:: #mythtv

Daily chat history

Current users (87):

abqjp, aloril, Anssi, anykey_, Beirdo, brfransen, brtb, CaCtus491, cattelan_away, cesman, Chutt, clever, coling, Cougar, damaltor, danielk22, dekarl, dlblog, Dr{Wh0}, ElmerFudd, foobum, foxbuntu, ghoti, gregL, GreyFoxx, Guest45851, highzeth, J-e-f-f-A, j-rod|afk, jams, jarle, jcarlos, joe_, jpabq, jpabq-, jstenback, justinh, jwhite, jya, k-man, kenni, knightr, kormoc, kurre2, kwmonroe, laga, lolcat`, mag0o, MavT, mrand, MythBuild, MythLogBot, mzanetti, noahric, peitolm, pheld, poptix, purserj, rhpot1991, RingZer0, rsiebert_, skd5aner, Slasher`, sphery, sraue, stuarta, stuartm, superm1, sutula, tgm4883, TheAsp, ThisNewGuy1, ToadP, tomimo, tris, Unhelpful, vallor, wagnerrp, wahrhaft, wseltzer, xavierh, xavierh_, XDS2010_, xris, ybot_, zCougar, _charly_
Tuesday, March 6th, 2012, 00:04 UTC
[00:04:27] cesman (cesman!~cesman@pdpc/supporter/professional/cesman) has quit (Quit: Konversation terminated!)
[00:07:17] knightr: xavierh_, stuartm, danielk22 the situation you guys described earlier today of having two words being translated in the same way breaking tests made on the translated words is possible but very improbable in the same screen... We might translate two different words the same way in two different screens but I can't see that happening in the same screen... It would mean that the original English text was confusing to begin with...
[00:08:44] knightr: That said I am not a fan of doing the kind of test xavierh_ described but I guess the person who did it that was didn't have much choice...
[00:10:27] xavierh_: knightr: I have been a bit picky, it is quite unlikely. However if it did happen, it would be difficult to debug
[00:12:51] eugo (eugo!~none@unaffiliated/eugo) has left #mythtv ()
[00:13:16] stuartm: jpabq: Am I'm right in remembering that you were responsible for the channel group code?
[00:14:39] knightr: xavierh_ you are right, I don't like that kind of test being done on the translated string, there is a risk, it might be small but there is a risk... I also find it weird to do a test on something that actually needs to be looked up, the overhead of doing so might be small, there is some... If the problem did happen it would like only affect some translations not all of them which would suggest something like that occurent...
[00:16:28] knightr: s/occurent/occured
[00:18:58] knightr: a similar problem occured in the past with one of the MythWeb translation which made us realized that some string weren't properly escaped before trying to use them... There was an apostrophe in the text of one of the translation which wasn't properly escaped to be displayed in a Javascript popup and for that translation that screen stopped working properly...
[00:19:09] stuartm: I've been wanting to go through and fix instances, but it's always been very low on my list of priorities because it works as it is
[00:19:11] knightr: s/string/strings
[00:19:37] sphery: stuartm: + (I think paul-h was the one who saw it through, eventually, but seemed to be mainly contributed patch + changes)
[00:19:55] stuartm: if someone wants to work up a patch, screen by screen (too many conflicts likely otherwise) then I'll look at it
[00:20:15] sphery: (meaning I think you're safe to go through and change anything you want related to channel groups)
[00:20:57] stuartm: sphery: it's more a question of who I assigned the bugs to
[00:21:20] sphery: and, fwiw, I like the idea of making the configuration screens for it available directly inside the guide, and not buried in settings code (and having a separate web-based configuration with backend http setup)
[00:21:25] stuartm: I didn't realise that it was so far back, I thought it had been more recent
[00:21:52] sphery: oh... thought this might be related to your idea of moving a lot of that setup stuff to guide area
[00:25:34] taylorr: jya: AvFormatDecoder::ProcessAudioPacket() assigns pts but doesn't use it
[00:25:50] wagnerrp: danielk22: i was updating my slave backend to see if i could get it to trigger the fault, and ended up accidentally restarting the wrong box
[00:25:54] wagnerrp: just getting into it now
[00:26:25] jya: taylorr: I didn't write that code, I only ever updated to any changes required by the audio class
[00:26:33] wagnerrp: danielk22: it is a segfault occurring upon teardown, but the question is... why is it shutting itself down in the first place
[00:27:18] taylorr: seems that removing the int64_t typecast for AV_NOPTS_VALUE causes the warning "avformatdecoder.cpp:3097:25: warning: comparison between signed and unsigned integer expressions [-Wsign-compare]"
[00:27:32] stuartm: damn it, I've lost the ticket which I was going to assign to whomever wrote/committed the channel group changes
[00:28:47] stuartm: E_TOOMANYTICKETS
[00:28:52] taylorr: odd that it complains that it's a comparison between signed and unsigned.... INT64_C should be signed
[00:29:11] sphery: stuartm: so, on #10312, we already have a non-mythui popup that's suppposed to appear, but it's not shown.
[00:29:59] stuartm: sphery: sorry, this is the one you tried to discuss with me earlier? (If forgot you wanted my input)
[00:31:11] sphery: yeah, just started looking into what was in there... I have an idea on what's broken and will probably need your input on whether it's fixable or not, so don't worry about it for now, and I'll get back to you with more
[00:31:29] stuartm: sphery: to your earlier question, yes, to do that properly with a UI (using mythui) it would probably need not just refactoring but changes to emit progress events which would drive the UI
[00:32:08] jya: taylorr: I'm guessing like 4007–4008 testing pkt->dts can be dismissed completelly, the same test is done again before actually using it
[00:33:50] stuartm: actually, probably not so much refactoring, you might be able to get a reasonable UI without significant changes but I'd need to check where in th frontend initialisation we do the schema checks etc
[00:35:52] stuartm: sphery: let me know when you've got it figured out, I'm off to bed now so it will be tomorrow before I respond
[00:36:30] stuartm: triaging is FUN
[00:36:54] jya: taylorr: yeah, I would completely remove anything to do with long long pts variable, at worse it does nothing, because it's not used anyway. Now , weither it should have been used as opposed to later use pkt->pts, I don't know for sure
[00:37:17] jya: stuartm: until you get the people whinging : why did you close my ticket !!
[00:38:40] xavierh_: What is the right way to update / fix conflict in local git ?
[00:38:58] taylorr: jya: just remove the declaration and it's one assignment, don't touch anything else
[00:39:06] xavierh_: I have done a pull, fix the files with conflicts, then what ?
[00:39:29] jya: taylorr: I agree…
[00:39:53] jya: but why was that test there in the first place ? it could have been for a good reason
[00:41:01] jya: taylorr: do you want me to do it, or you're doing it ?
[00:41:08] stichnot: isn't #7881 just asking for PlaybackBox::ShowRecGroupChanger() to consider both the recorded and record tables?
[00:41:46] taylorr: jya: go ahead, I'm in the middle of the other change I've been talking about
[00:41:49] Beirdo: xavierh_: Then commit locally (you may need to git add the file again)
[00:42:24] Dave123 (Dave123! has quit (Ping timeout: 252 seconds)
[00:42:31] jya: taylorr: I'm looking in the history of the code. this line was imported from the refactoring done by Jannau in sha1:2b539804
[00:42:43] jya: so looking what was there before
[00:42:52] xavierh_: Beirdo: : thx, wanted to be sure
[00:43:11] jpabq: stuartm, Was not me. However, if you need me to work on something in that area, I could give it a go. My company is moving offices this month, so during the week I am swamped, but could do something on the weekends.
[00:43:25] stuartm: jya: that's a sarcastic 'fun', as in not much fun at all, and what concerns me more is that I'll assign so many tickets to danielk22 for 0.25 that he'll feel put upon ;)
[00:43:46] taylorr: jya: yes, the refactor made it unnecessary
[00:44:04] xavierh_: Beirdo: : do I have to commit all the file modified by the pull too (i.e the one I never modifier myself) ?
[00:44:08] jya: made it unecessary or the refactor introduced a problem ?
[00:44:12] stuartm: jpabq: no need, already sorted and by sorted I mean I've lost track of which ticket it was so ...
[00:44:28] taylorr: jya: it's not needed anymore... I imagine he just forgot to remove it
[00:44:52] taylorr: jya: probably a good idea to make sure though
[00:44:58] jya: there's also the same in the GetFrame routine
[00:45:03] jya: it's assigned and not used
[00:45:41] stichnot (stichnot!chatzilla@mythtv/developer/stichnot) has quit (Ping timeout: 255 seconds)
[00:45:54] stuartm: unfortunately when it comes to all things EIT/Scanner/DVB/Recorder related danielk22 is the only one left around here who works in that area, that's not really fair on him and we need to draft in more people willing to deal with those bugs
[00:46:38] Beirdo: well, stuarta too, when he has time
[00:46:51] Beirdo: at least EIT-wise, IIRC
[00:47:22] Beirdo: xavierh_: you shouldn't need to, just make sure all the changes come in
[00:48:07] jya: looking at source code in github makes my firefox browser crawls..
[00:48:22] stuartm: he hasn't had the time for well over a year now, he can't swear that he'll have the time in the near future either (in so far as none of us can see the future)
[00:48:40] Beirdo: heh, yeah, so true
[00:49:04] xavierh_: Beirdo: seems that I messed up :( will try again tomorrow
[00:49:08] stuartm: that area of the code is too big for just two people anyhow
[00:49:24] Beirdo: if I had the requisite setup, I could help out with EIT stuff some
[00:49:43] Beirdo: but I doubt my OTA stuff would be of much use :)
[00:50:38] stuartm: I've some limited experience with the DVB/EIT code, but no time to work on it in addition to the other stuff I want to do :(
[00:51:27] Beirdo: yeah, I have some limited experience in the recorder (to get the HVR-2250 up and running in analog), but I think that's been reworked since, and yeah, much other stuff to do too
[00:51:56] xavierh_: stuartm: tell me what you want to do, I will do it. That should give you some time to do what you don't want :D
[00:53:27] Dave123 (Dave123! has joined #mythtv
[00:56:40] jya: taylorr: Ok, I went back to the original code before the split of ProcessFrame… the long long pts was declared there and only used for the video processing. When the function got split into ProcessVideo and ProcessAudio, all the local variables got duplicated, including pts, even though it never ever got used for audio
[00:56:48] jya: so it's definitely safe to remove
[00:58:45] joki (joki! has quit (Ping timeout: 265 seconds)
[00:59:35] joki (joki! has joined #mythtv
[00:59:45] xavierh_ (xavierh_! has quit (Quit: ChatZilla 0.9.88 [Firefox 10.0.2/20120216080748])
[01:03:48] highzeth (highzeth! has quit (Ping timeout: 244 seconds)
[01:04:09] highzeth (highzeth! has joined #mythtv
[01:05:18] taylorr: jya: cool, thanks
[01:05:49] danielk22: taylorr: Beirdo: jya: INT64_C() will just append LL systems where a long long is 64 bits and L on a system where long is 32 bits. The intent of it is to give you a value to set an int64_t with.
[01:05:56] taylorr: Beirdo, jya: how do you properly resolve merge conflicts with a cherry-pick?
[01:06:10] jya: you edit it manually ? :)
[01:06:16] taylorr: and then what?
[01:06:22] jya: when it failed
[01:06:33] jya: it tells you the command to complete the commit
[01:06:44] jya: like commit -c 1231231231 or something
[01:06:46] taylorr: something about "git add" and "git rm"
[01:07:12] taylorr: danielk22: when I removed the cast it gave a warning about unsigned versus signed which didn't make any sense to me
[01:07:17] jya: ah, it may just bit that you merge failed because it doesn't track enough name change
[01:07:29] jya: so increased the value for the name change and try your merge again
[01:07:35] jya: i've seen cases where it helped
[01:07:36] wagnerrp: danielk22: in ServerPool, im using 'delete' on the various sockets
[01:07:54] wagnerrp: im changing those to deleteLater()s on the off chance something is trying to signal them and failing
[01:16:43] kormoc (kormoc!~kormoc@mythtv/developer/kormoc) has quit (Quit: kormoc)
[01:18:37] noahric (noahric!~noahric@ has quit (Quit: noahric)
[01:24:26] danielk22: taylorr: Just leave it as is, no harm in it.
[01:26:47] danielk22: wagnerrp: If you changed how sockets are deleted that could do it. I'm not so sure deleteLater() will get you out of the trouble though.
[01:29:06] danielk22: wagnerrp: In the case of my mythfrontend backtrace it was shutting down because I told it to. The SBE stuff may be something else.
[01:29:39] wagnerrp: danielk22: the sockets should have only ever been what was sending the signal
[01:29:56] wagnerrp: unless they get something traveling through from the defined parent, they would have never received anything
[01:35:22] danielk22: wagnerrp: The socket sends the signal and then it is deleted before the signal handler finishes running. At least that is what was happening the last time we saw those same backtraces.
[01:40:08] joki (joki! has quit (Ping timeout: 240 seconds)
[01:40:36] joki (joki! has joined #mythtv
[01:45:19] Beirdo: taylorr: if that's *supposed* to be signed, simply set it to int64_t(-1)
[01:45:39] Beirdo: with the MSBit set, that's by definition an unsigned int
[01:46:15] Beirdo: and git mergetool is how I go to fix merge conflicts
[01:46:54] Beirdo: followed by git add -i to choose the missing parts to stage, then the commit that it gave on the failure
[01:47:41] taylorr: Beirdo: thanks, I'll have to check out "mergetool" next time
[01:48:13] Beirdo: it will run vim in 3 columns (at least does for me)
[01:49:04] Beirdo: left and right are the two that you are merging, and the middle is the merged version, and will highlight the changes, and makes it easy to find the conflict (same search for >>>> ==== <<<<)
[01:52:17] Dave123 (Dave123! has quit (Quit: Leaving)
[01:55:22] joki (joki! has quit (Ping timeout: 272 seconds)
[01:55:29] noahric (noahric!~noahric@ has joined #mythtv
[01:58:06] joki (joki! has joined #mythtv
[02:05:56] davide_ (davide_!~david@mythtv/developer/gigem) has quit (Remote host closed the connection)
[02:06:18] davide_ (davide_! has joined #mythtv
[02:06:18] davide_ (davide_! has quit (Changing host)
[02:06:18] davide_ (davide_!~david@mythtv/developer/gigem) has joined #mythtv
[02:08:22] joki (joki! has quit (Ping timeout: 276 seconds)
[02:12:37] Dave123 (Dave123! has joined #mythtv
[02:28:43] cesman (cesman! has joined #mythtv
[02:28:44] cesman (cesman!~cesman@pdpc/supporter/professional/cesman) has joined #mythtv
[02:28:44] cesman (cesman! has quit (Changing host)
[02:43:36] joki (joki! has joined #mythtv
[02:48:06] joki (joki! has quit (Client Quit)
[03:08:18] stichnot (stichnot!chatzilla@mythtv/developer/stichnot) has joined #mythtv
[03:14:12] dblain (dblain!~dblain@mythtv/developer/dblain) has quit (Ping timeout: 252 seconds)
[03:16:52] dblain (dblain! has joined #mythtv
[03:16:52] dblain (dblain! has quit (Changing host)
[03:16:52] dblain (dblain!~dblain@mythtv/developer/dblain) has joined #mythtv
[03:46:48] jpabq (jpabq!~jpabq@mythtv/developer/jpabq) has quit (Ping timeout: 252 seconds)
[03:46:49] jpabq_ is now known as jpabq
[03:47:39] jpabq- (jpabq-! has quit (Ping timeout: 245 seconds)
[03:49:13] stichnot: stuartm: executing a jump point during a playlist crashes the frontend :)
[03:50:16] abqjp (abqjp! has joined #mythtv
[03:52:01] jpabq- (jpabq-! has joined #mythtv
[04:09:02] sphery: stuartm: I went ahead and pushed a fix in cf9ee654f . If you want to take a different approach on the fix, I'll leave that to you, but figured this was a good enough fix since we have other non-mythui dialogs in the schema wizard
[04:13:28] TheAsp (TheAsp! has quit (Ping timeout: 246 seconds)
[04:13:43] TheAsp (TheAsp! has joined #mythtv
[04:33:36] joki (joki! has joined #mythtv
[04:33:38] dblain (dblain!~dblain@mythtv/developer/dblain) has quit (Ping timeout: 260 seconds)
[04:38:16] kormoc (kormoc!~kormoc@mythtv/developer/kormoc) has joined #mythtv
[04:42:08] joki (joki! has quit (Ping timeout: 245 seconds)
[05:09:29] jeff999 (jeff999!~jeff@ has joined #mythtv
[05:10:43] jeff999 (jeff999!~jeff@ has quit (Remote host closed the connection)
[05:16:49] wagnerrp: MythBuild: force master-freebsd-64bit now
[05:16:50] MythBuild: try 'force build [--branch=BRANCH] [--revision=REVISION] <WHICH> <REASON>'
[05:16:58] wagnerrp: MythBuild: force build master-freebsd-64bit now
[05:17:01] MythBuild: build forced [ETA 10m34s]
[05:17:01] MythBuild: I'll give a shout when the build finishes
[05:21:20] wagnerrp: danielk22: user reports #10405 no longer crashing
[05:21:37] wagnerrp: but thats likely because its actually running rather than trying to terminate and encountering the issue
[05:25:40] cesman (cesman! has joined #mythtv
[05:25:41] cesman (cesman!~cesman@pdpc/supporter/professional/cesman) has joined #mythtv
[05:25:41] cesman (cesman! has quit (Changing host)
[05:26:35] MythBuild: build #2091 of master-freebsd-64bit is complete: Failure [failed show versions] Build details are at . . . /builds/2091
[05:28:36] wagnerrp: MythBuild: force build master-freebsd-64bit now
[05:28:39] MythBuild: build forced [ETA 10m34s]
[05:28:39] MythBuild: I'll give a shout when the build finishes
[05:31:07] joki (joki! has joined #mythtv
[05:36:27] MythBuild: build #2092 of master-freebsd-64bit is complete: Success [build successful] Build details are at . . . /builds/2092
[05:59:06] Beirdo: OK, this seems to be new.
[05:59:42] Beirdo: delete the last recording on a list of episodes of a show... it goes to the next show rather than the previous episode of the same show
[06:00:09] Beirdo: that's a recent change in behavior best I can tell
[08:02:30] noahric (noahric!~noahric@ has quit (Quit: noahric)
[08:12:35] Goga777 (Goga777! has joined #mythtv
[08:13:22] jya: stuarta: what command line options are you using to build on the mac buildbot ?
[08:23:09] Beirdo: jya: . . . lders/
[08:24:35] jya: thanks ! ok… that's what I thought… I was looking at options being handled in the code, but not documented in anyways… they are all used in that script
[08:26:11] Beirdo: heh
[08:32:48] Goga777 (Goga777! has quit (Ping timeout: 260 seconds)
[08:33:28] Goga777 (Goga777! has joined #mythtv
[08:34:51] Goga777 (Goga777! has quit (Remote host closed the connection)
[08:37:26] natanojl (natanojl! has joined #mythtv
[08:37:33] joki (joki! has quit (Ping timeout: 245 seconds)
[08:41:25] sraue (sraue!~stephan@xbmc/staff/sraue) has quit (Ping timeout: 248 seconds)
[08:56:22] sraue (sraue!~stephan@xbmc/staff/sraue) has joined #mythtv
[08:56:32] peitolm: hi jya, that builds, do we not accept <esc> for quit on osx?
[08:58:47] jya: peitolm: this will depend on what you've mapped on the ESC key
[08:58:55] jya: it seems screwed up
[08:59:07] jya: ESC is assigned to more than one function
[08:59:17] jya: for me, esc leaves without asking for anything
[08:59:36] jya: got into the key binding editor, and make sure escape is assigned to only one key in the main meny
[08:59:38] jya: menu
[09:00:32] natanojl (natanojl! has quit (Read error: Operation timed out)
[09:03:28] peitolm: ta, it's always worked by default, and i don't think i ever changed it, but i'll check
[09:03:43] jya: me neither
[09:03:56] jya: I'm thinking something got screwed up in the settings during the upgrade
[09:04:03] jya: all my frontends got screwed up
[09:04:06] jya: I lodged a bug about it
[09:04:23] jya: stuartm assigned one to me, where you press Command-Q during playback
[09:04:37] jya: window closes, but audio still continue playing and you can't ever exit
[09:04:39] jya: very weird
[09:14:35] stuarta: morning all
[09:14:56] peitolm: i have a recording that ends suddenly, but haven't really looked into why it stops playing, nor verified it will play in other players
[09:18:43] xavierh: jay: someone talked about the ESC issue on this channel few day ago
[09:18:55] xavierh: jya: ^^^
[09:19:52] jya: xavierh: yes, about the ESC being improperly mapped to too many actions
[09:22:52] Goga777 (Goga777! has joined #mythtv
[09:23:39] xavierh: jya: are you sure this is not correct, I thought the ESC key what bind once in the global context and once in the main menu or something
[09:24:19] jya: xavierh: i;m not sure what it's *supposed* to be
[09:24:41] jya: but from one day to the next, all my frontends suddenly had ESC mapped to quit and quit with prompt
[09:25:01] jya: after editing the binding it's working again
[09:25:13] xavierh: what did you change in the binding ?
[09:25:42] joki (joki! has joined #mythtv
[09:26:32] jya: again: I didn't change anything… it all happened by itself
[09:26:59] jya: someone had seen the same issue, and found out that ESC was bound to more than one action in the main menu
[09:27:10] jya: and after editing the bindings it worked again
[09:31:26] xavierh: what action is it bind to ?
[09:34:05] stuartm: the binding of ESCAPE was changed by a db update, it seems that update is screwing up for some reason – iirc sphery was the author of that change
[09:35:06] jya: xavierh: I just wrote what it was: "but from one day to the next, all my frontends suddenly had ESC mapped to quit and quit with prompt"
[09:36:04] stuartm: it was supposed to bind to 'quit' if you had 'quit without prompts' selected in the old setting, but it's binding it even when that wasn't the case and it's leaving the other binding in place
[09:37:06] jya: stuartm: I find the default to be exit with prompt (like it was before) far more intuitive
[09:37:24] jya: you never expect your application to suddenly exit simply by pressing esc
[09:37:41] jya: especially, it you're pressing esc quickly to come back from sub-menus
[09:37:46] stuartm: jya: like I said, that shouldn't have changed, but it seems the DB update is going awry
[09:37:53] jya: I do that often: esc-esc-esc.. oh damn, I'm out
[09:38:29] joki (joki! has quit (Ping timeout: 245 seconds)
[09:38:38] stuartm: that said, I'm not sure why we need to allow exit without prompt, I doubt many people actually use it
[09:39:12] joki (joki! has joined #mythtv
[09:40:05] peitolm: i always do
[09:40:19] jya: peitolm: exit without prompt?
[09:40:26] peitolm: yep
[09:40:37] peitolm: if i close it, i'll mutter and re-open
[09:40:39] jya: I guess if you use a keyboard you are more in control
[09:40:48] peitolm: but if i'm running a kiosk, then yes, prompt
[09:41:04] jya: but with a remote, and one that don't register all, it became a habit of pressing the exit key on my remote a lot
[09:41:05] peitolm: but otherwise, i don't want to be asked to shutdown,
[09:41:15] jya: peitolm: that's a different config
[09:41:19] peitolm: sounds like we still need both
[09:41:21] jya: you can be asked or not
[09:41:30] jya: and then in the choice of action
[09:41:39] jya: you cna have quit, reboot, shutdown, you name it
[09:41:44] xavierh: jya: What I meant was "To what action is Esc bound to apart for ESCAPE ?"
[09:41:49] Goga777 (Goga777! has quit (Remote host closed the connection)
[09:42:02] jya: and as I said: quit *and* quit with prompt
[09:42:17] jya: if that's not the answer you want, I don't understand your question
[09:43:03] jya: key bindings -> main menu: you have two entries there
[09:43:09] jya: both were bound to Esc
[09:43:24] xavierh: jya: sorry did not realise that quit was actually a ACTION :)
[09:48:44] xavierh: So basicaly you can map a ky to exit without prompt regardless the setting ? and this key new action have been bound to esc by default. new keybinding are not defined in the db upgrade stuff, why is it different this time
[09:48:48] xavierh: ?
[09:52:52] jeff999 (jeff999!~jeff@ has joined #mythtv
[09:53:04] jeff999 (jeff999!~jeff@ has left #mythtv ()
[10:00:48] stuartm: xavierh: there is no setting anymore
[10:01:34] stuartm: the key binding replaced the setting, the db update didn't define the binding, it just enabled it for those who had previously used the setting
[10:07:04] Malard (Malard!Malard@xbmc/staff/malard) has quit (Read error: Connection reset by peer)
[10:09:41] xavierh: stuartm: strange, I did an update yesterday and did not notice that the setting have been removed
[10:09:56] xavierh: did not show as a comflict at least
[10:18:58] xavierh: stuartm: I tried to do an update yestarday and it wen horribly wrong (done a backup before). I did pull, but all the change made in master sinc my first checkout where marked as modified, I resold some conflict, commit the conclited files. but when I tried to generate the patch, it did include the change from master. Whas is the correct way ?
[10:19:30] stuartm: xavierh: this happened months ago
[10:20:01] stuartm: unless you were upgrading from an old version, which is probably what you meant
[10:20:49] stuartm: xavierh: I don't really understand git, so I can't help
[10:25:47] xavierh: stuartm: I probably converted the "exit prompt" 6 month ago, but I started from scratch 2 month ago, and just did not notice it
[10:26:26] xavierh: stuartm: I might end up doing a fresh checkout, apply my patch, fix it and generate a new patch :(
[10:28:50] stuartm: xavierh: the setting defining what appears in the exit menu is still there, just not the setting which decides whether it's shown or not
[10:30:21] xavierh: oh I see
[11:03:53] ThisNewGuy1 (ThisNewGuy1! has joined #mythtv
[11:04:28] kormoc (kormoc!~kormoc@mythtv/developer/kormoc) has quit (Ping timeout: 272 seconds)
[11:05:02] mike|2 (mike|2! has quit (Remote host closed the connection)
[11:05:48] mike (mike! has joined #mythtv
[11:06:14] mike is now known as Guest45851
[11:06:49] ThisNewGuy (ThisNewGuy! has quit (Ping timeout: 252 seconds)
[11:27:16] dblain (dblain! has joined #mythtv
[11:27:16] dblain (dblain! has quit (Changing host)
[11:27:16] dblain (dblain!~dblain@mythtv/developer/dblain) has joined #mythtv
[11:28:45] sphery: jya: It would be a lot easier to debug the issue if you'd expand upon, "but from one day to the next, all my frontends suddenly had ESC mapped to quit and quit with prompt". In the ticket you say, "A regression that only started recently. Not sure when." but the DB update you all are blaming occurred 7 months ago, so are you saying that you actually upgraded from 0.24-fixes (or, at least from DB schema 1279) to master (DB schema 1299) on that ...
[11:28:51] sphery: ... "one day to the next"?
[11:29:07] jya: sphery: let say that one day I noticed the problem.
[11:29:34] jya: every even, to turn off one of my frontend, I press esc on the main menu and select shutdown
[11:29:41] jya: every day I mean
[11:30:08] jya: and one day it simply exited (to restart automatically, I have a script running in loop to restart it in case it crashed)
[11:30:26] jya: i didn't pay much attention then, simply thinking it was something local
[11:30:44] jya: then I realised the same thing occurred on my 3 frontends.. no more prompt
[11:31:12] jya: it wasn't something that caught me as being wrong , so I didn't pay attention until later.
[11:31:44] jya_ (jya_!~jyavenard@mythtv/developer/jya) has quit (Quit: jya_)
[11:36:51] sphery: so, you can't provide any helpful information about what might have changed? you can't even say whether the change was you upgrading from 0.24 to 0.25? It just seems like if a DB update was broken, someone might have noticed in less than 7months
[11:37:11] jya: unfortunately, no… In the past 2 weeks would be my best guess
[11:37:30] jya: I'm pretty certain it didn't come from upgrading from 0.24 to 0.25
[11:37:57] jya: as I've seen it well after I had upgraded to master
[11:41:25] sphery: OK, that's helpful. I haven't been able to figure out any possible way, with any combination of settings, the db update could leave both exit and exitprompt bound. I'll see if I can figure out what might be doing it. Thanks.
[11:43:24] jya: On my frontend, I think the only configuration I ever changed when it comes to exiting, is making sure it offers me to shutdown
[11:43:59] jya: which I think is the default anyway
[11:44:06] xavierh: Can someone advise me on how to update my local git with local commited change? the only things that I manage to pull correctly so far, are my hair :)
[11:44:21] jya: xavierh: are these changes you want to keep or not ?
[11:44:43] xavierh: jya: yes I want to keep them, to update a patch on trac
[11:44:57] jya: when you did git pull, did it worked or did it show a conflict?
[11:45:39] jya: there are a few ways you can do it. One is do a merge and resolve the issues
[11:46:22] xavierh: I had some conflict wich I resolved by editing the files then I ahve done a git add on the conflicted files. The problem I am having is a lot of files I did not modified, are marked as modified since the pull
[11:46:33] jya: One is do a git diff > mypatch.txt so you have a copy of the diff file. You need to make sure all the files you want to save are being tracked by git. To do so you do git add on all the files you want to keep
[11:46:42] jya: git diff HEAD > file.diff and it will be there
[11:46:51] jya: the other option is stashing your changes
[11:50:06] jya: git stash
[11:50:06] jya: this will pile up all your changes, now you can do a git pull
[11:50:06] jya: to restore your changes, you do git stash pop
[11:50:06] jya: and there you manage the conflict
[11:50:06] xavierh: ok, I give it a go, thanks
[11:50:06] jya: git stash is probably the easiest to make sure your file are there. but to recover the change, you need to manage the conflicts by hand, which may not be trivial if you're not familiar with git conflicts
[11:50:06] jya: git diff option , saving your patch may be the easiest, but you have to be careful that the patch include changes to all the file you need
[11:50:06] jya: because git diff, will only diff files it is currently tracking
[11:50:06] jya: so if you created a new file, never added it to git, it won't be in your diff .. Does that make sense?
[11:52:12] xavierh: To sum up, I should make sure that all my change are commited, then I do git stash, git pull, git stash pop, resolve the conflict by hand, git add on conflictinf file. correct ?
[11:53:01] xavierh: jya: ^^^
[11:53:25] jya: not with git stash
[11:53:35] jya: I never talked about committing anything !
[11:53:49] jya: I said "add" if you want to use the plain diff method
[11:54:30] jya: with git stash, it doesn't matter that your file has been added or not. git stash will stack any changes you have made since the last commit, and put it in the stash stack
[11:55:01] xavierh: ok, I commited some file log time ago, so stash is not an option
[11:55:06] jya: I'm sure if you actually went to read one of the many git FAQ, you'll get your answers much more easily
[11:55:33] jya: git stash / diff : it's to save any work you have in progress
[11:55:41] jya: this isn't relevant to things you have committed
[11:55:55] jya: those will have to be resolved when you merge
[11:56:14] jya: when you do pull, if there's no conflict, it will automatically merge whatever is required
[11:56:17] xavierh: merge with pull ?
[11:56:33] jya: the only thing to remember, ever, is *NEVER* use --force
[11:56:52] jya: because that will screw things up badly for everyone later
[11:56:59] stuarta: to get more control, git fetch; git merge origin/master
[11:57:11] jya:
[11:57:13] stuarta: that's what git pull tries to do
[11:57:32] stuarta: or git rebase origin/master if you have local commits
[11:58:00] jya: stuarta: he has conflicts, that won't work
[12:11:33] dblain (dblain!~dblain@mythtv/developer/dblain) has quit (Ping timeout: 265 seconds)
[12:12:16] dlblog (dlblog! has quit (Ping timeout: 246 seconds)
[12:17:21] xavierh: I still don't understand why the file I did not touched but have been change since I clone master are marked as modified. This is my only issue
[12:20:53] ToadP (ToadP!~jboyle@ has quit (*.net *.split)
[12:20:53] kwmonroe (kwmonroe!~kwmonroe@ has quit (*.net *.split)
[12:20:53] jams (jams! has quit (*.net *.split)
[12:20:54] Captain_Murdoch (Captain_Murdoch!~cpinkham@mythtv/developer/CaptainMurdoch) has quit (*.net *.split)
[12:20:54] justinh (justinh! has quit (*.net *.split)
[12:20:54] jwhite (jwhite!jwhite@ has quit (*.net *.split)
[12:20:54] mag0o (mag0o! has quit (*.net *.split)
[12:20:54] dekarl (dekarl! has quit (*.net *.split)
[12:20:54] ybot_ (ybot_!quassel@nat/nokia/x-voutwarmabxvrdai) has quit (*.net *.split)
[12:23:49] ToadP (ToadP!~jboyle@ has joined #mythtv
[12:23:50] kwmonroe (kwmonroe!~kwmonroe@ has joined #mythtv
[12:23:50] jams (jams! has joined #mythtv
[12:23:50] Captain_Murdoch (Captain_Murdoch!~cpinkham@mythtv/developer/CaptainMurdoch) has joined #mythtv
[12:23:50] justinh (justinh! has joined #mythtv
[12:23:50] jwhite (jwhite!jwhite@ has joined #mythtv
[12:23:50] mag0o (mag0o! has joined #mythtv
[12:23:50] dekarl (dekarl! has joined #mythtv
[12:23:50] ybot_ (ybot_!quassel@nat/nokia/x-voutwarmabxvrdai) has joined #mythtv
[12:26:24] dlblog (dlblog! has joined #mythtv
[12:42:55] kwmonroe` (kwmonroe`!~kwmonroe@ has joined #mythtv
[12:45:30] kwmonroe (kwmonroe!~kwmonroe@ has quit (Ping timeout: 255 seconds)
[12:46:06] kwmonroe (kwmonroe!~kwmonroe@ has joined #mythtv
[12:48:38] kwmonroe` (kwmonroe`!~kwmonroe@ has quit (Ping timeout: 240 seconds)
[12:53:13] sraue (sraue!~stephan@xbmc/staff/sraue) has quit (Ping timeout: 246 seconds)
[13:01:49] sraue (sraue!~stephan@xbmc/staff/sraue) has joined #mythtv
[13:03:08] dblain (dblain! has joined #mythtv
[13:03:08] dblain (dblain! has quit (Changing host)
[13:03:08] dblain (dblain!~dblain@mythtv/developer/dblain) has joined #mythtv
[13:04:38] davide_ (davide_!~david@mythtv/developer/gigem) has quit (Remote host closed the connection)
[13:04:58] davide_ (davide_! has joined #mythtv
[13:04:58] davide_ (davide_!~david@mythtv/developer/gigem) has joined #mythtv
[13:04:58] davide_ (davide_! has quit (Changing host)
[13:41:10] danielk22: stuartm: I use exit without prompt, but I bind the exit action to alt-esc.. I think the real reason it exists though is for mythwelcome, where you want a bunch of escapes to bring you out of mythtv and into mythwelcome.
[14:02:38] kormoc (kormoc! has joined #mythtv
[14:02:40] kormoc (kormoc! has quit (Changing host)
[14:02:40] kormoc (kormoc!~kormoc@mythtv/developer/kormoc) has joined #mythtv
[14:10:05] stuartm: makes sense
[14:12:46] danielk22: wagnerrp: yeah, the logic error you fixed means there is no exit and hence socket shutdown race condition. The deleteLater() may be sufficient; the signals are sent in the same thread that will eventually delete the socket (assuming it does get deleted.)
[14:17:08] wagnerrp: danielk22: however that logic error should not have manifested in a slave backend
[14:17:30] wagnerrp: the only reason that occurred is because the user blanked out the default '::1' for BackendServerIP6
[14:17:48] wagnerrp: meaning the IP selection routine thought it was an unconfigured frontend
[14:17:59] wagnerrp: and instead listened on any available IPv6 address
[14:18:10] stuarta: could be worse
[14:19:51] xavierh: Do we have some sort of slider widget ? or only spinbox ?
[14:21:51] zombor (zombor!~zombor_@kohana/developer/zombor) has joined #mythtv
[14:28:49] knightr (knightr!~knightr@mythtv/developer/knightr) has quit (Ping timeout: 265 seconds)
[14:38:10] joki (joki! has quit (Ping timeout: 260 seconds)
[14:39:27] stuartm: no slider
[14:40:49] j-rod|afk is now known as j-rod
[14:48:48] joki (joki! has joined #mythtv
[14:49:38] jya: xavierh: what machine are you using? linux, pc, mac?
[14:51:52] xavierh: jya: linux, why ?
[14:52:03] xavierh: linux on a pc :)
[14:52:15] jya: just wondering in case you were on a case-insensitive file system…
[14:53:25] xavierh: I finally manage to merge, pull, resolf conflic & add then commit everything including files I did not change myself and done
[14:53:54] xavierh: I can't even understand what I wrote :D
[14:54:11] xavierh: bottom line is: it is done
[14:56:31] Yancho (Yancho! has joined #mythtv
[14:56:31] Yancho (Yancho! has quit (Changing host)
[14:56:31] Yancho (Yancho!~mpulis@unaffiliated/yancho) has joined #mythtv
[14:57:25] stuartm: ugh, R-Pi shipping date has just slipped back three weeks
[14:59:59] stuarta: :(
[15:02:30] stuartm: or not, their website says the email was a mistake
[15:05:06] jams: it's funny they had an email like that ready to go
[15:07:17] stuartm: apparently they updated the shipping date for new orders on their system and accidentally applied it to existing orders as well
[15:07:48] stuartm: which triggered an automated email to everyone with an open order telling them their dispatch date had changed
[15:09:22] xavierh: stuartm: some settings used to use sliders, they are using spinbox now
[15:13:38] xavierh: stuartm: are you going to leave standardsetting classes in mythfrontend or are you going to move the abstracted classes + transcient implementation to libmythui ?r
[15:18:12] stuartm: xavierh: I haven't really decide
[15:18:14] stuartm: d
[15:29:03] brfransen (brfransen!~brfransen@ has quit (Quit: brfransen)
[15:29:04] xavierh: I believe jams wanted to update is program to the new UI.
[15:29:15] xavierh: jams: ^^^
[15:29:17] jams: yes i did
[15:29:19] jams: do
[15:29:20] brfransen (brfransen!~brfransen@ has joined #mythtv
[15:29:58] wagnerrp: has anyone been contacted by this GO!COSS! group? (
[15:30:00] jams: it makes heavy usage of the old qt settings and will need to be updated
[15:30:06] jya: stuarta: I have finally understand precisely why qt failed to compile from source unless mysql was compiled in 32 bits… and why so much of qt failed to compile as well
[15:30:23] wagnerrp: i just commented out a line they added to the wiki about their pre-built media centers
[15:30:26] jya: s/understand/understood
[15:30:37] stuarta: you were right the first time
[15:30:38] xavierh: jams: will you link to libmythui?
[15:30:50] wagnerrp: as they are installing mythtv on a Cortex A8, which is of minimal use without hardware decoding support (which we dont have)
[15:30:54] stuarta: -have
[15:31:07] stuarta: anyway...
[15:31:39] jams: xavierh- yes i will
[15:31:43] jams: do
[15:32:13] jya: it's the same reason why I had so much trouble compiling mythzoneminder
[15:33:27] jya: we pass as configure line : "-qt-sql-mysql -mysql_config "$PREFIX/bin/mysql_confi" , unless you build it as m32, this will output something like -L.osx-packager/lib/ -arch x86_64 -lmysql
[15:35:19] jya: problem this is added to the LIBS like so LIBS+=-arch x86_64. LIBS already contains -arch x86_64. qmake doesn't know how to do with LDFLAGS on the LIBS. so it thinks you're added the lib -arch x86_64 -arch x86_64, it simplifies as -arch x86_64 -arch (it removes the 2nd x86_64),, and now you have an invalid compile line with a -arch without argument
[15:36:20] jya: anyhow, now that I've figured it out, we can now pass -qtsrc <version> like -qtsrc 4.7.4 ; it will retrieve 4.7.4 source code and compile it
[15:36:32] jya: tested with 4.8.0 and 4.7.4 so far
[15:37:29] ** jya off to bed **
[15:37:32] stuarta: fun
[15:42:21] joki (joki! has quit (Quit: c ya!)
[15:48:21] joki (joki! has joined #mythtv
[15:52:46] joki (joki! has quit (Ping timeout: 246 seconds)
[15:53:51] joki (joki! has joined #mythtv
[16:02:29] joki (joki! has quit (Ping timeout: 252 seconds)
[16:02:29] Guest45851 (Guest45851! has quit (Read error: Connection reset by peer)
[16:03:07] Guest45851 (Guest45851! has joined #mythtv
[16:07:25] stichnot (stichnot!chatzilla@mythtv/developer/stichnot) has quit (Ping timeout: 260 seconds)
[16:11:47] rsiebert_ (rsiebert_! has joined #mythtv
[16:14:52] joki (joki! has joined #mythtv
[16:15:19] rsiebert (rsiebert! has quit (Ping timeout: 276 seconds)
[16:28:09] joki (joki! has quit (Ping timeout: 252 seconds)
[16:35:35] sphery: jpabq-: you around? had some questions about #10403 (and since j y a is sleeping, thought you might be able to give me some info on your system/usage). Specifically, did you just upgrade to master right before you lost the exit prompt or have you run any version of master since July 21, 2011 where the exit prompt worked properly?
[16:37:24] joki (joki! has joined #mythtv
[16:47:19] Jordack (Jordack! has joined #mythtv
[16:50:17] sphery: kenni: You said you saw #10403 when upgrading a frontend from 0.24-fixes to master. I'm assuming you actually ran either mythfrontend or mythtv-setup from 0.24 on that frontend at some point, right? (I'm also assuming "upgraded from 0.24-fixes to git master" involved upgrading all frontends and backends, right--not ignoring schema versions and running frontends and/or backends against wrong DB versions?)
[16:52:16] sphery: stuarta: and, FWIW, I'm setting up Ctrl+Esc as EXIT and not mapping it to a remote button, and setting up Alt+Esc as EXITPROMPT and putting that on my remote--then when I'm doing maintenance, I can Ctrl+Esc to exit mythfrontend when the TV is off without having to hope I'm selecting the right button in the "Are you sure" prompt
[16:52:22] sphery: er, stuartm ^^^
[16:57:25] danielk22: wagnerrp: FYI I'm not seeing the segfault on exit anymore.
[16:57:44] wagnerrp: perhaps the delete/deleteLater() was it
[16:59:36] taylorr: j-rod: what are the odds that Mauro will send the hd-pvr picture controls patch to ? ;)
[17:00:09] j-rod: taylorr: :\
[17:00:30] joki (joki! has quit (Ping timeout: 260 seconds)
[17:00:30] j-rod: thought he'd actually done something about that already, but it seems not
[17:00:44] j-rod: I did get a note from greg kh about the imon bits going into 3.2-stable at least
[17:00:50] j-rod: will have to ping mauro again
[17:02:20] taylorr: j-rod: thanks! he seems to respond to you and not me
[17:02:24] xavierh: stuartm: I was planning to convert mythgallery setting to MythUI i.e. I will need to access my classes ... ?
[17:06:43] cesman (cesman! has joined #mythtv
[17:06:43] cesman (cesman! has quit (Changing host)
[17:06:43] cesman (cesman!~cesman@pdpc/supporter/professional/cesman) has joined #mythtv
[17:13:38] stichnot (stichnot!chatzilla@mythtv/developer/stichnot) has joined #mythtv
[17:17:57] joki (joki! has joined #mythtv
[17:25:49] stichnot (stichnot!chatzilla@mythtv/developer/stichnot) has quit (Quit: ChatZilla 0.9.88 [Firefox 10.0.2/20120215223356])
[17:32:23] stuartm: heh, the spectrum visualiser is still very buggy
[17:32:31] stuartm: and goom crashes :/
[17:35:55] joki (joki! has quit (Ping timeout: 276 seconds)
[17:36:26] joki- (joki-! has joined #mythtv
[17:36:46] joki- is now known as joki
[17:50:29] joki (joki! has quit (Ping timeout: 260 seconds)
[17:53:46] stichnot (stichnot!chatzilla@nat/intel/x-fjgrkjxutbywzhge) has joined #mythtv
[17:53:46] stichnot (stichnot!chatzilla@nat/intel/x-fjgrkjxutbywzhge) has quit (Changing host)
[17:53:46] stichnot (stichnot!chatzilla@mythtv/developer/stichnot) has joined #mythtv
[17:53:58] joki (joki! has joined #mythtv
[18:05:16] knightr (knightr!~knightr@mythtv/developer/knightr) has joined #mythtv
[18:05:31] davide_ (davide_! has joined #mythtv
[18:05:32] davide_ (davide_! has quit (Changing host)
[18:05:32] davide_ (davide_!~david@mythtv/developer/gigem) has joined #mythtv
[18:08:59] noahric (noahric!~noahric@ has joined #mythtv
[18:42:17] kenni: sphery: Hmm, thinking about it, I *think* the 0.24 frontend started up once while the backend was 0.25...I'm not completely sure, but I think I restarted the frontend system once (which caused mythfrontend 0.24 to run automatically at boot), before the frontend was upgraded to 0.25.
[18:43:23] sphery: but the 0.24 frontend exited because of schema version mismatch, right?
[18:43:55] sphery: meaning you never forced mythfrontend to connect to a wrong-versioned DB by compiling it to ignore DB version checks, right?
[18:47:01] kenni: Yep, it gave some error message about the version being incompatible, AFAIR. And no, I've absolutely not forced it to do anything :) My frontends are stock MythBuntu with daily auto-build.
[18:47:31] sphery: I asked about running 0.24 because the only possible way I've found to get a failure resulting in a key binding collision for exit/exitprompt is if you never run any version of 0.24 on a host that was a 0.23 frontend--and even then it's only possible that the collision will occur on a single host in the network and it might not even be possible because of the way data is returned (so it's theoretical)... but I can't imagine that all of the ...
[18:47:37] sphery: ... people affected skipped from 0.23 to master and happened to hit the unlikely issue I've found
[18:48:12] sphery: kenni: so how many hosts do you have in your mythtv system? Also, any chance you could pastebin the output of: select * from settings where value in ('AllowQuitShutdown', 'NoPromptOnExit');
[18:48:24] sphery: and thanks for the info, too
[18:51:15] kenni: that frontend is only 6 months old and has only been running 0.24(-fixes). I have 2 x MythBuntu 11.04 frontends with auto-builds enabled and 1 x backend which I compile myself on SL6.2.
[18:51:48] kenni: T410 is my test-machine for translations
[18:55:59] sphery: and which one had the collision?
[18:56:36] kenni: sphery: mc2
[19:00:15] kenni: sphery: I have to go for an hour or so, if you want me to test something, just post it and I'll follow up when I get back.
[19:02:06] sphery: weird... That's all the questions I have for now. Thanks for the info. I can't see any possible way that could happen, unless there was a MySQL failure. On this particular upgrade, I allowed it to continue even with sql failures because a failure wouldn't be catastrophic, but I don't see any reason you could have a SQL failure there.
[19:21:45] stuartm: does anyone know of any free iptv (rtp) services which I can test against in the UK?
[19:21:54] stuartm: highly unlikely I know
[19:25:43] Beirdo: stuartm: thank you for that last commit, I was grumbling to myself about that a while back :)
[19:25:59] Beirdo: Why do I keep ending up at the main menu when I come to the TV... ooooh )
[19:26:01] Beirdo: :)
[19:27:07] stichnot: same here. :) I was thinking mythfrontend had crashed and my watcher had restarted it
[19:31:37] stuartm: next couple of commits would have left you in no doubt as to the reason, but I take your point
[19:32:05] stuartm: damn, just interrupted the timer I was testing to reply to that
[19:36:27] Beirdo: I like the feature though. It was just unexpected, and I will likely be one to set it to 0
[19:36:49] joki (joki! has quit (Ping timeout: 260 seconds)
[19:36:58] joki- (joki-! has joined #mythtv
[19:37:11] joki- is now known as joki
[19:42:34] stuartm: ok, I just need to do something about the default theme for the standby screen now ...
[19:42:59] joki- (joki-! has joined #mythtv
[19:43:23] joki (joki! has quit (Ping timeout: 245 seconds)
[19:43:23] joki- is now known as joki
[19:44:21] stuartm: I had a vague notion that we could do some other 'power saving' stuff when the frontend enters standby mode, slow down the UI redraw timer etc
[19:44:53] clever: stuartm: have you looked at the powertop program?
[19:44:53] stuartm: well, actually so far that's the only one I've come up with and I'm not even sure whether it would work
[19:45:12] stuartm: clever: yes and I can't see how I'd glean anything useful from it
[19:45:41] clever: it might help once you get under 1% cpu usage, when top starts to become less usefull
[19:46:00] clever: wakeups/sec would be more usefull then running_time/sec
[19:46:03] stuartm: I know usb wakeups happen a lot, but they are kinda necessary for usb devices and they have nothing to do with mythfrontend
[19:46:30] clever: just a quick check shows me that plugin-container is the worst right now
[19:46:35] ** clever goes to murder flash... **
[19:46:45] stuartm: tbh the savings to be had are so vanishingly small that it's only worthwhile for a battery powered device
[19:47:44] clever: let me see what my master backend says
[19:50:54] clever: mythbackend is causing 60 wakeups/sec
[19:51:09] clever: though it doesnt give much info on why
[19:51:23] clever: yeah, oprofile might be better, ran when you know its supposed to be idle
[19:53:31] clever: but its still a very small fraction of a gain, [Rescheduling interrupts] <kernel IPI> is causing 80% of the wakeups, no point in fixing myth yet
[19:57:28] clever: hmmm, that might have been nfs activity
[19:59:19] joki (joki! has quit (Read error: Operation timed out)
[19:59:25] skd5aner: well – seems the announcement for 0.25 beta was rather popular – 11,921 views on the release notes – I could be wrong, but I thought that was around 5000 when I looked about a month ago
[20:02:01] joki (joki! has joined #mythtv
[20:05:14] sphery: stuartm: FWIW, I'd love to limit the UI refresh timer to something more reasonable than 70/s always. I've been running this on my 0.24 system since I upgraded to 0.24 long ago), and the only down side is that the pulsing/animation is slower, but it's much easier on system/resource usage:
[20:05:45] stuartm: skd5aner: when I checked the other day we were a little behind on the release notes, are you able to catch up or would you like to put out a request for help?
[20:05:52] sphery: it's not time-based animation, nor does it take into account current screen refresh rate, but IMHO, it's better than a flat 70/s
[20:05:57] skd5aner: I would like someone to clarify – someone added the following line to "Special Notices and Instructions" area of the release notes, and I'd like to validate that it should indeed be called out – "After upgrading, remove the vdpaubuffersize filter using mythfrontend Setup, TV Settings, Playback, Playback Profiles (page 3), Edit, Custom profiles (2nd page) Custom filters:. Check all three VDPAU profiles (High Quality, Normal and
[20:05:57] skd5aner: Slim.)"
[20:06:42] stuartm: sphery: if we change it then we'll have to go through existing themes and correct the animation rates for those too
[20:06:50] skd5aner: stuartm: I'm about 2 weeks behind, but I've been reading and I know there's only about a half-dozen or so I need to add... I've actually just finished up a project for work that's been taking up my time and have a little window right now – should be up to date within the next couple of hours
[20:06:54] skd5aner: stuartm: thanks for asking
[20:07:07] SteveGoodey (SteveGoodey! has joined #mythtv
[20:07:39] sphery: skd5aner: yeah, it needs to be done, but should probably explicitly say the vdpau profile names, rather than assuming the user will figure it out... i.e. VDPAU Normal, VDPAU High Quality, and VDPAU Slim
[20:07:42] stuartm: skd5aner: ok, thanks for working on it :)
[20:07:43] skd5aner: back to my question above... does that setting /need/ to be removed? what happens if they don't?
[20:08:33] sphery: stuartm: or we just let animations cycle more slowly until themer updates them, if they want to
[20:08:36] skd5aner: sphery: can't a dbupdate take care of that?
[20:08:52] sphery: skd5aner: it could have if someone had done a db update to fix it
[20:08:53] stuartm: skd5aner: the dead key stuff for text input should get special mention, it's likely that affected a lot of people from 0.21 onwards
[20:08:54] skd5aner: sphery: surely doing it for the user woudl be way more desireable than asking the use to do it
[20:09:20] skd5aner: sphery: well.... I nominate you :D... I mean, I think that's good enough to classify as a bug-fix
[20:09:33] skd5aner: sphery: will it cause issues if a user doesn't remove the setting?
[20:10:12] skd5aner: stuartm: ok... would you like me to word something up when I get there or would you like to provide the entry for the dead key stuff?
[20:10:13] sphery: yes, but they were editing settings that weren't advertised and that were meant more for debugging than use, so...
[20:10:22] stuartm: we shouldn't be requiring a user to follow any special instructions after upgrading, since most won't ever read the release notes to see those instructions
[20:11:02] sphery: anyway, feel free to tell the guy who changed the meaning of the setting that he should have done a db update for it
[20:11:15] sphery: I'm just the messenger, so please point your gun somewhere else
[20:11:29] skd5aner: stuartm: agreed – but sometimes it's nice to call things out – for example, I might call out that all themes will now be obtained via the download manager... and in the last release, we needed to call out the need for folks to rescan their audio devices
[20:12:01] skd5aner: sphery: what happens if a user does nothing? is it ignored, or does it break something?
[20:12:07] stuartm: skd5aner: whatever you come up with is fine, the commit message covers it all iirc but I want to draw attention to it because it's not obvious that it's fixed unless you try it (and why would you when it's not worked for several releases)
[20:13:15] Beirdo: more for packagers than anything... Logging changes.
[20:13:16] skd5aner: np – thanks for letting me know – I might have skipped it otherwise – it might be worthwhile to send out a request to the other devs if there's anything else critical enough to call out there... anything they would expect a user would need to do or really need to know about as part of the upgrade (you get the gist)
[20:13:21] sphery: skd5aner:
[20:13:23] stuartm: sphery: did you happen to know who that was without having to look it up?
[20:13:50] skd5aner: stuartm: either I can, or if you'd like to (people might be more responsive if they see it coming from you vs me)
[20:14:04] stuartm: oh, Mark
[20:14:07] skd5aner: my guess is there's probably not much to call out, but...
[20:14:17] skd5aner: doesn't hurt to ask
[20:15:37] skd5aner: Beirdo: ok, good point – would you mind providing me a bullet point about what people/packagers would need to change? I know the re-write is covered in the release notes themselves, but if things will "Break" without intervention, that's would be good to call out in this section
[20:15:40] natanojl (natanojl! has joined #mythtv
[20:16:14] Beirdo: I think the mythbuntu guys have it solved their way already. Sure, I'll write one up later today
[20:16:43] stuartm: the random exit from recordings playlists being fixed might be worth a mention, although I'm only 99% sure that's fixed, I don't know for certain
[20:16:44] sphery: stuartm: so the end effect--as far as I understand what was said by Mark--is that now that we have automatic dynamic buffer sizing, MythTV uses what it thinks it needs and that setting now allows users to say to use more than MythTV thinks it needs... Since before it told MythTV to use exactly that number, those who set it have it way too high--and will likely have performance and/or memory issues when using the 0.24 values they set
[20:16:55] skd5aner: Beirdo: ty sir
[20:17:44] danielk22: sphery: If that's the case we should either rename the value or clear it out of people's settings..
[20:18:18] skd5aner: stuartm: sorry, forgot to put your name on the message I sent earlier: re:"np – thanks for letting me know..." – just wanted to make sure you knew I was talking to you
[20:18:26] noahric (noahric!~noahric@ has quit (Quit: noahric)
[20:18:37] sphery: I don't even know how that setting is implemented, since it's buried somewhere in the playback profiles stuff
[20:19:09] stuartm: sphery: as a filter I believe
[20:19:27] ** skd5aner is 60% sure he has that filter set **
[20:20:04] stuartm: those undocumented things you type as a comma separated list into a field labelled just as "Custom filters"
[20:20:32] skd5aner: correct
[20:20:39] stuartm: those things that should instead be selected from a list of checkboxes with full helptext
[20:20:56] skd5aner:
[20:22:41] skd5aner: Why not just add a line that says to ignore "vdpaubuffersize" in 0.25, if not just remove it from whatever row in the db it resides if it's there?
[20:23:04] ** skd5aner would volunteer, but would be warry of doing it wrong **
[20:23:54] skd5aner: I would think the former would be preferred, that way if someone enters it AFTER an upgrade, it doesn't matter
[20:24:00] skd5aner: it's just ignored
[20:26:08] danielk22: anyone object if I just rename it to vdpaubuffercount ?
[20:26:45] skd5aner: wouldn't it just be ignored? . . . ut_vdpau.cpp
[20:27:11] danielk22: That way any existing vdpaubuffersize lsettings are ignored and anyone that still needs the setting can use the new name.
[20:27:12] skd5aner: I mean, I could be totally wrong, but looking from line 1183 on, it wouldn't fit in any of those statements...
[20:27:42] skd5aner: or, duh... it's there
[20:28:50] skd5aner: danielk22: sorry if this is a dumb question, but if set will that filter attempt to overwrite the dynamic setting that was added this dev cycle?
[20:28:58] skd5aner: or does it "conflict"?
[20:29:39] danielk22: skd5aner: I believe it used to set the raw number of buffers and now it sets the number to use above what is required by the codec.
[20:29:45] sphery: stuartm: So, it turns out breaks DB upgrades. The Schema Wizard and upgrade code was designed so that UpgradeTVDatabaseSchema() was the only entry point for schema initialization and/or upgrade, and since you made InitializeDatabase() (now InitializeMythSchema()) non-static and called it directly in mythfrontend and mythtv-setup, the schema wizard doesn't know that the DB is brand new. This ...
[20:29:53] sphery: ... broke the schema wizard's code to not do DB backups and not ask for permission to upgrade the schema on initial schema creation. I have no idea what issues you were fixing there, so was hoping you could look at it to try to come up with a plan to fix it. (Note, also, the problem is visible in the logs when the backup code uses StorageGroup to figure out where to write and the prepare() fails with a DBError and also because we never ...
[20:29:57] sphery: ... output, "Database appears to be empty/new!".)
[20:31:09] skd5aner: danielk22: yea, that's basically what I was getting... so, it seems to me that having it set doesn't really "break" anything explicitely – it does what it did before, but it actually just over-rides it trying to automatically use an appropriate number of buffers – correct?
[20:31:11] sphery: danielk22: I think changing the name is a great approach since it will prevent issues with all the people who add random values referred to in old posts Google finds for them in the list archives
[20:31:31] skd5aner: danielk22: so, it's advised to have it unset, but having it set just reverts back to the previous behavior of having an explicit buffer count
[20:31:49] skd5aner: danielk22: does that sound right?
[20:34:43] sphery: skd5aner: as I said (here and in the post I linked) and danielk22 also said, setting it seems to use the value specified plus what mythtv thinks is appropriate--so you end up with too many buffers and you have resource issues on the graphics card
[20:34:56] skd5aner: if that is the case, I think the question would be, would there be any reasonable expectation for anyone to ever explicitly set the buffer size going forward if mythtv tries to set an appropriate size dynamically based on the content? If the answer is "no" – then I would just remove the "if (name.contains("vdpaubuffersize"))" section entirely
[20:35:10] sphery: danielk22: as for a new name, perhaps something like extravdpaubuffers (to indicate it's actually extra in excess of what we think we need)
[20:35:27] skd5aner: sphery: an additive effect? hmmm, that's bad
[20:35:53] sphery: skd5aner: in theory there's no reason at all for any user to set it with current code--unless we have a bug in the code that tries to decide how many buffers are needed
[20:36:38] sphery: so, removing it may be premature, for now, especially with all the broken-/poorly-encoded videos people try to play with MythTV :)
[20:36:40] skd5aner: yea – which is why I would propose removal over rename
[20:37:27] sphery: either way works for me, but I wouldn't argue with keeping it around at least one more release to see how it goes
[20:37:30] stuartm: sphery: if the schema is new then then DBSchemaVer will be 1280, no? i.e. whatever the schema creation initialises it to
[20:37:33] kormoc (kormoc!~kormoc@mythtv/developer/kormoc) has quit (Ping timeout: 244 seconds)
[20:38:26] sphery: stuartm: yes, but it could be 1280 even when not new... granted, when I do a db rollup (when I'm reasonably sure no one is going to do another db update), the new schema will be initialized to the current schema version
[20:38:48] sphery: so only people who run master after 0.25 will see this issue
[20:38:53] stuartm: so abstract out the 'InitialSchemaVer', if (DBSchemaVer == InitialSchemaVer) { dont_prompt = true }
[20:39:23] sphery: and since schema wizard needs to be completely rewritten, I don't mind if we leave it as is for now
[20:40:08] sphery: (I prefer as is over changing the new db stuff to try to detect a 1280 db as new)
[20:40:38] sphery: especially since that would probably cause bugs when I do the rollup to current schema version
[20:40:55] stuartm: fwiw, this all worked as it should back when I made that initialisation change, so I'm not sure what's changed since that time
[20:43:20] sphery: more than anything, my concern was that the schema wizard code (which I don't know, except for a couple parts I've fixed in recent years) seems to be designed around the assumption that it knows, but since it took me a year to investigate why it was asking me to upgrade an initial schema and we haven't had any reports of issues, I'm guessing that it's not that big a deal
[20:43:33] kormoc (kormoc!~kormoc@mythtv/developer/kormoc) has joined #mythtv
[20:43:55] stuartm: obviously the schema upgrade won't even try to run if there are no upgrades beyond InitialSchemaVer, if Upgrade always starts at Initial + 1 then there would be no problem
[20:44:29] stuartm: this all makes sense in my head, I'm aware I'm probably not conveying it very well
[20:45:13] sphery: stuartm: so I guess the main question is, do we really need to initialize the schema, then do 100lines of other stuff, then do db upgrade? why can't we just do a UpgradeTVDatabaseSchema() where you do the InitializeMythSchema()?
[20:45:53] sphery: UpgradeTVDatabaseSchema() will initialize and upgrade the schema (so that the 100lines of code actually knows what to expect of the db schema and data)
[20:46:06] stuartm: sphery: I did a lot of testing starting from an empty db back when I made that change, it always worked perfectly and without prompting that I recall
[20:47:30] stuartm: sphery: I can't remember the reason why I split the schema creation from the upgrade, but I wouldn't have done it on a whim
[20:49:35] stuartm: I can only assume that what I said in that commit was relevant, namely that "These changes should ensure that the user is always prompted for language/locale as the first thing we do ... it guarentees that all subsequent dialogs are in the chosen language"
[20:50:30] stuartm: so the user is never shown a prompt of any kind which is in English, a language they likely won't understand
[20:50:43] RingZer0 (RingZer0!~ringzer0@unaffiliated/ringzer0) has joined #mythtv
[20:50:44] stuartm: but beyond that, I can't give specific
[20:50:46] stuartm: s
[20:51:18] sphery: but perhaps your "early settings" stuff makes that no longer required (the code you added to save settings to a list until you can dump them to the db once we have a valid db/schema)?
[20:51:19] stuartm: if the 'upgrade' process prompts at any point that would be the reason for splitting them
[20:52:16] RingZer0: hey guys, I have an apple tv that I have just jailbroke and installed xbmc with mythbox frontend. however, the first requirement in the readme is mythtv. is there a repo I can import via root/ssh, or how is this done?
[20:52:39] stuartm: sphery: iirc the early settings stuff just stores settings until they can be written to the DB, it doesn't ensure that code reading settings get those instead
[20:52:45] stuartm: could be remembering that incorrectly
[20:52:52] skd5aner: RingZer0: wrong channel, try #mythtv-users
[20:53:06] skd5aner: RingZer0: but even there, they probably won't be able to help much with the xbmc piece
[20:53:29] RingZer0: sorry
[20:53:32] skd5aner: np
[20:53:41] stuartm: RingZer0: you probably want #xbmc, since their mythtv integration has nothing to do us
[20:54:11] skd5aner: but he still has to install mbe somewhere it sounds like
[20:54:13] skd5aner: so...
[20:54:25] skd5aner: but, I digress
[20:54:28] RingZer0: actually my xbmc works great, I just need MythTV installed on my darwin :)
[20:54:39] stuartm: they want to integrate with mythtv, they can field the support requests ;)
[20:54:55] ** RingZer0 reads channel topic a bit late. **
[20:54:56] skd5aner: RingZer0, join #mythtv-users and start there – this channel is for development talk only
[20:55:10] RingZer0: well, maybe i'll like it enough to join the community :D
[20:55:27] kth (kth!~kth@unaffiliated/kth) has joined #mythtv
[20:55:32] RingZer0: make the 'wife' happy, lol, thanks guys.
[20:55:54] danielk22: stuartm: you reported 2 years ago in #7486 that the patch didn't work..
[20:58:06] stuartm: so I did, but a couple of other users reported otherwise and since I was only able to test it with DVB-T I'm not sure whether my observation was valid or not
[20:58:34] stuartm: I can't remember anything about it 2 years later :/
[20:58:54] danielk22: stuartm: The patch doesn't seem to add any NIT filtering as you might expect. It seems to be pretty specific to a specific cable provider's munging of NIT tables.
[21:00:08] stuartm: << Work in progress, but a taste of what can be done now mythmusic uses mythui
[21:00:27] sphery: stuartm: yeah, looks like early settings stuff doesn't help, and actually causes problems since it can't write the db schema version, so it's probably best to leave it as is and fix it by replacing the schema wizard with one that works with mythui
[21:01:34] stuartm: danielk22: ok, we can push that to 0.26 and have stuarta look at it :)
[21:02:40] danielk22: heh
[21:04:18] Captain_Murdoch (Captain_Murdoch!~cpinkham@mythtv/developer/CaptainMurdoch) has quit (Ping timeout: 255 seconds)
[21:07:47] stuartm: if we can get it fixed for 0.25 then great, but if it's going to take up too much time, time which could be better spent on two or more other bugs instead then ...
[21:09:15] noahric (noahric!~noahric@ has joined #mythtv
[21:09:22] stuartm: two years ago I still remembered enough about the DVB spec and our channel scanning code to look at the ticket and test the patch, today I remember only the most basic aspects of DVB and nothing of the the scanning code so I'm not going to be much help
[21:17:51] Captain_Murdoch (Captain_Murdoch! has joined #mythtv
[21:17:52] Captain_Murdoch (Captain_Murdoch! has quit (Changing host)
[21:17:52] Captain_Murdoch (Captain_Murdoch!~cpinkham@mythtv/developer/CaptainMurdoch) has joined #mythtv
[21:20:06] Beirdo: still got two plugins warnings for unused parameter "back" in synaethesia and bumpscope
[21:20:20] Beirdo: anyone planning on murdering those?
[21:21:27] Beirdo: I think I may look at reimplementing (should we still need it) the pxsup2dast.c stuff for a future release
[21:25:08] joki (joki! has quit (Ping timeout: 272 seconds)
[21:35:11] SteveGoodey (SteveGoodey! has quit (Remote host closed the connection)
[21:35:29] joki (joki! has joined #mythtv
[21:40:54] zombor (zombor!~zombor_@kohana/developer/zombor) has quit (Ping timeout: 260 seconds)
[21:45:23] danielk22: stuartm: The one in my queue that bugs me is #9830. It just looks like we never get any data after a channel change on a PVR-150/350, but I haven't a clue why.
[21:47:59] xavierh_ (xavierh_! has joined #mythtv
[21:51:22] stuartm: danielk22: could it be a change external to mythtv, in IVTV?
[21:57:11] stuartm: a quick get suggests the last changes to our ivtv code were in 2007, so it would be strange for that to suddenly start failing unless the drivers are doing something differently
[21:57:50] stuartm: the ticket was opened before the rec2 merge, so that doesn't leave many suspects
[21:59:50] danielk22: stuartm: It's possible, something that ivtv depends on could have changed.
[22:00:13] danielk22: The big problem is that I don't even have any analog source to test with anymore.
[22:00:44] joki (joki! has quit (Ping timeout: 244 seconds)
[22:01:28] stuartm: neither do I, I might still have the PVR-150 stored away somewhere, but nothing I can connect to it
[22:01:47] jams: i can probably set something up to test the ivtv theory
[22:01:59] jams: i have an analog channels still
[22:02:48] jams: if that would be of value?
[22:03:14] stichnot (stichnot!chatzilla@mythtv/developer/stichnot) has quit (Ping timeout: 260 seconds)
[22:03:16] danielk22: jams: Yeah, anything to isolate the problem would be useful.
[22:03:50] jams: alright..when is the last known good version of mythtv to start the test?
[22:04:04] danielk22: 0.23 :)
[22:04:27] danielk22: I still had analog then...
[22:04:39] jams: works for me.. i will get working with .23 then update myth and only myth
[22:05:02] xavierh_: stuartm: I am going to move standardsetting classes to libmyth. I know you like to remove libmyth. But I need to have access to them from the plugins too. You will still be able to move them somewhere else later on.
[22:05:11] danielk22: jams: thx!
[22:10:23] xavierh_ (xavierh_! has quit (Read error: Connection reset by peer)
[22:13:24] xavierh_ (xavierh_! has joined #mythtv
[22:17:20] natanojl: stuartm: I seem to have the same edition of LOTR and the same problem as Otto reported in #9429. The non-English subtitles seem to be offset 4 steps, i.e. the menu shows 1: English, 2: Danish, 3: Swedish etc, but selecting 6: Icelandic gets you Danish. Selecting 7: Norwegian gets you Swedish. VLC shows the same behavior while mplayer works ok
[22:17:23] stuartm: jams: great!
[22:18:42] stuartm: natanojl: that's good news
[22:19:18] stuartm: that someone is able to reproduce and possibly do some debugging for us I mean
[22:20:31] stuartm: I suspect it's an ffmpeg bug, dvdnav seems to report the tracks correct, but we rely on libav instead and it appears from the logs to have trouble
[22:20:33] aloril (aloril!~aloril@ has quit (Ping timeout: 265 seconds)
[22:25:07] stuartm: natanojl: I may ask you to test some debugging patches, but maybe not before 0.25 is released as I've got other tickets to work through first
[22:32:28] Jordack (Jordack! has quit ()
[22:33:51] aloril (aloril! has joined #mythtv
[22:35:46] natanojl: stuartm: I too suspect ffmpeg given that VLC shows the same behavior.
[22:37:25] kth (kth!~kth@unaffiliated/kth) has quit (Quit: Leaving.)
[22:39:07] stichnot (stichnot!~chatzilla@mythtv/developer/stichnot) has joined #mythtv
[22:40:16] stuartm: a resync might fix it, there may be one for 0.26, failing that we could think about using the track list from dvdnav instead of the one from libav but I've no idea how much work that might involve
[22:41:29] stuartm: clearly if dvdnav can determine the track order correctly then ffmpeg should be able to as well, so getting the issue fixed upstream is preferable
[22:44:28] joki (joki! has joined #mythtv
[22:46:01] skd5aner: stichnot: ping
[22:48:04] stuartm: I wonder if it's purely an enumeration bug, I'll have to look at the logs again
[22:49:57] stichnot: skd5aner: available in 15 minutes
[22:50:12] skd5aner: stichnot: k, thx – just ping me back when free
[22:51:23] j-rod is now known as j-rod|afk
[22:51:28] natanojl: stuartm: First chance for trying any debugging patch might be in the weekend since I've been quite busy at work lately.
[22:52:21] stuartm: natanojl: won't be before then, I'm not going to have the time
[22:53:29] davide_: danielk22: for analog debugging, it might not work ideally or at all for channel changing, but don't you have an stb with an rf output?
[22:55:58] Dave123 (Dave123! has quit (Quit: Leaving)
[22:58:05] MavT (MavT! has joined #mythtv
[22:58:07] skd5aner: stuartm: add the deadkey changesets now... not sure they're significant enough to call out in the special notes section – unless I'm missing something (happy to add it if you feel otherwise)?
[22:59:49] MaverickTech (MaverickTech! has quit (Ping timeout: 260 seconds)
[23:00:43] sphery: skd5aner: it's probably worth calling out that they're supported directly in the text boxes rather than requiring use of the on-screen keyboard, just so users who need them don't keep using the on-screen keyboard and its compose functionality--since, like stuartm said, they're unlikely to expect it to work, and would probably just keep doing it the hard way
[23:01:00] jams: stuartm danielk22 . I have one pvr-150 working with myth .24 and changing channels with it. Anything else I should setup before compiling .25 beta?
[23:01:18] skd5aner: sphery: gotcha... I'll figure out a way to highlight it – thx
[23:02:28] skd5aner: jams: I've got a PVR-250 and PVR-500 on .24 too, if it helps danielk22 and stuartm to test something
[23:03:07] stuartm: jams: 0.24.0, 0.24.1 or 0.24.2?
[23:03:18] jams: well i only tested .24 because I had the install handy was going to start on .23 but .24 was preinstalled
[23:03:20] jams: let me check
[23:04:08] jams: i'm going with .24.1
[23:04:09] stuartm: if 0.24.2 is fine then we can rule out any commits which were backported
[23:04:21] joki (joki! has quit (Ping timeout: 248 seconds)
[23:04:43] jams: but I can certainly test .24.2 will just take a bit to compile
[23:05:21] stichnot: skd5aner: back now.
[23:05:30] stuartm: skd5aner: there's a ticket open for breakage with the PVR-350 and possibly PVR-150 when changing channels, it just stops streaming data
[23:05:34] natanojl: stuartm: #9407 affects me as well btw
[23:06:05] jams: stuartm MythTV Version  : g294968b
[23:06:05] jams: MythTV Branch  : 0.24-fixes
[23:06:06] jams: Network Protocol : 63
[23:06:06] jams: Library API  : 0.24.20110505–1
[23:06:14] stuartm: jams: maybe better to jump to 0.25, faster at least if we just want to rule in/out ivtv
[23:06:16] jams: where does that fall on the release schedule ?
[23:06:19] skd5aner: stichnot: quick q for you, trying to summarize your recent commits related to the OSD font settings...
[23:06:33] jams: k..downloading now
[23:06:37] stuartm: jams: 0.24.0 ?
[23:06:50] jams: could be
[23:07:02] jams: checking out .25 now
[23:07:05] skd5aner: stichnot: is this an accurate statement for the release notes, "* FreeMono is now set as the default font for CEA-608, CEA-708, teletext, and SRT subtitles ({{gitcommit|25fe663}, {{gitcommit|6202c0d}}, {{gitcommit|6f24f3b}})"
[23:07:39] stuartm: jams: actually no, that's pretty recent, 0.24.2
[23:07:53] stuartm: Date: Wed Jan 4 19:05:15 2012 +0000
[23:08:35] skd5aner: stichnot: I got a litte lost with the back and forth around that time frame – just wanted to make an accurate statement :)
[23:08:57] stichnot: skd5aner: yes, that sounds good.
[23:09:03] skd5aner: cool, thanks
[23:09:06] stuartm: natanojl: is harder to fix, I don't know the dvdnav/highlight code well enough
[23:09:32] jams: stuartm alright then. proceeding with a fresh .25 checkout now
[23:09:43] skd5aner: stichnot: from one of your commits, I thought it actually pointed to FreeSans or something
[23:09:51] skd5aner: stichnot: (as the default)
[23:10:16] stichnot: skd5aner: early on that was true. I'll try to summarize here:
[23:10:30] natanojl: stuartm: I don't know it at all :) Anyway, time for some sleep. Good night
[23:10:43] jams: stuartm, while we are waiting do you think I need more then one card defined?
[23:10:54] jams: can add a hdhomerun easy enough
[23:12:13] sphery: jams / stuartm : FWIW, says he thinks it was backported (and specified an actual commit number)
[23:12:28] stuartm: jams: the way it's described on the ticket I don't think that's necessary
[23:13:58] stichnot: skd5aner: The existing default for all (except 708) was FreeSans, so many users had that set in their DB. FreeSans looks terrible for teletext, since each character is individually placed onto a grid. 608 is supposed to use a monospace font, but the alignment issues aren't so critical. We decided to wipe the slate clean and change the name of the setting, using FreeMono as default. Note...
[23:13:59] stichnot: ...that originally, the 708 default font was hard-coded as Droid Sans Mono.
[23:14:11] stuartm: sphery: right, but earlier in that same ticket he says 0.24 worked fine and the problem only started when he upgraded to 0.25
[23:15:09] natanojl (natanojl! has quit (Ping timeout: 252 seconds)
[23:16:05] skd5aner: stichnot: k – when I update the release notes here (in about ~30 mins I'll be done and will submit my changes) take a look and make sure you're happy with the wording – don't want to misinterpret/misrepresent the changes, but I think I've got it accurate :)
[23:16:06] stuartm: and although he talks about that commit, he doesn't actually connect it to the issue he's seeing, he's rambling a lot :(
[23:16:19] skd5aner: stichnot: I also mentioned your little note about it in the special notes section at the top... you'll see what I mean
[23:16:26] stichnot: skd5aner: I think it would be nice to point out in the release notes that CEA-708 captions are now fully functional in 0.25, thanks mostly to the timing fixes, and this hasn't been the case in 0.24 or earlier.
[23:16:48] skd5aner: thanks – I'll be sure to explicitly say so
[23:16:49] stichnot: ok, I'll have a look first thing.
[23:17:12] stuartm: increasing a tuning timeout, which is what that commit does, shouldn't make tuning more likely to fail
[23:17:59] joki (joki! has joined #mythtv
[23:18:37] sphery: stuartm: yeah, looking at code, it seems kLiveTVOpenTimeout is 10s, now, so no clue what he meant
[23:19:34] stuartm: it could have knock-on affects if tuning has failed, namely other timeouts hit and instead of exiting cleanly from playback it gets much messier but those are the after affects not the cause
[23:20:55] stuartm: I've noticed that we don't handle errors in livetv too gracefully, it can get pretty confused if things don't happen exactly as they were expected to
[23:21:38] sphery: stichnot: quick question... This and its related confuses me... As I grep the code, it looks like the video playback window is the only consumer of EXIT_TO_MENU event, and it seems that you've just added code to send EXIT_TO_MENU to all listeners registered with MythObservable right before we send the event to the video playback window... So, it ...
[23:21:44] sphery: ... looks like it's just spamming that message for each screen popped off the stack--maybe just seems to work because it changes timing on a race condition?
[23:22:04] sphery: note that you can see the start of where it's sent right after you send it in the 2nd changeset
[23:22:29] joki (joki! has quit (Ping timeout: 244 seconds)
[23:22:57] sphery: . . . ow.cpp#L1439 is the current code
[23:38:00] stichnot: sphery: As far as I could tell, there is no screen called "video playback window" in the system, so that particular event is never sent.
[23:38:16] jams: compile finished. Just a note. mythfrontend segfaults when trying to load plugins from .24
[23:38:30] stichnot: I didn't want to clean it up any more at this point in the release.
[23:39:21] jams: stuartm ^^
[23:40:52] sphery: stichnot: Ah, so that may be the real problem... Maybe just need to change conditional to use the right name for the playback window (assuming we still have a specific window for playback). If nothing else, we probably want to get rid of the bad conditional (but that could wait 'til after 0.25)...
[23:41:36] stichnot: sphery: also, ExitToMainMenu() is only called once at the beginning of jumppoint execution. The rest of it is handled by the topmost screen firing off an escape key event and propagating that event through the rest of the screens.
[23:41:39] sphery: jams: yeah, that's been happening for several releases, now... haven't made time to look into it (and figure if nothing else, it helps users clean up their installations :)
[23:41:43] jams: stuartm, danielk22 with .25 channel changing with a pv150 is very iffy. sometimes it works, sometimes it takes a really long time (5+ seconds) should take less then 2. Finally sometimes it outright fails with "video frame buffering failed to many times"
[23:42:34] jams: sphery- hehe sure..just was odd. The first time it told me which plugin it failed, after that it just segfaulted with no reason
[23:42:45] stichnot: I want to overhaul the whole mechanism for 0.26, so I try to minimize the changes (i.e. risk) in the meantime.
[23:43:55] jams: when i get the frame buffering error. reentering live-tv puts me on the correct channel (the one I tried to change to)
[23:45:17] stichnot: sphery: I believe the actual name of the video playback window is "TV", but for whatever reason it is not a child/descendant of the main window, which is why the event had to be blasted to the world.
[23:46:50] sphery: stichnot: OK. I thought the postEvent(this, key); was resulting in the method being called recursively until the screen was mainmenu, but I admit I don't know that code. It does make sense to minimize changes now, and we do need to fix that with the PopScreen() approach, so it will be OBE "soon", I hope.  :)
[23:47:51] sphery: ("soon" becomes much more possible, too, now that xavierh_ has got a good approach and patches for the settings conversion)
[23:48:40] stichnot: yeah, I'll be very happy not to have to deal with the legacy stuff.
[23:50:07] xavierh_: me too, my path regarding input event hit the legacy code, gave me some motivation :)
[23:51:11] zombor (zombor!~zombor_@kohana/developer/zombor) has joined #mythtv
[23:54:03] xavierh_: s/path/patch
[23:55:38] zombor (zombor!~zombor_@kohana/developer/zombor) has quit (Remote host closed the connection)
[23:58:14] davide_ (davide_!~david@mythtv/developer/gigem) has quit (Remote host closed the connection)
[23:58:15] Guest29306 (Guest29306! has quit (Quit: WeeChat 0.3.7)

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