| Tuesday, July 2nd, 2013, 00:17 UTC | ||
| [00:17:06] | ggervasio (ggervasio!~gervasio@c-98-234-64-25.hsd1.ca.comcast.net) has left #mythtv () | |
| [00:33:23] | wagnerrp: | stuarta: because no one ever set up logrotate |
| [01:02:05] | NightMonkey (NightMonkey!~NightrMon@pdpc/supporter/professional/nightmonkey) has quit (Quit: Body blow! Body blow!) | |
| [01:03:15] | J-e-f-f-A (J-e-f-f-A!~J-e-f-f-A@unaffiliated/j-e-f-f-a) has quit (Quit: Oh No!!!! ;-)) | |
| [01:22:53] | danielk221: | stichnot: Yeah, as you discovered nothing wrong with QRunnable, just a MThreadPool to run it. |
| [01:34:50] | amessina (amessina!~amessina@2001:470:c1dc:7779:d6be:d9ff:fe8d:7c1e) has joined #mythtv | |
| [01:52:01] | jya: | anyone with experience on running gdb on another machine and debugging remotely? |
| [01:52:24] | jya: | trying to debug the linux frontend from my mac, as I find that gdb graphic frontends on linux suck |
| [02:04:12] | jya: | wagnerrp: what's the user command with mythutil to send a message to a frontend? |
| [02:05:22] | jya: | ah found it.. --message |
| [02:06:24] | wagnerrp: | it's a bit odd to use |
| [02:06:39] | wagnerrp: | you have to give it --message_text and optionally --timeout |
| [02:06:52] | wagnerrp: | but you can only know that by running --print-template |
| [02:14:46] | jya: | hum..whenever I run mythutil I see the message: ASSERT failure in QCoreApplication::sendEvent: "Cannot send events to objects owned by a different thread. Current thread 109e100a0. Receiver '' (of type 'LoggerThread') was created in thread 109a31d90", file kernel/qcoreapplication.cpp, line 532 |
| [02:14:57] | jya: | (using Qt compiled in full debugging mode) |
| [02:16:01] | jya: | so is --message required if you give a --message_text? |
| [02:16:31] | wagnerrp: | i believe so. the syntax is really bad |
| [02:16:35] | jya: | I don't see --message_text anywhere on the list of options |
| [02:17:10] | wagnerrp: | right, because it's not a defined input |
| [02:17:33] | wagnerrp: | the messenger scans through the command line arguments on its own |
| [02:17:58] | jya: | i see.. going to have a look, and see what can be improved… just looking at the help, I have no idea on how to send a message :) |
| [02:18:13] | wagnerrp: | run 'mythutil --print-template' |
| [02:18:23] | wagnerrp: | it defines the two additional arguments you can use |
| [02:18:37] | jya: | yes, I see that.. but surely, I don't have to provide xml to a command line to send a message no? |
| [02:18:51] | wagnerrp: | no, just the string, and it applies the string to those locations |
| [02:19:50] | jya: | how can you even remember how that works :) |
| [02:20:25] | wagnerrp: | for example, you run 'mythutil --message --message_text=blah' |
| [02:20:51] | jya: | why not --message="blah" ? |
| [02:21:11] | wagnerrp: | and it sends '<mythmessage version="1"><text>blah</text><timeout>0</timeo ut></mythmessage>' to the backend |
| [02:21:30] | wagnerrp: | i believe it was intended to be extensible, and support more arguments |
| [02:22:45] | wagnerrp: | don't ask me, i didn't set up the behavior |
| [02:23:19] | jya: | because I compiled Qt in debugging, it's full of asserts… looks like it doesn't like how we've been using the logger thread at all… never so it because I always run with --nodblog |
| [02:24:04] | J-e-f-f-A (J-e-f-f-A!~J-e-f-f-A@unaffiliated/j-e-f-f-a) has joined #mythtv | |
| [02:25:08] | sl1ce (sl1ce!~johnathan@pool-100-0-73-123.bstnma.fios.verizon.net) has joined #mythtv | |
| [02:31:16] | amessina (amessina!~amessina@2001:470:c1dc:7779:d6be:d9ff:fe8d:7c1e) has quit (Quit: Konversation terminated!) | |
| [02:32:33] | amessina (amessina!~amessina@2001:470:c1dc:7779:d6be:d9ff:fe8d:7c1e) has joined #mythtv | |
| [02:36:31] | jya: | pretty neat this commandlinearg |
| [02:36:33] | _nyloc_ (_nyloc_!~quassel@p4FE4C49B.dip0.t-ipconnect.de) has joined #mythtv | |
| [02:41:07] | nyloc (nyloc!~quassel@p4FE4DC1B.dip0.t-ipconnect.de) has quit (Ping timeout: 276 seconds) | |
| [02:43:26] | danielk221: | jya: I've spent many hours debugging on the gdb command line. Here is my ~/.gdbinit http://pastebin.com/RCtVJMeB |
| [02:43:37] | jya: | thanks…. |
| [02:43:52] | jya: | I wrote some macro allowing to print the content of some Qt objects ; like qtstring |
| [02:44:18] | jya: | ah you have that already there |
| [02:44:22] | danielk221: | jya: It is mostly copied from the KDE gdb macros. It has print methods for a few of the basic Qt objects. |
| [02:44:42] | jya: | this is great. |
| [02:45:11] | danielk221: | qs4 Is my favourite :) |
| [02:45:28] | jya: | I have a great deal of trouble debbugging using gdb multi-threaded stuff… sometimes a step through makes me jump on another thread, and I have to start right from the start |
| [02:45:57] | jya: | I called mine qprint :) |
| [02:46:53] | jya: | how does the fs macro work? |
| [02:48:14] | danielk221: | I believe finish takes you out of the thing you've stepped into. But I rarely do line by line step debugging these days. |
| [02:49:19] | danielk221: | The idea is you've gone into some function, but decide it is not interesting after all so you want to resume your stepping at where the function was called. |
| [02:51:35] | danielk221: | "set history save" <-- this is awesome but does leave .gdb_history files everywhere you debug from. |
| [02:53:14] | jya: | ok |
| [02:54:14] | sl1ce (sl1ce!~johnathan@pool-100-0-73-123.bstnma.fios.verizon.net) has quit (Remote host closed the connection) | |
| [02:54:21] | peper03 (peper03!~peper03@mythtv/developer/peper03) has quit (Ping timeout: 256 seconds) | |
| [02:55:18] | sl1ce (sl1ce!~johnathan@pool-100-0-73-123.bstnma.fios.verizon.net) has joined #mythtv | |
| [02:55:40] | peper03 (peper03!~peper03@mythtv/developer/peper03) has joined #mythtv | |
| [02:56:22] | sl1ce (sl1ce!~johnathan@pool-100-0-73-123.bstnma.fios.verizon.net) has quit (Client Quit) | |
| [02:58:08] | sl1ce (sl1ce!~johnathan@pool-100-0-73-123.bstnma.fios.verizon.net) has joined #mythtv | |
| [02:58:24] | sl1ce (sl1ce!~johnathan@pool-100-0-73-123.bstnma.fios.verizon.net) has quit (Client Quit) | |
| [03:07:23] | stichnot (stichnot!~stichnot@adsl-68-125-54-123.dsl.pltn13.pacbell.net) has joined #mythtv | |
| [03:07:23] | stichnot (stichnot!~stichnot@mythtv/developer/stichnot) has joined #mythtv | |
| [03:07:23] | stichnot (stichnot!~stichnot@adsl-68-125-54-123.dsl.pltn13.pacbell.net) has quit (Changing host) | |
| [03:08:52] | stichnot: | jya: in gdb you can freeze all the other threads with "set scheduler-locking on" to avoid the annoyance of step jumping you into another thread. But be sure to "set scheduler-locking off" before resuming or restarting the app. |
| [03:11:14] | jya: | danielk221: the printq4map macro, how do you use it? when I type: printq4map cmdline.m_namedArgs ; I get 48 items in the map, Missing argument 1 in user function. |
| [03:12:05] | jya: | oh, I see that there's a little doc |
| [03:17:20] | jya: | wagnerrp: in a MythCommandLineParser; is it possible to see the top argument name? I'd like to share the code between --message and --notification; but I can't see a way to distinguish which option triggered the call to the code |
| [03:17:55] | wagnerrp: | yeah, i think so... need to look |
| [03:21:09] | wagnerrp: | no |
| [03:21:30] | wagnerrp: | running '--help "keyword"' shows some extended help for that argument, and lists the specific argument |
| [03:21:52] | wagnerrp: | but the parser loop passes that in directly |
| [03:21:55] | wagnerrp: | it doesn't seem to be stored |
| [03:22:21] | jya: | bugger... |
| [03:22:31] | jya: | oh well, I can extend that I guess |
| [03:22:33] | wagnerrp: | rather, it's stored as the value to the '--help' keyword |
| [03:24:31] | jya: | you wrote that class didn't you? |
| [03:24:37] | wagnerrp: | yeah |
| [03:24:41] | jya: | mind if I add a GetCaller() method? |
| [03:25:21] | wagnerrp: | sure |
| [03:26:33] | jya: | finally some documented code :) |
| [03:26:45] | wagnerrp: | although if they are going to have different behavior, shouldn't they also have separate arguments and help text? |
| [03:27:04] | jya: | what do you mean different behaviour? |
| [03:27:22] | wagnerrp: | well if you need to know which one was used, you're going to do something different based off that, right? |
| [03:27:32] | jya: | the aim here is simply to know which top-level commands was used to call the callback function |
| [03:28:13] | jya: | wagnerrp: I will only return a qstring, and from that point on I can do some code for ityeah.. basically instead of <message/> it's going to be <notification/> |
| [03:28:18] | jya: | that's the only difference |
| [03:28:23] | wagnerrp: | ah |
| [03:28:55] | jya: | when the notification is more matured, and can display a confirmation button, I guess we could move the message to notification.. but in the mean time |
| [03:34:24] | jya: | hmn… actually, it's not MythCommandLinerParser needing moditication.. |
| [03:36:13] | wagnerrp: | yeah, you can just make the change in CommandLineArg::Set |
| [03:36:59] | wagnerrp: | ugh... weather alert on my phone |
| [03:37:45] | wagnerrp: | only slightly less annoying than waking up at 2:30 in the morning to an amber alert (kidnapped child) on my phone |
| [03:37:59] | wagnerrp: | first time i had heard it, i was out of town in a hotel... i thought it was the fire alarm at first |
| [03:38:26] | wagnerrp: | wait a second, i do store it in m_usedKeyword |
| [03:39:36] | wagnerrp: | jya: looks like all you need to do is add a getter for CommandLineArg::m_usedKeyword |
| [03:40:17] | fetzerch (fetzerch!~quassel@unaffiliated/fetzerch) has quit (Ping timeout: 248 seconds) | |
| [03:41:31] | fetzerch (fetzerch!~quassel@unaffiliated/fetzerch) has joined #mythtv | |
| [03:41:34] | jya: | wagnerrp: actually, I think it's simply in the map already, cause mythutil does toBool(key) and if true call the function. |
| [03:42:12] | wagnerrp: | that key is the internal tag, not the one used on the command line |
| [03:42:16] | jya: | so doing a toBool on the command name should do the trick… I had tried that but it didn't work … must be something else |
| [03:42:23] | jya: | ahhh that's why |
| [03:42:26] | jya: | oh well, that will do |
| [03:42:49] | wagnerrp: | to allow one CommandLineArg to handle multiple different keywords, but be referenced with just one tag |
| [03:43:23] | wagnerrp: | like '-h' and '--help', with the internal tag 'showhelp' |
| [03:44:35] | wagnerrp: | but yeah, a one-liner in the header for m_usedKeyword will get you what you want |
| [03:58:25] | jya: | hum… I added utilMap["notification"] = &SendMessage; |
| [03:58:48] | jya: | yet, toBool("notification") always returns false, must be missing something obvious.. time for cold pizza anyway |
| [04:06:24] | wagnerrp: | jya: did you add anything here? http://code.mythtv.org/cgit/mythtv/tree/mytht . . . ser.cpp#n124 |
| [04:07:45] | wagnerrp: | you probably want to change that --message line to something like 'QStringList() << "--message" << "--notification"' |
| [04:08:43] | wagnerrp: | see http://code.mythtv.org/cgit/mythtv/tree/mytht . . . er.cpp#n2247 |
| [04:12:07] | rkulagow (rkulagow!43b0814a@gateway/web/freenode/ip.67.176.129.74) has quit (Ping timeout: 250 seconds) | |
| [04:12:42] | amessina (amessina!~amessina@2001:470:c1dc:7779:d6be:d9ff:fe8d:7c1e) has quit (Quit: Konversation terminated!) | |
| [04:25:51] | joki (joki!~joki@p54865484.dip0.t-ipconnect.de) has quit (Ping timeout: 252 seconds) | |
| [04:30:39] | ** wagnerrp goes to bed ** | |
| [04:32:45] | joki (joki!~joki@p54865236.dip0.t-ipconnect.de) has joined #mythtv | |
| [05:00:10] | amessina (amessina!~amessina@2001:470:c1dc:7779:d6be:d9ff:fe8d:7c1e) has joined #mythtv | |
| [05:01:52] | amessina (amessina!~amessina@2001:470:c1dc:7779:d6be:d9ff:fe8d:7c1e) has quit (Client Quit) | |
| [05:04:24] | amessina (amessina!~amessina@2001:470:c1dc:7779:d6be:d9ff:fe8d:7c1e) has joined #mythtv | |
| [05:22:08] | bill6502 (bill6502!~bill@205.178.26.43) has left #mythtv () | |
| [05:34:02] | jya: | wagnerrp: oh, yes, I got that covered already... |
| [05:43:21] | lomion0815 (lomion0815!~markus@88-117-33-183.adsl.highway.telekom.at) has joined #mythtv | |
| [05:55:14] | toeb: | q |
| [05:59:17] | brfranse_ (brfranse_!~brfransen@64.179.141.163) has joined #mythtv | |
| [06:00:34] | skd5aner (skd5aner!~skd5aner@50-90-5-146.res.bhn.net) has quit (Read error: Connection reset by peer) | |
| [06:00:34] | brfransen (brfransen!~brfransen@64.179.141.163) has quit (Read error: Connection reset by peer) | |
| [06:00:57] | whoDat (whoDat!~cal@cpe-75-179-59-17.neo.res.rr.com) has quit (Ping timeout: 268 seconds) | |
| [06:01:12] | skd5aner (skd5aner!~skd5aner@50-90-5-146.res.bhn.net) has joined #mythtv | |
| [06:02:11] | whoDat (whoDat!~cal@cpe-75-179-59-17.neo.res.rr.com) has joined #mythtv | |
| [06:03:04] | SteveGoodey (SteveGoodey!~steve@host86-145-229-173.range86-145.btcentralplus.com) has joined #mythtv | |
| [06:03:06] | toeb: | jya: i did test the notifications with vdpau and it works, on issue i have is, that i can't dissmiss the notification with escape/back those exit playback or dismiss the osd not the notification |
| [06:03:55] | jya: | toeb: yes.. in the OSD , it is only displayed, there's no interaction with it, it will disappear when it times out |
| [06:04:45] | toeb: | ok, then this is the next issue i'm having, how does it timeout? It feels quit random :-) |
| [06:04:57] | jya: | it's not an interactive item per say.. it can be dismissed outside playback, but it's a non-intended side effect |
| [06:05:19] | jya: | it depends on what you sent in the notification message |
| [06:05:20] | toeb: | and sometimes it doesent disapear... |
| [06:05:21] | jya: | default is 5s |
| [06:05:41] | toeb: | ok mayby i'm not patient enough.. |
| [06:05:45] | jya: | anything <=5 is, 5s |
| [06:05:59] | jya: | can set a notification for less than that.. may change the behaviour there though |
| [06:06:45] | jya: | can -> can't |
| [06:07:37] | jya: | I'll probably change so if you set a duration of -1, it will stay forever until dismissed. once I find out how I can grab key presses during playback |
| [06:08:32] | toeb: | ok mayby one could add another Notification event type "dismiss" to remove it.. |
| [06:09:13] | jya: | if you want to control how a notification is going to be displayed |
| [06:09:30] | jya: | then you need to register, which give you a unique id for the notification screen |
| [06:09:41] | jya: | when you unregister, the screen is dismissed |
| [06:10:03] | jya: | proper usage if you want to control what happens to your notification, is to register first. |
| [06:10:26] | jya: | otherwise, you explicitely tell the notification center: here it is, deal with it. that's it |
| [06:10:51] | toeb: | yep i did, but for the entier time the frontend is running, i already was wondering why you would want to unregister |
| [06:11:03] | jya: | well, there you go |
| [06:11:09] | jya: | for example, for playback |
| [06:11:13] | jya: | in airplay |
| [06:11:30] | jya: | airplay will register, get an id on how to later control the notification |
| [06:11:50] | jya: | you set the id of each notiification you send using that registered id |
| [06:12:08] | jya: | now I can update the playback location, always in the same screen |
| [06:12:30] | jya: | otherwise, every second an update is sent, you'll get a new notification to pop-up |
| [06:12:41] | dblain (dblain!~dblain@mythtv/developer/dblain) has quit (Ping timeout: 256 seconds) | |
| [06:13:10] | toeb: | yes i got this to work with the id and the NEW and UPDATE types |
| [06:13:41] | jya: | you could actually send a new to the same screen |
| [06:14:00] | jya: | NEW tells the notification center that the content is new, so all fields will be refreshed. |
| [06:14:18] | jya: | so say you send a new image notification, followed by a new text notification |
| [06:14:30] | jya: | the image won't show during the new text one |
| [06:14:40] | jya: | while with update, the existing content is untouched |
| [06:15:06] | toeb: | ok |
| [06:17:12] | jya: | it's really designed to be very simple to use |
| [06:17:16] | jya: | I hope I've achieved that |
| [06:17:58] | toeb: | so i'm using the MythMediaNotification because i like to have the progress bar, but i'm using it more like the MythProgress dialog. e.g I do not really want to show the duration but a progress independend of time... |
| [06:18:11] | SteveGoodey (SteveGoodey!~steve@host86-145-229-173.range86-145.btcentralplus.com) has quit (Quit: Konversation terminated!) | |
| [06:18:20] | jya: | I see... |
| [06:18:23] | stichnot (stichnot!~stichnot@mythtv/developer/stichnot) has quit (Ping timeout: 240 seconds) | |
| [06:18:39] | toeb: | yes, interaction with the notification center is fine, now that i get how the ID is meant to be used :-) |
| [06:18:52] | jya: | hmmm.. problem is that the progress widget requires the duration to calculate the percentage of the progress bar |
| [06:19:38] | jya: | do you just want a progress bar? could just create a new type of notification |
| [06:20:07] | jya: | MythProgressNotification; like mythmedia but no duration value |
| [06:20:17] | jya: | having said that… you could provide 100 |
| [06:20:23] | jya: | show like 100% |
| [06:20:54] | jya: | can add a boolean to show a % instead of a duration |
| [06:20:59] | jya: | would that do? |
| [06:21:12] | toeb: | yes this would work for me |
| [06:21:50] | jya: | ok, Ill extend mythmedianotification to be set as % |
| [06:21:54] | J-e-f-f-A (J-e-f-f-A!~J-e-f-f-A@unaffiliated/j-e-f-f-a) has quit (Ping timeout: 252 seconds) | |
| [06:22:15] | toeb: | thanks |
| [06:22:41] | jya: | though, we could make this themable |
| [06:23:16] | jya: | so a theme decide if you show duration in % or in hh:mm:ss |
| [06:23:33] | jya: | looking forward what you're going to use it for:) |
| [06:24:11] | J-e-f-f-A (J-e-f-f-A!~J-e-f-f-A@pool-173-48-114-173.bstnma.fios.verizon.net) has joined #mythtv | |
| [06:24:12] | J-e-f-f-A (J-e-f-f-A!~J-e-f-f-A@pool-173-48-114-173.bstnma.fios.verizon.net) has quit (Changing host) | |
| [06:24:12] | J-e-f-f-A (J-e-f-f-A!~J-e-f-f-A@unaffiliated/j-e-f-f-a) has joined #mythtv | |
| [06:25:06] | toeb: | well i think it is pretty application specific wether its actually a duration or a multi step process.. |
| [06:27:08] | whoDat (whoDat!~cal@cpe-75-179-59-17.neo.res.rr.com) has quit (Ping timeout: 256 seconds) | |
| [06:28:33] | toeb: | jya: https://github.com/tobser/mythteatime |
| [06:28:39] | whoDat (whoDat!~cal@cpe-75-179-59-17.neo.res.rr.com) has joined #mythtv | |
| [06:29:22] | toeb: | still a work in progress... |
| [06:30:39] | jya: | toeb: for SetId to do anything, you must call SetParent too |
| [06:31:03] | toeb: | oh ok :-) |
| [06:31:34] | jya: | it's my poor-man safety system, so you need to provide the same address as you gave during registration |
| [06:31:53] | jya: | aim is to prevent someone writing to someone's else notification |
| [06:32:08] | jya: | check mythraopconnection.cpp like 1766 |
| [06:36:43] | jya: | hum…. looking at your code; does this even compile? |
| [06:37:00] | jya: | you pass NULL, instead of a QImage& |
| [06:37:04] | jya: | that's not going to work |
| [06:37:51] | jya: | you also don't have to perform a new there, imho, it would be more efficient to work on the heap instead. |
| [06:46:36] | Cougar (Cougar!~cougar@176.124.247.42) has quit (Ping timeout: 252 seconds) | |
| [06:48:22] | Cougar (Cougar!~cougar@2a03:5880:104:10:a9b5:eff2:81f8:f10c) has joined #mythtv | |
| [07:04:02] | MaverickTech (MaverickTech!~MaverickT@111.86.233.220.static.exetel.com.au) has joined #mythtv | |
| [07:07:07] | jwhite (jwhite!~jwhite@75-146-153-89-minnesota.hfc.comcastbusiness.net) has quit (Ping timeout: 264 seconds) | |
| [07:10:41] | FabriceMG (FabriceMG!~Thunderbi@LCaen-156-54-30-212.w80-11.abo.wanadoo.fr) has joined #mythtv | |
| [07:29:16] | toeb: | Well it did compile and it seemed to me as it was working as intended.. but as i said, its just a quik test... |
| [07:31:01] | toeb: | i also think you check for QImage and create one if you receive none.. but i 'm not sure.. |
| [07:31:33] | jya: | toeb: I made a mistake with the duration… I inaverdently called the duration parameter for the playback length, the same as the duration for the life of the notification |
| [07:31:44] | jya: | I've changed it to SetLength |
| [07:35:28] | Merlin83b (Merlin83b!~Daniel@2a00:1ee0:3:1337:24d1:410b:87b6:1a93) has joined #mythtv | |
| [07:35:42] | jwhite (jwhite!~jwhite@75-146-153-89-minnesota.hfc.comcastbusiness.net) has joined #mythtv | |
| [07:36:38] | jya: | toeb: for your use, a MythPlaybackNotification is probably what you want.. provided you don't use images |
| [07:40:52] | toeb: | ok, ill try to follow your adwise... |
| [07:41:04] | jya: | I've added a new constructor for it |
| [07:41:28] | jya: | MythPlaybackNotification(int length, int position, QString text) |
| [07:42:27] | jya: | if text is defined, it will use that text to display the length *duration) of the progress bar |
| [07:45:46] | toeb: | thanks, that's probably what i need. |
| [07:53:17] | toeb: | Do these strings in the metadata actually mean something? Can't guess what "minm" should stand for |
| [07:53:22] | jya: | having 2nd thoughts… the progress bar should take a value between 0 and 1; and not bother with duration, length etc... |
| [07:53:37] | jya: | toeb not sure to be honnest, that's what Apple uses for iTunes |
| [07:53:41] | jya: | they may well do |
| [07:54:15] | toeb: | ok, this at least explains where they come from :-) |
| [07:54:19] | jya: | and the text to display for the duration |
| [07:54:30] | jya: | simpler that way, and less parameters |
| [07:58:03] | toeb: | probably depends on how it is used... for me its more "my progress is at step 3 of 7" and telling it like MythPlaybackNotification(7,3, "bla 3/7") would be easeyer for me... but your way is fine too |
| [08:40:25] | dblain (dblain!~dblain@mythtv/developer/dblain) has joined #mythtv | |
| [08:47:55] | gregL (gregL!~greg@cpe-74-76-105-205.nycap.res.rr.com) has quit (Ping timeout: 264 seconds) | |
| [08:51:15] | gregL (gregL!~greg@cpe-74-76-105-205.nycap.res.rr.com) has joined #mythtv | |
| [09:00:31] | jya: | well, you could just provide MythPlaybackNotification((float)3/7, "blah 3/7") |
| [09:00:36] | jya: | easy as |
| [09:06:26] | jya: | but I'll keep the positiion, length type so it will keep easy ... |
| [09:20:38] | toeb: | sure, what ever you prefer, i'm fine with both |
| [09:21:01] | jya: | I'll ping you when I commit my changes |
| [09:23:12] | toeb: | no rush, i'm not sure when i'll get around to it... |
| [09:23:41] | stichnot (stichnot!~stichnot@mythtv/developer/stichnot) has joined #mythtv | |
| [11:35:33] | purserj_ (purserj_!~purserj@hosting.collaborynth.com.au) has quit (Ping timeout: 240 seconds) | |
| [11:48:17] | sraue (sraue!~stephan@xbmc/staff/sraue) has quit (Ping timeout: 248 seconds) | |
| [11:52:06] | purserj (purserj!~purserj@hosting.collaborynth.com.au) has joined #mythtv | |
| [12:01:35] | sraue_ (sraue_!~stephan@178-191-73-134.adsl.highway.telekom.at) has joined #mythtv | |
| [12:31:15] | dekarl (dekarl!~dekarl@p4FCEFEB3.dip0.t-ipconnect.de) has quit (Ping timeout: 240 seconds) | |
| [12:32:45] | sraue_ is now known as sraue | |
| [12:32:47] | sraue (sraue!~stephan@178-191-73-134.adsl.highway.telekom.at) has quit (Changing host) | |
| [12:32:47] | sraue (sraue!~stephan@xbmc/staff/sraue) has joined #mythtv | |
| [12:35:43] | dekarl (dekarl!~dekarl@p4FE85478.dip0.t-ipconnect.de) has joined #mythtv | |
| [12:37:14] | jya_ (jya_!~jyavenard@mythtv/developer/jya) has quit (Quit: jya_) | |
| [12:39:37] | jya_ (jya_!~jyavenard@mythtv/developer/jya) has joined #mythtv | |
| [13:21:39] | stichnot: | I was wondering – why does our configure script do anything special for distcc and ccache, when you can accomplish the same just by prepending /usr/lib/distcc and/or /usr/lib/ccache to your $PATH? |
| [13:29:43] | jya_: | i think it's historical… there's conflict between the two IRC |
| [13:36:20] | stuarta: | maybe to give you the ability *not* to use it if you have it installed? |
| [13:40:19] | jya_: | that too :) |
| [13:40:45] | jya_: | aren't all binaries in /usr/bin regardless? including ccache |
| [13:46:31] | IReboot (IReboot!~doug@CPE10bf48e67915-CM78cd8e7e342d.cpe.net.cable.rogers.com) has joined #mythtv | |
| [13:47:57] | stuarta: | yes they are these |
| [13:48:00] | stuarta: | +days |
| [13:48:39] | stichnot: | If you don't want to use ccache, don't put /usr/lib/ccache in your PATH :) |
| [13:49:53] | stichnot: | At work, I spent many weeks annoyed that building LLVM doesn't "support" ccache, until a coworker kindly pointed out /usr/lib/ccache |
| [14:05:49] | jya_: | stichnot: on ubuntu at least (and same on my mac), if ccache is installed, it's always in the path |
| [14:06:07] | jya_: | yavenard@htpc:~$ which ccache |
| [14:06:17] | jya_: | /usr/bin/ccache |
| [14:07:13] | jya_: | I use ccache, and certainly don't have /usr/lib/ccache in my path |
| [14:14:03] | stichnot: | ccache is definitely in the PATH, but modifying a Makefile to prepend ccache to everything is the problem |
| [14:18:25] | lomion0815 (lomion0815!~markus@88-117-33-183.adsl.highway.telekom.at) has quit (Remote host closed the connection) | |
| [14:45:03] | jya_ (jya_!~jyavenard@mythtv/developer/jya) has quit (Quit: jya_) | |
| [14:57:24] | danielk22 (danielk22!~danielk@exchange.wgen.net) has joined #mythtv | |
| [15:05:18] | stichnot (stichnot!~stichnot@mythtv/developer/stichnot) has quit (Ping timeout: 246 seconds) | |
| [15:32:50] | stichnot (stichnot!~stichnot@207.198.105.24) has joined #mythtv | |
| [15:32:50] | stichnot (stichnot!~stichnot@mythtv/developer/stichnot) has joined #mythtv | |
| [15:32:50] | stichnot (stichnot!~stichnot@207.198.105.24) has quit (Changing host) | |
| [15:47:13] | FabriceMG (FabriceMG!~Thunderbi@LCaen-156-54-30-212.w80-11.abo.wanadoo.fr) has quit (Quit: FabriceMG) | |
| [16:09:58] | rsiebert (rsiebert!~quassel@e179131117.adsl.alicedsl.de) has joined #mythtv | |
| [16:11:02] | rsiebert_ (rsiebert_!~quassel@g231187065.adsl.alicedsl.de) has quit (Read error: Connection reset by peer) | |
| [16:21:34] | Beirdo: | stupid stupid irssi |
| [16:22:27] | ** stuarta waves to Beirdo ** | |
| [16:22:40] | Beirdo: | Heya :) |
| [16:25:52] | SteveGoodey (SteveGoodey!~steve@host86-145-229-173.range86-145.btcentralplus.com) has joined #mythtv | |
| [16:31:16] | stichnot (stichnot!~stichnot@mythtv/developer/stichnot) has joined #mythtv | |
| [16:32:58] | gary_buhrmaster (gary_buhrmaster!~gtb@2001:470:80e4:3::36) has joined #mythtv | |
| [16:34:13] | stichnot: | jpabq: do you think #11470 is now fixed after the recorder seektable fixes? |
| [16:34:13] | ** MythLogBot http://code.mythtv.org/trac/ticket/11470 ** | |
| [16:35:21] | jpabq: | stichnot: Almost definitely. I actually sent warpme an email last week, but have not gotten a response. I am tempted to mark it as closed, and see if he complains. |
| [16:35:47] | jpabq: | The sample I used to 'fix' came from him, and it now works for me.... |
| [16:35:59] | stichnot: | ok, sounds good. |
| [16:40:09] | stichnot: | peper03: You might as well grab #11292 and add to it or close it as you see fit. I don't expect to have much time for it before the various freeze dates. |
| [16:40:09] | ** MythLogBot http://code.mythtv.org/trac/ticket/11292 ** | |
| [16:41:24] | jpabq: | I would love to close #11352 as well. I was never able to reproduce the leak, and valgrind indicated that if there is a leak, that it is in Qt. |
| [16:41:24] | ** MythLogBot http://code.mythtv.org/trac/ticket/11352 ** | |
| [16:53:23] | Beirdo: | So... after a nearly 1yr break... I come back just before freeze? :) Doesn't that just figure :) |
| [16:53:27] | Merlin83b (Merlin83b!~Daniel@2a00:1ee0:3:1337:24d1:410b:87b6:1a93) has quit (Read error: Connection reset by peer) | |
| [16:54:20] | jpabq: | stichnot: Enough people are complaining about LiveTV, that it would be nice to say "It is fixed in 0.27". Personally, if you are willing to commit to making LiveTV work better, I think we should delay 0.27 to give you that chance....even if that means delaying yet another month. |
| [16:54:50] | jpabq: | I would really like to convince users that 0.27 is worth upgrading to, so we stop having people file bug reports against 0.25 |
| [16:54:55] | SteveGoodey: | Hey, beirdobot is back, good! |
| [16:59:42] | jpabq: | Beirdo: what magic are planning to add for 0.28 ;-) |
| [16:59:50] | jpabq: | are you |
| [17:05:32] | stichnot: | jpabq: I have no confidence that I could "fix" live TV in any medium length of time. The main problem is the inability to reproduce the most common failures. When I can reproduce them, they get fixed. :) |
| [17:06:21] | jpabq: | Fair enough ;-) |
| [17:06:25] | stichnot: | With that said, a kind user here loaned me an HVR-1950, which I'll set up and try out when I get back. |
| [17:07:20] | stichnot: | My biggest pain point atm is that my backend really doesn't like it when the garage gets to 110+ degrees... |
| [17:07:30] | exoon (exoon!~exoon@p4FD392FE.dip0.t-ipconnect.de) has joined #mythtv | |
| [17:09:18] | jpabq: | Heh, I put up with the fan noise of having my mythbackend in my office. Unlike last week, it is not supposed to break 100 here, this week. |
| [17:09:18] | Beirdo: | heh, when is 0.28? :) |
| [17:09:28] | stichnot: | I replaced the stock CPU cooler with "Cooler Master Hyper 212 EVO", to no effect, so now my suspicions are on the northbridge/pch which just has a heatsink |
| [17:09:46] | Beirdo: | I might be able to get some gpu stuff actually "completed" in a few months if I'm not too busy with other things, of course |
| [17:10:12] | stichnot: | a northbridge cooler with fan should be arriving today, hopefully that will fix it |
| [17:10:20] | jpabq: | Beirdo: Freeze for 0.27 went into effect last night. I don't believe that anyone has proposed a timeline for 0.28 |
| [17:11:00] | Beirdo: | OK :) |
| [17:11:14] | Beirdo: | it's about the right time for a freeze, I guess :) |
| [17:13:15] | Beirdo: | ugh, and apparently, for a meeting I'd rather sleep through :) |
| [17:13:36] | jpabq: | Beirdo: Someone I know proposed using audio "fingerprinting" as a technique for identifying commercials. Maybe using something like http://acoustid.org/ . Sounds like an interesting idea to me. The trick would be coming up with the fingerprint DB. |
| [17:14:03] | Beirdo: | Yeah, that could easily be added to the framework I was working on, I think |
| [17:15:06] | Beirdo: | I'll have the dev box up real soon, and then I can try to refresh that part of my brain along with everything else |
| [17:15:36] | Beirdo: | I think I had it to a point where I had a whole pile of audio and video events to correlate to one another and to determine likely transitions |
| [17:15:51] | Beirdo: | but then life got in the way and all :) |
| [17:15:56] | jpabq: | My biggest confusion with the current commflagger, is that it will decide a commercial starts on a frame which is not completely black, and pass over a frame which is. Sometime I may need to go in and figure out why. |
| [17:16:31] | Beirdo: | yeah, it's just odd. If I can get my framework working, it should theoretically work better on CPU-only setups too |
| [17:16:40] | Beirdo: | but.. gotta make it not vaporware |
| [17:16:40] | Beirdo: | heh |
| [17:21:29] | jhall_ (jhall_!~jhall@2001:468:d01:9c:226:b9ff:fe2d:482e) has quit (Remote host closed the connection) | |
| [17:27:48] | stichnot: | jpabq: Is that still happening after the seektable fixes? |
| [17:28:58] | jpabq: | stichnot: Are you talking about commflag? If so, then I don't know. I have not actually watched much TV lately. |
| [17:30:02] | stichnot: | yeah, commflag. I've watched a small amount of TV but I haven't done much editing so I don't know how the precision is holding up, and the commskip buttons are accurate enough |
| [17:30:35] | taylorr: | jpabq: should I backport the latest changes to 0.26 for the h264 parser and dtvrecorder to 0.25? |
| [17:31:37] | exoon (exoon!~exoon@p4FD392FE.dip0.t-ipconnect.de) has quit (Ping timeout: 246 seconds) | |
| [17:32:20] | stichnot: | Commflag is supposed to choose the centerpoint of the blank region as the mark, so if it is all the way into the non-blank region, it could be because the frame numbering is off, e.g. by a keyframe interval. |
| [17:32:22] | jpabq: | Ah — I had forgotten that this stuff had been ported all the way back to 0.25. Yes, http://code.mythtv.org/trac/changeset/09b7873 . . . a860e/mythtv should be backported to fix that problem. The changes to H264Parser are probably not necessary, though. |
| [17:33:14] | Beirdo: | Hmm, whether to get a HDHomeRun Prime or a Ceton InfiniTV... |
| [17:35:01] | jpabq: | stichnot: What I noticed, is it starting a commerical block on a transition that didn't go completely black, and even still had people talking — so it started the comm skip several seconds before it was supposed to. Then, when the commercial block was over, there is a nice completely black frame with silence, but the commflagger did not use it, and instead waited several seconds into the show before resuming playback. |
| [17:35:31] | Beirdo: | part of the problem with the current commflagger is that it utterly ignores audio (last I looked) |
| [17:35:49] | Beirdo: | which does tend to lead to silliness like that in some cases :) |
| [17:36:32] | stichnot: | jpabq: so it actually sounds like the start/end were shifted by a consistent amount. |
| [17:37:04] | jpabq: | Beirdo: Mark Spieth wrote a patch to commflag based on silence, but when I tested it, it had issues. |
| [17:37:52] | gary_buhrmaster: | jpabq: Did not someone suggest it was probably time to cut a final 0.25.4? release, and then stop back-porting to 0.25 fixes? |
| [17:38:01] | Beirdo: | yeah, it's been such a black-magic thing for so long |
| [17:38:35] | stichnot: | Also, lvr has a patch that gives visual feedback from audio levels for the cutlist editor, but it only works for 1- or 2-channel audio |
| [17:39:16] | stichnot: | I wonder if jya has some nice new tools for this kind of audio analysis |
| [17:39:25] | jpabq: | stichnot: this was just an example. On some stations, it actually works VERY well. On others it works very porely. For example, with ABC it pretty consistently misses the end of the comm block, and I have to rewind — sometimes as much as 90 seconds, even though there was a nice solid black frame with silence at the end of the comm block. |
| [17:39:45] | Beirdo: | He likely does |
| [17:40:25] | Beirdo: | I know I have GPU code ready to detect the RMS audio level so you can find sections significantly above and below "normal" for a recording |
| [17:40:38] | Beirdo: | but no visualization of it all, of course |
| [17:41:08] | Beirdo: | I was testing by dumping graphs to disk IIRC |
| [17:41:50] | stichnot: | jpabq: any chance this involved changing frame rates? |
| [17:41:53] | Beirdo: | I'm sure JYA has something far more elegant available for his testing |
| [17:42:15] | jpabq: | It would help a LOT, if the commflagger keyed off of the Content Rating icon which seems to be shown after every commercial block. |
| [17:42:37] | Beirdo: | that could be factored in for US recordings at least |
| [17:43:04] | jpabq: | stichnot: possible. I have not investigated properly. |
| [17:44:52] | taylorr: | jpabq: ok, I'm testing the dtvrecorder change on 0.25 and will push later.... I thought the h264 parser commit fixed some issues? |
| [17:46:00] | jpabq: | taylorr: the changes to H264Parser make it follow the current spec more tightly. If you want to backport them, then that is fine. I just don't really expect a functional difference. |
| [17:46:34] | taylorr: | jpabq: ok, unless it is known to fix anything then I'll leave it out for now |
| [17:49:41] | jpabq: | stichnot: If you are going to leave that computer in the garage, you may have to go with a water block. I also have the 212 EVO, mostly to quiet down the cooling, but it is not going to compete with that kind of ambient tempts. Corsair and Cooler Master apparently make some nice closed-loop water coolers. |
| [17:50:08] | Beirdo: | hook up a swamp cooler :) |
| [17:51:32] | jpabq: | Actually, thinking about it, a closed-loop water cooler may not do the trick either — it is not going to be able to cool it below ambient, and if your ambient is already 100 degrees F, you are in trouble. |
| [17:52:02] | jheizer__: | Open the case side and put a box fan on it? Even if it is just for a few days. |
| [17:52:10] | jheizer__: | help the chipset HS |
| [17:52:14] | jheizer__ is now known as jheizer | |
| [17:52:27] | stichnot: | jpabq: It appears that the CPU is staying cool, both with the stock cooler and with the 212 EVO. The PCH heatsink is hot to the touch, even to my heat-insensitive fingers, so I think that's it. |
| [17:52:56] | stichnot: | The other possibilities are the RAM (but it doesn't feel hot at all), and possibly the power supply might be flaking under the heat |
| [17:52:58] | jpabq: | I believe they make water blocks for the PCH as well... |
| [17:54:03] | stichnot: | Newegg is delivering http://www.newegg.com/Product/Product.aspx?It . . . tualParent=1 today, with luck that will cure it |
| [17:54:50] | stichnot: | I had zero problems with my old Core 2 Duo system in the same garage 2–3 years ago, so I believe this is doable |
| [17:55:44] | jpabq: | That should work. Probably good that you don't care about noise. |
| [17:55:59] | stichnot: | jheizer: I am using a box fan today :) |
| [17:56:45] | jheizer: | Ah ok. Was trying to go with the simple idea. Hopefully that HS helps. |
| [17:57:06] | jheizer: | May also try taking some can air to the PSU to clean it out good. |
| [17:57:45] | jheizer: | Having a cool basement has it's perks. :) |
| [17:58:22] | stichnot: | yeah, I should do that to the whole system. There's a startling amount of spider webs inside the case. |
| [17:59:55] | jheizer: | Haha a server and a hotel |
| [18:01:11] | stichnot: | I expect the hotel is more inviting in the winter |
| [18:03:25] | NightMonkey (NightMonkey!~NightrMon@173-164-139-193-SFBA.hfc.comcastbusiness.net) has joined #mythtv | |
| [18:03:26] | NightMonkey (NightMonkey!~NightrMon@173-164-139-193-SFBA.hfc.comcastbusiness.net) has quit (Changing host) | |
| [18:03:26] | NightMonkey (NightMonkey!~NightrMon@pdpc/supporter/professional/nightmonkey) has joined #mythtv | |
| [18:06:31] | NightMonkey (NightMonkey!~NightrMon@pdpc/supporter/professional/nightmonkey) has quit (Client Quit) | |
| [18:43:10] | NightMonkey (NightMonkey!~NightrMon@64.124.185.45) has joined #mythtv | |
| [18:43:11] | NightMonkey (NightMonkey!~NightrMon@64.124.185.45) has quit (Changing host) | |
| [18:43:12] | NightMonkey (NightMonkey!~NightrMon@pdpc/supporter/professional/nightmonkey) has joined #mythtv | |
| [18:52:07] | gigem: | Beirdo: I have an HDHR Prime and a Ceton InfiniTV4. Both work very well. The only knock on the Ceton is the coax connector is pretty fragile. Basically, if you ever need to move the cable or system, disconnect the coax first. I broke the connector on my first one within a couple of days. Fortunately, Ceton knew about the problem and sent an advanced replacement fre of charge. Hopefully, they're using |
| [18:52:09] | gigem: | better connectors on newer cards. Anyway, if I were buying now, I'd probably go with the InfiniTV6 for the better bang for the buck. |
| [18:52:32] | Beirdo: | Yeah, I was considering the InfiniTV6 |
| [18:52:42] | Beirdo: | now that I'm with Comcast, might as well |
| [18:53:10] | Beirdo: | cool, thanks for the input, I'll probably be ordering one in a couple days |
| [18:55:11] | gigem: | yw. |
| [19:51:15] | stichnot: | sphery: how are you planning to fix up the navigation in editing? |
| [20:43:17] | SteveGoodey (SteveGoodey!~steve@host86-145-229-173.range86-145.btcentralplus.com) has quit (Quit: Konversation terminated!) | |
| [20:51:20] | dblain (dblain!~dblain@mythtv/developer/dblain) has quit () | |
| [21:07:43] | gary_buhrmaster (gary_buhrmaster!~gtb@2001:470:80e4:3::36) has quit (Ping timeout: 264 seconds) | |
| [21:15:30] | bas-t (bas-t!~tycho@52484E89.cm-4-1b.dynamic.ziggo.nl) has joined #mythtv | |
| [21:19:59] | bas-t (bas-t!~tycho@52484E89.cm-4-1b.dynamic.ziggo.nl) has quit (Client Quit) | |
| [22:06:18] | jya: | Beirdo: nice to see you bock… wondering were you were... |
| [22:06:52] | jya: | is having you back got anything to do with blackberry declaring a big loss this quarter ? what did you do this time ? |
| [22:08:28] | jya: | as far as analysing the audio levels, could use a similar method as I did for the vuvuzela filter… I'll ask my DSP engineer some quick way to determine audio levels… actually something that I'm sure would help is some kind of detectiong of audio compression if thats possible. The thing that make the ad appear louder |
| [22:20:55] | gigem: | jya: That's not supposed to happen anymore in the States. I believe broadcasters can be fined now if their perceived audio levels, including commercials, change to wildly. |
| [22:22:05] | jya: | do they actually state "perceived" ? remember watching a program last year here explaining again why ads really aren't any louder |
| [22:40:42] | Seeker`: | I can't remember if they introduced something that limited it being louder, or limited the amount of compression they can apply |
| [22:43:32] | wagnerrp: | it had to sound roughly the same level as the show it is interspersed in |
| [22:55:53] | stichnot: | jya: I'd like a routine such that given timecode T and delta D, it would return some representation of audio levels (maybe broken down by frequency range) for the time range [T-D..T+D]. |
| [22:59:31] | jya: | what do you call audio levels? |
| [23:00:01] | jya: | just to be clear |
| [23:06:13] | NightMonkey (NightMonkey!~NightrMon@pdpc/supporter/professional/nightmonkey) has quit (Ping timeout: 246 seconds) | |
| [23:07:00] | NightMonkey (NightMonkey!~NightrMon@64.124.185.45) has joined #mythtv | |
| [23:07:00] | NightMonkey (NightMonkey!~NightrMon@pdpc/supporter/professional/nightmonkey) has joined #mythtv | |
| [23:07:00] | NightMonkey (NightMonkey!~NightrMon@64.124.185.45) has quit (Changing host) | |
| [23:24:32] | stichnot: | jya: (sorry I had to step away) I don't really know/care exactly what the audio levels mean. E.g. I would like some values that I would display as an equalizer style histogram, primarily to indicate whether there is audio around the current frame or silence |
| [23:25:38] | jya: | just that you have this "perceived" audio level and the actual dB level… but something telling you if it's silent or close to, would be much easier |
| [23:26:43] | jya: | the commflagger already receive audio, so plugging that there wouldn't be too complicated I think… when you get an audio frame, with the timestamp, could measure the audio level there, up to you to do whatever you want |
| [23:27:30] | jya: | audio received is a signed 16 bits integer.. did you try simply doing a small average on the samples received? |
| [23:27:39] | jya: | wonder if something that simple would do the trick |
| [23:27:58] | jya: | or calculate the amplitude and measure the difference between min and max |
| [23:28:01] | jya: | that should do |
| [23:28:12] | jya: | even easier |
| [23:29:25] | wagnerrp: | not unsigned? |
| [23:29:48] | jya: | only unsigned value used as audio are 8 bits |
| [23:30:07] | jya: | which is then extended to be within -128 , +127 |
| [23:32:27] | stichnot (stichnot!~stichnot@mythtv/developer/stichnot) has quit (Ping timeout: 246 seconds) | |
| [23:45:32] | danielk221: | jya: The CALM act specifies that broadcasters apply an ATSC standard which basically tells you that all commercials and promos must have an average loudness of -24 dB as given by ITU-R BS.1770.{1,2,3}, and the overall long form program must have an average human dialogue loudness of -24 dB. |
| [23:45:58] | jya: | ok… so that's not apparent loudness then, that's actual audio levels |
| [23:46:31] | danielk221: | Not exactly. BS.1770 applies a perceptual filter. |
| [23:46:53] | jya: | how do you measure in dB apparent level? |
| [23:47:17] | danielk221: | Levels in 3000 Hz get a greater weight than levels at 16kHz. |
| [23:47:57] | danielk221: | BS.1770.1 is a free download and is short read for a broadcast standard. |
| [23:49:50] | danielk221: | Basically a bunch of fancy apparent level algorithms were submitted to the ITU but this freely available one performed about as well as the best ones. It is by no means perfect, you can cheat by 3dB fairly easily.. but the things people were complaining most about show up as huge dB changes with this algorithm. |
| [23:55:26] | danielk221: | Beirdo: Welcome back! |
| [23:58:47] | stichnot (stichnot!~stichnot@207.198.105.24) has joined #mythtv | |
| [23:58:47] | stichnot (stichnot!~stichnot@mythtv/developer/stichnot) has joined #mythtv | |
| [23:58:47] | stichnot (stichnot!~stichnot@207.198.105.24) has quit (Changing host) | |
IRC Logs collected by
BeirdoBot.
Please use the above link to report any bugs.