:: #mythtv

Daily chat history

Current users (88):

aloril, Anssi, anykey_, Beirdo, brfransen, brtb, CaCtus491, cattelan_away, cesman, Chutt, clever, coling, Cougar, damaltor, danielk22, Dave123, dekarl, dlblog, Dr{Wh0}, ElmerFudd, foobum, foxbuntu, ghoti, gregL, GreyFoxx, highzeth, J-e-f-f-A, j-rod|afk, jams, jarle, jcarlos, joe_, joki, jpabq, jpabq-, jpabq_, jstenback, justinh, jwhite, jya, k-man, kenni, knightr, kurre2, kwmonroe, laga, lolcat`, mag0o, MaverickTech, mike|2, mrand, MythBuild, MythLogBot, mzanetti, NightMonkey, peitolm, pheld, poptix, purserj, rhpot1991, RingZer0, rsiebert, skd5aner, Slasher`, sphery, sraue, stuarta, stuartm, superm1, sutula, tester68, tgm4883, TheAsp, ThisNewGuy1, tomimo, tris, Unhelpful, vallor, wagnerrp, wahrhaft, wseltzer, xavierh, xavierh_, XDS2010_, xris, ybot_, zCougar, _charly_
Wednesday, March 7th, 2012, 00:02 UTC
[00:02:39] sraue (sraue!~stephan@xbmc/staff/sraue) has quit (Ping timeout: 260 seconds)
[00:03:19] davide (davide! has joined #mythtv
[00:06:30] joki (joki! has joined #mythtv
[00:18:55] joki (joki! has quit (Ping timeout: 276 seconds)
[00:21:53] stichnot: stuartm, natanojl: re #9407. It looks like there were a number of changes to SubtitleScreen::DisplayDVDButton() after the ticket was opened. Does the problem actually still exist in master?
[00:25:01] knightr: stuarm, shouldn't 1d9ebefaa have translation calls (... of ...) ? (I noticed 31a6a77b1 too but I saw that you fixed it in f1937eaaa). I could add the tr() but I would have to get that reviewed before committing and since it's quite possible that you are still modifying this piece of code I kinda prefer not to touch it...
[00:26:53] jpabq (jpabq!~jpabq@mythtv/developer/jpabq) has quit (Quit: jpabq)
[00:26:54] abqjp is now known as jpabq
[00:26:55] jpabq (jpabq! has quit (Changing host)
[00:26:55] jpabq (jpabq!~jpabq@mythtv/developer/jpabq) has joined #mythtv
[00:52:02] joki (joki! has joined #mythtv
[00:59:27] jya: stuarta: on your mac mini, what does sysctl hw.cpu64bit_capable returns please ?
[01:03:38] stichnot (stichnot!~chatzilla@mythtv/developer/stichnot) has quit (Ping timeout: 240 seconds)
[01:09:17] jya: just catching up on the discussions here.. I agree that rather than break upgrades for people who set up vdpaubuffersize, we hsould rename the settings if it now has a different meaning now. At least there will be no conflict and no particular operation to perform after the upgrade
[01:10:14] ** wagnerrp suggests jya catch up on commits too **
[01:10:14] wagnerrp: :)
[01:10:28] jya: ah it's been done? excellent
[01:10:30] wagnerrp: specifically... a9bf1d341
[01:10:37] jya: who said we can't make decisions uh ?
[01:10:43] wagnerrp: about 5 hours ago
[01:13:00] jya: actually, looking at markk's original commit, I don't think vdpaubuffersize meaning was changed.. it overrides what was automatically detected, not add to it
[01:13:25] joki (joki! has quit (Ping timeout: 260 seconds)
[01:14:02] jya: so I don't think sphery 's recollection was good there
[01:14:12] wagnerrp: so if we only have 99 hosts hitting smolt, is that all the people actually running 0.25?
[01:14:46] jya: wagnerrp: is smolt always run by itself? because I don't think I ever called it
[01:14:57] wagnerrp: i dont know if it is enabled by default
[01:15:48] jya: I would… but I'm a big brother's fan anyway
[01:16:47] wagnerrp: im still waiting for it to get moved to the new server on alcor, so we can actually see those new statistics
[01:17:22] wagnerrp: (plus i need to write up that ELD parsing bit)
[01:18:02] jya: wagnerrp: I would just run the output of that command I gave you…
[01:18:55] jya: cat /proc/asound/card*/eld*
[01:18:59] wagnerrp: whether on the client or the server end, it still has to get parsed somewhere if we want to do meaningful display of the information
[01:19:24] wagnerrp: otherwise, 95 bytes of binary data is just gibberish
[01:19:34] jya: now that wouldn't tell us which card has an ELD nor which device has one
[01:19:54] jya: and that also works on all machines, not just the one with ALSA > 1.0.25
[01:20:11] sraue (sraue!~stephan@xbmc/staff/sraue) has joined #mythtv
[01:20:13] wagnerrp: is it possible that would only be visible if the tv is on?
[01:21:13] jya: it will only be visible if the machine was able to read the EDID at the time the video drivers initialised (e.g. when X started)
[01:21:41] jya: once it's read, it doesn't matter if the TV is turned off or not. I believe it only tries to read it once
[01:21:49] wagnerrp: ah, so if i force fed the nvidia drivers an EDID with that blanked out, it would not exist
[01:22:15] wagnerrp: on the other hand, if i force fed the nvidia drives the original EDID, it should show up whether the TV is on or not?
[01:22:17] jya: wagnerrp: yes.. it would be in the CustomEDID info you're loading
[01:22:31] jya: ELD is part of EDID
[01:22:34] wagnerrp: i still have a copy of the original floating around
[01:22:52] wagnerrp: yeah, i didnt know if ALSA was manually pulling a second copy of it
[01:22:59] wagnerrp: or if it was all done by X
[01:24:13] jya: sphery: yep… in regards to vdpaubuffersize, I don't see why anyone would have had problems upgrading keeping the value as is… The actual meaning of vdpaubuffersize (and now vdpaubuffercount) is that it's the minimum size of the buffer. If the vdpau code detects that it needs to be X buffers instead, it adds (X-vdpaubuffer size).. so really, danielk22 change was not necessary. it would have been all good
[01:24:32] jya: wagnerrp: having said that. 've noticed since I upgraded to 1.0.25
[01:24:42] jya: that if I suspend my machine and restart it.
[01:25:05] jya: if the TV is off at the time of the PC powering back on from sleep
[01:25:14] jya: then I'll get no audio anymore (and no ELD)
[01:25:29] wagnerrp: even with a stored EDID?
[01:25:49] jya: I believe that the reason is that by default, ubuntu/mythbuntu, before going to sleep unload the alsa kernel module, to reload it later. I think this force reading the ELD again
[01:26:13] jya: wagnerrp: yep, even with a stored EDID. in was actually so annoying that behavour that I uninstalled 1.0.25 and put back 1.0.24
[01:27:11] jya: because I suspend my frontend several times a day, but I use the setup once only.. so I know what priority to give :)
[01:27:16] Anssi: ELD is not part of EDID... ELD is an hda-intel interchange format between audio driver <=> display driver
[01:27:33] Anssi: but the data in it is generated from EDID, yes
[01:27:44] jya: Anssi: what I meant by that is that the ELD info is found in the EDID
[01:27:50] Anssi: just nitpicking here :)
[01:28:47] jya: glad you're monitoring this channel ; do you have a filter on edid, eld and alsa ? :)
[01:29:41] Anssi: no, this was pure chance :)
[01:29:54] noahric (noahric!~noahric@ has quit (Quit: noahric)
[01:30:42] jya: Anssi: according to intel doc, the ELD binary data is found in the EDID CEA-861-B extension
[01:31:49] jya: in any case, too many problems with the released 1.0.25 IMHO, no audio with nvidia cards , doesn't work after suspend… it's a big of a disaster that release
[01:33:40] Anssi: yep, the imporant parts like SADs have the same format as EDID so they are straight-copied
[01:34:45] jya: you know… feel free to use mythtv's ELD reader class code if you ever want XBMC to catch up with us :)
[01:34:48] wagnerrp: jya: it seems you have to go through the frontend setup wizard at least once to enable smolt
[01:35:12] jya: wagnerrp: yeah, and I doubt people upgrading would bother doing so
[01:35:23] jya: unless you're curious
[01:35:29] wagnerrp: it wont run automatically unless it has already been forced once, by using the setup wizard
[01:37:00] jya: the first time mythfrontend is run (after an upgrade), I would prompt the user if they are okay to help us by uploading their data, with YES as the answer by default and run smolt then
[01:37:41] jya: for the machine that have the python plugin installed that is (mac doesn't)
[01:38:49] wagnerrp: thats the other thing im concerned about
[01:38:50] stuartm (stuartm!~stuartm@mythtv/developer/stuartm) has quit (Remote host closed the connection)
[01:39:12] wagnerrp: the automatic operation is disabled if mythtv was compiled without the bindings
[01:39:28] wagnerrp: it does not check to see if they were independently installed
[01:39:55] wagnerrp: these distros that are splitting anything and everything out into separate packages
[01:40:03] Anssi: jya: heh, I might if I get around to it :)
[01:40:10] wagnerrp: i dont know if mythtv is compiled with the bindings enabled
[01:41:13] jya: wagnerrp: at least for ubuntu, while the bindings are packaged separately, they are the result of a single compilation of mythtv
[01:42:08] jya: they just run scripts extracting whatever they need
[01:42:23] jya: I even think that the python binding is a required dependency
[01:42:26] jya: let me check
[01:43:19] jya: yep.. if I try to remove the python bindings, it will remove everything frontend related
[01:43:39] jya: doesn't touch the backend related packages though
[01:43:54] jya: aren't the python scripts used from the backend? (like the metadata)
[01:44:12] wagnerrp: frontend and backend, yes
[01:44:28] jya: ok, so theres a bug in their package then
[01:44:39] jya: python should also be made a dependency of the backend too
[01:46:08] Chutt (Chutt! has quit (Ping timeout: 240 seconds)
[01:58:07] Chutt (Chutt! has joined #mythtv
[02:03:30] joki (joki! has joined #mythtv
[02:06:07] jya: wagnerrp stuartm: shepherd have implemented the channel icons properly, and it is now returned in the xmltv… would the data in the xmltv override any channel icon settings stored in mysql ?
[02:06:09] davide (davide! has quit (Remote host closed the connection)
[02:06:31] wagnerrp: dont really know anything about that
[02:06:32] davide (davide! has joined #mythtv
[02:06:50] wagnerrp: and youre going to be waiting a couple hours before you get any response from stuartm
[02:08:11] jya: and he's not connected, so probably won't read it
[02:08:30] wagnerrp: heh, well there is that too
[02:10:17] wagnerrp: lousy election primaries, cluttering up my recordings
[02:11:16] wagnerrp: its ohio, they use paper ballots so it takes forever to see any districts reporting in
[02:11:56] wagnerrp: anything before about 11PM is just going to be the same exact absentee numbers they've been repeating for the past five hours, so why even bother?
[02:22:10] joki (joki! has quit (Ping timeout: 272 seconds)
[02:24:32] joki (joki! has joined #mythtv
[02:30:45] zombor_ (zombor_!~zombor_@kohana/developer/zombor) has joined #mythtv
[02:31:14] zombor_ (zombor_!~zombor_@kohana/developer/zombor) has quit (Remote host closed the connection)
[02:32:09] joki (joki! has quit (Ping timeout: 260 seconds)
[02:43:57] help (help! has joined #mythtv
[02:44:22] help is now known as Guest10049
[02:44:43] Guest10049 (Guest10049! has quit (Client Quit)
[02:45:25] dblain (dblain!~dblain@mythtv/developer/dblain) has quit (Ping timeout: 265 seconds)
[02:45:40] dblain (dblain! has joined #mythtv
[02:45:41] dblain (dblain! has quit (Changing host)
[02:45:41] dblain (dblain!~dblain@mythtv/developer/dblain) has joined #mythtv
[02:52:54] mike (mike! has joined #mythtv
[02:53:20] mike is now known as Guest50256
[02:55:45] Guest45851 (Guest45851! has quit (Ping timeout: 244 seconds)
[02:56:31] ToadP (ToadP!~jboyle@ has quit (Read error: Connection reset by peer)
[02:57:54] ToadP (ToadP!~jboyle@ has joined #mythtv
[03:08:57] Dave123 (Dave123! has joined #mythtv
[03:21:28] skd5aner: danielk22: you around?
[03:24:04] skd5aner: danielk22: I had a ticket,#9845 ( – was curious if that was the same as the fix you applied for #10416? my ticket never got any love because it was owned by janne around the time he left then transferred to markk :/
[03:24:38] skd5aner: danielk22: anyway – wasn't sure if it was a dupe or unique issue – either way, you mind taking a quick look at the backtrace and see if it's something you might be willing to look at in your queue?
[03:25:15] skd5aner: danielk22: oops, I actually meant a dupe of #9868, which is a different livetv segfault –
[03:25:34] skd5aner: but maybe look at both real quick and see if you see anything :)
[03:43:31] joki (joki! has joined #mythtv
[03:52:14] joki (joki! has quit (Ping timeout: 252 seconds)
[03:58:00] noahric (noahric!~noahric@ has joined #mythtv
[04:25:57] stichnot (stichnot!~chatzilla@mythtv/developer/stichnot) has joined #mythtv
[04:30:01] joki- (joki-! has joined #mythtv
[04:31:22] joki- is now known as joki
[04:41:52] sphery: jya: it's not my recollection, it's my reading of exactly what he said
[04:42:06] sphery: he may have changed it since, but see the post I linked for links to his discussion in here
[04:44:07] joki (joki! has quit (Ping timeout: 276 seconds)
[04:47:26] sphery: jya: and even if the meaning hasn't changed, users shouldn't be overriding the new dynamic sizing, with some random value that made sense to them before, so it makes sense to change the name
[04:48:19] sphery: i.e. they should use the new, smarter code that does a better job of using resources effectively
[04:49:07] jya: sphery: ok… but no matter what settings you've set in vdpaubuffersize, if it worked before it would have worked after. it just sets the minimum size, and if there aren't enough, it adds only what is required
[04:49:25] jya: but yeah, getting rid of the old settings make sense.
[04:49:45] jya: plus, that show our clear split with the original author :)
[04:55:16] sphery: jya: also, as far as channel icons... if present in the xml data, they are available for mythfilldatabase to download, but it will only do so for channels that do not have icons specified in the database
[04:56:39] sphery: (so it won't change icons that users have customized)
[05:18:03] joki (joki! has joined #mythtv
[05:22:30] joki (joki! has quit (Ping timeout: 265 seconds)
[05:32:31] joki (joki! has joined #mythtv
[05:41:11] jya: Beirdo: I think it's unfortunate that you remove the --logfile command
[05:41:41] jya: I still use it , because I like to run two sets of myth at once for development purposes
[05:42:35] wagnerrp: jya: just use two separate log directories
[05:43:23] wagnerrp: or log them same to the same directory, and filter the files using the version stamp at the top of the file
[05:44:38] wagnerrp: or ignore the files all together, and use the database logs
[05:46:03] wagnerrp: Beirdo: speaking of which, did you intend to drop the default database log verbosity?
[05:46:26] Beirdo: hmm?
[05:46:42] Beirdo: you mean like default to off?
[05:47:28] wagnerrp: i mean the log level
[05:47:51] wagnerrp: there was mention of only logging a subset of the file/syslog content to the database a while back
[05:47:55] Beirdo: the there is only one log level
[05:48:06] Beirdo: no plans of ever changing that
[05:48:16] wagnerrp: maybe i just imagined that
[05:48:17] wagnerrp: ignore me
[05:48:42] Beirdo: the whole reason for database logging is to use the log viewer sphery's working on
[05:48:45] Beirdo: :)
[05:53:42] wagnerrp: Beirdo: also... switching smolt to alcor so we start collecting some of the mythtv specific stuff?
[05:56:10] Beirdo: Have to talk to xris, I hope to get some of that (if not all) moved over this weekend
[05:57:01] xris: still waiting for beirdo to tell me the mail server is ready to go so I can move www.  :P
[05:57:10] Beirdo: hehe
[05:57:17] xris: then jams can re-set-up smolt on alcor
[05:57:32] Beirdo: OK, so we want web, then mailinglists last?
[06:03:33] xris: web == mailing lists (due to mailman running on the same host/db)
[06:03:45] xris: need to have it all ready.
[06:03:51] Beirdo: ahhh, right
[06:03:51] xris: I can sync all the files in a matter of minutes.
[06:04:06] Beirdo: forgive me, work has flushed too many useful details from my brain
[06:04:06] Beirdo: heh
[06:04:29] Beirdo: I'll triplecheck the mailserver, but I'm 95% sure it's ready now
[07:01:41] ToadP (ToadP!~jboyle@ has quit (Ping timeout: 244 seconds)
[07:06:00] ToadP (ToadP!~jboyle@ has joined #mythtv
[07:10:49] dlblog (dlblog! has quit (Ping timeout: 260 seconds)
[07:10:58] ToadP (ToadP!~jboyle@ has quit (Ping timeout: 252 seconds)
[07:11:20] dblain (dblain!~dblain@mythtv/developer/dblain) has quit (Ping timeout: 252 seconds)
[07:18:52] dlblog (dlblog! has joined #mythtv
[07:19:05] pheld (pheld! has quit (Quit: Leaving.)
[07:43:51] noahric (noahric!~noahric@ has quit (Quit: noahric)
[07:53:19] gregL (gregL! has quit (Read error: Connection reset by peer)
[08:04:04] jeff999 (jeff999! has joined #mythtv
[08:05:59] jeff999 (jeff999! has left #mythtv ()
[08:10:19] gregL (gregL! has joined #mythtv
[08:21:45] joki (joki! has quit (Ping timeout: 252 seconds)
[08:23:32] joki (joki! has joined #mythtv
[08:32:08] dblain (dblain! has joined #mythtv
[08:32:08] dblain (dblain!~dblain@mythtv/developer/dblain) has joined #mythtv
[08:32:08] dblain (dblain! has quit (Changing host)
[08:52:10] joki (joki! has quit (Ping timeout: 252 seconds)
[08:52:37] joki (joki! has joined #mythtv
[08:57:11] pheld (pheld! has joined #mythtv
[09:03:17] joki (joki! has quit (Ping timeout: 248 seconds)
[09:03:38] joki- (joki-! has joined #mythtv
[09:04:20] joki- is now known as joki
[09:11:32] xavierh_ (xavierh_! has quit (Remote host closed the connection)
[09:15:22] joki (joki! has quit (Ping timeout: 255 seconds)
[09:18:17] ** stuarta yawns **
[09:19:31] stuarta: jya: $ sysctl hw.cpu64bit_capable
[09:19:31] stuarta: hw.cpu64bit_capable: 0
[09:19:39] jya: no error?
[09:19:58] stuarta: nope, command successfully completes, return code 0
[09:20:00] jya: and if you do sysctl hw.cpu64bit_capable ; echo $?
[09:20:05] stuarta: nope, command successfully completes, return code 0
[09:20:08] joki (joki! has joined #mythtv
[09:20:11] jya: cool , thanks
[09:20:27] stuarta: automagic++
[09:22:03] jya: that it return 0 is annoying :)
[09:22:13] jya: I mean the error code
[09:22:50] stuarta: why? it's correct, it tells you what you ask for and the sysctl command completes successfully
[09:23:40] stuarta: it's pretty simple to parse the output of sysctl
[09:25:05] jya: stuarta: because according to my googling, it normally returns an exit code of 1
[09:25:09] jya: so I had coded that
[09:30:17] stuarta: sometimes what people write, that google finds, isn't correct
[09:30:25] kormoc (kormoc!~kormoc@mythtv/developer/kormoc) has quit (Ping timeout: 252 seconds)
[09:30:44] jya: stuarta: I'm trying to compile 0.24 against Qt framework… its even more a pain than 0.25… like I get the qtWebKit framework not found trying to compile libkerneldeint.. there are so many rubbish cross depednencies
[09:31:09] stuarta: i'd give up on 0.24
[09:31:18] jya: I'm about ito...
[09:32:10] stuarta: btw, in 3hrs i'm off to the airport and won't be back online till monday
[09:32:46] jya: cool! so I can commit some changes and if it breaks the autobuild that let me 4 days to fix them
[09:32:52] stuarta: :-p
[09:33:27] ** stuarta looks forward to beer in Munich **
[09:36:02] jya: hum… I like Munich.. love the food there
[09:36:22] stuarta: meat, beer.
[09:36:27] stuarta: er bread
[09:36:32] jya: must be the only French guy liking German food.. or maybe it's just the girls seving it
[09:37:08] joki- (joki-! has joined #mythtv
[09:37:39] joki (joki! has quit (Ping timeout: 252 seconds)
[09:37:39] joki- is now known as joki
[09:44:33] stuartm (stuartm! has joined #mythtv
[09:44:33] stuartm (stuartm! has quit (Changing host)
[09:44:34] stuartm (stuartm!~stuartm@mythtv/developer/stuartm) has joined #mythtv
[10:07:40] joki (joki! has quit (Ping timeout: 272 seconds)
[10:15:42] joki (joki! has joined #mythtv
[10:20:49] joki (joki! has quit (Ping timeout: 276 seconds)
[10:56:40] joki (joki! has joined #mythtv
[11:00:53] joki (joki! has quit (Ping timeout: 252 seconds)
[11:05:02] Guest50256 (Guest50256! has quit (Remote host closed the connection)
[11:05:53] mike|2 (mike|2! has joined #mythtv
[11:07:33] natanojl (natanojl! has joined #mythtv
[11:08:59] natanojl: stichnot: yes, I reproduced it on master
[11:18:18] knightr (knightr!~knightr@mythtv/developer/knightr) has quit (Ping timeout: 260 seconds)
[11:19:05] knightr (knightr!~knightr@mythtv/developer/knightr) has joined #mythtv
[11:20:10] joki (joki! has joined #mythtv
[11:24:38] joki (joki! has quit (Ping timeout: 240 seconds)
[11:25:11] joki (joki! has joined #mythtv
[11:29:58] joki (joki! has quit (Ping timeout: 260 seconds)
[11:36:01] stuartm: knightr: please feel free to fix simple translation related issues like that
[11:38:07] stuartm: jya: currently yes, we always prefer the icon defined in the xml since xmltv data will often push updated icons when the channel logo changes
[11:38:54] jya: sounds like a worthy option: get the latest TV icons :)
[11:41:21] stuartm: stichnot, sphery: the tv play window is called "Playback" – tv_play.cpp ln 1198
[11:49:26] stuartm: jams: ok, so that confirms the problem is MythTV in 0.25, I don't suppose you'd be willing to bisect to find the problem commit?
[11:50:14] zombor (zombor!~zombor_@kohana/developer/zombor) has joined #mythtv
[11:52:51] zombor (zombor!~zombor_@kohana/developer/zombor) has quit (Remote host closed the connection)
[11:54:32] cecil (cecil! has joined #mythtv
[11:54:32] cesman (cesman!~cesman@pdpc/supporter/professional/cesman) has quit (Read error: Connection reset by peer)
[11:55:40] joki (joki! has joined #mythtv
[11:58:43] davide (davide! has quit (Remote host closed the connection)
[12:00:39] davide (davide! has joined #mythtv
[12:15:19] joki (joki! has quit (Ping timeout: 252 seconds)
[12:17:12] joki (joki! has joined #mythtv
[12:20:01] cecil_ (cecil_! has joined #mythtv
[12:20:28] cecil (cecil! has quit (Ping timeout: 244 seconds)
[12:21:23] xavierh: stuartm: good news, MythMusic settings are already using MythUI. I still think we should use as much as possible standardsetting as much as possible to lower the number of screen to theme + add some consitency. Lower priority, but what are you thought on this ? Also I converted mytharchive and mythgalerry settings to MythUI
[12:21:38] joki (joki! has quit (Ping timeout: 240 seconds)
[12:34:12] joki (joki! has joined #mythtv
[12:48:38] pheld: why is myth moving towards DB-storage for logs? and if one does wouldn't it at least be preferable to use partitoned (by date) tables?
[12:49:52] wagnerrp: the logs arent being moved to the database, but rather multiplexed to the database
[12:50:23] wagnerrp: as for partitioning, im not sure how large the table would have to grow before that would start to be a real concern
[12:53:41] joki (joki! has quit (Ping timeout: 248 seconds)
[12:55:24] knightr: stuartm, I have to leave for work now so I can't fix it before leaving but I can't touch anything that's not a translation, is not in themestringtool or is not between quotes (and I was given OK to change that only a week ago) without review... My guess is people don't actually realize I am a programmer since I do translation stuff (and I have been programming professionnaly for more than 18 years now) but these are the limits that were (and st
[12:55:24] knightr: ill are, the only thing that changed a week ago is that I was given the OK to change strings) to me....
[12:56:36] stuartm: knightr: I'm giving you permission
[13:00:20] stuartm: knightr: we impose these restrictions to prevent access creep where people who were given access to do one thing end up working on ever more things as time goes on, sometimes without the necessary ability to do so without breaking things
[13:00:27] knightr: stuartm, I really need to go now or else I'll arrive late for work but your OK will only be good for this time.. The last time (a week ago) to have the OK to do it without asking each time it needed the OK of a few devs (just like the votes that sometimes go on -developers....)
[13:00:55] stuartm: knightr: we'll discuss amending the restrictions
[13:01:38] knightr: stuartm, Thank you and Goodbye, ttyl...
[13:02:34] stuartm: but there will still be a line somewhere since we assign code access based on a history of submitted patches, I've no doubt you're more than capable but we don't have examples of prior work to reference :)
[13:03:41] stuartm: wagnerrp: do we cull old entries from the DB after a preset duration? e.g. a week, or two?
[13:05:56] davide (davide! has quit (Remote host closed the connection)
[13:06:18] davide (davide! has joined #mythtv
[13:06:46] stuartm: seems like it based on my logging table, it only goes back two weeks
[13:09:10] Jordack (Jordack! has joined #mythtv
[13:09:41] joki (joki! has joined #mythtv
[13:11:38] pheld: with logging you have to prepare for abnormal situations. what do you do in a sitiuation where a log-storm causes the daily delete to take more than 24hrs?
[13:14:08] wagnerrp: looks like logs are chopped at 2 weeks and 140k lines
[13:14:17] pheld: not to mention DB-locking and I/O performance-issues related to parallell deletion and insertion into the same tables.
[13:14:22] wagnerrp: applications other than mythbackend and mythfrontend are cleaned at 1 week
[13:14:57] wagnerrp: the logger is asynchronous, so locking up that table for deletes will not block the rest of the application
[13:15:32] pheld: my years from TNM-development tells me DB-logging is at best suspect. It
[13:15:59] pheld: 's ok for aggregated data and stats, but generally bad for raw data
[13:16:10] wagnerrp: sphery, Beirdo ^^^^
[13:19:13] wagnerrp: stuartm: re fd1800a11 discussion, would a descriptive error in the logs be sufficient?
[13:20:03] pheld: anyway, it didn't take me long to discover the nodblog-option so I'm happy ;)
[13:20:26] wagnerrp: although technically at that point, it would be cerr on the terminal, rather than the logs
[13:21:09] wagnerrp: i dont know if stderr is something that would likely be captured by init scripts
[13:26:22] stuartm: it's not, generally if something fails to start via an init script that info goes to /dev/null and you're left scratching your head
[13:27:49] stuartm: if you're experienced enough to check running the app manually and happy to be editing the init script you'll be fine, although it's an inconvenience, but there will be users who won't catch on that quickly
[13:28:01] wagnerrp: well im adding a 'SetRemoved()' to the command line parser, so we can at least provide that
[13:28:32] wagnerrp: IMHO, if youre experienced enough to be compiling from source and writing your own init scripts, you can check by running the app manually and edit the init
[13:28:48] wagnerrp: if youre using some distro's packages, that distro should provide a properly functioning init
[13:52:33] j-rod|afk is now known as j-rod
[13:58:46] wagnerrp: stuartm: whats the restriction on ABI changes during freeze?
[14:04:36] jams: stuartm, sure i can start a bisect. Shouldn't be hard todo that in the background during the day. Never used git bisect before seems like a good time to learn
[14:07:59] stuartm: wagnerrp: if they can be justified then no restriction
[14:11:50] danielk22: Beirdo: Please revert the --logfile removal commit. We don't want to have any feature removal commits right now.
[14:16:04] wagnerrp: danielk22: well that would largely remove the need for what i was about to commit
[14:17:25] wagnerrp:
[14:18:12] wagnerrp: although that could be used for the various other options we may have removed in this cycle
[14:21:37] zombor (zombor!~zombor_@kohana/developer/zombor) has joined #mythtv
[14:22:13] joki- (joki-! has joined #mythtv
[14:22:40] joki (joki! has quit (Ping timeout: 260 seconds)
[14:22:40] joki- is now known as joki
[14:26:56] joki (joki! has quit (Read error: Operation timed out)
[14:31:20] Jordack (Jordack! has quit ()
[14:31:58] danielk22: wagnerrp: I think it still might make sense to apply Beirdo's commit post freeze. But then I don't think we should silently allow --logfile, it's better to exit early and get people to adjust their init scripts.
[14:33:59] wagnerrp: well its going to exit early already
[14:34:12] wagnerrp: except right now, it would just tell people '--logfile is an unknown option'
[14:34:16] joki (joki! has joined #mythtv
[14:34:28] wagnerrp: this patch would make it say '--logfile has been removed, use --logpath instead'
[14:38:00] danielk22: wagnerrp: ah, ok. That's definitely an improvement.
[14:40:03] danielk22: Beirdo: ^^^ Maybe open a ticket with both patches so it can be applied after feature freeze in one go?
[14:41:25] damaltor (damaltor! has quit (Remote host closed the connection)
[14:41:57] wagnerrp: im actually thinking of going forward with my half of it, and using it for other command line options that have been removed over this cycle
[14:42:30] wagnerrp: even if --logfile remains for the initial release
[14:43:58] danielk22: wagnerrp: I think that's a great idea. We should probably drop those after a couple release cycles so can you put a comment for each as to when it was dropped (i.e. // option dropped in 0.25)
[14:44:14] wagnerrp: can do
[14:45:08] wagnerrp: (that patch actually lists 0.25 as the removal point for --logfile)
[14:45:58] danielk22: wagnerrp: We had talked about it, but think forgot to do it before the freeze.
[14:53:23] j-rod is now known as j-rod|afk
[14:53:49] skd5aner: danielk22: sorry to ping you again, but did you see my comment in the backlog about #9868 and #9845?
[14:54:18] danielk22: skd5aner: Yeah, I added them to my queue. I haven't looked at them yet.
[14:55:21] seeker (seeker!~seeker@unaffiliated/seeker) has joined #mythtv
[14:55:49] skd5aner: not sure if #9868 is the same (or related) to your recently open and closed #10416... thanks for adding to your queue, just didn't want them to get lost and I noticed you were looking at similiar tickets in the area so thought it might be fresh on the top of your mind prior to .25 – thanks :)
[14:55:53] skd5aner: danielk22: ^
[14:56:09] jams: danielk22- so during the bisect i ran into the problem where myth compiled without ivtv support. I'm assuming this has to do with v4l1 removed from the kernel?
[14:56:48] danielk22: jams: Yeah, I assume so.
[14:57:02] jams: was there an easy fix for that?
[14:57:36] danielk22: jams: Yes, just put the v4l1 headers in your /usr/include
[14:58:32] wagnerrp: arent we including the v4l1 headers in our own source?
[14:58:32] danielk22: MythTV will use the v4l2 calls first anyway.
[14:59:31] danielk22: wagnerrp: We were then we dropped them because v4l was reliably installed on systems, and then shortly thereafter v4l1 disappeared.
[15:01:49] xavierh: Is Terra still the default theme ?
[15:09:12] danielk22: xavierh: I believe so.
[15:09:53] damaltor (damaltor! has joined #mythtv
[15:10:34] taylorr: --logfile would indicate a file where --logpath would indicate a path, they are not the same
[15:11:06] taylorr: so are not allowing the user to specify a filename now?
[15:11:44] damaltor (damaltor! has quit (Remote host closed the connection)
[15:11:48] wagnerrp: taylorr: thats the general idea, as specifying a particular file has some bad side effects
[15:12:00] wagnerrp: namely only that particular application gets logged
[15:12:16] wagnerrp: anything it starts, such as a commflag, transcode, or filldatabase task does not
[15:12:47] damaltor (damaltor! has joined #mythtv
[15:12:52] wagnerrp: if you specify a path, those applications generate their own filename based off application name, starttime, and PID
[15:13:32] taylorr: ah, sounds good
[15:16:26] skd5aner: wagnerrp: whatever happens in this space ultimately before the 0.25 release, let me know if it's import enough to specifically call out in the "special notes/instructions" area of the release notes
[15:16:44] skd5aner: (re: --logfile changes)
[15:33:25] natanojl (natanojl! has quit (Ping timeout: 252 seconds)
[15:34:31] natanojl (natanojl! has joined #mythtv
[15:38:49] natanojl (natanojl! has quit (Ping timeout: 252 seconds)
[15:39:11] natanojl (natanojl! has joined #mythtv
[15:40:40] gregL (gregL! has quit (Quit: Leaving)
[15:43:21] xavierh: danielk22: the positionning of the widgets are particulary messed up for plugins
[15:43:35] natanojl (natanojl! has quit (Ping timeout: 252 seconds)
[15:43:56] natanojl (natanojl! has joined #mythtv
[15:48:21] natanojl (natanojl! has quit (Ping timeout: 252 seconds)
[15:49:09] natanojl (natanojl! has joined #mythtv
[15:49:13] danielk22: xavierh: We talked about switching to another default. We could still do it but we'd probably need to do another beta then.
[15:51:44] xavierh: danielk22: It is mainly the buttons such as Ok, Cancel and Finish which are on top of the bottom bar (with the date/time), I suspect they have not been implemeted in terra and using default-wide instead. Is is possible to partially reimplement a dialog to just move some element, or will I need to to a complete coly of the default ?
[15:53:38] xavierh: danielk22: I don't know the state of other theme, probably best to stick with it at the moment. but it does need some polishing before 0.25. It does give a very bad impression for a first time user
[15:54:08] xavierh: Will have a look tonight and send some patches
[15:54:27] natanojl (natanojl! has quit (Quit: Bye)
[15:55:03] jeff999 (jeff999! has joined #mythtv
[15:55:47] jpabq_ (jpabq_!~jpabq@mythtv/developer/jpabq) has joined #mythtv
[15:56:49] xavierh: Also I installed mythmusic for the first time on this computer, It does not seems to have any default value set (ex: audio device is blank when it should be default according to the description)
[16:02:29] stichnot (stichnot!~chatzilla@mythtv/developer/stichnot) has quit (Ping timeout: 248 seconds)
[16:02:58] stuartm: xavierh: I'm going to try and fix some of those screens in Terra before the release, but my heart isn't really in it so I've not made a start on it yet
[16:10:42] xavierh: stuartm: I know you don't like your baby anymore :P If you like I can do some of it and send patches. nothing fancy just some alignement and style consistancy accross setting pages
[16:11:51] rsiebert (rsiebert! has joined #mythtv
[16:13:46] stuartm: xavierh: that would be welcome, it will save me a lot of time
[16:15:43] rsiebert_ (rsiebert_! has quit (Ping timeout: 276 seconds)
[16:16:11] skd5aner: stuartm: release notes up-to-date
[16:17:52] stuartm: skd5aner: great, thanks
[16:17:55] xavierh: stuartm: If I wanted to add some logging on the standard output to tell me what file is loaded when I enter a screen, where about would that be ?
[16:18:31] stuartm: xavierh: -v gui
[16:18:35] jams: should have picked a faster computer todo the bisect on
[16:20:16] stuartm: xavierh: with "-v gui" you'll get log entries like "XMLParseBase: Loading window watchrecordings from /usr/local/share/mythtv/themes/carbon/recordings-ui.xml"
[16:20:23] xavierh: stuartm: ah cool, thx
[16:24:51] gregL (gregL! has joined #mythtv
[16:29:34] j-rod|afk is now known as j-rod
[16:30:41] Jordack (Jordack! has joined #mythtv
[17:17:06] wagnerrp: anyone see any reason to replace the old 'mythfrontend --get-setting' option?
[17:17:30] wagnerrp: frankly, i dont even remember consciously removing it when i redid the command line parser
[17:17:38] wagnerrp: and no one has come forward to complain about its loss
[17:19:31] noahric (noahric!~noahric@ has joined #mythtv
[17:24:31] xavierh: Is there a way to inherit from an other window (theme),there is few setting screen in Terra with a title label, cancel and save button which I like to fix once for all.
[17:24:35] xavierh: ?
[17:31:46] xavierh: I tried from , no success
[17:32:14] danielk22: wagnerrp: I wasn't aware of it. If re-introduced I think it makes more sense in mythutil.
[17:32:48] wagnerrp: yeah, i wasnt planning on putting it back into mythfrontend
[17:33:14] wagnerrp: more, should i implement it in mythutil, or just set the removed comment to say to use the services api
[17:33:45] stichnot (stichnot!~chatzilla@ has joined #mythtv
[17:33:45] stichnot (stichnot!~chatzilla@ has quit (Changing host)
[17:33:45] stichnot (stichnot!~chatzilla@mythtv/developer/stichnot) has joined #mythtv
[17:41:59] sphery: wagnerrp: I'm for (and think the reason it was removed was because of) the services api reference
[17:43:22] seeker_ (seeker_!~seeker@unaffiliated/seeker) has joined #mythtv
[17:44:09] sphery: danielk22: fwiw, I think the --logfile should stay removed for 0.25, since it's really more of a bug fix than a feature removal--it's fixing MythTV to properly log output from all applications rather than throwing away all log output from child processes (that users will need to debug)
[17:44:39] seeker_ (seeker_!~seeker@unaffiliated/seeker) has quit (Client Quit)
[17:45:34] seeker (seeker!~seeker@unaffiliated/seeker) has quit (Ping timeout: 245 seconds)
[17:47:35] danielk22: sphery: Very rarely have I needed to see the output of child processes to debug, maybe 1 case per 1000.
[17:48:21] stichnot: mythcommflag seems a good example, no?
[17:49:23] sphery: yeah, that's what I mean--when their commercial detection or previews or metadata or whatever isn't working, they won't have any information to tell them why because they have no logging for it
[17:49:49] stichnot: but to play devil's advocate, why wouldn't you just derive a --logpath from --logfile?
[17:50:39] sphery: that was my original plan, but Beirdo made the valid argument that we weren't given permission to write other files in that directory--only the file specified
[17:51:00] sphery: i.e. user may be upset if we create 100 log files in a directory they hadn't set aside for mythtv logs
[17:54:04] stichnot: so the other question would be, why not use the parent --logfile for child processes?
[17:54:35] wagnerrp: so were not back with the confusing logs weve had all the way up through 0.24
[17:55:03] wagnerrp: where you have multiple applications, and multiple instances of the same application, all running synchronously and piling into the same log file
[17:55:05] danielk22: stichnot: that's what we used to do; but I think that is technically difficult using the current mechanism and it can make the logs difficult to read.
[17:55:10] Beirdo: if you want that behavior, use syslog, and put them all inot one file
[17:55:29] stichnot: ok, that all makes sense. Done playing devil's advocate. :)
[17:55:41] wagnerrp: plus you still have the issue that the -HUP signals are not propagated through the child processes
[17:55:51] sphery: yeah, we've had plenty of, "Master backend locks up with invalid protocol version" reports and such
[17:55:55] wagnerrp: so if you rotate the logs, at most one application would continue logging after that event
[17:56:59] Beirdo: The whole point of removing the borked --logfile option was to make it necessary to update the init scripts. That will be highlighted in the changelogs. Mythbuntu users already have a changed setup for that using syslog.
[17:57:26] sphery: and one user on the list already admitted that the change inspired him to fix his broken config :)
[17:57:28] Beirdo: it's up to the packagers to stay current with the package, not for us to stay stagnant for the sake of old init scripts
[17:57:41] sphery: well said
[17:58:00] danielk22: logdir is the way to go, I just think at this point we should be telling the distro's to update their scripts, then April 3rd we kill logfile.
[17:58:42] Beirdo: then we will have prolonged pain. I don't see the difference.
[17:58:51] Beirdo: anyways, I'll mull it over on the bus to work
[17:58:54] sphery: danielk22: do we actually need the v4l1 headers in master? We restored the headers for 0.24-fixes in , but I thought you said they weren't needed for master?
[18:00:01] danielk22: sphery: I don't believe we need them in master, jams is looking for a regression that may have occurred before we patched master not to need them.
[18:00:29] wagnerrp: Beirdo: more often than not, people who run release versions dont submit logs
[18:00:35] danielk22: sphery: We're trying to figure out when PVR-150 channel changing broke and whether it was us or something upstream.
[18:00:49] wagnerrp: and if they do, we tell them to upgrade, try again, and provide us new logs from the new version if it still fails
[18:00:53] sphery: ah, ok... that's good--if we do need them and don't provide them in 0.25, we'd have a problem. thanks for the explanation
[18:01:34] sphery: wagnerrp: but more often than not, people expect that the start script they used for 0.25 will work for 0.25-fixes :)
[18:02:01] wagnerrp: true... if they upgrade, their distro should upgrade their init script along with it
[18:02:03] sphery: I think it's better to do it now than after release
[18:05:02] jams: danielk22- this error and similiar ones in upnp keep popping up Any suggestions? For the ones in upnp I commented, but since this is in tv_play figuered it wasn't safe to do so.
[18:07:02] davide (davide! has quit (Remote host closed the connection)
[18:07:25] davide (davide! has joined #mythtv
[18:08:11] danielk22: jams: See if it runs, mythdeque wasn't originally designed to hold QStrings.
[18:10:07] jams: so go ahead and comment and see what happens?
[18:11:48] stichnot: danielk22: a week or two ago, you asked me what CEA-708 fonts broadcasters are using, and I said font tags 0 (default), 1 (mono serif), and 3 (mono sans). For the record, it's actually only 0 and 3. The fact that 1 was showing up in the logs was a 708 decoder error. Most broadcasters use 0, and Fox and some PBS stations use 3.
[18:11:59] jams: btw the channel changing problem is before the 4vl internal headers were added back.
[18:16:39] kormoc (kormoc!~kormoc@mythtv/developer/kormoc) has joined #mythtv
[18:20:23] danielk22: Beirdo: After thinking about it a bit I'm withdrawing my revert request. I think there should have been a heads up first, but it is still the right thing to do.
[18:22:14] danielk22: jams: That's good news. If it happened before the header changes and before the rec2 merge we should be able to isolate this fairly well as there weren't that many changes made aside from those two since 0.23.
[18:23:46] jams: well thats good. The version where you added the headers back compiled without isses and failed changing channels right away. Anything before that is having trouble compiling for one reason or another on my test machine
[18:25:02] jams: wonder if it's my QT version thatsmaking this difficult. it's qt 4.7.3
[18:28:02] danielk22: jams: I'm confused. In master we didn't add the headers back into the source.
[18:28:28] danielk22: jams: Are you saying it works with early 0.24 and breaks when we added the headers?
[18:29:21] jams: somewhere between .24 and adding the headers it's broken
[18:30:25] danielk22: jams: ok. switch over to bisecting master.. It should work shortly after 0.24 was cut and then break sometime between then and now..
[18:30:54] jams: thats what i'm working on..except i'm running into compile errors
[18:32:26] danielk22: Can you post a pastebin?
[18:33:08] jams: yeah..let me start over to undo all my mucking around
[18:33:12] jams: give me a few minutes
[18:34:54] andreax (andreax! has joined #mythtv
[18:37:37] Beirdo: danielk22: Yeah, sorry for the lack of heads-up. I could have communicated the nature of that change a lot better. Sorry.
[18:40:40] danielk22: Beirdo: BTW I was just reading about the archive mysql DB storage engine. Could we use that to speed up logging to the DB. Supposedly it has "Incredible insert speeds"
[18:41:45] Beirdo: Hmm, worth investigating... post release :)
[18:42:16] Beirdo: of course /dev/null has incredible insert speeds too
[18:43:57] Beirdo: yay. and the day's meetings begin
[18:45:45] danielk22: Beirdo: It looks like ubuntu doesn't compile it in.. it also doesn't support delete which could be a problem...
[18:47:28] wagnerrp: just a heads up to everyone, github will require you to re-confirm your SSH keys before allowing you to do anything
[18:47:39] wagnerrp: as of about 20 minutes ago
[18:48:05] jams: danielk22- clean checkout and removed previously installed libmyth* before compiling.
[18:49:32] kormoc: danielk22, the csv engine might be a better choice then the archive engine
[18:50:26] danielk22: jams: what is the revision #
[18:50:43] kormoc: danielk22, it's fast to insert, you can delete, and is entirely readable from a terminal
[18:51:06] NightMonkey (NightMonkey!~NightMonk@pdpc/supporter/professional/nightmonkey) has joined #mythtv
[18:51:19] jams: danielk22- it's in the pastebin, but 6ff252dfe3da25c39378335c558ffb5022b85afe
[18:51:29] jams: the one right before jannu removed the headers
[18:53:15] jams: oops guess i put revision where jannu removed the headers in the pastebin
[19:05:37] danielk22: jams: Ok, I'm compiling. It looks like there are a few places we're doing a QString(0) instead of QString(). I'll make a patch that allows this to compile.
[19:06:48] jams: thanks, is this related to the QT version?
[19:12:11] danielk22: jams: more likely the g++ version. Even Qt 4.0 doesn't have a QString(int) constructor.
[19:19:09] danielk22: jams: On second thought looks like some were remnants from Qt3 which accepted NULL as a string initializer. Maybe this was secretly allowed in earlier Qt4 revisions.
[19:20:05] danielk22: jams:
[19:20:21] jams: i don't mind switching compiling an older qt. but wasn't really going to change out gcc
[19:23:02] jams: applied and compiling
[19:26:37] Beirdo: wagnerrp: how long do we plan to keep the "Removed" arguments around?
[19:26:48] Beirdo: like a release cycle or something?
[19:27:21] wagnerrp: any reason not to leave them in indefinitely?
[19:27:29] Beirdo: accumulated cruft
[19:27:37] wagnerrp: i suppose
[19:27:43] Beirdo: same reason we don't keep all of the schema updated indefinitely
[19:27:50] Beirdo: updates rather
[19:27:52] wagnerrp: well... we do
[19:27:59] wagnerrp: or at least we have up until now
[19:28:00] Beirdo: ummm, no we don't :)
[19:28:09] Beirdo: we did, yeah
[19:28:12] wagnerrp: and thats only because those old schema updates are broken
[19:28:37] Beirdo: anyways, I'd suggest a lifetime of 2 or so release cycles would do well
[19:28:51] Beirdo: after that, they don't really need to still be around
[19:29:00] wagnerrp: ive got no real opinion one way or another
[19:29:10] Beirdo: it's a good idea though :)
[19:29:54] danielk22: wagnerrp: Even in 0.23 we only upgrade from '1027' onward.. We've just been lazy about trimming since Bruce left the project.
[19:30:50] wagnerrp: heh... "only 1027 onward"
[19:31:14] Beirdo: that's around 0.14
[19:31:16] wagnerrp: if you have to update more than 250 versions, youre on your own?
[19:32:07] Beirdo: nah, just find an older version that lets you upgrade
[19:32:13] Beirdo: upgrade in steps
[19:32:18] wagnerrp: well if we get smolt enabled on a majority of users' systems
[19:32:33] wagnerrp: well get some idea of what people are running, and a timeframe to moonlight this kind of stuff
[19:32:38] Beirdo: or... blow it all away and start anew, after all, it's 0.14 -> 0.25!
[19:32:58] Beirdo: no, we'll see what people using 0.25 are using :)
[19:33:12] Beirdo: the people using ancient crap won't be in smolt
[19:33:51] wagnerrp: right, but the people using ancient crap wont have the new command line parser, and removed argument definitions either
[19:34:10] Beirdo: hhehe
[19:34:12] Beirdo: so true
[19:34:36] Beirdo: the smolt stuff, is that on backends and frontends, or frontends only?
[19:35:14] wagnerrp: yes
[19:35:17] xavierh_ (xavierh_! has joined #mythtv
[19:35:32] Beirdo: ummm
[19:35:36] wagnerrp: :)
[19:35:41] Beirdo: syntax error.
[19:36:02] wagnerrp: there is a mechanism to call it through the services api
[19:36:11] wagnerrp: but the auto-update is only performed by the frontend
[19:36:18] wagnerrp: and only when the frontend starts up
[19:36:18] Beirdo: Hmm
[19:36:33] wagnerrp: and only once per month at the most frequent
[19:36:33] Beirdo: it would be good to get backend setups too for the capture cards
[19:37:00] wagnerrp: i believe each frontend should be sending duplicates of all that stuff
[19:37:11] Beirdo: hmm, OK
[19:37:14] wagnerrp: since it pulls it through the database, and not the local sysfs
[19:37:22] Beirdo: not the ideal way, but cool.
[19:37:47] wagnerrp: at such time as we add in the methods for scanning for hardware into the services api
[19:37:52] wagnerrp: we can start shifting things over
[19:37:53] Beirdo: yeah
[19:37:57] jams: Beirdo- from my point of view the integration is not complete. it was deemed good enough as a POC and thats where it stayed.
[19:38:16] Beirdo: yup, agreed. It's a good start :)
[19:38:39] wagnerrp: right, its all still nearly untested
[19:39:05] wagnerrp: waiting for data to actually start coming in before we can decide how to display it and tweak the data sets
[19:39:22] ** wagnerrp eyes an idle alcor **
[19:39:26] Beirdo: makes sense
[19:40:21] Beirdo: haha
[19:40:23] jams: yeah i have to redo the alcor setup it got resynced out of existence. So now i'm waiting for everything else to be complete
[19:40:34] Beirdo: tried to view my hardware profile...
[19:40:41] Beirdo: 500 internal error
[19:41:10] wagnerrp: been a while since you updated?
[19:41:18] Beirdo: ooooh, it wanted the pub_ one
[19:41:22] wagnerrp: site seems to be working find here
[19:41:26] wagnerrp: fine
[19:41:36] Beirdo: works fine with the correct UUID :)
[19:42:05] Beirdo: EHCI Host Controller -> unknown
[19:42:08] Beirdo: heh OK.
[19:42:30] Beirdo: yeah, we'll polish it more over time, I'm sure
[19:42:31] wagnerrp: yeah, the hardware id databases could stand to be updated (if there are updates available)
[19:42:33] jams: sometimes that 500 error is due to being idle to long and a db conneciton needs to be re-established
[19:43:36] Beirdo: but yeah, that's definitely only my frontend hardware
[19:43:37] wagnerrp: at the moment, all that extra mythtv data is just being discarded by the server, correct?
[19:43:39] Beirdo: we'll get there :)
[19:43:55] jams: wagnerrp- yeah
[19:46:50] jams: danielk22- patch applied/compiled and I can change channels
[19:47:19] jeff999 (jeff999! has quit (Remote host closed the connection)
[19:47:23] jams: that means it's somewhere between jannu removing the v4l headers and your 4vl1 fix about a year later :)
[19:47:54] wagnerrp: jams: i thought different frontends were supposed to have different profiles
[19:48:13] jams: wagnerrp- they do..each machine in a mythsystem/cluster has it's own uuid
[19:48:23] sphery: jams: Qt 4.7 only works with changesets applied at the end of (not the patches in there--that were basically what you're doing, just making things compile, even though it breaks some stuff)
[19:48:28] jams: there is one shared uuid that is unique to each mythsystem
[19:49:05] sphery: (that fixing the QString(NULL) brokenness)
[19:49:31] sphery: if you can compile with Qt 4.6, it works without changes
[19:49:35] wagnerrp: jams: two of my frontends managed to randomly generate the same exact UUID
[19:50:16] jams: sphery- good to know, looks like for my purpose the patch daniel provided works well enough
[19:50:35] stuartm: would an upgrade from 0.14 or thereabouts even work? Yes the schema should get upgraded but there's probably quite a lot of newish metadata that we've been collecting over the years and even newer code that depends on the existance of certain information that won't be in the database for someone upgrading from such an old version
[19:50:49] wagnerrp:
[19:50:58] jams: wagnerrp- that is highly unlikely it happened by chance as the client even calls back to the smolt server to make sure the uuid is not in use by another machine.
[19:51:29] wagnerrp: i just freshly ran smolt for the first time on 'myth2'
[19:51:47] wagnerrp: i found it odd that one of my frontends was reporting it had a PVR-150
[19:52:11] jams: wonder if the code that iam lin daro added to regenrate a uuid caused it
[19:54:29] sphery: stuartm: fwiw, there are many db updates that don't behave the same, now, with current Qt and MySQL as they did when the code was written
[19:54:40] stuartm: Beirdo: fwiw, my proposal is that we keep schema upgrades for two or three major releases, so you can upgrade from 0.22 to 0.25, but not 0.21 to 0.25 – if old 'removed' arguments are kept for a similar length of time ...
[19:55:04] Beirdo: sounds reasonable
[19:55:30] sphery: due to things changed underneath MythTV--like Qt changing to always use prepared statements, regardless of what the Qt code asks for, in Qt 4.5
[19:56:31] Beirdo: is there a way to run a smolt scan from the command line
[19:56:42] stuartm: the question of 2 vs 3 really comes down to how often we release, at the current pace 2 seems more than reasonable, that allows someone to upgrade just one every 2+ years and still be ok
[19:56:59] wagnerrp: /usr/share/mythtv/hardwareprofile/ i believe
[19:57:15] Beirdo: yeah 1–2 years back sounds reasonable overall
[19:57:21] Beirdo: ah, cool, thanks
[19:57:27] sphery: stuartm: and you're definitely right that someone upgrading from 0.14 would have to go back through setup--and likely delete all capture cards and re-create due to changes to the underlying kernel/drivers and such, not to mention going through frontend setup
[19:58:54] sphery: (and, yeah, I'll push that change soon--probably tomorrow at the latest--I was just caught up figuring out some other db update bugs and wanted to get them all figured out before pushing)
[20:00:02] stuartm: yeah, I'm willing to bet that it would break in lots of small and obscure ways too, ways in which we can't even imagine because we don't even have recordings made with 0.14 any more
[20:01:19] jams: stuartm, you bet. I have recording from .18 that don't play correctly anymore
[20:02:28] sphery: stuartm: yeah, that too--forgot about recordings themselves
[20:03:08] noahric (noahric!~noahric@ has quit (Quit: noahric)
[20:05:38] jarle (jarle! has quit (Remote host closed the connection)
[20:08:54] jarle (jarle! has joined #mythtv
[20:10:01] stuartm: all it takes is for 0.15 say to have started storing the recording length in the database when the recording completes, then in 0.22 someone decides to check whether a recording is playback based on the stored length, it wouldn't affect anyone running 0.22 because they haven't kept recordings from years ago but someone running 0.14 today and upgrading would find the frontend won't let them play anything – that's not a real example but I'd be
[20:10:02] stuartm: very surprised if there aren't real instances where similar problems would arise
[20:10:44] stuartm: bah – s/is playback/is playable/
[20:11:22] ** Beirdo laughs **
[20:11:36] Beirdo: I ran the smolt stuff on my backend box.
[20:11:53] Beirdo: HDPVR Killer Device v1.0 shows up
[20:11:57] Beirdo: *snork*
[20:30:32] davide_ (davide_! has joined #mythtv
[20:30:33] davide_ (davide_! has quit (Changing host)
[20:30:33] davide_ (davide_!~david@mythtv/developer/gigem) has joined #mythtv
[20:31:10] davide (davide! has quit (Ping timeout: 276 seconds)
[20:31:35] stuartm: yeah, we discussed the potential for some embarrassing USB devices to show up in smolt reports, I don't think anyone considered more sinister items ;)
[20:34:11] andreax (andreax! has quit (Read error: Connection reset by peer)
[20:44:25] tester68 (tester68! has joined #mythtv
[20:50:40] gregL_ (gregL_! has joined #mythtv
[20:54:56] Jordack (Jordack! has quit (Quit: What does a canibal do after dumping his girlfriend? Wipes his but)
[21:01:04] noahric (noahric!~noahric@ has joined #mythtv
[21:01:06] stichnot_ (stichnot_!~chatzilla@mythtv/developer/stichnot) has joined #mythtv
[21:03:10] foobum (foobum! has quit (Ping timeout: 272 seconds)
[21:03:14] stichnot (stichnot!~chatzilla@mythtv/developer/stichnot) has quit (Ping timeout: 260 seconds)
[21:03:17] stichnot_ is now known as stichnot
[21:05:31] foobum (foobum! has joined #mythtv
[21:06:08] joki (joki! has quit (Ping timeout: 240 seconds)
[21:06:18] joki- (joki-! has joined #mythtv
[21:06:29] joki- is now known as joki
[21:11:43] foobum (foobum! has quit (Ping timeout: 245 seconds)
[21:14:01] foobum (foobum! has joined #mythtv
[21:28:49] skrock (skrock! has joined #mythtv
[21:31:01] noahric (noahric!~noahric@ has quit (Quit: noahric)
[21:41:12] gregL_ (gregL_! has quit (Quit: Leaving)
[21:41:18] zombor (zombor!~zombor_@kohana/developer/zombor) has quit (Ping timeout: 260 seconds)
[21:44:28] skd5aner: I have a few recordings from 2005/2006 still arond – what version would that have been .17, .18?
[21:48:37] skd5aner: jams: How come when I look at this profile (which is probably Beirdo's), the link for the HVR-2250 shows up as N/A – . . . 4dc9f321bc0d
[21:49:29] MaverickTech (MaverickTech! has joined #mythtv
[21:51:07] MavT (MavT! has quit (Ping timeout: 276 seconds)
[21:53:11] noahric (noahric!~noahric@ has joined #mythtv
[21:55:48] Beirdo: yup, that's mine
[22:06:04] skrock (skrock! has quit (Quit: WeeChat 0.3.7)
[22:30:26] j-rod is now known as j-rod|afk
[22:37:16] xavierh_ (xavierh_! has quit (Ping timeout: 276 seconds)
[22:41:37] stuartm: fwiw, it just goes to prove that these stats aren't really anonymous, you can determine who they belong to if you know enough about their setup
[22:42:42] wagnerrp: well that raises the question, should recent updates be listed?
[22:42:59] wagnerrp: youre not likely going to be guessing a UUID
[22:43:43] wagnerrp: so unless you have access to the database, removing that 'recent' page would make it anonymous
[22:44:14] danielk22: You can't really make statistics like this anonymous. But as long as we don't claim they are anonymous I don't see the harm.
[22:44:15] stuartm: is there even a need to be able to access unique reports? I believe we're only interested in the aggregated totals
[22:44:57] wagnerrp: not really
[22:53:29] xavierh_ (xavierh_! has joined #mythtv
[22:53:53] cecil_ (cecil_! has quit (Quit: Konversation terminated!)
[22:55:26] superm1: the page that asks you to submit stats isn't even in the mythtv-setup binary though right? I only saw it when browsing the output of the backend on port 6544
[22:55:51] jams: i use them for debugging. instead of asking for what hardware I ask for the pub_uuid. Up to the person if they want to give it out or not
[22:57:00] jams: superm1- it's only in the frontend right now
[22:57:35] superm1: jams: oh, what was that I saw on the backend's http server output then?
[22:57:58] superm1: but the thing in the frontend, it's still not going to nag you – people would need to go look for the option and opt in at that point
[22:58:21] stuartm: superm1: it's accessible there too, that's where I first saw it, I don't remember ever seeing it in the frontend settings
[22:58:46] jams: unless it moved it was part of the startup wizard
[22:59:06] superm1: oh i haven't done a startup of a new frontend since jumping to 0.25 yet
[22:59:09] jams: i have not used it in sometime so can't say for certain
[22:59:22] jams: but your right it needs a better home
[22:59:35] stuartm: hmm, Setup Wizard is last in the Setup menu, shouldn't it be first?
[23:00:50] jams: would think..but well I'm not going to say a whole lot about that.
[23:03:23] stuartm: and shouldn't you be prompted to go through the setup wizard when you first setup a frontend? If we're going to make it a feature of frontend setup ...
[23:08:50] superm1: if you want the data to be useful it really should be somewhere that lots of people will see it and have the choice to participate
[23:09:44] stuartm: I think we won't see this done properly until 0.26
[23:12:24] stuartm: I've just hit an issue with the setup wizard that means I wouldn't push it on users for 0.25 – there's no warning that testing video playback initiates a download, and no way to abort that download so if you're on a really slow connection (or worse, dial-up) then you're really stuck
[23:13:31] jams: don't forget ppl that pay by bit
[23:14:11] stuartm: yup, no-doubt they'd just yank the cable from the wall ;) (or kill the frontend)
[23:14:38] kormoc (kormoc!~kormoc@mythtv/developer/kormoc) has quit (Ping timeout: 245 seconds)
[23:16:16] stuartm: a 76MB download to test HD is slow enough on my 10Mbit connection, I'd hate to be out in the country or worse somewhere where 1Mbit is still considered to be top tier broadband
[23:17:14] stuartm: we probably need a prompt there, "Testing HD video playback requires downloading a 70MB sample video, this may take some time, do you wish to continue?"
[23:17:41] stuartm: and allowing the download to be aborted shouldn't be difficult
[23:18:42] stuartm: we can make the Setup Wizard one of the focuses of 0.26, work out the kinks and ensure it operates smoothly
[23:20:06] stuartm: ooh, the HD sample _isn't_ Big Buck Bunny ... ok now I'm curious
[23:20:40] xavierh_: Talking about the wizard, pressing "Previous" does not work, Unless I mess up. Did anyone else notice ?
[23:21:11] xavierh_: The current screen diseapear, but the previous one is not rendered
[23:25:27] jams: xavierh_ it works here, but i'm also not on the latest code right now
[23:26:00] stuartm: xavierh_: yeah ... those screens were implemented in a kinda funky way, not really how I would have done it and not in a way that works with mythui
[23:26:05] wagnerrp: stuartm: have you read the internal list today?
[23:26:43] stuartm: wagnerrp: I thought I had, did I miss something?
[23:27:31] wagnerrp: i sent an email last night specifically asking about running the wizard the first time the frontend is run, so the user is prompted to enable smolt
[23:27:56] wagnerrp: with some discussion about the wizard being semi-broken
[23:28:06] stuartm: xavierh_: instead of a true back/forward with a state machine it just hides the first screen when you move to the second, only hiding screens was never implemented for mythui
[23:28:10] wagnerrp: and more discussion about delaying the smolt stuff for a couple days
[23:28:34] wagnerrp: well, less discussion, more one or two replies
[23:28:46] stuartm: wagnerrp: ah right, yeah I did read it but only with half an eye
[23:28:50] xavierh_: stuartm, jams: works for you james but not for you stuartm ?
[23:30:33] stuartm: xavierh_: I've not tested, I only remember a ticket being opened about mythui being 'broken' and an argument about it when said that mythui wasn't broken, the implementation of the setup wizard was broken
[23:32:27] xavierh_: stuartm: I was asking as I noticed it while I am fixing Terra, I was wondering if it was my fault (don't see how to be honest)
[23:33:22] jams: stuartm, danielk22 The major problem commit is the one surrounded by ####
[23:33:34] jams: the merge of mythtv-rec2
[23:34:46] jams: anything after that fails to work
[23:34:59] stuartm: fwiw, 'state machine' wasn't quite what I meant before, I simply meant that it doesn't keep a record of the state of each setting so that we could move backward without having to leave each screen in the stack
[23:36:25] stuartm: the latter was important to iamlindoro because he's based his theme of the setup wizard around transparent popup dialogs and didn't want the old dialog to remain in the background behind the new one
[23:37:31] jams: guess that means the rec2 branch will need to be bisected
[23:41:18] stuartm: jams: that's interesting, the timeline on the ticket is fuzzy since it doesn't report dates just vague '6 month ago' references but I thought last night I'd established the ticket was opened before the rec2 merge
[23:42:41] xavierh_: hmm, button and dialog are made of images :( not easy to fix this
[23:43:43] xavierh_: I wish we could use svg ;(
[23:43:56] jams: stuartm, I think there is more then one problem with live-tv. after the rec2 merge it flat out refuses to work. Before that channel changes work, but they can be iffy.
[23:44:14] stuartm: xavierh_: change the dialog to use one of the Terra dialog backgrounds, they are all bigger
[23:44:20] jams: i can probably keep going back to figure out when they were last truly stable
[23:45:45] jams: rec2 merge was may 11th 2011
[23:46:46] jams: ticket 9830 was opened 9 months ago
[23:47:36] jams: which would be june 2011
[23:47:46] stuartm: xavierh_: popups/largemenubackground.png maybe?
[23:48:27] wagnerrp: huh... segfault in mythfrontend
[23:48:54] xavierh_: stuartm: yep, I am looking at them. Out of curiosity, does Qt support svg ?
[23:48:55] stuartm: jams: hmm, either I got my dates mixed up or there were some further rec2 merges made after June
[23:48:59] stuartm: not that it matters now
[23:49:28] stuartm: xavierh_: yes, but not in the same way it supports other image types, it operates very differently
[23:49:33] jams: yeah..will see what danielk wants to do next.
[23:50:14] stuartm: xavierh_: you can't just load an svg like you would any other image
[23:50:43] noahric (noahric!~noahric@ has quit (Quit: noahric)
[23:51:15] xavierh_: stuartm: I wanted to do my own themes, but I am rubbish at image editing :)
[23:51:33] stuartm: xavierh_: I do everything in svg and then export to png
[23:52:04] stuartm: all my image work and mockups are done with Inkscape, the same applies to most other themers
[23:52:52] cesman (cesman! has joined #mythtv
[23:52:53] cesman (cesman! has quit (Changing host)
[23:52:53] cesman (cesman!~cesman@pdpc/supporter/professional/cesman) has joined #mythtv
[23:53:24] xavierh_: Why did I not think of this :)
[23:54:07] wagnerrp: anyone know what paulh was talking about in reference to brokenness in the setup wizard?
[23:54:22] stuartm: one of the reasons I no longer like Terra is that it was image-heavy, it makes new screens and maintenance very hard work
[23:54:40] wagnerrp: theres loss of changes when you go backwards and then forwards again
[23:54:45] wagnerrp: if thats what he was referring to
[23:55:01] stuartm: wagnerrp: what xavierh_ and I were talking about earlier, Back/Forwards not really working properly
[23:55:19] xavierh_: stuartm: I completly understand. I am not sure that any of the background would fit. Do you still have your svg files ?
[23:55:22] wagnerrp: just where you lose the settings stored for the forward screens when you go backwards?
[23:55:26] stuartm: add that to my issues with the video download component
[23:55:47] xavierh_: wagnerrp: when you do "Previous" the screen disapear
[23:56:17] wagnerrp: oh, its not compatible with most themes
[23:56:18] stuartm: wagnerrp: for some people pressing back you get stuck without any screens visible
[23:56:31] xavierh_: with Terra
[23:56:59] xavierh_: May fault really, I change my mind :D
[23:57:04] xavierh_: My
[23:57:22] stuartm: xavierh_: I do, it's a real mess though, I have my own way of working with Inkscape ;)
[23:57:47] wagnerrp: works fine for me on arclight, mythcenter, and terra
[23:58:15] xavierh_: stuartm: I won't judge you ;) I just need one of the dialog to start with
[23:59:06] xavierh_: wagnerrp: strange, will have to clean the theme and try again
[23:59:39] wagnerrp: stuartm: well anyway, if the wizard is out, what are your thoughts on breaking the smolt stuff into a separate dialog and allowing that to pop up independently
[23:59:56] stuartm: xavierh_: if you put the list in the 'top' section of large menu background, the text edit below the line and then the three buttons horizontally below that? I'm sure there's room to do that, it just might not look great
[23:59:57] wagnerrp: a certain amount of time after updating to or installing 0.25

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