:: #mythtv

Daily chat history

Current users (74):

aarcane_, aloril_, amessina, Anssi, Beirdo, brfransen, caelor, Captain_Murdoch, Chutt, clever, coling, Cougar, dblain, dekarl1, eharris, ElmerFudd, fetzerch, ghoti, Gibby, gregL, GreyFoxx, Guest85778, jams, jarle, jarryd, jheizer, joe____, johanbr, joki, jpharvey, jst, jwhite, jya, kc, kurre2, kwmonroe, laga_, MavT, moparisthebest, MythBuild, MythLogBot, nephyrin, Nothing4You, nvzn_, peper03, poptix, purserj, rhpot1991, robink, rsiebert_, Seeker`, seld, Sharky112065, sjennings, skd5aner, sl1ce, SmallR2002, sphery, sraue, stuarta, stuartm, superm1, taylorr, tgm4883, toeb, tonsofpcs, tris, wagnerrp, wahrhaft, wolfgang1, XDS2010, xris, _charly_, _nyloc_
Friday, October 18th, 2013, 00:00 UTC
[00:00:53] stichnot: The biggest example for me is the .srt subtitle delay, which is usually OK but some recordings it's off by a few seconds and has to be readjusted every time you resume playback.
[00:02:19] stichnot: Or, some broadcast is windowboxed and needs an "adjust fill" setting
[00:02:46] stichnot: Audio sync is another big one, but the adjustment may be TV-specific
[00:03:34] stichnot: And if you were previously listening to a particular audio track, you'd probably want that to continue.
[00:09:58] jya: stuartm: isn't there something that also needs to be done which is removing the channel icon information and redownloading it again? I know that there was something like that to be done in shepherd as the original file wasn't stored in a storage group
[00:11:43] stuartm: there was nothing that needed doing if the icons had been inserted by a mythtv app, but yes, some third parties were inserting absolute urls to the icons and even storing them in different location to the one used by myth
[00:12:22] stuartm: in which case the affected users would need to remove the existing icon paths and re-insert
[00:13:02] stuartm: shepherd (per usual) was one of those apps acting directly on the database instead of going through mfdb
[00:14:25] stuartm: I've already ranted once about shepherd today :)
[00:16:55] stuartm: this just isn't funny, QTScript will parse practically any date format you can imagine except for the one the specification says that it _must_ handle, the International Standard format and the very format we've standardised on for the services API
[00:19:49] jya: stuartm: they fixed that a while back after I wrote to them about it last year
[00:20:28] stuartm: jya: changes apparently haven't filtered down to the packages I'm using them
[00:20:37] stuartm: then
[00:21:43] jya: packaging shouldn't matter... seeing the shepherd auto update itself nicely
[00:22:14] jya: got an email from the cubox guy, they are sending me a prototype board to port myth... they say they have openGL available through X
[00:22:24] jya: that should give a headstart compare to the pi
[00:22:35] stuartm: jya: sorry, I thought you were talking about the QTScript issue
[00:22:52] jya: ah no, I was still on shepherd :)
[00:27:37] stichnot (stichnot!~stichnot@mythtv/developer/stichnot) has quit (Ping timeout: 272 seconds)
[00:37:01] jheizer_: bill6502, ah, good call on the log so I'll assume it was expected. Still for my "Photographs" defined in .27 to try to implement against it vs a VM of master it returns recordings still.
[00:51:59] NightMonkey (NightMonkey!~NightrMon@pdpc/supporter/professional/nightmonkey) has quit (Quit: Body blow! Body blow!)
[00:52:46] rsiebert_ (rsiebert_! has joined #mythtv
[00:55:33] nascar (nascar! has joined #mythtv
[00:55:51] rsiebert (rsiebert! has quit (Ping timeout: 245 seconds)
[01:43:17] stichnot (stichnot!~stichnot@mythtv/developer/stichnot) has joined #mythtv
[01:43:57] skd5aner: stuartm: I had full relative paths to my icons in the db. I had to parse it via SQL to remove everything before the last "/" and reinsert just the filename. However, I think the majority of mine were entered manually because the buit-in grabber grabs the low-quality icons from lyngsat
[01:44:23] skd5aner: and, I manually downloaded all of the HQ icons
[01:44:34] skd5aner: (like, 5 years ago)
[01:44:48] skd5aner: I just manually suppliment as needed
[01:45:27] skd5aner: I spent hundreds of hours working on approving submissions to the mythtv icon service so that it would be good for others, but I myself don't even use it... heh :)
[01:45:53] skd5aner: it's probably been 2 years since I've went in to that system though and tried to approve icons for the US
[01:46:02] skd5aner: I bet no one else has since then either
[01:47:22] moparisthebest_ (moparisthebest_! has joined #mythtv
[01:48:14] moparisthebest (moparisthebest! has quit (Ping timeout: 264 seconds)
[01:57:11] jya: stuartm: what was wrong with the "Add channelIcons to the list of "special" storage group?
[01:58:16] ** jya a tad too many reverts on a fixes branch to my liking :( why were they committed in the first place so quickly? **
[02:01:01] skd5aner: jya: lots of discussion on that in today's backlog...
[02:01:51] skd5aner: high level summary, it's already considered a "built-in" storage group, which, from what I understand, is different than a "special" storage group
[02:02:41] skd5aner: built-in ones do not need to be defined, whereas specials ones must be created and stored in the db, at least, that was my understanding
[02:03:29] skd5aner: I wrote the patch, and it worked in my case, but the wiki gave erronous instructions. It said to create a channelicon storage group, which isn't actually necessary/recommended...
[02:04:01] wagnerrp: sounds like we need some sanity check to prevent the creation of built-in groups in the database
[02:04:28] wagnerrp: make the UI refuse to do it
[02:04:34] skd5aner: wagnerrp: probably – and stuartm recommended potentially a schema update that would remove any that were created based on the bad instructions
[02:14:52] _nyloc_ (_nyloc_! has joined #mythtv
[02:15:14] dblain (dblain!~dblain@mythtv/developer/dblain) has joined #mythtv
[02:19:02] nyloc (nyloc! has quit (Ping timeout: 240 seconds)
[02:29:02] nascar (nascar! has quit (Quit: ChatZilla [Firefox 24.0/20130911155223])
[02:39:19] peper03 (peper03!~peper03@mythtv/developer/peper03) has quit (Ping timeout: 260 seconds)
[02:39:52] peper03 (peper03!~peper03@mythtv/developer/peper03) has joined #mythtv
[02:45:29] dgeary2_ (dgeary2_! has quit (Quit: Ex-Chat)
[03:06:35] joki (joki! has quit (Ping timeout: 272 seconds)
[03:10:48] joki (joki! has joined #mythtv
[03:47:06] fetzerch (fetzerch!~quassel@unaffiliated/fetzerch) has quit (Ping timeout: 245 seconds)
[03:48:42] fetzerch (fetzerch!~quassel@unaffiliated/fetzerch) has joined #mythtv
[04:02:25] jheizer_ (jheizer_! has quit (Quit: Leaving)
[04:42:14] dblain: stuartm: Why would you need to parse a ISO date? (although you should be able to). If you are trying to call a service function, you should be able to just pass a QDateTime directly to it. Guess I'd need to understand what you're trying to accomplish before I can offer help.
[04:42:42] dblain: Also, how did you get QScript to crash the backend? I'd like to be able to protect that from happening/
[05:39:13] stichnot (stichnot!~stichnot@mythtv/developer/stichnot) has quit (Ping timeout: 272 seconds)
[06:07:36] FabriceMG (FabriceMG! has joined #mythtv
[06:22:13] seld_ (seld_! has quit (Ping timeout: 245 seconds)
[06:41:16] SteveGoodey (SteveGoodey! has joined #mythtv
[07:01:39] cbovy (cbovy!~cbovy@ has joined #mythtv
[07:40:42] stichnot (stichnot!~stichnot@mythtv/developer/stichnot) has joined #mythtv
[09:29:10] stuartm: dblain: I was working in QTScript/ECMA, I wasn't aware you could create a QDateTime object instead of using Date?
[09:30:05] stuartm: dblain: the services stuff calls throw() all over the place, but it's not being caught hence the backend crash
[09:38:10] stuarta: that's not great
[09:42:52] stuartm: dblain: in my case because it couldn't parse the iso date format, GetProgramGuide() was being called with an invalid date from the QTScript –
[09:52:58] dekarl1 (dekarl1! has joined #mythtv
[09:55:17] dekarl (dekarl! has quit (Ping timeout: 261 seconds)
[10:45:05] stuartm: dblain: services API allows various arguments to be optional, but the actual methods don't – e.g. GetProgramGuide() must be called with all five arguments from QTScript – can that restriction be lifted?
[10:47:34] stuartm: mostly for reasons of consistency
[10:49:30] stuartm: consistency in usage – we don't need one set of documentation; and behaviour – if defaults for optional values get changed in the future it applies to everything
[11:01:46] SteveGoodey (SteveGoodey! has quit (Quit: Konversation terminated!)
[11:07:32] stuartm: dblain: I'm having trouble determining what the default value for optional boolean arguments is?
[11:44:37] knightr (knightr!~Nicolas@mythtv/developer/knightr) has quit (Read error: Connection reset by peer)
[12:29:19] knightr (knightr!~Nicolas@mythtv/developer/knightr) has joined #mythtv
[12:38:04] Jordack (Jordack! has joined #mythtv
[13:09:53] FabriceMG (FabriceMG! has quit (Ping timeout: 272 seconds)
[13:28:04] stichnot: jpabq: For #11882, this "fixes" playback:
[13:28:04] ** MythLogBot **
[13:30:18] stichnot: The issue seems to be that AvFormatDecoder::PreProcessVideoPacket() is waiting to see the first actual keyframe in the file before allowing AvFormatDecoder::GetFrame() to return success.
[13:31:05] stichnot: This is essentially defined by the condition (pkt->flags & AV_PKT_FLAG_KEY), which is not happening in the problematic recordings.
[13:33:54] stichnot: So I guess ffmpeg must have another way of determining keyframes in these recordings.
[13:45:16] moparisthebest_ is now known as moparisthebest
[13:50:48] FabriceMG (FabriceMG! has joined #mythtv
[14:03:44] TandyUK2 (TandyUK2! has joined #mythtv
[14:04:49] TandyUK (TandyUK! has quit (Ping timeout: 246 seconds)
[14:08:15] TandyUK (TandyUK! has joined #mythtv
[14:11:19] TandyUK2 (TandyUK2! has quit (Ping timeout: 272 seconds)
[14:13:44] FabriceMG (FabriceMG! has quit (Quit: FabriceMG)
[14:21:29] TandyUK (TandyUK! has quit (Quit: I wonder what this 'X' does?)
[14:32:34] jpabq: stichnot: thanks. I will take a look at this tomorrow.
[14:40:27] stichnot: jpabq: thanks, I may also get time to look more in the meantime.
[14:48:57] stichnot (stichnot!~stichnot@mythtv/developer/stichnot) has quit (Ping timeout: 272 seconds)
[14:48:58] cbovy (cbovy!~cbovy@ has quit (Ping timeout: 272 seconds)
[14:58:25] cbovy (cbovy!~cbovy@ has joined #mythtv
[15:26:37] dblain: stuartm: The reason I used throw() through-out the services is because I needed to be able to use the return type for the data, but needed to return errors as well. (a null return tells nothing of why it failed). Exceptions seemed like the simplest solution.
[15:29:15] dblain: As for the optional parameters. I'll have to see if it's possible to provide default values in the function prototypes and still have the metadata generated correctly. Since they are considered slots as far as Qt is concerned, I'm not sure if it will handle it correctly.
[15:30:02] dblain: BTW: it would be easy enough to make all parameters required in the REST calls! (not that I think that is a good idea)
[15:33:22] SteveGoodey (SteveGoodey! has joined #mythtv
[15:36:24] dblain: The default for boolean arguments should be 'false'
[15:37:48] cbovy (cbovy!~cbovy@ has quit (Ping timeout: 240 seconds)
[15:47:14] stichnot (stichnot!~stichnot@mythtv/developer/stichnot) has joined #mythtv
[15:47:51] sphery: stichnot: FWIW, I replied to the "Hauppauge PVR questions" thread with my concerns about storing (or, really, exposing) the card ID for recordings. Basically, if you store a cardinput.displayname in recorded, it's fine with me (but doesn't make sense in oldrecorded), but please make we emphasize that value is only meaningful long term with external system configuration to make it so. Sorry for taking so long to get back to you. (Not even ...
[15:47:58] sphery: ... sure if you've already done something for it.)
[15:49:10] sphery: and if you're super motivated, feel free to add a capturecard.displayname and store that, too, and use it in backend status page and upcoming recordings and wherever else we expose cardid.  :)
[15:54:26] stuartm: dblain: if you put the example I posted earlier into a .qsp and then fetch it, the exceptions generated aren't caught but I couldn't figure out where or why (backtrace didn't help)
[15:55:08] stuartm: dblain: thanks for the answers otherwise :)
[15:57:56] stuartm: dblain: there's overlap between the Channel::GetChannelInfo() class and ServiceUtil::FillChannelInfo(), I was going to eliminate the latter as it doesn't populate all the fields in the Channel object (most notably 'visible')
[15:59:41] stuartm: any objections?
[16:01:58] dblain: stuartm: none from me.
[16:06:55] stichnot: sphery: np. I don't have any immediate intentions to do something about this item, but I think I understand the issues.
[16:11:55] stichnot (stichnot!~stichnot@mythtv/developer/stichnot) has quit (Ping timeout: 272 seconds)
[16:12:47] natanojl (natanojl!~jonatan@mythtv/developer/natanojl) has joined #mythtv
[16:14:59] seld (seld! has joined #mythtv
[16:24:30] stichnot (stichnot!~stichnot@ has joined #mythtv
[16:24:31] stichnot (stichnot!~stichnot@mythtv/developer/stichnot) has joined #mythtv
[16:24:31] stichnot (stichnot!~stichnot@ has quit (Changing host)
[16:30:04] lapion (lapion! has joined #mythtv
[16:30:18] lapion (lapion! has left #mythtv ("Ex-Chat")
[16:48:05] NightMonkey (NightMonkey!~NightrMon@pdpc/supporter/professional/nightmonkey) has joined #mythtv
[16:49:58] robink (robink!~quassel@unaffilated/robink) has quit (Quit: No Ping reply in 180 seconds.)
[16:50:21] robink (robink!~quassel@unaffilated/robink) has joined #mythtv
[17:55:10] cbovy (cbovy!~cbovy@ has joined #mythtv
[17:57:02] cbovy (cbovy!~cbovy@ has quit (Client Quit)
[19:38:23] joki (joki! has quit (Ping timeout: 272 seconds)
[19:42:03] jams_ is now known as jams
[20:07:31] joki (joki! has joined #mythtv
[20:12:47] stichnot (stichnot!~stichnot@mythtv/developer/stichnot) has joined #mythtv
[20:24:07] gigem (gigem!~david@mythtv/developer/gigem) has quit (Quit: WeeChat 0.4.1)
[20:25:05] gigem (gigem!~david@mythtv/developer/gigem) has joined #mythtv
[20:47:50] SteveGoodey (SteveGoodey! has quit (Quit: Konversation terminated!)
[21:29:03] natanojl (natanojl!~jonatan@mythtv/developer/natanojl) has quit (Ping timeout: 240 seconds)
[21:51:39] nvzn (nvzn!~nvzn@unaffiliated/nvzn) has joined #mythtv
[21:54:47] nvzn_ (nvzn_!~nvzn@unaffiliated/nvzn) has quit (Ping timeout: 260 seconds)
[22:06:32] Jordack (Jordack! has quit ()
[22:17:25] Tobbe5178 (Tobbe5178! has quit (Read error: Connection reset by peer)
[22:44:32] nvzn_ (nvzn_!~nvzn@unaffiliated/nvzn) has joined #mythtv
[22:47:23] nvzn (nvzn!~nvzn@unaffiliated/nvzn) has quit (Ping timeout: 248 seconds)
[23:57:50] stuartm: somewhere we're losing the strict starttime ordering of programs in GetProgramGuide() – whether called via the service API or via QTScript the ordering of the programs changes and they are rarely in starttime order
[23:59:01] stuartm: ProgramList (AutoDeque) is an ordered container, so is QVariantList (QList), so I'm not sure where it all goes wrong unless it's the final step where we serialise the data

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