:: #mythtv

Daily chat history

Current users (78):

aca20031, aloril, Anssi, Beirdo_, bobweaver, brfransen, cesman, Chutt, clever, coling, Cougar, danielk22, danielk221, dblain, dekarl, ElmerFudd, fetzerch, foobum, foxbuntu`, gary_buhrmaster, ghoti, Gibby, gigem, gregL, GreyFoxx, IReboot, J-e-f-f-A, jams_, jarle, jarryd, jhall_, jheizer, joe_____, joki, jpabq, jpabq_, jpharvey, jst, jwhite, jya, kormoc, KungFuJesus, kurre2, kwmonroe, laga, Merlin83b, moparisthebest_, mrand, MythBuild, MythLogBot, neufeld, poptix, purserj, rhpot1991, rsiebert_, seld, Sharky112065, skd5aner, sl1ce_1g, SmallR2002, sphery, sraue, stuartm, superm1, tgm4883, Tobbe5178, toeb, tonsofpcs, tris, unforgiven512, wagnerrp, wahrhaft_, whoDat_, wolfgang2, XDS2010_, xris, _charly_, _nyloc_
Wednesday, June 12th, 2013, 00:38 UTC
[00:38:29] SteveGoodey (SteveGoodey! has joined #mythtv
[00:40:26] jr3us^: stichnot: I am indeed using master as built for mythbuntu.
[00:40:34] NightMonkey (NightMonkey!~NightrMon@pdpc/supporter/professional/nightmonkey) has quit (Quit: Body blow! Body blow!)
[00:42:13] jr3us^: also, I have witnessed 2 different scenarios; 1 scenario is where it will start playing analog, but be very pokey, in the sense that it plays about 5 fps, with no audio, and then after a minute or 2, the screen goes black, and I have to kill the mythfrontend process.
[00:43:00] jr3us^: the other scenario involves it never displaying video, but rather only displays black screen. Again, I have to kill the front end to regain control.
[00:44:34] jr3us^: I am using nvidia and vdpau normal for tv settings. This is a mac mini 2010 with the nvidia graphics running mythbuntu 12.04.
[00:46:37] jr3us^: The video 'card' is a Hauppauge HVR-950Q, and it does load the firmware correctly per the syslog. Let me know if you want me to set any logging, etc. for running another test.
[00:49:11] jr3us^: I will check in tomorrow. and attempt to create logs as requested.
[00:50:15] jr3us^ (jr3us^! has quit (Quit: Leaving)
[01:24:51] nyloc (nyloc! has quit (Read error: Operation timed out)
[01:25:30] nyloc (nyloc! has joined #mythtv
[01:26:28] joki (joki! has quit (Ping timeout: 240 seconds)
[01:27:31] dekarl (dekarl! has quit (Read error: Operation timed out)
[01:28:51] joki (joki! has joined #mythtv
[01:29:46] dekarl (dekarl! has joined #mythtv
[01:30:53] stichnot (stichnot! has joined #mythtv
[01:30:53] stichnot (stichnot!~stichnot@mythtv/developer/stichnot) has joined #mythtv
[01:30:53] stichnot (stichnot! has quit (Changing host)
[02:14:39] _nyloc_ (_nyloc_! has joined #mythtv
[02:15:13] nyloc (nyloc! has quit (Ping timeout: 246 seconds)
[02:36:28] jya_ (jya_!~jyavenard@mythtv/developer/jya) has quit (Quit: jya_)
[02:36:42] SteveGoodey (SteveGoodey! has quit (Quit: Konversation terminated!)
[02:47:56] jya_ (jya_!~jyavenard@mythtv/developer/jya) has joined #mythtv
[03:17:34] taylorr: stichnot: I saw your livetv changes – do they fix any issue or was it just to clean up the design?
[03:30:59] Captain_Murdoch: any git experts around that know the best way (or even the MythTV way) :) to merge in a commit from someone's fork of our github repo? for instance the patch referenced in #11590
[03:30:59] ** MythLogBot **
[03:31:45] sraue (sraue!~stephan@xbmc/staff/sraue) has quit (Ping timeout: 252 seconds)
[03:32:35] Captain_Murdoch: is that what "git am" is for?
[03:32:49] fetzerch (fetzerch!~quassel@unaffiliated/fetzerch) has quit (Ping timeout: 248 seconds)
[03:34:42] fetzerch (fetzerch!~quassel@unaffiliated/fetzerch) has joined #mythtv
[03:36:16] sraue (sraue!~stephan@xbmc/staff/sraue) has joined #mythtv
[03:38:33] Captain_Murdoch: nevermind, figured it out. "git am" was much easier than the first answers I was finding regarding setting up a remote link to merge in the commit.
[03:45:00] gary_buhrmaster: Captain_Murdoch: Yes, I have been (recently) appending the version that a simple git am will use.
[03:45:44] gary_buhrmaster: Captain_Murdoch: (at the request of one of the other developers, since I was creating so many tickets.)
[03:45:57] Captain_Murdoch: thanks. :) made it easy
[03:47:44] gary_buhrmaster: Captain_Murdoch: I *can* provide the git way to pull in patches from other repos, but this is just easier. For some of my previous tickets (before I did it in the ticket), just take the url, and apply a '.patch' at the end, and you will get a 'git am' usable patch.
[03:47:45] sraue (sraue!~stephan@xbmc/staff/sraue) has quit (Ping timeout: 248 seconds)
[03:47:47] Captain_Murdoch: only thing that would have made it easier was if I didn't have to ammend/edit the commit message to add the "Fixes #11590" so it would auto-close your ticket. :)
[03:48:20] Captain_Murdoch: yeah, based on what I was finding when I was trying to learn how to merge a remote commit, this seems much easier.
[03:48:29] joki (joki! has quit (Read error: Operation timed out)
[03:49:12] Captain_Murdoch: the "Fixes #" is a catch-22 though and it's probably easier for me to edit the commit message than for you to go back and edit after you've opened the ticket.
[03:49:32] gary_buhrmaster: Captain_Murdoch: That must be a MythTV commit hook. And, of course, it is impossible (for me, a lowly ticket creator) to be able to know the ticket number that the git commit will be...
[03:50:49] Captain_Murdoch: yeah, it's the trac ticket #, but it's a catch-22, you have to have the commit to make the ticket, so you can't reference the track ticket # in the commit message unless you go back and amend. I just ran your patch through "git am" then did a "git commit --amend" to add the "Fixes #" line then I pushed.
[03:51:03] Captain_Murdoch: well, insert a test compile in the middle there. :)
[03:51:26] gary_buhrmaster: Captain_Murdoch: I suppose I could create a ticket, then go back and do the git commit, and then go back to the ticket and append, but that would somewhat break my personal workflow (where I create the patch, commit it, compile (and test where I can)), then create the ticket.
[03:51:45] Captain_Murdoch: too much work IMHO, easier for the committer to just amend and add the line.
[03:51:53] gary_buhrmaster: Captain_Murdoch: We are on the same page (and it is a twisty maze of passages, all different).
[03:52:08] Captain_Murdoch: especially if you have 10–15 you're pushing to your repo and opening tickets for.
[03:52:17] Captain_Murdoch: 10–15 at a time :)
[03:53:04] gary_buhrmaster: Captain_Murdoch: If there is a better way for me to be submitting these tickets/patches, I can try to adopt.
[03:53:50] joki (joki! has joined #mythtv
[03:53:58] Captain_Murdoch: stuart m probably knows best if he's been dealing with most.
[03:54:05] gary_buhrmaster: Captain_Murdoch: btw, if the committers like the git-am patch format, (in my copious free time) I can go try to update the wiki pages so that others will follow the same practice.
[03:55:23] gary_buhrmaster: Captain_Murdoch: Since stuartm is pushing coverity fixes (and I fully support the efforts, that is why I have been submitting patches), he has committed most.
[03:55:29] Captain_Murdoch: does "git am --interactive" let you edit the commit message? that might be easier than "git commit --amend".
[03:56:40] Captain_Murdoch: I think git-am patch format is nice for things like this that are simple, easy to confirm. for larger fixes, it might not be best since you don't want to commit before testing, especially if you have to edit something if there's a bug in a feature patch.
[03:59:59] gary_buhrmaster: Captain_Murdoch: re: --interactive. I do not know. Never tried it, and my git experience never needed it. You have located a git knowledge hole.
[04:01:03] sraue (sraue!~stephan@xbmc/staff/sraue) has joined #mythtv
[04:01:27] Captain_Murdoch: that's well past my knowledge of git, hence my original question. :) learning though.
[04:01:41] gary_buhrmaster: Captain_Murdoch: re: large fixes. Agreed, although until you push, it has not yet happened.
[04:02:27] Captain_Murdoch: yeah, I just wouldn't want to have to revert all the time just to edit something just because applying it via git-am was easier than via patch.
[04:02:55] Captain_Murdoch: will patch ignore the stuff at the beginning of a git-am formatted patch?
[04:03:25] Captain_Murdoch: if so, then it's moot. just use git-am format and it gives the best of both worlds.
[04:05:17] gary_buhrmaster: Captain_Murdoch: patch will try to ignore "stuff" at the head/tail, but I have occasionally seen it get confused (although that was a long time ago, and maybe it has gotten better).
[04:06:20] Captain_Murdoch: I was thinking it would skip lines until it found the diff header lines, but maybe it only will so many lines before getting confused.
[04:07:30] gary_buhrmaster: Captain_Murdoch: I admit that once I started with source version systems, a long time ago (rcs, cvs, svn, and yes, visual source safe, and finally git), and often had commit access, I stopped doing things the hard way with mailing patches all around, and lost track of the edge cases.
[04:08:47] gary_buhrmaster: Captain_Murdoch: And of course git-am formatted let me mail patches around (all over again).
[04:24:42] gary_buhrmaster: Captain_Murdoch: and thinking pedantically, you would not want to trust my commits with a 'fixes #' line anyway (I could be evil, and intentionally trying to confuse your trac system).
[04:32:52] Captain_Murdoch: fixes #1–12000.
[04:33:02] Captain_Murdoch: doh! :)
[04:34:46] Captain_Murdoch: have you touched the mythfrontend ones? I was going to fix a few more of those "uninitialized pointer field" ones before heading offline.
[04:40:14] gary_buhrmaster: Captain_Murdoch: No, I have mostly been leaving those alone, partially out of not taking the time to really look at them.
[04:43:53] gary_buhrmaster: Captain_Murdoch: And partially for not having spent enough time to fully understand the code/flow to insure (to my mind at least) that the patch is appropriate. btw, while you (again) should not trust me, I do make sure all my patches compile (at least once) before I submit them.
[04:45:02] Captain_Murdoch: right now I'm only touching things I'm confident in. :)
[04:46:15] gary_buhrmaster: Captain_Murdoch: but I have never managed to get a Windows builder working, and my FreeBSD system is currently turned off. So my "it compiles" means Linux gcc most of the time (I also compile with clang because of its static-analysis tool finding thngs cppcheck and coverity do not).
[04:46:29] Captain_Murdoch: have a few I've looked at that I have questions about since it seems like it picks up one uninitalized variable but not others. had the same thing with the 'leakage' fixes in transcode. I think there were other paths that could leak that it didn't see. makes me wonder if it stops at the first type of error in a class or something.
[04:47:55] gary_buhrmaster: Captain_Murdoch: In terms of some uninitialized variable but not others, the (abbreviated) report on the trac server (I have been told) only reports the first issue, and you need to go into the coverity site to see the entire report.
[04:48:59] Captain_Murdoch: ok, thx. makes sense based on the one I just looked at but didn't touch.
[04:49:00] gary_buhrmaster: Captain_Murdoch: And while I do not know about coverity, I know with scan-build (the clang/llvm static analysis) it does not seem to report variables that have no viable path to use (or at least it thinks there are none).
[04:49:23] Captain_Murdoch: time to get off the computer for a while, have to be back on in ~2 hours for a maintenance window.
[04:50:26] gary_buhrmaster: Captain_Murdoch: That is another one of the reasons I have left some of the "uninitialized" variables alone, since I do not have access to the coverity reports, it would be playing wack-a-mole (fix the one reported, the next pops up). I hate wack-a-mole.
[04:51:22] sphery: Captain_Murdoch: I think stua rtm was saying that the actual Coverity report (on their site) from which he's pulling the information lists all the variables, but the summary or whatever he's using only has the first
[04:51:38] sphery: ah, just saw that's what gary_buhrmaster said.
[04:52:00] gary_buhrmaster: sphery: Yeah, once in a while I happen to get something right :-)
[04:52:44] sphery: I suppose I should have read the last 5 lines before speaking up--but I felt so close to caught up...
[04:53:11] gary_buhrmaster: sphery: Or, more correctly, I happen to be able to parrot what I have been told.
[04:53:23] sphery: hehe, that's all I ever do :)
[05:03:56] garybuhrmaster (garybuhrmaster! has joined #mythtv
[05:04:20] garybuhrmaster (garybuhrmaster! has quit (Client Quit)
[05:06:59] gtb (gtb! has joined #mythtv
[05:08:04] gtb (gtb! has quit (Client Quit)
[05:13:41] unforgiven512 (unforgiven512! has quit (Quit: ZNC -
[05:14:26] unforgiven512 (unforgiven512! has joined #mythtv
[05:27:37] stichnot: taylorr: my recent live TV changes were just cleanups/simplifications. I wasn't able to reason out any actual race conditions of any real likelihood.
[05:30:50] stichnot: If the frontend somehow waits 2+ minutes at a DUMMY recording for the actual recording to start, the auto-expirer might delete the DUMMY recording and prevent the frontend from ever moving to the actual recording, but you probably only see that if you have the frontend stopped for 2+ minutes in the debugger.
[05:59:54] dekarl: Captain_Murdoch: gary_buhrmaster, between downloading the .patch from github and applying with git am you can simply edit the message to add "Fixes #abc", no need for funky amending or rebasing with reword (both work well, too, though)
[06:00:31] whoDat_ (whoDat_! has quit (Ping timeout: 264 seconds)
[06:04:08] SteveGoodey (SteveGoodey! has joined #mythtv
[06:05:03] stichnot (stichnot!~stichnot@mythtv/developer/stichnot) has quit (Ping timeout: 256 seconds)
[06:12:16] FabriceMG (FabriceMG!~Thunderbi@ has joined #mythtv
[06:14:05] SteveGoodey (SteveGoodey! has quit (Quit: Konversation terminated!)
[06:14:53] whoDat_ (whoDat_! has joined #mythtv
[06:38:21] stichnot (stichnot! has joined #mythtv
[06:38:21] stichnot (stichnot!~stichnot@mythtv/developer/stichnot) has joined #mythtv
[06:38:21] stichnot (stichnot! has quit (Changing host)
[06:58:49] Captain_Murdoch: dekarl, you mean save it to an actual file and bring up an editor manually? :) I just used something like "wget -O – ${URL} | git am"  :)
[07:06:30] dekarl: Captain_Murdoch: maybe, extra points for adding a "vi -" in there :D
[07:15:52] Captain_Murdoch: would be nice if git am would let you edit the commit message. can always make an alias or wrapper script to do it.
[08:06:29] dgeary2 (dgeary2!~debian@ has joined #mythtv
[08:07:56] jya_ (jya_!~jyavenard@mythtv/developer/jya) has quit (Quit: jya_)
[08:20:58] Gibby_ (Gibby_!~Gibby@ has quit (Ping timeout: 245 seconds)
[08:20:59] stuartm: Captain_Murdoch: I *think* --interactive just lets you choose which hunks of the patch actually get applied
[08:21:10] Gibby (Gibby!~Gibby@ has joined #mythtv
[08:21:45] stuartm: what we need is a git am --append-message "" {....}
[08:22:17] stuartm: of course no such thing exists, so it's git commit --amend or nothing :(
[08:29:06] stuartm: fwiw, the 'pull' process isn't so bad, especially for multiple commits, just once call "git fetch" then git cherry-pick --signoff {sha1} for each commit
[08:29:34] stuartm: it's actually easier than patches as you don't have to download and save them first
[08:32:06] stuartm: you only need to re-run fetch to update that copy
[08:32:40] stuarta: stuartm: you can edit the commit message before you push
[08:33:00] stuartm: stuarta: aye, that's what we're doing with git commit --amend
[08:33:43] stuartm: CM was just looking for a one-step solution which would apply the patch (pick the commit) and edit the message in one
[08:34:04] stuartm: I guess you could script that using the git alias stuff
[08:34:42] stuartm: actually, that's worth a try ...
[09:11:41] jya_ (jya_!~jyavenard@mythtv/developer/jya) has joined #mythtv
[09:35:39] jya_ (jya_!~jyavenard@mythtv/developer/jya) has quit (Quit: jya_)
[09:42:08] jya_ (jya_!~jyavenard@mythtv/developer/jya) has joined #mythtv
[09:53:30] jya_ (jya_!~jyavenard@mythtv/developer/jya) has quit (Quit: jya_)
[10:03:10] coling_ is now known as coling
[10:24:15] jya_ (jya_!~jyavenard@mythtv/developer/jya) has joined #mythtv
[10:29:57] paul-h (paul-h!~Paul@ has joined #mythtv
[10:32:52] amessina (amessina!~amessina@2001:470:c1dc:7779:d6be:d9ff:fe8d:7c1e) has quit (Remote host closed the connection)
[10:33:42] amessina (amessina!~amessina@2001:470:c1dc:7779:d6be:d9ff:fe8d:7c1e) has joined #mythtv
[10:34:01] gary_buhrmaster (gary_buhrmaster!~gtb@2001:470:80e4:3::36) has quit (Ping timeout: 240 seconds)
[10:35:56] jya_ (jya_!~jyavenard@mythtv/developer/jya) has quit (Quit: jya_)
[10:38:56] jya_ (jya_!~jyavenard@mythtv/developer/jya) has joined #mythtv
[10:39:00] gary_buhrmaster (gary_buhrmaster!~gtb@2001:470:80e4:3::36) has joined #mythtv
[11:05:16] IReboot (IReboot! has joined #mythtv
[11:32:27] rhpot1991 (rhpot1991!~rhpot1991@ubuntu/member/rhpot1991) has quit (Ping timeout: 260 seconds)
[11:33:08] dgeary2 (dgeary2!~debian@ has quit (Quit: Ex-Chat)
[11:36:07] rhpot1991 (rhpot1991!~rhpot1991@2001:4968:202:3:d10a:34a0:5de9:ccbb) has joined #mythtv
[11:36:08] rhpot1991 (rhpot1991!~rhpot1991@2001:4968:202:3:d10a:34a0:5de9:ccbb) has quit (Changing host)
[11:36:08] rhpot1991 (rhpot1991!~rhpot1991@ubuntu/member/rhpot1991) has joined #mythtv
[11:46:59] pvr4me (pvr4me! has joined #mythtv
[11:55:44] skd5aner (skd5aner! has quit (Read error: Connection reset by peer)
[11:56:11] skd5aner (skd5aner! has joined #mythtv
[12:08:52] MythBuild: build #993 of master-linux-64bit-icc is complete: Failure [4failed compile core] Build details are at . . . c/builds/993 blamelist: Paul Harrison < >
[12:19:10] paul-h: danielk22: ^ one of the tests went tits up *** glibc detected *** ././libmythbase/test/test_mythsystem/test_mythsystem: corrupted double-linked list: 0x0000000000ac36c0 ***
[12:23:19] danielk221: paul-h: Thanks. There is a reason I
[12:23:58] danielk221: was hesitant to look at mythsystem and am doing it cautiously...
[12:35:49] dblain (dblain!~dblain@mythtv/developer/dblain) has quit ()
[12:42:40] dmfrey (dmfrey! has joined #mythtv
[12:42:46] aloril_ (aloril_! has quit (Read error: Connection reset by peer)
[12:44:39] aloril (aloril! has joined #mythtv
[12:46:32] MythBuild: build #994 of master-linux-64bit-icc is complete: Success [3build successful] Build details are at . . . c/builds/994
[12:58:09] paul-h: stuartm: in IdleScreen coverity is complaining m_upcomingList isn't initialised but it isn't used anywhere as far as I can see. Safe to remove it?
[13:00:19] jya_ (jya_!~jyavenard@mythtv/developer/jya) has quit (Quit: jya_)
[13:02:20] jya_ (jya_!~jyavenard@mythtv/developer/jya) has joined #mythtv
[13:03:12] dmfrey (dmfrey! has quit (Remote host closed the connection)
[13:03:41] dmfrey (dmfrey! has joined #mythtv
[13:06:15] amessina (amessina!~amessina@2001:470:c1dc:7779:d6be:d9ff:fe8d:7c1e) has quit (Quit: Konversation terminated!)
[13:26:59] MavT (MavT! has quit (Ping timeout: 256 seconds)
[13:32:09] bobweaver:
[13:41:31] jheizer__ (jheizer__! has quit (Ping timeout: 256 seconds)
[13:42:03] rhpot1991 (rhpot1991!~rhpot1991@ubuntu/member/rhpot1991) has quit (Quit: Coyote finally caught me)
[13:42:32] rhpot1991 (rhpot1991!~rhpot1991@2001:4968:202:3:d10a:34a0:5de9:ccbb) has joined #mythtv
[13:42:33] rhpot1991 (rhpot1991!~rhpot1991@2001:4968:202:3:d10a:34a0:5de9:ccbb) has quit (Changing host)
[13:42:33] rhpot1991 (rhpot1991!~rhpot1991@ubuntu/member/rhpot1991) has joined #mythtv
[13:42:42] rhpot1991 (rhpot1991!~rhpot1991@ubuntu/member/rhpot1991) has quit (Client Quit)
[13:42:52] rhpot1991 (rhpot1991!~rhpot1991@2001:4968:202:3:d10a:34a0:5de9:ccbb) has joined #mythtv
[13:42:53] rhpot1991 (rhpot1991!~rhpot1991@2001:4968:202:3:d10a:34a0:5de9:ccbb) has quit (Changing host)
[13:42:53] rhpot1991 (rhpot1991!~rhpot1991@ubuntu/member/rhpot1991) has joined #mythtv
[13:49:17] dmfrey (dmfrey! has quit (Remote host closed the connection)
[13:49:25] bas-t (bas-t! has joined #mythtv
[13:49:46] dmfrey (dmfrey! has joined #mythtv
[14:02:44] dmfrey (dmfrey! has quit (Remote host closed the connection)
[14:03:13] dmfrey (dmfrey! has joined #mythtv
[14:06:30] jheizer (jheizer! has joined #mythtv
[14:21:37] dmfrey (dmfrey! has quit (Remote host closed the connection)
[14:22:08] dmfrey (dmfrey! has joined #mythtv
[15:08:24] dblain (dblain!~dblain@mythtv/developer/dblain) has joined #mythtv
[15:13:14] pvr4me (pvr4me! has quit (Quit: pvr4me)
[15:22:35] jpabq (jpabq!~quassel@mythtv/developer/jpabq) has quit (Disconnected by services)
[15:22:40] jpabq_ is now known as jpabq
[15:22:49] jpabq (jpabq! has quit (Changing host)
[15:22:49] jpabq (jpabq!~quassel@mythtv/developer/jpabq) has joined #mythtv
[15:23:36] exoon (exoon! has joined #mythtv
[15:23:59] jpabq_ (jpabq_!~quassel@mythtv/developer/jpabq) has joined #mythtv
[15:36:06] FabriceMG (FabriceMG!~Thunderbi@ has quit (Quit: FabriceMG)
[15:38:21] stuartm: paul-h: it can be removed for now, I added it with the intention of copying mythwelcome and allowing the display of the upcoming scheduled recordings, but I'm unlikely to work on that any time soon
[15:44:54] stuartm: we're down to 114 coverity defects and 111 cppcheck defects, getting there ...
[15:49:35] stuarta: i'll see if i can setup a coverity builder later
[15:50:23] stuarta: whilst deciphering kernel dumps
[16:06:33] gary_buhrmaster: stuartm: I agree that cherry-picking is easy, and is usually my preferred approach, but the flexibility of git means everyone can do it their way (which is also why I tend to learn new things about git on a regular basis; I never did it that way before).
[16:09:27] rsiebert_ (rsiebert_! has joined #mythtv
[16:10:49] rsiebert (rsiebert! has quit (Read error: Operation timed out)
[16:19:05] stichnot (stichnot!~stichnot@mythtv/developer/stichnot) has joined #mythtv
[16:19:43] stichnot: jpabq: I want to try calling something like SendMythSystemRecEvent("REC_STARTED_WRITING", pginfo) as soon as the recorder actually starts writing data to the file. Where is the right place for this?
[16:21:00] stichnot: The purpose is to get more predictable/repeatable srt synchronization when using neufeld's
[16:25:43] stichnot: And if I can get near-perfect synchronization, it should be possible to get these captions for in-progress recordings and Live TV (WAF issue for me...)
[16:27:01] stuartm: you can get srt subtitles for livetv?
[16:27:28] jpabq: stichnot: I will look up the exact place for you, but as soon as the signal monitor is happy, myth will start writing the file.
[16:29:57] jpabq: stichnot: I have tinkered with an 'external' recorder module. I have a proof of concept working, where a an external 'black box' app can be controlled by Myth, and feed myth a mpts file to 'record'. One idea I had for such a mechanism, is to create an external app which reads from the hdpvr and reads from /dev/vbiX and muxes the captions into the H.264 before feeding it into Myth.
[16:34:35] Steve-Goodey (Steve-Goodey! has joined #mythtv
[16:41:00] stichnot: stuartm: To get srt subtitles via neufeld's script for in-progress recordings (including live TV), we would need to add a way to re-fetch the srt file, or fetch updates, from the RingBuffer.
[16:42:03] stichnot: BTW, if no one objects, one of these days I will add .srt files to the set of files deleted by the delete/auto-expire code. They're cluttering up my system. :)
[16:42:48] stichnot: jpabq: I was thinking about myth only starting to write once it sees the appropriate keyframe or whatever. Is that logic part of the signal monitor?
[16:43:21] stuartm: stichnot: I'm just surprised that someone is typing out subtitles in near real-time in srt format, I mean it's done here by broadcasters but even so they are slightly delayed and often full of errors
[16:43:51] bas-t (bas-t! has quit (Quit: Ex-Chat)
[16:43:52] tafy (tafy! has quit (Read error: Operation timed out)
[16:44:08] stuartm: or are they converting them from closed captions that the HDPVR can't capture?
[16:44:10] stichnot: jpabq: That "external" recorder module with caption muxing would be great.
[16:44:33] stichnot: stuartm: the latter
[16:45:28] stichnot: stuartm: The problem is that there is no standard for inserting cc608/cc708 captions into the component (or HDMI) output.
[16:45:31] exoon (exoon! has quit (Quit: Verlassend)
[16:45:55] stuartm: stichnot: that sucks
[16:46:29] stichnot: neufeld's script leverages the analog output of the STB to grab the cc608 captions, uses ccextractor to get an srt file, and tries to fix up the synchronization at the end.
[16:46:29] stuartm: does the HDPVR even support burning them into the image?
[16:46:58] stichnot: In the past, if I wanted captions, I had to configure the STB to burn them into the image.
[16:47:58] stichnot: And I want captions for myself but not the young kids, so I rigged up a complex stateful script to reconfigure the STB via the IR blaster as needed
[16:48:29] stichnot: stuartm: does that make sense? I have a feeling we may be misunderstanding each other
[16:49:51] stichnot: stuartm: and by Live TV, I don't necessarily mean a show that is being real-time captioned (same experience here for delays and humorous mistakes)
[16:51:16] natanojl (natanojl! has joined #mythtv
[16:54:18] stichnot: jpabq: To finish my thought, I would love the external recorder module because it would eliminate all need for synchronization as well as any special hacks for in-progress recordings / live TV.
[17:02:34] NightMonkey (NightMonkey!~NightrMon@pdpc/supporter/professional/nightmonkey) has joined #mythtv
[17:04:47] sphery: stuartm: I think the approach he's now using relies on both an HD-PVR (for video) and a PVR-x50 (for extracting closed captions, which are written to an .srt file)
[17:06:11] sphery: stuartm: specifically
[17:07:16] sphery: the "open captions" approach (having the captions enabled in the STB so they're overlayed on the image that's recorded by the HD-PVR) is the only "single-card" approach for captions with high-definition analog recording
[17:08:22] gary_buhrmaster: sphery, stuartm, stichnot: Was there not a dvb driver that could "fake it" from some source contents? It may have been a work in progress (and never finished), but I have this vague recollection of something.
[17:11:09] dmfrey (dmfrey! has quit (Remote host closed the connection)
[17:11:38] dmfrey (dmfrey! has joined #mythtv
[17:13:30] stichnot: gary_buhrmaster: I don't have any recollection of that, fwiw
[17:14:57] gary_buhrmaster: stichnot: Here is a reference for something (I did a "google search")
[17:16:14] gary_buhrmaster: stichnot: So I am not (completely) crazy :-) (I know, asserting facts not in evidence...)
[17:18:22] Jordack (Jordack! has joined #mythtv
[17:21:25] jpabq: stichnot: I would be tempted to put the SendMythSystemRecEvent("REC_STARTED_WRITING", pginfo) in SignalMonitor::run *after* the while (!exit) loop completes.
[17:24:04] jpabq: Although, you may want to call the event something like "REC_GOOD_SIGNAL" to indicate what the event really is saying. Writing should start immediately after the signalmonitor completes, but there still would not actually be a guaranty.
[17:25:30] jpabq: If you really want it to symbolize REC_STARTED_WRITING, it would have to go in TFW, but that would probably add some inefficiencies.
[17:26:31] stichnot: jpabq: thanks, I'll give that a try. This would be proof-of-concept testing only, no commits without more discussion/consensus
[17:26:37] dmfrey (dmfrey! has quit (Remote host closed the connection)
[17:27:09] dmfrey (dmfrey! has joined #mythtv
[17:27:14] jpabq: If you want to emit it after the first keyframe, then that would be in dtvrecroder. You have to handle both code paths though — mpeg and H.264.
[17:28:12] jpabq: stichnot: I kinda like the idea of muxing it into the H.264 as well. I don't have much experience with ffmpeg, so don't know how hard that will be...
[17:31:14] neufeld: jpabq: one problem, you can't mux a .srt text stream with an H.264 transport stream. I tried, ffmpeg silently discards the .srt file during the conversion to TS. I opened a bug, ffmpeg closed it
[17:31:34] jpabq: If you want to do it at the first keyframe (which probably would be the most accurate), then it should go in DTVRecorder::HandleKeyframe() and DTVRecorder::HandleH264Keyframe() as part of the "if (_first_keyframe < 0)" block
[17:31:48] KungFuJe1us (KungFuJe1us!~adam@ has quit (Read error: Operation timed out)
[17:31:57] KungFuJesus (KungFuJesus!~adam@ has joined #mythtv
[17:32:00] neufeld: jpabq: it might be possible with dvb captions, I never got far enough to figure that out, I just went with a separate .srt file in the recording directory.
[17:32:29] stichnot (stichnot!~stichnot@mythtv/developer/stichnot) has quit (Ping timeout: 255 seconds)
[17:33:50] jpabq: neufeld: Whenever I got around to playing with it, I was going to try and use libzvbi. That lib should allow a more "raw" extraction of the data. I was guessing that the output from libzvbi would probably have to be converted before being able to mux it, though.
[17:35:35] jpabq: neufeld: I did successfully "mux" the h.264 and .srt into a mkv at one point.
[17:38:42] neufeld: jpabq: yeah, you can make a .mkv, but not a transport stream. I prefer to keep the transport stream in my recordings, so that the cutlist editor still works and for mytharchive, which handles MPEG-TS differently.
[17:38:51] jpabq: neufeld: I would love to be able to preserve the positioning information. srt format does not seem to keep that info, and the CC just end up at the bottom of the screen, even if that means their covering other info.
[17:41:03] jpabq: With recent kernels, the set-fmt-sliced-vbi=cc ioctls have to go against the /dev/vbi device instead of the /dev/video device.
[17:41:23] neufeld: jpabq: correct, .srt loses the position data. My script can easily be adjusted to make any of ccextractor's supported output formats, but I found only .srt files were being spotted and used by the player.
[17:43:00] jpabq: neufeld: if there is another format that preserves the position, then sticknot would be the one to talk to about that. He could probably add support for it.
[17:45:42] neufeld: I think we looked at that a bit. If I recall, the caption support is "whatever ffmpeg supports". Possible alternatives were MS .sami and McPoodle's 'dvdraw'.
[17:46:04] neufeld: I don't think either was supported by ffmpeg
[17:46:35] ephemer0l (ephemer0l!~ephemer0l@unaffiliated/ephemer0l) has quit (Ping timeout: 256 seconds)
[17:46:48] neufeld: hmm, I see some .sami patches from a year ago in ffmpeg, maybe I should investigate again
[18:09:13] bobweaver: Hello there I am looking for some c++/qt help I want to add a way to scan for the backend on start up on my qml app. using a c++ plugin from my main.cpp but I do not know where to look or how One would use it. I have looked in libmythsercies and libmythunpn and others. But I do not know how I can return a string. from some function that I can not find . Thanks
[18:15:44] SteveGoodey (SteveGoodey! has joined #mythtv
[18:18:26] peper03 (peper03! has joined #mythtv
[18:43:37] natanojl: bobweaver: I don't know much about it but MythContextPrivate::UPnPautoconf in mythtv/libs/libmyth/mythcontext.cpp might give you some ideas
[18:43:59] bobweaver: thanks natanojl
[18:44:32] natanojl: np
[18:52:57] dmfrey (dmfrey! has quit (Remote host closed the connection)
[18:53:35] dmfrey (dmfrey! has joined #mythtv
[18:59:54] whoDat_: sphery: i Just bought an HD-PVR. ditching svideo. lol
[19:00:20] dekarl (dekarl! has quit (Ping timeout: 240 seconds)
[19:04:26] dekarl (dekarl! has joined #mythtv
[19:18:41] SteveGoodey (SteveGoodey! has quit (Quit: Konversation terminated!)
[19:21:26] Steve-Goodey (Steve-Goodey! has quit (Quit: Konversation terminated!)
[20:03:14] aloril (aloril! has quit (Ping timeout: 255 seconds)
[20:07:56] dmfrey (dmfrey! has quit (Remote host closed the connection)
[20:08:27] dmfrey (dmfrey! has joined #mythtv
[20:10:54] Sharky112065 (Sharky112065! has quit (Ping timeout: 264 seconds)
[20:13:24] stichnot (stichnot!~stichnot@mythtv/developer/stichnot) has joined #mythtv
[20:14:38] Sharky112065 (Sharky112065! has joined #mythtv
[20:15:57] aloril (aloril! has joined #mythtv
[20:19:07] Sharky112065 (Sharky112065! has quit (Ping timeout: 264 seconds)
[20:19:25] peper03 (peper03! has quit (Quit: Konversation terminated!)
[20:21:40] Jordack (Jordack! has quit (Ping timeout: 260 seconds)
[20:30:44] taylorr: jpabq: fyi, I finally figured out why I've always got such random results for channel change times... I added get_signal() to my channel change script a long time ago to check my satellite signal level.... seems that it acts up sometimes causing the channel change script to hang... just thought I'd pass that info along
[20:33:16] stuarta (stuarta!~stuarta@mythtv/developer/stuarta) has quit (Quit: router firmware upgrade)
[20:34:39] stichnot (stichnot!~stichnot@mythtv/developer/stichnot) has quit (Ping timeout: 251 seconds)
[20:38:27] stuarta (stuarta! has joined #mythtv
[20:38:27] stuarta (stuarta!~stuarta@mythtv/developer/stuarta) has joined #mythtv
[20:38:27] stuarta (stuarta! has quit (Changing host)
[20:42:05] Sharky112065 (Sharky112065! has joined #mythtv
[20:43:51] dmfrey (dmfrey! has quit (Remote host closed the connection)
[20:44:23] dmfrey (dmfrey! has joined #mythtv
[20:49:12] gary_buhrmaster: stuartm: re: coverity numbers. I think that you may be close to being under 100 with the recent (and the few pending) fixes. And resolving less than a hundred starts to feel achievable (one of those human nature things, I have been told; like a price of 99 seems so much cheaper than 100).
[20:50:19] stichnot (stichnot!~stichnot@ has joined #mythtv
[20:50:19] stichnot (stichnot!~stichnot@mythtv/developer/stichnot) has joined #mythtv
[20:50:19] stichnot (stichnot!~stichnot@ has quit (Changing host)
[20:50:55] danielk22: My experience is that resolving the problems gets tougher the closer you get to zero.
[20:53:17] gary_buhrmaster: danielk22: Yeah, mine too, but I am trying to remain positive.....
[20:53:55] dmfrey (dmfrey! has quit (Quit: Ex-Chat)
[20:54:29] stuarta (stuarta! has joined #mythtv
[20:54:29] stuarta (stuarta!~stuarta@mythtv/developer/stuarta) has joined #mythtv
[20:54:29] stuarta (stuarta! has quit (Changing host)
[20:56:31] danielk22: :)
[21:04:04] natanojl (natanojl! has quit (Ping timeout: 240 seconds)
[21:27:44] stuartm: maybe, maybe not, the very worst bugs were cherry picked early on (last year mostly), what's left is more minor stuff that was avoided because it's dull and boring
[21:29:24] stuartm: that said, there are still a fair few issues outstanding that haven't been tackled because they were in areas of code which aren't widely understood
[21:32:09] stuartm: e.g. 1023889 – out of bounds read, you have to know what is supposed to be happening there for a proper fix
[21:33:49] stuartm: some of the freemheg stuff, because unless you're prepared to wade through the specs it's something of a guessing game
[21:40:57] stichnot: Same for 700317 – I have a feeling that latin1 should be conditionally assigned somewhere in the method, but I'm not sure how to go from the teletext spec to latin1
[21:46:13] stichnot (stichnot!~stichnot@mythtv/developer/stichnot) has quit (Ping timeout: 276 seconds)
[22:02:30] stichnot (stichnot!~stichnot@mythtv/developer/stichnot) has joined #mythtv
[22:05:59] jr3us (jr3us! has joined #mythtv
[22:08:40] jr3us: stichnot: should I continue pursuing the live analog tv problems I described yesterday? if so let me know how I should setup logging to capture info
[22:14:36] coling (coling! has quit (Ping timeout: 260 seconds)
[22:14:45] stichnot (stichnot!~stichnot@mythtv/developer/stichnot) has quit (Read error: Operation timed out)
[22:15:01] Gibby (Gibby!~Gibby@ has quit (Ping timeout: 246 seconds)
[22:16:19] coling (coling! has joined #mythtv
[22:17:03] Gibby (Gibby!~Gibby@ has joined #mythtv
[22:49:07] danielk221 (danielk221!~danielk22@ has quit (Ping timeout: 264 seconds)
[22:49:14] paul-h (paul-h!~Paul@ has quit (Quit: Konversation terminated!)
[22:58:15] kormoc (kormoc!~kormoc@mythtv/developer/kormoc) has quit (Ping timeout: 256 seconds)
[22:58:45] kormoc (kormoc!~kormoc@mythtv/developer/kormoc) has joined #mythtv
[22:59:36] jya_ (jya_!~jyavenard@mythtv/developer/jya) has quit (Quit: jya_)
[23:00:38] jya_ (jya_!~jyavenard@mythtv/developer/jya) has joined #mythtv
[23:04:00] danielk221 (danielk221! has joined #mythtv
[23:21:58] jr3us: I'll check back in awhile
[23:22:11] jr3us (jr3us! has left #mythtv ()
[23:43:14] gregL (gregL! has quit (Read error: Connection reset by peer)
[23:44:38] gregL (gregL! has joined #mythtv

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