:: #mythtv

Daily chat history

Current users (71):

alan`, aloril, amessina, andreaz, Anssi, brfransen, caelor_, Casper82, cesman, ChanServ, Chutt__, clever, coling, dblain, dekarl, ElmerFudd, fetzerch, gary_buhrmaster, ghoti, Gibby, gigem, gregL, GreyFoxx_, J-e-f-f-A, jab416171, jams_, jheizer_, jnylen, jpabq, jpharvey, jst, jwhite, jya, Kevin`, knightr__, kormoc, kurre2, Merlin83b, moparisthebest, MythBuild, MythLogBot, nephyrin, peper03, poptix, pppingme, purserj_, rich0, robink, Seeker`, seld, Sharky112065, sheedy-away, skd5aner, sl1ce, sphery, sraue, stuarta, stuartm, suffice, superm1, taylorr, tgm4883, toeb, tonsofpcs, tris, unforgiven512, wagnerrp, XDS2010_, xris, zentec_, _charly__
Wednesday, February 4th, 2015, 00:03 UTC
[00:03:31] joki (joki! has joined #mythtv
[00:10:08] arescorpio (arescorpio! has joined #mythtv
[00:45:49] brian|lfs (brian|lfs! has joined #mythtv
[00:46:14] brian|lfs (brian|lfs! has left #mythtv ("Leaving")
[02:11:20] sheedy-away is now known as sheedy
[03:33:02] arescorpio (arescorpio! has quit (Excess Flood)
[03:39:07] Roklobsta (Roklobsta! has joined #mythtv
[03:48:00] andreaz (andreaz! has quit (Quit: Leaving.)
[04:11:57] sheedy is now known as sheedy-away
[04:16:13] Roklobsta (Roklobsta! has quit (Ping timeout: 250 seconds)
[04:53:13] fetzerch (fetzerch!~quassel@unaffiliated/fetzerch) has quit (Ping timeout: 264 seconds)
[07:20:00] fetzerch (fetzerch!~quassel@unaffiliated/fetzerch) has joined #mythtv
[07:24:46] Roklobsta (Roklobsta! has joined #mythtv
[07:40:30] Tobbe5178 (Tobbe5178!~asdf@2001:2002:d9d4:ce7e:5c87:3658:281e:a35d) has joined #mythtv
[07:43:45] Roklobsta (Roklobsta! has quit (Ping timeout: 252 seconds)
[07:45:14] Roklobsta (Roklobsta! has joined #mythtv
[07:57:49] Roklobotomy (Roklobotomy! has joined #mythtv
[08:01:37] Roklobsta (Roklobsta! has quit (Ping timeout: 255 seconds)
[08:09:43] xris (xris!~xris@mythtv/developer/xris) has quit (Ping timeout: 255 seconds)
[08:52:46] stuarta: morning all
[09:18:34] Merlin83b (Merlin83b! has joined #mythtv
[09:39:06] dekarl1 (dekarl1!~dekarl@mythtv/developer/dekarl) has joined #mythtv
[09:41:27] dekarl (dekarl!~dekarl@mythtv/developer/dekarl) has quit (Ping timeout: 264 seconds)
[09:43:05] andreaz (andreaz! has joined #mythtv
[09:43:05] andreaz (andreaz! has quit (Client Quit)
[10:24:47] dekarl1 is now known as dekarl
[10:29:33] dekarl: tgm4883: stuartm: pushing all three arches at the same time sounds like it avoids cases where an amd64 backend is connected to an armhf frontend and we change the protocol/database version. With >100 known master installs (according to smolt) it would be nice to avoid the surprise
[10:47:12] xris (xris! has joined #mythtv
[11:09:54] dekarl: stuartm, is that on purpose that 4:3 preview images in the "last 10 recordings" are showing fully in the preview bar, but zoomed in the main view? (aka without pillar box)
[11:11:13] dekarl: the image is fetched with height=371, so I'm wondering if either it should be pillar boxed or the call changed to width=660
[11:13:58] stuartm: dekarl: that sounds like a bug, but a baffling one, the height/width arg in no way tells the code to crop images
[11:14:44] stuartm: could it be that they are actually being stretched? That I can believe
[11:14:56] dekarl: no, already tested that :)
[11:17:35] stuartm: well no idea, images are shown correct here, but then we've not had any true 4:3 since the 90s iirc
[11:17:42] dekarl:
[11:18:25] dekarl: notice the area above the head in the small vs large view
[11:18:55] dekarl: the station logo is a bit hard to see in the small view, but missing (out of image) in the large view
[11:18:56] stuartm: dekarl: ohh, I see where you're talking about – yes, that's probably an intentional thing of the galleria widget
[11:19:18] stuartm: maybe configurable though
[11:19:59] dekarl: if its on purpose then that is perfectly ok. using more of the potentially scarce screen space for actual content
[11:20:13] dekarl: next thing to figure out, which is the "last 10" showing 3
[11:20:19] dekarl: w/which/why/
[11:20:38] stuarta: auto censorship ;-p
[11:20:45] dekarl: its started out with 2, a reload added the third one
[11:21:27] dekarl: maybe because the images are generated when the page is first loaded? who knows. (mythical censoship, we need to promote that feature :)
[11:26:46] stuartm: the 'overview' page mostly predates my work, all I've done is to add the frontend status bit to the bottom, and even that is a work in progress
[11:30:21] dekarl: stuartm, btw. notice that both the remote frontend and the local frontend on the backend are online in the picture, when mythfrontend is not running on the backend...
[11:31:00] dekarl: myth-fe1 being the RFE, mythmaster being the MBE (I know, very imaginative hostnames)
[11:33:30] dekarl: starting and stopping a frontend on mythmaster doens't do anything to the status, it works nicely for myth-fe1 (changes between offline/online as I stop/start mythfrontend on that host)
[11:39:59] stuartm: dekarl: hmm, well that should be impossible, it tracks clients that have opened a monitor/playback socket to the master, and the master can't open sockets to itself
[11:40:46] Roklobotomy (Roklobotomy! has quit (Ping timeout: 255 seconds)
[11:40:56] stuartm: would need to look at the backend logs with -v network to figure out what's going wrong there
[11:43:11] Roklobsta (Roklobsta! has joined #mythtv
[11:49:48] dekarl: what does mythweb do?
[11:50:52] stuarta: lurks quietly in the corner feeling unloved
[12:03:36] stuartm: dekarl: it might open a monitor socket, hadn't considered that
[12:04:55] stuartm: guess I need to beef up the logic, though the only way I can think of doing so reliably would require a major change in the protocol, requiring frontends to announce (ANN) themselves as frontends
[12:05:28] stuartm: dekarl: I'd be surprised if mythweb kept that socket open when it wasn't be used
[12:06:45] dekarl: its offline now... maybe mythcommflag while recording?
[12:07:01] ** dekarl scratches head **
[12:07:18] stuarta: that's more likely than mythweb
[12:07:31] stuarta: it makes connections as you navigate the pages
[12:10:01] stuartm: hmm, think I could extend the protocol with a new token without breaking older client, our frontends would explicitly announce themselves as frontends while other clients, including non-frontends would continue to use the old PLAYBACK/MONITOR tokens
[12:11:17] Roklobotomy (Roklobotomy! has joined #mythtv
[12:11:18] dblain_ (dblain_!dblain@ has joined #mythtv
[12:11:18] dblain_ (dblain_!dblain@mythtv/developer/dblain) has joined #mythtv
[12:11:18] dblain_ (dblain_!dblain@ has quit (Changing host)
[12:13:52] zentec_ (zentec_!~zentec@ has joined #mythtv
[12:15:38] GreyFoxx_ (GreyFoxx_! has joined #mythtv
[12:20:55] Roklobsta (Roklobsta! has quit (*.net *.split)
[12:20:55] dblain (dblain!dblain@mythtv/developer/dblain) has quit (*.net *.split)
[12:20:55] jwhite (jwhite! has quit (*.net *.split)
[12:20:55] GreyFoxx (GreyFoxx! has quit (*.net *.split)
[12:20:55] jab416171 (jab416171! has quit (*.net *.split)
[12:20:56] zentec (zentec!~zentec@ has quit (*.net *.split)
[12:21:14] peper03 (peper03!~peper03@mythtv/developer/peper03) has quit (Ping timeout: 265 seconds)
[12:22:29] peper03 (peper03!~peper03@mythtv/developer/peper03) has joined #mythtv
[12:27:11] dekarl: stuartm, hmm the "only 3 of 10 last recordings" is funny
[12:27:58] dekarl: meh, didn't capture the mouse hovering over the img src path, it pops up the preview image, as expecte
[12:28:14] jwhite (jwhite! has joined #mythtv
[12:28:46] jab416171 (jab416171! has joined #mythtv
[12:29:03] jab416171 (jab416171! has quit (Max SendQ exceeded)
[12:29:06] dekarl: "width: 216px" one level higher looks fishy
[12:29:28] jab416171 (jab416171! has joined #mythtv
[12:30:22] dekarl: if I change that, then all 7 thumbnails get their spot reserved in one row. (7? why that ;)
[12:30:56] dekarl: but that is likely just part of the sympton, not of the cause
[12:36:45] dekarl: wrt the Online indicater, just started mythtranscode and its on again, so it looks like that was it
[12:37:10] stuarta: makes sense
[12:37:33] stuartm: dekarl: ok, testing a patch now which changes the protocol so the backend knows the difference between a frontend and other clients
[12:37:53] stuartm: backwards compatible with third party clients
[12:38:06] stuartm: but I'll still need to bump the protocol
[12:38:29] dekarl: sounds like the same could have happend with clients like kodi, too. so appreciate it
[12:39:45] stuartm: well I was happy for third party frontends to be listed there, but I'll let them decide whether to add support for the new token
[12:40:26] stuartm: I still need to add similar tracking for slaves
[12:41:20] dekarl: hmm, IIUIC then other clients understand different means of control should not be handled identically. e.g. does Kodi obey our play command?
[12:41:29] stuartm: true
[12:41:55] dekarl: lets add stuff like "tell Kodi/My TV to play this recording via generic UPNP protocol" later
[12:44:18] stuartm: yeah, never enough time, I'd love to get upnp renderer and control support into MythTV, but I can't do that and finish the webfrontend :/
[12:45:24] dekarl: reminds me to come up with a "unfinished projects that are almost there" list so others can find the low hanging fruit
[12:49:15] stuarta: good idea
[13:19:45] dekarl: stuartm, the synopsis over here usually contains quotation marks (at least for the recordings that break the "last ten" view) which appears to not be properly escaped
[13:21:11] stuarta: dekarl: if you use your web browsers developer tools you should see the errors in the console
[13:24:15] stuartm: dekarl: noted, I'll fix
[13:25:02] dekarl: stuarta, nice idea. I see CORS related to mixing backend web frontend and frontend web frontend resources
[13:26:23] stuarta: dekarl: i spend much of my time with the dev tools open when testing the site migrations
[13:26:25] dekarl: some deprecation warnings, I guess I'll let stuartm have a look at them first. oddly no mention of the HTML code errors
[13:27:08] dekarl: either way, lukely using the source helped, too ;)
[13:27:44] stuartm: I thought I'd made exceptions for the frontend with CORS, but perhaps not
[13:31:22] stuartm: ah, hmm, might be that I add the master server IP to the allowed origins, but not the hostname
[13:34:54] dekarl: Hmm, how do we get the FQDN? The links to the frontend web frontend would look nicer with the FQDN, too.
[13:35:07] dekarl: "whatever a reverse lookup reveals"?
[13:35:47] dekarl: thinking about it, with multiple IPs (e.g. v4 and v6) that may be a can of worms for little gain
[13:36:20] dekarl: for the backend itself it could be "whatever you called me" and the ip for remote frontends
[13:36:21] stuarta: no, reverse lookup is unlikely to work
[13:40:28] ** dekarl loves interlacing **
[13:41:18] dekarl: stuartm, removing the quotation marks unbreaks that image for galleria, so its that
[13:41:47] stuarta: oooo, you've got ghosts
[13:42:11] dekarl: its the new digital ghosts ;)
[13:49:23] stuartm: dekarl:
[13:55:15] dekarl: uhh, downloading the raw view gives me a patch with ^M that doesn't apply... the joys of working with text ;)
[13:57:17] stuartm: bah, that's why we don't use
[13:59:13] stuartm: can't understand how it's the biggest pastebin site when you can't get the raw, unmodified pastes back out
[13:59:42] stuartm: and is not working, again
[13:59:54] stuarta:
[14:00:33] stuarta: if you use fedora there is even an fpaste command so you can do shit like `echo fish | fpaste`
[14:00:45] stuartm:
[14:01:42] stuartm: stuarta: handy, just installed fpaste
[14:02:17] stuartm:
[14:03:30] stuartm: will have to put a wrapper around it that provides some args just for diffs
[14:03:41] dekarl: I applied it, but now the page is broken :)
[14:03:52] dekarl: it ends where the first image should be inserted
[14:04:05] stuartm: dekarl: what's the error from the backend log
[14:04:41] dekarl: backend? isn't that frontend code?
[14:04:50] dekarl: s/frontend/browser/
[14:04:57] stuartm: no, backend code
[14:05:03] stuartm: nevermind, just noticed what the issue is
[14:06:12] dekarl: no error in the logfile
[14:06:31] stuartm:
[14:06:35] Jordack (Jordack! has joined #mythtv
[14:07:07] Steve-Goodey (Steve-Goodey! has joined #mythtv
[14:07:19] dekarl: ok, the include was what made me think its code for the browser
[14:07:52] stuartm: dekarl: yeah, thinko had me add it as a client side include instead of server side
[14:08:38] dekarl: works well
[14:08:50] dekarl: ty
[14:08:52] stuarta: \o/
[14:10:18] stuartm: dekarl: ok, will push when I'm not wrestling this frontend announce change
[14:13:40] dekarl: next up, how do you determine if a channel has an icon? it appears its not the contents of the "icon" column in the channel table
[14:14:03] dekarl: I'm trying to sort out why I'm sometimes seeing the question mark and sometimes nothing (404)
[14:14:22] dekarl: I already merged my two MYTHCONFDIRs
[14:14:33] stuarta: is that good or bad?
[14:14:38] stuartm: dekarl: certainly should be in current master
[14:14:53] stuartm: the contents of the icon column
[14:15:10] stuartm: strictly speaking, whether it's empty or not
[14:15:10] dekarl: I'm on master as of last night
[14:15:18] stuartm: " " wouldn't be an empty string
[14:15:32] dekarl: stuartm, ok need to look for stuff like that (and where it might be coming from)
[14:18:41] dekarl: does it work with png icons, or only jpg?
[14:19:09] dekarl: nvm, some of the pngs are working
[14:22:14] dekarl: oh my
[14:22:43] dekarl: looks like mfdb stores the 404 html page into the icon file if an icon is missing
[14:23:08] dekarl: surprisingly the backend web frontend doesn't show the proper icon even though there is a file
[14:23:11] moparisthebest (moparisthebest!~mitb@unaffiliated/moparisthebest) has quit (Ping timeout: 256 seconds)
[14:36:58] moparisthebest (moparisthebest!~mitb@gateway/tor-sasl/moparisthebest) has joined #mythtv
[14:38:49] toeb (toeb! has joined #mythtv
[14:43:36] stuartm: it's trying to display an image, it won't display html ;)
[14:48:47] MythBuild: build #1318 of master-win8-msvc-2010–32bit is complete: Failure [4failed Configure and Build] Build details are at . . . /builds/1318 blamelist: Stuart Morgan < >
[15:03:14] stuartm: gigem: what did we decide on for the card/tuner term? Was it just 'input'?
[15:23:10] MythBuild: build #1319 of master-win8-msvc-2010–32bit is complete: Success [3build successful] Build details are at . . . /builds/1319
[15:41:04] gigem: stuartm: Input or Recording Input if more context is needed.
[16:06:00] superm1: tgm4883: right now i don't think it will matter that much, particularly for -fixes
[16:06:15] superm1: probably will matter more if someone put a change for ABI in master
[16:18:27] tgm4883: superm1: I made it just look all builds being successful
[16:18:49] tgm4883: I'll revisit it if it becomes an issue
[16:18:52] superm1: okay
[16:19:03] superm1: is everything building succesffully on arm then?
[16:20:02] tgm4883: yes and no. Everything builds on ARM, but it's not 100% successful 100% of the time
[16:20:07] tgm4883: due to the builders
[16:20:18] dekarl: note to self, download like this to avoid the MYTHCONFDIR fun: sudo su – mythtv -c mythfilldatabase
[16:23:17] dekarl: stuartm, are the "next 5 conflicts" of a recording rule ment to show any conflicts or just conflicts that are related to the recording rule?
[16:35:23] stuartm: dekarl: just conflicts relating to that rule
[16:35:43] superm1: tgm4883: huh well that''s not a good solution then i don't think
[16:35:57] superm1: if the builders fail sometimes it would be better to still promote 32/64 bit intel builds
[16:36:18] stuarta: the first build died on a segfault in qemu
[16:36:24] dekarl: stuartm, then its not working :)
[16:36:38] stuarta: feature!
[16:37:03] tgm4883: superm1: that gets a bit sticky. Technically, I'm not moving binaries, I move the source package and any built packages against it
[16:37:41] stuartm: dekarl: oh? it _was_ working when I wrote it, but I had to force a deliberate conflict to test, don't see any in real world usage as I've got enough tuners and there are enough repeats of everything
[16:46:10] stuartm: dekarl: it should be any conflicts of that rule, not conflicts caused by that rule
[16:48:28] dekarl: stuartm, basically "this stuff is in the way"?
[16:48:52] stuarta: hmmm nice, oneall are prepared to sponsor us on their pro plan for a year
[17:04:13] stuartm: dekarl: no the opposite – the following programmes matching this rule WON'T be recorded
[17:08:01] dekarl: . . . eaa11700493b
[17:08:58] dekarl: looks like the filter didn't make it into . . . dvr.cpp#L625
[17:16:51] skd5aner (skd5aner! has quit (Read error: Connection reset by peer)
[17:17:05] skd5aner (skd5aner! has joined #mythtv
[17:48:07] gary_buhrmaster: stuarta: Re: oneall. And after the year? (or, in a different vernacular: "the first hit is free"?) Yeah, I am the suspicious type.
[18:43:50] stuartm: dekarl: good catch
[18:52:43] dekarl: stuartm: ty
[18:53:50] dekarl: building now
[18:54:25] stuartm: well we've fixed a few WebFrontend related issues now, so that's good
[19:20:58] superm1: tgm4883: oh i see, that's annoying then
[19:21:34] tgm4883: superm1: I'm going to watch it for a few days and if it becomes a problem then we'll try and fix it
[19:22:27] tgm4883: the failures are definitely random though. They failed yesterday when I went to build them but succeeded during the daily build time slot. No changes were made to mythtv code nor packaging code during that span
[19:25:26] superm1: so stuff like that how will we get around?
[19:25:37] natanojl (natanojl!~jonatan@mythtv/developer/natanojl) has joined #mythtv
[19:29:32] tgm4883: well, we'd need to check that the i386 and amd64 builds are both done successfully. Once we know those are done successfully, check armhf that it's not in a "building" or "waiting to build" state (not sure those are the correct names). If armhf is in one of those states, check back later. If it's not, then copy the packages to release PPA, but doing this
[19:29:33] tgm4883: means that if the armhf package failed to build then ARM users could get some breakage here
[19:30:04] tgm4883: I had a prototype of that check working yesterday, but I hadn't coded the armhf check yet
[19:50:16] superm1: oh because it moves armhf too
[19:50:17] superm1: dang
[19:52:39] tgm4883: superm1: well not so much because it moves armhf too (users still wouldn't be able to download it), but more because libmyth (and a few others) build with the i386 packaging and are available for all arches
[20:28:05] dekarl: oh my, POTM is a InternetTV player with 100+ channel database... but without source code or database easily to find. I think we can do better ;)
[20:28:40] dekarl: been thinking about a channel database that connects tuning parameters (stream urls) with icons and guide data on and off
[20:34:54] stuartm: dekarl: conflict list look any better now?
[20:35:39] dekarl: aye
[20:38:02] dekarl: does not show conflicts on program without conflicts, and does show the expected conflicts for programs with conflicts
[20:38:58] dekarl: a bit confusing to see the conflict twice, but thats something for another day
[20:39:14] dekarl: (conflicts can also be next showings)
[20:39:48] dekarl: and due to the translation all (will record, other time, other channel, conflict) all look the same IIRC
[20:43:00] dekarl: after a restart the frontend list is empty. starting a frontend then reloading the page makes it appear like before
[20:44:22] dekarl: the play recording on frontend x isn't showing
[20:47:02] gregL (gregL! has quit (Remote host closed the connection)
[20:48:13] stuartm: dekarl: yeah, the frontend list isn't currently persistent between backend restarts, and I haven't yet added the 'magic' to add frontends to the list when they come online for the first time after a backend restart
[20:49:29] stuartm: the latter is just a matter of client side javascript, the former is a more general issue at the backend, need to add a table to store a list of clients/slaves etc
[20:49:39] gregL (gregL! has joined #mythtv
[20:50:42] stuartm: I wanted to do that for a while, since it makes clearing up old settings/records easier – if you no longer have a frontend, then deleting it through a ui will remove all the associated settings/config
[20:52:17] Steve-Goodey (Steve-Goodey! has quit (Ping timeout: 246 seconds)
[20:58:16] stuartm: I could just refresh the page when new frontends appear, that's the easiest way to do it and good enough in the short term – modern browsers can detect only the bits that actually changed so you don't see the transition like you used to do
[21:07:46] stuarta: gary_buhrmaster: you aren't the only one.
[21:28:56] Tobbe5178 (Tobbe5178!~asdf@2001:2002:d9d4:ce7e:5c87:3658:281e:a35d) has quit (Read error: Connection reset by peer)
[21:32:36] stuartm: particularly when it's a product or service that once your setup to use, is extremely disruptive to cancel – telling our users that they can't access their forum/wiki accounts because we used a third party to process the login ...
[21:38:00] stuarta: yep, time to go back to exploring simple oauth solutions. mediawiki has one already, so if we can get that for the forum, ppl can use the same account against both
[21:38:52] Roklobotomy (Roklobotomy! has quit (Quit: KVIrc 4.2.0 Equilibrium
[21:48:30] Jordack (Jordack! has quit (Quit: Did you hear about the kidnapping at school? Turned out to be ok, he woke up)
[21:52:39] stuartm: "2015-02–04 14:18:15.792027 W ChannelBase[3]: Setting start channel 'Please add' failed, selected to 'Please add' on input 'MPEG2TS' instead."
[21:54:43] stuarta: bit odd
[21:55:28] stuartm: starting channel isn't validated apparently, doesn't even seem to restrict it to an existing channel number
[21:56:27] stuartm: because of the whole ATSC subchannel nonsense we allow string characters in the field, but they aren't limited to '_' or '.'
[21:58:22] stuartm: . . . ce.cpp#L2884
[22:08:00] brfransen (brfransen! has quit (Read error: Connection reset by peer)
[22:08:46] brfransen (brfransen! has joined #mythtv
[22:18:12] stuartm: unclear how the label ends up in the database, or even whether it's something that happens on all setups or only this one user's
[22:20:00] stuartm: value = (value.isEmpty()) ? label : value;
[22:20:07] stuartm: well that's dumb
[22:36:28] Ahmuck (Ahmuck!~chatzilla@ has joined #mythtv
[22:36:45] Ahmuck (Ahmuck!~chatzilla@ has left #mythtv ()
[23:03:34] sl1ce (sl1ce! has quit (Quit: Konversation terminated!)
[23:03:48] sl1ce (sl1ce! has joined #mythtv
[23:10:41] natanojl (natanojl!~jonatan@mythtv/developer/natanojl) has quit (Ping timeout: 250 seconds)
[23:37:23] andreaz (andreaz! has joined #mythtv
[23:58:21] joki (joki! has quit (Ping timeout: 256 seconds)

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