: stuartm: I've been working on the enum support. a couple of questions: 1) Did you expect us to serialize using enum name or value (I was expecting to use name, but it will break backwards compat). 2) To use Q_ENUM, all enums must be defined in a QObject derived class... should they be in each class that they are used or do I create a class that only contains all the enum definitions (I'm
[18:09:44] dblain: leaning toward the latter).
[18:11:53] natanojl (natanojl!~jonatan@mythtv/developer/natanojl) has joined #mythtv
[18:13:51] dekarl (dekarl! has joined #mythtv
[18:19:01] stuartm: dblain: I wish I could find the logs relating to the discussion we had a while back as I can't now remember exactly what was proposed
[18:19:03] dekarl: stuartm, stuarta, wagnerrp, what happened to the plan of just removing support for analogue frame grabbers (while keeping analogue hardware encoder support)?
[18:21:37] stuartm: dblain: I suspect we'd moved away from using Q_ENUM since I remember discussing the ability to reference by it's name e.g. kRecordAll but also provide a simple mechanism to get a translated string representation for a UI
[18:24:09] stuartm: dblain: most current enum usage converts the enum value to a string, so we break backwards compatibility either way
[18:26:20] dblain: stuartm: I figured out a why to use the built in translation tools to provide a Displayable description, so Q_ENUM is still a viable option.
[18:27:11] dblain: stuartm: I figured the best approach is to use the enum name as coded for all serialization (no translation), then provide a means to get its display text which is translated.
[18:27:30] stuartm: oh cool
[18:27:54] dblain: I just committed a change to support Q_ENUM in the wsdl/xsd output. The xsd for a enum provides enum name, value & translated description.
[18:27:57] stuartm: if you think that's best then that's what we'll do :)
[18:29:07] dblain: I still think I need to provide a service that will return the same info the xsd does in a more service like way (that way if the caller wants it in json they could and wouldn't be forced to parse xsd xml)
[18:29:20] dekarl: dreamcat4: you do realize that the same patent holders claim that their patents cover mpeg audio layer 1,2 and 3... so twolame must have the same restriction that lame has, or someone got to do their due dilligence... The same goes for e.g. AC-3 support in ffmpeg, if you restrict one, you got to restrict the other...
[18:30:03] dblain: stuartm: as for the second question... place enums in each of the datacontract classes or create a new class that contains just enums?
[18:30:38] dblain: (the more I think about it, the more flexible a new class becomes).
[18:30:57] stuartm: well as far as where the enums get defined, I personally prefer the former, although it can go either way, it's your call
[18:31:19] stuartm: qt itself defines all enums in the QT class
[18:32:07] stuartm: and although we aren't following the QT style guidelines so I don't know why I brought that up :)
[18:32:20] dblain: I like having them part of the class that uses them, but it means an enum is specific to that class (not sure how many enums are shared). The one I've been testing with is recordinginfo.RecordingDupMethodType
[18:32:58] stuartm: few are shared, but that doesn't mean we want to limit ourselves
[18:33:03] dekarl: dreamcat4: maybe you can split the dstribution into softwarepatents aye / nay so the rest of the world can move on :) at least that worked when crypto was restricted for US based projects
[18:33:55] dblain: stuartm: true, both approaches can be used together. I will focus on keping the enum definition with the class, unless there is a good re-use reason not to.
[18:34:08] stuartm: dekarl: mp2 is supposed to be free of patents now, it's older than mp3 which loses it's patent cover next year
[18:35:01] dekarl: stuartm: i just looked it up and some say 2015, others say 2017... depends on the patent holder...
[18:35:19] stuartm: wikipedia states that mp2 is patent free (yes, not an authoritative source but still)
[18:37:12] stuartm: dekarl: anyway, we'd not have to use twolame since ffmpeg can encode natively, so no problematic dependencies for packaging
[18:38:11] stuartm: the issue here after all is not that mythtv potentially contains patent protected stuff, but that if you're building mythtv packages you don't want to allow have to build and maintain all it's dependencies
[18:38:38] dekarl: stuarm, of course ffmpeg needs to get the same restriction :)
[18:39:21] dekarl: I understood is that the Makefile of lame contains a RESTRICTED=patent FUD, see
[18:39:30] stuartm: yes, but ffmpeg is internal and would form part of a single mythtv package which can be hosted on a private server by an individual
[18:40:27] dekarl: you can host lame packages for freebsd on your private server, too. The issue is enabling freebsd packages on the main project's servers
[18:41:19] stuartm: dekarl: it's a moot point really, since at the very least we all agree that the option to disable lame is a good idea
[18:41:46] stuartm: and having it use a different audio codec is an even better idea
[18:42:03] stuartm: assuming we don't just rip out the framegrabber support, which is probably the best idea
[18:42:57] dekarl: aye, given our ressources (and how much time this has already taken) I'm for removing it. Just allow mythmusic to build without somehow.
[18:43:14] stuartm: mythmusic can already be built without lame
[18:43:20] stuartm: it's only the backend that couldn't
[18:49:51] dekarl: btw, I got "MP3 is patent restricted until 2017" from here . . . PEG-2_H_264/ also says "AudioMPEG claims that their patents cover MPEG-1 layers 1,2 and 3."
[18:51:37] dekarl: and wrt the containter, it appears that we have to remux the elementary streams for UPNP anyway because of the players only supporting a limited list of combinations of codecs and containers... (that's back from when I streamed to a PS3)
[18:51:40] stuartm: stuff I was reading earlier says that claims to patents beyond 2015 relate to stuff that was present in the ISO standard and therefore public domain
[18:52:28] dekarl: that's the point with software patents.. so much FUD and scare tactics :(
[19:00:15] stuartm: dekarl: DLNA "Home Network Devices" must support mpeg2 in MPEG PS and TS with mp2 audio
[19:00:43] stuartm: allegedly
[19:01:13] dekarl: is that wrt the frame grabber NUV replacement?
[19:01:43] stuartm: aye, or just a sane default for mythtranscode in general
[19:02:12] dekarl: Ok, but what about all the other codecs that come down a DTV pipe? like H.264 + E-AC3?
[19:02:52] dekarl: I vaguely remember fixing buts wrt lossless cut for recordings that only had these codecs.
[19:02:57] dekarl: s/buts/bugs/
[19:04:42] stuartm: not sure what DLNA says about H.264, but we're only talking about the defaults for a lossy transcode
[19:05:03] dekarl: ok
[19:05:11] stuartm: we don't want a default that then has to be transcoded again to get it into a format we can use for upnp etc
[19:05:15] dekarl: i was mixing that up then
[19:05:45] dekarl: anything works better then our NUV :)
[19:07:21] stuartm: right, if we're replacing our proprietary nuv we want to go for something that's as universal as possible, hence why I'm pushing the standard mpeg combo since there's practically no device or software that can't play that
[19:10:13] stuartm: if users _choose_ to alter the default to something like 10bit VC1 in a MKV container then that's up to them, when they find nothing will play that combination they can't blame us
[19:11:35] dekarl: sounds like a plan. what is the migration path for existing recordings? reencode to the new default or just remux to a different container?
[19:11:48] dekarl: bbl
[19:12:57] stuartm: haven't given migration of old recordings much thought, obviously we'll keep the ability to play mjpeg in nuv for as long as possible, that bit isn't causing much trouble yet
[19:14:40] dreamcat4: just to say: ffmpeg will work for FreeBSD binary pkgs problem because it has seperate options on/off for everything licence-restricted
[19:15:22] dreamcat4: so please do, by all means (at least it should not cause FreeBSD users any issue)
[19:46:16] dekarl: I think we have to agree to disagree on the license knobs. If the MP3 patents are a problem, so are the other MPEG patents (e.g. MPEG 4 Part 2), and the ports license knobs have nothing to do with our internal ffmpeg, but looking at the Makefiles, the knobs for other MPEG codecs are simply missing. So lets rip out the lame requirement.
[20:11:54] jheizer_ (jheizer_!~jheizer@ has joined #mythtv
[20:14:08] Warped (Warped! has quit (Quit: ChatZilla [Firefox 30.0/20140605174243])
[20:15:03] jheizer (jheizer!~jheizer@ has quit (Ping timeout: 256 seconds)
[20:15:42] stuarta: dekarl: stuartm i don't think there should be any migration plan for existing nuv files, leave em be, if somebody wants them in a format so other devices can play them, then they can (or should have already) transcoded them
[20:16:38] stuartm: stuarta: well atm we don't support transcoding to other containers, so you can't transcode from nuv to something more portable
[20:16:42] andreaz (andreaz! has joined #mythtv
[20:17:32] stuarta: i believe we file that under "shit happens"
[20:17:40] stuartm: hence minimally we should add that support, which is what we're proposing to do anyway :)
[20:17:50] stuarta: heh
[20:19:24] stuartm: sorta positive feedback loop there, we drop transcoding to nuv and add support for transcoding to other containers, in the process users gain the ability to transcode from nuv to a different container ... win, win
[20:21:00] stuarta: \o/
[20:28:56] sl1ce_1g (sl1ce_1g! has quit (Remote host closed the connection)
[20:30:16] sl1ce (sl1ce! has joined #mythtv
[20:37:08] dblain: Anyone know how to de-stringize a parameter in a macro?
[20:41:05] stuarta: duct tape?
[21:04:20] stichnot (stichnot!~stichnot@mythtv/developer/stichnot) has quit (Ping timeout: 260 seconds)
[21:09:25] SteveGoodey (SteveGoodey! has quit (Quit: Konversation terminated!)
[21:10:46] dblain: heh, duct tape :) ... Trying to figure out how to use QT_TRANSLATE_NOOP (or anything else that will allow translation of enum value names) while defining an enum, so we don't have to duplicate the enum names in multiple places..
[21:25:32] dekarl: dblain: someone was asking for a list of states that the frontend service can return (I'm not finding the question atm) I wanted to reply with a link to the WSDL, but its just a string... can the new MYTH_ENUMS help with that?
[21:28:18] dekarl: oh, I just saw at the wiki what that returns... was only thinking of the <String key="state">WatchingPreRecorded</String> part
[21:29:37] Jordack (Jordack! has quit ()
[21:41:23] natanojl (natanojl!~jonatan@mythtv/developer/natanojl) has quit (Ping timeout: 255 seconds)
[22:07:11] cesman (cesman!~cesman@pdpc/supporter/professional/cesman) has quit (Quit: Konversation terminated!)
[22:19:17] Tobbe5178 (Tobbe5178! has quit (Read error: Connection reset by peer)
[22:20:28] stuarta: ugh why is master rendering the video colours wrong on this new install
[22:20:34] stuarta: 0.27 is fine
[22:23:53] moparisthebest (moparisthebest! has quit (Remote host closed the connection)
[22:25:06] stuarta: bah, why are you picking OpenGL1 !!!
[22:36:00] J-e-f-f-A (J-e-f-f-A!~J-e-f-f-A@unaffiliated/j-e-f-f-a) has quit (Quit: Oh No!!!! ;-))
[22:39:20] J-e-f-f-A (J-e-f-f-A! has joined #mythtv
[22:47:07] stuarta: well now at least they are both consistently crap, using the opengl renderer give pinkish video. clearly a wrong colourspace or something like that
[22:53:45] knightr: dblain, . . . anslate-noop
[22:56:09] knightr: the choice between the two depends on whether you want to give a specific translation context (it's usually the class name under most circumstances or specify one...)
[22:56:37] knightr: don't play with QT_TRANSLATE_NOOP3, it's not worth the trouble...
[23:09:35] dekarl: I'm wondering why ProgInfo has "QString stars" (containing a float) when its base class already has "float stars". Just remove it?
[23:33:30] rsiebert_ (rsiebert_! has joined #mythtv
[23:36:22] rsiebert (rsiebert! has quit (Ping timeout: 250 seconds)
[23:43:56] stichnot (stichnot!~stichnot@ has joined #mythtv
[23:43:56] stichnot (stichnot!~stichnot@mythtv/developer/stichnot) has joined #mythtv
[23:43:56] stichnot (stichnot!~stichnot@ has quit (Changing host)
[23:54:35] arescorpio (arescorpio! has joined #mythtv

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