MythLogBot@irc.freenode.net :: #mythtv-users

Daily chat history

Current users (68):

mengoshmink, MythLogBot, mad_enz, Valen, bpmedley, clueminati, ink0gnito, aloril, ghoti, illuminated, Muzer, troyt, Maliuta, Moscherkobold, shurd, gregl, jm|laptop, jpabq, Scopeuk, EtherMan, Floppe, G, knowledgejunkie, clever, frobnic, GreyFoxx, Heliwr, tonsofpcs, afisher, LedHed, p71, pppingme, sutula, AndyCap, niska, SleePy, tris, ChanServ, GWG, libsci, BLZbubba, kusznir, mkbloke, Panic, rhpot1991, _charly__, Azelphur, ijc, jya, lotia, mack-, IonTau, blahdodo, amessina, trumee, Maliuta1, vesper11, gigem_, croppa_, esteban_, brfransen, Guest87029, cybrNaut, sphery, nutron|w, quicksilver, doppo_, Guest76513
Sunday, August 23rd, 2020, 01:02 UTC
[01:02:06] cybrNaut: where is the commercial flagging data stored?
[01:03:05] cybrNaut: when I do "select cutlist from recorded", it's apparently all booleans.
[01:03:31] cybrNaut: same for "select editing from recorded"
[01:05:47] cybrNaut: i would expect the commercial markers to be in a separate table, but nothing in https://www.mythtv.org/wiki/Database seems to hold that data
[01:10:49] jm|laptop: cybrNaut: https://www.mythtv.org/wiki/Recordedseek_table ?
[01:17:07] cybrNaut: thanks!
[01:40:47] cybrNaut: hmm.. i don't think this contains the commercial markers
[01:43:19] cybrNaut: I dumped the rows for a single recording, and it was hundreds of rows of output (yet only two slices of commercials are in this recording). The marker "type" was either a "9" or "33", nothing that would correspond
[01:46:18] cybrNaut: this is my query: 'select * from recordedseek where chanid = "1234" and starttime like "2020-08–22_07_00_00" and (type = "4" or type = "5");'
[01:52:02] cybrNaut: in fact, if I remove the chanid and starttime criteria, there is still zero output. not a single MARK_COMM_START or MARK_COMM_END in the entire table
[02:01:38] mkbloke: cybrNaut: silly question, but is the mythcommflag job running correctly? have you tried running mythcommflag on the command line? you could then add -v <level> to it to ger more details, which *might* help you out?
[02:05:52] cybrNaut: if I press "e" during playback followed by "z", it shows red segments covering the commercials.
[02:07:44] cybrNaut: i try mythcommflag, but if it were failing then wouldn't the problem manifest in some way? Playback avoids commercials just fine in the cases where the commercials are marked.
[02:08:18] cybrNaut: i mean "i /could/ try mythcommflag.."
[02:08:54] mkbloke: I'm guessing that means it's working, although I don't the the FE to know the specifics of comm skipping within it. it's possible a manual com flag with appropriate verbosity could give you the info you're looking for though...
[02:09:46] mkbloke: s/the the/use the/
[02:10:30] cybrNaut: maybe, but it's not ideal because sometimes i manually edit the commercial markers, and running mythcommflag would lose those edits
[02:11:46] mkbloke: I think you can run mythcommflag against a particular recording. you could just use a test recording for the purpose?
[02:13:43] cybrNaut: i'll have to try it b/c i'm out of options
[02:23:30] cybrNaut: i've noticed that the built-in transcode operation does not cut out commercials (even though playback does), so maybe there is a problem. It seems transcoding relies on the seek table and playback gets the commercial flags from somewhere else
[02:31:31] mkbloke: I'm not entirely familiar with how it works, as use neither comm flag nor transcode, but yes, I *think* flagging is not cutting.
[02:32:24] mkbloke: s/as use/as I use/
[02:32:32] cybrNaut: so i ran "mythcommflag --loglevel debug -v -f $some_file" and after reaching 100% it dumped a bunch of errors like "2020-08–22 22:29:02.903306 E decoding error End of file (-541478725)"  — but edit mode still correctly shows the red segments
[02:34:10] cybrNaut: the Recordedseek table has lots of "type 9" and "type 33" rows, corresponding to MARK_GOP_BYFRAME and MARK_DURATION_MS
[02:34:48] cybrNaut: but not a single MARK_CUT_END, MARK_CUT_START, MARK_COMM_START, or MARK_COMM_END row
[02:35:43] cybrNaut: the verbose debug output of mythcommflag doesn't give that info either
[02:35:54] cybrNaut: i wonder if it's being embedded in the TS file
[02:36:17] mkbloke: You might be better off posting to the mythtv-users mailing list on this unless anybody else here can help you.
[02:38:03] cybrNaut: i think i've tried that before and the anti-spam measures made it too hard.. but I appreciate the suggestion. (same problem using the forums, and same problem posting bug reports)
[02:38:55] cybrNaut: irc seems to be the only open free-world option for mythtv
[02:40:01] cybrNaut: maybe i'll try reading the code to see what mythcommflag does when it detects a commercial
[02:46:18] cybrNaut: https://code.mythtv.org/doxygen/mythtv_2progr . . . .html#l00545
[02:46:55] cybrNaut: i see a lot of "if (useDB)" statements, which implies that there's a non-database approach
[02:49:24] mkbloke: that *could* be related to the --skipdb option
[02:50:01] mkbloke: according to this: https://github.com/MythTV/mythtv/blob/master/ . . . g/README.txt the ints you are looking for (4,5) are used...
[02:52:40] cybrNaut: thanks for the link.. looks like an interesting read. indeed 4 and 5 mark types are missing from my seek table
[02:56:31] cybrNaut: interesting.. if I supply the --outputfile option, the output file shows a list with marks 4 and 5.
[03:15:01] Steve-Goodey (Steve-Goodey!~steve@2a00:23c5:7d83:6501:6032:9caa:16d4:6ffa) has joined #mythtv-users
[03:15:01] Mode for #mythtv-users by ChanServ!ChanServ@services. : +v Steve-Goodey
[03:31:14] bill6502 (bill6502!~bill@205.178.24.168) has joined #mythtv-users
[03:31:44] bill6502: cybrNaut: recordedmarkup
[03:32:06] bill6502: 4 and 5 are the result of mythcommflag
[03:32:39] bill6502: 0 and 1 are the cut list
[03:33:04] bill6502: mythcommflag doesn't set the cutlist
[03:34:20] bill6502: you do after pressing e, and then likely z to load the likely commercial starts/ends
[03:34:58] bill6502: exit the e(dit) mode and save the results and you have a cutlist
[03:37:41] bill6502: mythutil has a bunch of switches. Try: mythutil --help getcutlist for starters
[03:38:24] Steve-Goodey (Steve-Goodey!~steve@2a00:23c5:7d83:6501:6032:9caa:16d4:6ffa) has quit (Quit: Konversation terminated!)
[04:26:13] illuminated_ is now known as illuminated
[04:29:27] bill6502 (bill6502!~bill@205.178.24.168) has left #mythtv-users ()
[04:54:05] mkbloke: bill6502++
[05:30:00] peterben1 (peterben1!~pi@2601:183:100:50e:5e48:98fb:7c6e:ebaf) has quit (Quit: WeeChat 2.3)
[05:31:51] peterbennett (peterbennett!~pi@mythtv/developer/peterbennett) has joined #mythtv-users
[05:31:51] Mode for #mythtv-users by ChanServ!ChanServ@services. : +v peterbennett
[07:27:03] Steve-Goodey (Steve-Goodey!~steve@2a00:23c5:7d83:6501:6032:9caa:16d4:6ffa) has joined #mythtv-users
[07:27:03] Mode for #mythtv-users by ChanServ!ChanServ@services. : +v Steve-Goodey
[08:18:25] jya (jya!sid396171@gateway/web/irccloud.com/x-nxjzaijlugeikask) has quit (Ping timeout: 240 seconds)
[08:20:24] jya (jya!sid396171@gateway/web/irccloud.com/x-gfejzorjptdxthnv) has joined #mythtv-users
[09:07:26] RokLobsta (RokLobsta!~Dirkka@121-200-12-5.79c80c.mel.nbn.aussiebb.net) has quit (Remote host closed the connection)
[09:24:22] user798645 (user798645!~User@user-10mt7in.cable.mindspring.com) has joined #mythtv-users
[10:32:44] markk__ (markk__!~mark@host86-161-98-194.range86-161.btcentralplus.com) has joined #mythtv-users
[11:48:45] Warped (Warped!~Warped@unaffiliated/warped) has quit (Ping timeout: 272 seconds)
[12:12:00] markk__ (markk__!~mark@host86-161-98-194.range86-161.btcentralplus.com) has quit (Ping timeout: 256 seconds)
[12:15:34] Warped (Warped!~Warped@unaffiliated/warped) has joined #mythtv-users
[13:28:01] cybrNaut: bill6502: thanks! that's quite helpful
[14:07:47] bpmedley (bpmedley!~bpm@2601:246:8201:8e0:8b:cc94:4a72:8f4f) has quit (Ping timeout: 260 seconds)
[15:16:48] SteveGoodey (SteveGoodey!~steve@host86-149-198-216.range86-149.btcentralplus.com) has joined #mythtv-users
[15:16:48] Mode for #mythtv-users by ChanServ!ChanServ@services. : +v SteveGoodey
[16:08:59] user798645 (user798645!~User@user-10mt7in.cable.mindspring.com) has quit (Ping timeout: 240 seconds)
[17:01:51] jm|laptop (jm|laptop!~jamiem@null.jamiem.com) has quit (Quit: Bye.)
[17:02:07] jm|laptop (jm|laptop!~jamiem@null.jamiem.com) has joined #mythtv-users
[18:00:41] cybrNaut: mythutil seems to work on TS files but not MPG files. I don't know why mpg files get made (why doesn't mythtv produce a TS file consistently for all recordings?) Anyway, when something like "mythutil --getcutlist --chanid 1234 --starttime 20200723180000" refers to an mpg-formatted recording, the output is simply "Cutlist: ", yet "e >> z" during playback reveals that there are cut segments.
[18:00:59] cybrNaut: TS files have no problem
[18:28:50] bpmedley (bpmedley!~bpm@2601:246:8201:8e0:3d48:b765:d72c:1117) has joined #mythtv-users
[19:06:57] peterbennett: cybrNaut: When does mythtv produce mpg files? AFAIK those were only produced by mythtv in an older version.
[19:19:52] cybrNaut: in my case ~3% of the files are .mpg for some reason.
[19:28:46] cybrNaut: the only pattern that I see is that all mpg files were recorded in July
[19:33:07] cybrNaut: well, it's clearly a date separation. everything recorded before july 26th is mpg, and everything after is TS. I'm almost certain I did not upgrade versions on july 26th. Maybe I changed a setting that had that effect.
[20:24:43] Steve-Goodey (Steve-Goodey!~steve@2a00:23c5:7d83:6501:6032:9caa:16d4:6ffa) has quit (Quit: Konversation terminated!)
[20:28:13] gigem (gigem!~david@072-181-169-230.res.spectrum.com) has joined #mythtv-users
[20:28:13] gigem (gigem!~david@mythtv/developer/gigem) has joined #mythtv-users
[20:28:13] gigem (gigem!~david@072-181-169-230.res.spectrum.com) has quit (Changing host)
[20:28:13] Mode for #mythtv-users by ChanServ!ChanServ@services. : +v gigem
[20:36:30] gigem_ (gigem_!~david@47.189.228.62) has joined #mythtv-users
[20:36:39] gigem_ (gigem_!~david@47.189.228.62) has quit (Changing host)
[20:36:39] gigem_ (gigem_!~david@mythtv/developer/gigem) has joined #mythtv-users
[20:36:40] Mode for #mythtv-users by ChanServ!ChanServ@services. : +v gigem_
[21:07:16] SteveGoodey (SteveGoodey!~steve@host86-149-198-216.range86-149.btcentralplus.com) has quit (Quit: Konversation terminated!)
[22:32:00] RokLobsta (RokLobsta!~Dirkka@121-200-12-5.79c80c.mel.nbn.aussiebb.net) has joined #mythtv-users
[23:50:52] IonTau (IonTau!~IonTau@203-217-53-25.dyn.iinet.net.au) has joined #mythtv-users
[23:52:39] Warped (Warped!~Warped@unaffiliated/warped) has quit (Ping timeout: 272 seconds)
[23:56:20] RokLobsta (RokLobsta!~Dirkka@121-200-12-5.79c80c.mel.nbn.aussiebb.net) has quit (Remote host closed the connection)

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