| Tuesday, January 28th, 2020, 00:05 UTC | ||
| [00:05:32] | hampton: | I just committed the patch to use 'auto *' or 'const auto *' where appropriate. |
| [00:10:47] | peterbennett: | I don't know why dynamic_cast does not work in that case. I think if dynamic_cast was not working at all that the frontend would not run. |
| [02:39:59] | gigem: | peterbennett: mythbuild.sh does remove mythinstall when the fresh option is used. |
| [02:41:31] | gigem: | peterbennett: Have you been getting bounces from email to markspieth? I've gotten two now stating "There was a temporary problem delivering your message...". |
| [02:43:22] | markspieth: | that's strange! I'll have to look into that. |
| [02:53:04] | gigem: | markspieth: Do you an email other than mark@digivation.com ? I can forward you one of the bounces? |
| [02:54:52] | gigem: | fyi to any shield users out there. The 8.0.2 experience update from nvidia changes the MENU key to bring up the shield's settings menu by default. If you use that key in mythtv, you have to go into the shield settings and the MENU behavior to send MENU to the application. |
| [02:55:52] | gigem: | s/shield settings and the MENU/shield's settings and change the MENU/ |
| [03:02:45] | markspieth: | sent via email. |
| [03:03:27] | markspieth: | note I still use my airmouse. no menu behaviour change there. |
| [03:25:45] | mad_enz (mad_enz!~mad_enz@2606:6d00:135:9d84:9cf:32ea:3e7b:cae2) has quit (Remote host closed the connection) | |
| [03:26:38] | mad_enz (mad_enz!~mad_enz@2606:6d00:135:9d84:888a:73ea:bbe1:d23a) has joined #mythtv | |
| [03:26:55] | mad_enz (mad_enz!~mad_enz@2606:6d00:135:9d84:888a:73ea:bbe1:d23a) has quit (Remote host closed the connection) | |
| [04:27:59] | gigem: | markspieth: I forwarded one of the bounce messages to the address you sent. |
| [04:55:29] | peterbennett: | gigem: markspieth: markspieth2: I assume that Mark's email address is digivation.com.au, but somehow I have had digivation.com cc'ed on some emails and those get bounced. Probably a typo on my part. |
| [05:01:10] | markspieth: | correct |
| [05:04:12] | markspieth: | OT if you have seen 'Better off Ted' there are references to digivation (the us one). very funny. |
| [05:06:04] | markspieth: | all my investigations point to dynamic cast should be working. qt and myth have the correct type info. need that test project now. |
| [05:06:36] | peterbennett: | I don't know better off ted |
| [05:07:03] | peterbennett: | I am wondering if all dynamic casts are failing or just that one. |
| [05:07:17] | peterbennett: | Gotta go now — goodnight |
| [05:11:00] | markspieth: | there are others like delete show before exit watching. delete group when deleting last show. |
| [06:50:00] | peterbennett (peterbennett!~pi@mythtv/developer/peterbennett) has quit (Quit: WeeChat 2.3) | |
| [07:00:59] | peterbennett (peterbennett!~pi@mythtv/developer/peterbennett) has joined #mythtv | |
| [07:00:59] | Mode for #mythtv by ChanServ!ChanServ@services. : +v peterbennett | |
| [07:03:39] | markspieth22 (markspieth22!~yaaic@mythtv/developer/markspieth) has joined #mythtv | |
| [07:03:39] | Mode for #mythtv by ChanServ!ChanServ@services. : +v markspieth22 | |
| [07:06:17] | markspieth (markspieth!~yaaic@mythtv/developer/markspieth) has quit (Ping timeout: 264 seconds) | |
| [07:32:04] | markk_ (markk_!~mark@host109-151-194-111.range109-151.btcentralplus.com) has joined #mythtv | |
| [08:07:11] | Steve-Goodey (Steve-Goodey!~steve@2a00:23c5:7d90:bc01:9be6:831c:4ed8:495d) has joined #mythtv | |
| [08:07:11] | Mode for #mythtv by ChanServ!ChanServ@services. : +v Steve-Goodey | |
| [08:39:06] | Steve-Goodey: | Forum hit a new record? Most users ever online was 579 on Tue Jan 14, 2020 8:23 am |
| [08:44:15] | stuarta: | \o/ |
| [08:52:31] | Merlin83b (Merlin83b!~Daniel@2001:8b0:bbd6:bd9c:e9db:f3d6:6e37:eedc) has joined #mythtv | |
| [09:01:11] | jheizer (jheizer!~jon@c-73-51-91-251.hsd1.il.comcast.net) has quit (Ping timeout: 268 seconds) | |
| [09:20:05] | markk_: | can someone remind me what our custom mpegts demuxer (in ffmpeg) does? I seem to remember it handles stream changes properly/better. I'm looking at #13557 – playback is a mess with our own demuxer but enabling 'Use FFmpeg's original MPEG-TS muxer' and it plays perfectly. |
| [09:20:05] | ** MythLogBot https://code.mythtv.org/trac/ticket/13557 ** | |
| [09:20:33] | stuarta: | i believe it has todo with subtitles |
| [09:26:34] | markk_: | hrm – strange. ffprobe, mediainfo and our demuxer find 6 streams for that file – but using the ffmpeg demuxer internally it only finds 4 |
| [09:28:19] | markk_: | ah – and an old sample from another ticket that doesn't play well – also plays fine with ffmpeg demuxer – and same finnish channel. |
| [09:29:54] | markk_: | and likewise – only 4 of 6 streams detected |
| [09:38:34] | enyc_ (enyc_!~enyc@muddle.enyc.org.uk) has joined #mythtv | |
| [09:44:22] | stuarta: | what does ffprobe spit out for that sample? |
| [09:50:18] | SteveGoodey (SteveGoodey!~steve@host86-150-60-252.range86-150.btcentralplus.com) has joined #mythtv | |
| [09:50:18] | Mode for #mythtv by ChanServ!ChanServ@services. : +v SteveGoodey | |
| [09:56:19] | stuarta: | blimey, trac released 1.4 back in august |
| [09:56:33] | stuarta: | still not sure i'd bother moving to that |
| [09:56:45] | markk_: | stuarta: I've just dumped a load of data into the ticket #13557 |
| [09:56:45] | ** MythLogBot https://code.mythtv.org/trac/ticket/13557 ** | |
| [10:00:45] | stuarta: | so the main difference is in the dvb_subtitle streams |
| [10:06:08] | markk_: | stuarta: the main problem is that our demuxer does not detect the video size/format correctly. which leads to failure of hardware decoding, followed by numerous errors as we fallback to software decode with a guessed video size, which is then updated to the correct size... not a pretty sight:) |
| [10:07:04] | enyc_ is now known as enyc | |
| [10:07:34] | markk_: | it is also strange that ffprobe does a better job of finding streams than using the same demuxer in our own code. |
| [10:08:51] | stuarta: | scan duration? |
| [10:09:48] | markk_: | possibly – but I've already maxed out max_analyze_duration |
| [10:09:56] | markk_: | 60 seconds anyway |
| [10:10:36] | stuarta: | another thought, is there some bad feedback coming from the dvb_subtitle streams which is throwing off the video analysis? |
| [10:10:54] | stuarta: | i'm not really sure *how* that would happen tho, or even if it's possible |
| [10:20:48] | markk_: | stuarta: no idea – not really an area I know much about. Is our demuxer based on the FFmpeg version? is it kept in sync. no idea:) |
| [10:26:43] | stuarta: | AFAIK yes |
| [10:27:07] | stuarta: | peterbennett is the one to ask about keeping those in sync |
| [10:55:07] | markk_ (markk_!~mark@host109-151-194-111.range109-151.btcentralplus.com) has quit (Ping timeout: 265 seconds) | |
| [11:02:35] | markk_ (markk_!~mark@host109-151-194-111.range109-151.btcentralplus.com) has joined #mythtv | |
| [11:05:58] | SteveGoodey (SteveGoodey!~steve@host86-150-60-252.range86-150.btcentralplus.com) has quit (Quit: Konversation terminated!) | |
| [11:21:22] | markk_: | markspieth2: would it be possible to add libfftw3 to the android build? swap you for 4k support:) |
| [11:24:16] | markk_: | warpme_: do you have any idea about how to deal with #13538. The only solution I can think of at the moment is to disable opengle by default – which I don't really want to do. I'm assuming you are doing something different in your builds as I don't think you've had that problem. |
| [11:24:16] | ** MythLogBot https://code.mythtv.org/trac/ticket/13538 ** | |
| [11:24:32] | markk_: | opengl es |
| [11:56:41] | markspieth22 (markspieth22!~yaaic@mythtv/developer/markspieth) has quit (Ping timeout: 264 seconds) | |
| [12:13:10] | markk_: | warpme_: are you still seeing issues with deinterlacing in master? Just tried to recreate your problem but seems to be working here. |
| [12:18:16] | markspieth (markspieth!~yaaic@mythtv/developer/markspieth) has joined #mythtv | |
| [12:18:16] | Mode for #mythtv by ChanServ!ChanServ@services. : +v markspieth | |
| [12:26:39] | mad_enz (mad_enz!~mad_enz@2606:6d00:135:9d84:bd31:9dae:faef:a273) has joined #mythtv | |
| [13:06:20] | markspieth2: | markk_: sure. Only used for mythmusic in the past. |
| [13:52:36] | jheizer (jheizer!~jon@c-73-51-91-251.hsd1.il.comcast.net) has joined #mythtv | |
| [13:52:36] | Mode for #mythtv by ChanServ!ChanServ@services. : +v jheizer | |
| [15:24:30] | hampton: | What version of android should I be using? I installed the latest sdk yesterday, and installed the ndk side-by-side builds. I have android studio 191, and an android-ndk link pointing to android-sdk-linux/ndk/20.1.5948944. Trying to run the makelibs.sh script fails while compiling icu with a couple of undeclared identifiers. |
| [15:24:41] | ** hampton is trying to update his old android build setup. ** | |
| [15:25:47] | peterbennett: | hampton: I renamed the ~/android and created a new one from the latest android studio |
| [15:26:19] | peterbennett: | I have updated the readme in the git package android directory |
| [15:27:24] | hampton: | OK, I'll give that a try. |
| [15:29:41] | peterbennett: | hampton: Also remove the build64, libs64, mythinstall64 directories |
| [15:30:08] | peterbennett: | I am using android studio 3.5.2. |
| [15:31:06] | peterbennett: | Install the ndk from in there, latest ndk 21.0.6113669 |
| [15:31:42] | peterbennett: | android studio by default now installs in Android, create a link from android to Android |
| [15:31:45] | hampton: | I downloaded 3.5.3 yesterday, although the file is labeled ide-191.6010548. |
| [15:32:29] | hampton: | API: 29 is what studio is offering to install |
| [15:35:55] | peterbennett: | gigem: markspieth: markspieth2: hampton: Changing setenv – I am changing it to use a file called buildrc in the packaging/android directory for local parameters such as your keystore and setup of make.inc |
| [15:45:05] | gigem: | peterbennett: k. |
| [15:46:17] | hampton: | k |
| [15:46:43] | stuarta: | sounds like a cunning plan to me |
| [15:48:03] | markk_ (markk_!~mark@host109-151-194-111.range109-151.btcentralplus.com) has quit (Ping timeout: 260 seconds) | |
| [15:50:34] | gigem: | stuarta: Oh, and your recollection matches mine regarding our demuxer. It's mainly for some forms or rare subtitles. It used to also be for stream changes too but I think ffmpeg has provisions for that now. |
| [15:52:30] | hampton: | ok, running "makelibs.sh all" again. |
| [16:06:59] | peterbennett: | gigem: markspieth2: hampton: OK I pushed the setenv.sh changes. remove setenv.sh from your ~/android and create packaging/android/buildrc for amny overrides you had in setenv.sh. |
| [16:08:02] | peterbennett: | markspieth2: I changed mythbuild to remove IGNOREDEFINES. Please add the IGNOREDEFINES to your buildrc file. Thank you! |
| [16:19:38] | markk_ (markk_!~mark@host109-151-194-111.range109-151.btcentralplus.com) has joined #mythtv | |
| [16:26:46] | hampton: | ok, libs built this time. myth compile running. will check completion in a couple of hours. |
| [16:26:48] | hampton: | thanks peterbennett |
| [16:27:53] | markspieth (markspieth!~yaaic@mythtv/developer/markspieth) has quit (Ping timeout: 264 seconds) | |
| [16:29:43] | markspieth (markspieth!~yaaic@mythtv/developer/markspieth) has joined #mythtv | |
| [16:29:43] | Mode for #mythtv by ChanServ!ChanServ@services. : +v markspieth | |
| [16:31:16] | markk_: | gigem: I've just done a test of livetv using the ffmpeg demuxer – and it doesn't work well:( |
| [16:47:22] | markspieth (markspieth!~yaaic@mythtv/developer/markspieth) has quit (Ping timeout: 252 seconds) | |
| [16:51:58] | markspieth (markspieth!~yaaic@mythtv/developer/markspieth) has joined #mythtv | |
| [16:51:58] | Mode for #mythtv by ChanServ!ChanServ@services. : +v markspieth | |
| [16:52:08] | ooshlablu (ooshlablu!~ooshlablu@98.217.124.239) has quit (Remote host closed the connection) | |
| [16:53:07] | ooshlablu (ooshlablu!~ooshlablu@98.217.124.239) has joined #mythtv | |
| [17:27:23] | peper03 (peper03!~peper03@mythtv/developer/peper03) has quit (Ping timeout: 265 seconds) | |
| [17:30:10] | gigem: | markk_: What does it do? |
| [17:31:59] | peper03 (peper03!~peper03@mythtv/developer/peper03) has joined #mythtv | |
| [17:31:59] | Mode for #mythtv by ChanServ!ChanServ@services. : +v peper03 | |
| [17:34:16] | peper03_ (peper03_!~peper03@mythtv/developer/peper03) has joined #mythtv | |
| [17:34:16] | Mode for #mythtv by ChanServ!ChanServ@services. : +v peper03_ | |
| [17:37:32] | peper03 (peper03!~peper03@mythtv/developer/peper03) has quit (Ping timeout: 265 seconds) | |
| [17:44:52] | SteveGoodey (SteveGoodey!~steve@host86-150-60-252.range86-150.btcentralplus.com) has joined #mythtv | |
| [17:44:53] | Mode for #mythtv by ChanServ!ChanServ@services. : +v SteveGoodey | |
| [17:45:05] | Merlin83b (Merlin83b!~Daniel@2001:8b0:bbd6:bd9c:e9db:f3d6:6e37:eedc) has quit (Quit: Leaving) | |
| [17:54:12] | markk_: | gigem: missing audio streams (no audio!), unknown subtitle streams, video corruption and, strangely, it often starts paused |
| [17:54:28] | markspieth (markspieth!~yaaic@mythtv/developer/markspieth) has quit (Ping timeout: 252 seconds) | |
| [17:54:42] | markspieth (markspieth!~yaaic@mythtv/developer/markspieth) has joined #mythtv | |
| [17:54:42] | Mode for #mythtv by ChanServ!ChanServ@services. : +v markspieth | |
| [19:07:33] | markk_ (markk_!~mark@host109-151-194-111.range109-151.btcentralplus.com) has quit (Ping timeout: 268 seconds) | |
| [19:18:20] | warpme_ (warpme_!uid391875@gateway/web/irccloud.com/x-cfxaiaqsfylmnztg) has quit () | |
| [19:18:35] | warpme_ (warpme_!uid391875@gateway/web/irccloud.com/x-nbqcnunzpalitwig) has joined #mythtv | |
| [19:21:50] | markk_ (markk_!~mark@host109-151-194-111.range109-151.btcentralplus.com) has joined #mythtv | |
| [19:33:20] | gigem: | markk_: That is weird. The same channel works fine as a recording? |
| [19:46:07] | markspieth (markspieth!~yaaic@mythtv/developer/markspieth) has quit (Ping timeout: 252 seconds) | |
| [19:47:51] | markspieth (markspieth!~yaaic@mythtv/developer/markspieth) has joined #mythtv | |
| [19:47:51] | Mode for #mythtv by ChanServ!ChanServ@services. : +v markspieth | |
| [19:55:23] | Overlordz (Overlordz!~thor@50.35.126.95) has joined #mythtv | |
| [20:07:22] | markk_ (markk_!~mark@host109-151-194-111.range109-151.btcentralplus.com) has quit (Ping timeout: 265 seconds) | |
| [20:28:44] | peterbennett: | Live TV on android does not work at all – blank screen is all you see |
| [20:29:02] | ** stuarta sighs ** | |
| [20:29:19] | stuarta: | how we going to start the branching process in this state :( |
| [20:31:49] | peterbennett: | The android build is not constrained by Ubuntu release, so no worries. |
| [20:31:55] | stuarta: | true |
| [20:32:03] | stuarta: | although it would be "nice" to have it functional |
| [20:33:07] | peterbennett: | A great many changes have been going in lately, quite unlike recent prior releases. |
| [20:35:51] | stuarta: | aye |
| [20:36:05] | stuarta: | it's a double edged sword |
| [20:36:06] | hampton: | so I've build and installed the apk (yay!) but I can't get it to connect to my backend (boo). Updating the hostname in the configurations doesn't take. It always stays as 'localhost'. :-( |
| [20:36:17] | stuarta: | we need to get the branch cut, so it can start being packaged for ubuntu |
| [20:36:43] | stuarta: | because a release ver change is valid now, whilst later, release ver changes are not, but bugfix updates are |
| [20:37:03] | stuarta: | we actually get nearly 2 months bugfix time |
| [20:38:21] | hampton: | I don't plan on committing anything other than bugfixes until after the release is branched. |
| [20:38:34] | stuarta: | i suspect that'll be friday |
| [20:38:41] | stuarta: | i'm out all day tomorrow at a conference |
| [20:38:54] | peterbennett: | hampton: The frontend should go into the DB config page. |
| [20:39:03] | hampton: | interesting or boring? |
| [20:39:07] | ** stuarta beats his openshift cluster with a big stick ** | |
| [20:39:11] | peterbennett: | hampton: What do you see on the android? |
| [20:39:19] | stuarta: | hampton: interesting. openshift commons london |
| [20:40:05] | hampton: | peterbennett: yes, I change the name in the settings and after clicking <OK> it still shows as localhost. |
| [20:42:42] | mad_enz (mad_enz!~mad_enz@2606:6d00:135:9d84:bd31:9dae:faef:a273) has quit (Read error: Connection timed out) | |
| [20:43:20] | mad_enz (mad_enz!~mad_enz@2606:6d00:135:9d84:bd31:9dae:faef:a273) has joined #mythtv | |
| [20:46:34] | hampton: | I tried changing each of the text fields. None take. I also can't exit the DB Settings page now. I get a popup asking if I want to 1) save/exit, 2) exit w/o saving, or 3 cancel. No matter what I choose I stay in the settings page. I could exit the settings page earlier. |
| [20:47:19] | hampton: | I did got back to the home page and then back into myth before it stopped working. Might that be part of it? |
| [20:47:43] | gigem: | peterbennett: I mentioned the live tv problem yesterday. |
| [20:48:50] | gigem: | stuarta: The current android problems are most likely due to the new build process and not in mythtv proper. Though, if we can't figure out the build problems, we might have to make some code changes to work around the issues. |
| [20:49:55] | markk_ (markk_!~mark@host109-151-194-111.range109-151.btcentralplus.com) has joined #mythtv | |
| [20:50:56] | hampton: | Exiting the settings works again now that I force quit and restarted. Still can't get any of the text box changes to take. |
| [20:54:28] | hampton: | Seems like I can exit the settings page as long as I haven't changed anything. If I change even a checkbox I can't exit the db settings. |
| [20:54:29] | hampton: | bbiab |
| [20:56:59] | Steve-Goodey (Steve-Goodey!~steve@2a00:23c5:7d90:bc01:9be6:831c:4ed8:495d) has quit (Quit: Konversation terminated!) | |
| [21:08:12] | peterbennett: | hampton: I fixed that with the static_cast. do you have the latest source? |
| [21:15:03] | markspieth2: | peterbennett: great! good solution |
| [21:16:55] | markspieth2: | I had another idea. there are 2 c++ libs and qt and myth may be using different ones with incompatable typeinfo. Testing this now. |
| [21:20:10] | hampton: | peterbennett: no, that's from my working branch. I'll switch to master and try again. |
| [21:29:50] | hampton: | It works now. Sweet. |
| [21:30:09] | stuarta: | \o/ |
| [21:34:38] | hampton: | Is there an easy to get the logging output from myth on android? |
| [21:40:19] | peterbennett: | hampton: adb logcat |
| [21:41:00] | peterbennett: | hampton: To filter the log – adb logcat mfe:D *:S |
| [21:41:49] | markk_ (markk_!~mark@host109-151-194-111.range109-151.btcentralplus.com) has quit (Ping timeout: 265 seconds) | |
| [21:42:41] | peterbennett: | "adb logcat mfe:D *:S" |
| [21:42:48] | gigem: | hampton: You can also use the frontend, telnet^Wcontrol socket to change the verbose mask. |
| [21:43:08] | gigem: | markspieth2: That would explain things. Why are there two libs? |
| [21:44:44] | peterbennett: | gigem: Live TV playback has input set to "DUMMY" and plays nothing. I don't know why. |
| [21:46:14] | peterbennett: | It is possible there is another dynamic_cast failing somewhere in LiveTV |
| [21:48:48] | peterbennett: | Videoloop has m_isDummy set and as a result does nothing |
| [21:49:36] | gigem: | peterbennett: Something is failing somewhere. The problem is choosing the input for live TV is an insidious mess, IMO, and is hard to follow. I cleaned it up some to try to get the starting channel to be more predictable. It still doesn't work right all the time and I don't know why. |
| [21:50:58] | markspieth (markspieth!~yaaic@mythtv/developer/markspieth) has quit (Ping timeout: 252 seconds) | |
| [21:54:38] | markspieth (markspieth!~yaaic@mythtv/developer/markspieth) has joined #mythtv | |
| [21:54:38] | Mode for #mythtv by ChanServ!ChanServ@services. : +v markspieth | |
| [21:56:26] | gigem: | peterbennett: Your -dev post is interesting. In the tv_play case, both types are defined in Qt. In the settings case, the base class is defined in Qt but the specific type is defined by us. That distinction could be important. |
| [22:00:29] | stuarta: | iirc it is. i've fixed errors before where there are duplicate class names. confuses moc |
| [22:08:06] | stuarta: | it's been quite a while since i've seen tht |
| [22:47:15] | SteveGoodey (SteveGoodey!~steve@host86-150-60-252.range86-150.btcentralplus.com) has quit (Quit: Konversation terminated!) | |
| [23:03:25] | markspieth: | gigem: not sure. removed that too. was in the clang section in configure. |
| [23:04:45] | markspieth: | also instrumented and event in custom event is typeid qevent. need to instrument the sender in mythdialogbox to check end to end. |
| [23:07:07] | markspieth: | may be an include precedence too for /c++/v1/ |
| [23:08:52] | markspieth: | I also put in a pragma message into all 4 system typeinfo headers but none came up |
IRC Logs collected by
BeirdoBot.
Please use the above link to report any bugs.