| Tuesday, June 3rd, 2014, 00:27 UTC | ||
| [00:27:31] | wahrhaft (wahrhaft!~quassel@cpe-76-181-2-191.columbus.res.rr.com) has quit (Ping timeout: 240 seconds) | |
| [01:22:38] | jya_ (jya_!~jyavenard@mythtv/developer/jya) has quit (Quit: jya_) | |
| [01:24:32] | jya_ (jya_!~jyavenard@mythtv/developer/jya) has joined #mythtv | |
| [01:52:10] | dblain: | stuartm: no deserializers in git yet. I was starting to add complex parameters which would require the deserializers, but its been such a long time since I looked at it, I would most likely need to start over. |
| [01:54:38] | jya_ (jya_!~jyavenard@mythtv/developer/jya) has quit (Ping timeout: 240 seconds) | |
| [01:56:18] | jya_ (jya_!~jyavenard@mythtv/developer/jya) has joined #mythtv | |
| [02:00:04] | brfransen (brfransen!~brfransen@64.179.169.226) has quit (Ping timeout: 276 seconds) | |
| [02:00:47] | brfransen (brfransen!~brfransen@64.179.169.226) has joined #mythtv | |
| [02:28:50] | peper03 (peper03!~peper03@mythtv/developer/peper03) has quit (Ping timeout: 252 seconds) | |
| [02:29:52] | peper03 (peper03!~peper03@mythtv/developer/peper03) has joined #mythtv | |
| [02:30:50] | unforgiven512 (unforgiven512!~unforgive@oxymorphone.unforgivendevelopment.com) has quit (Quit: ZNC - http://znc.in) | |
| [02:34:08] | unforgiven512 (unforgiven512!~unforgive@oxymorphone.unforgivendevelopment.com) has joined #mythtv | |
| [02:43:38] | ghoti (ghoti!~paul@scratch.it.ca) has quit (Ping timeout: 240 seconds) | |
| [03:16:05] | wahrhaft (wahrhaft!~quassel@cpe-76-181-12-141.columbus.res.rr.com) has joined #mythtv | |
| [03:42:08] | fetzerch (fetzerch!~quassel@unaffiliated/fetzerch) has quit (Ping timeout: 252 seconds) | |
| [03:43:25] | fetzerch (fetzerch!~quassel@unaffiliated/fetzerch) has joined #mythtv | |
| [03:52:24] | arescorpio (arescorpio!~arescorpi@105-24-245-190.fibertel.com.ar) has quit (Excess Flood) | |
| [03:58:03] | jya_: | hmmm… my vdpau fix should have gone into devel/027candidates… |
| [04:31:27] | kc (kc!~Casper@unaffiliated/kc) has joined #mythtv | |
| [06:11:49] | FabriceMG (FabriceMG!~Thunderbi@217.112.59.207) has joined #mythtv | |
| [06:29:25] | ghoti (ghoti!~paul@scratch.it.ca) has joined #mythtv | |
| [06:34:06] | rsiebert (rsiebert!~quassel@f052000195.adsl.alicedsl.de) has joined #mythtv | |
| [06:36:57] | rsiebert_ (rsiebert_!~quassel@g225052152.adsl.alicedsl.de) has quit (Ping timeout: 252 seconds) | |
| [06:38:29] | SteveGoodey (SteveGoodey!~steve@host86-134-63-27.range86-134.btcentralplus.com) has joined #mythtv | |
| [07:49:53] | ghoti (ghoti!~paul@scratch.it.ca) has quit (Ping timeout: 255 seconds) | |
| [07:55:42] | stuarta: | morning all |
| [08:12:28] | stuartm: | dblain: ok thanks, looking to pass a QVariantMap between the fe and be, using the existing json serializer makes most sense so I'll have to write a deserializer for that type |
| [08:20:36] | Merlin83b (Merlin83b!~Daniel@office.34sp.com) has joined #mythtv | |
| [08:44:44] | jya: | Feeling like tagging a 0.27.2 with the new translations in, and the fixes I believe makes it important |
| [08:45:16] | jya: | though the playback issues I’m seeing are in the latest XBMC, and no one seems to report the problem |
| [08:46:06] | jya: | need to make the announcement for 0.27.1 first :) |
| [08:46:30] | jya: | I like the natural sort I added to master… really keen on backporting that too |
| [08:47:52] | jya: | stuartm: do you know how to push the changes to the main web site? is it just a matter of logging in and running hte publish script? |
| [08:53:13] | stuartm: | jya: you need to clone the www repo into your home directory and then run the script, probably with sudo, I'm happy to do that bit for you |
| [08:53:36] | stuartm: | this is one reason why I'd like to get it set up to run as a post-commit hook |
| [08:57:41] | jya: | stuartm: where do you clone, on alcor or home ? |
| [08:57:49] | jya: | stuarta: gave me sudo access there |
| [08:58:15] | stuartm: | alcor |
| [08:58:20] | stuarta: | stuartm: i have some plans for that |
| [08:58:48] | stuartm: | publish script doesn't have any connectivity, it needs to be run on the server |
| [08:59:09] | stuartm: | all it does is copy the files to the correct location and restart the httpd server |
| [08:59:29] | stuarta: | stuartm: i'll tell you what I had in mind |
| [08:59:44] | stuarta: | as i need to reset up the www repo's anyway |
| [08:59:48] | jya: | to add a new news, just need to add a NNN.php ? |
| [09:00:38] | stuartm: | jya: you can, but the recommended way is to run the newpost.pl script which creates the template for you |
| [09:00:46] | stuarta: | make it so we have *.staging.mythtv.org and different branches for live and staging, so you can push your changes first to check them, and then push them to live |
| [09:01:41] | jya: | easier for me to create the file here. |
| [09:01:54] | stuartm: | it fills out the metadata for the news post and then opens an editor for you to write in the actual post – you still need to add an html markup e.g. <p></p> around paragraphs |
| [09:02:36] | stuartm: | stuarta: cool |
| [09:02:49] | jya: | does the publish script push the data to the git server , or I can make my changes locally, push and then run publish ? |
| [09:03:14] | stuartm: | jya: push first, then run publish on alcor |
| [09:03:30] | stuartm: | publish script doesn't commit or push changes |
| [09:04:22] | stuartm: | which is why it really should be run as a post-commit hook on alcor, you just make the changes locally, commit, push and they are automatically published |
| [09:04:26] | jya: | ah so I have to reclone on alcor ( or pull) after pushing my changes |
| [09:04:36] | stuartm: | yup |
| [09:04:53] | jya: | and the last news creates automatically appear on the main page? |
| [09:05:11] | stuartm: | once published, yes |
| [09:05:35] | stuartm: | the NNN.php are loaded automatically by php on the server |
| [09:06:05] | stuarta: | it's a bit odd the way that all works |
| [09:15:37] | stuartm: | it's a bit better than what we had for several years which was a crippled install of phpNuke that only a couple of people could access, crippled because the install got hacked and dramatic changes were made to prevent that happening again |
| [09:16:39] | stuartm: | I don't doubt that it could be improved, especially the news posting bit, but auto-running the publish script would be a start |
| [09:17:10] | stuartm: | possibly auto-commiting/pushing from newpost.pl too |
| [09:17:38] | stuarta: | makes more sense to do it in the post commit |
| [09:18:40] | stuarta: | that way everyone can make their updates locally and just push and it's done |
| [09:18:50] | stuartm: | newpost.pl should also do a pull --rebase to ensure that everything is up to date before it creates the news posting |
| [09:19:34] | stuartm: | that would cut down the number of steps required :) |
| [09:20:17] | stuarta: | yep, i want it simple to use for all the devs, jya's recent experience proves just how crap this is at the moment |
| [09:20:46] | jya: | i’m still calling to use wordpress. |
| [09:20:54] | jya: | it would have taken 2 minutes to do all of this |
| [09:21:11] | stuartm: | we should sort out the perms so that most people have access, at least to post news and edit webpages |
| [09:21:13] | jya: | and we could easily change the look and feel of everything and keep the same content |
| [09:23:24] | stuartm: | the only reason we didn't use something like wordpress before was the phpnuke experience, these big CMS systems are the focus for a lot of script kiddies and need continual security upgrades to avoid being hit in drive by attacks |
| [09:23:45] | stuarta: | wordpress at the time was especially vunerable |
| [09:24:19] | stuarta: | i'll investigate various different options and see what I come up with |
| [09:24:29] | jya: | I’m not sure that’s the main reason… When I see any attempt to discuss our web site ends up being arguments over the use of apache vs something else no-one ever heard of… We’re doomed to have a crap web site |
| [09:24:49] | stuartm: | xris wanted something that was a lot more secure, and ssh keys are about as secure as it gets without two factor authentication with one-time passwords generators ;) |
| [09:25:11] | stuarta: | static content! |
| [09:25:27] | jya: | stuartm: you can perfectly have ssl keys to authenticate with wordpress |
| [09:25:51] | stuartm: | jya: there was no discussion (well not much) the last time the website was redesigned because we were still under the benign dictatorship, democracy brought chaos :( |
| [09:25:51] | jya: | there are millions of sites using wordpress these days.. but somehow it’s not suitable for us… go figure |
| [09:26:29] | stuartm: | jya: it was riddled with security holes at the time, site after site running wordpress being hacked |
| [09:26:41] | stuarta: | jya: what i'm trying to say is it was a decision that was made several years back. we are now in a position to review that, and the webserver being used is completely independent of that |
| [09:26:46] | jya: | stuartm: I clearly recall when I had someone to go over our web site architecture, define on how things could be organised… it again finished in apache vs other, wordpress blah blah |
| [09:26:58] | stuartm: | it may have changed since, but we're talking about a few years ago now |
| [09:27:49] | jya: | it’s one of the many things were no one appear to be able to make a decision, and so nothing is done and one year later, when the issue is raised again we find ourselves exactly at the same point |
| [09:28:08] | jya: | and that’s not just for our web site I’m afraid |
| [09:28:10] | stuartm: | jya: there was a much more recent series of arguments around the time we switched to git, xris was looking for alternatives to trac and that turned into a free for all on all the server software |
| [09:28:38] | stuarta: | jya: i'm going to take the lead on working out where to go with all this. the whole lot needs a broom putting through it |
| [09:29:00] | jya: | this morning, i assigned myself a few trac components as no-one appear to look over bugs or the owner of the bugs were no longer active. |
| [09:29:00] | ** stuarta seems to have inherited primary sysadmin role ** | |
| [09:29:17] | jya: | I couldnt’ something as simlpe as select all the components I wanted and assigned them to me |
| [09:29:40] | stuartm: | jya: no-one is saying we shouldn't use wordpress now, BUT it's not a 5 minute job, we'd need to theme it etc and that takes a lot of time away from development |
| [09:29:54] | jya: | stuarta: no offence but you appear to be a bit in dark age when it comes to server choices :) |
| [09:30:10] | stuartm: | ... |
| [09:30:13] | stuarta: | jya: no offense, but apache is the dark ages ;-P |
| [09:31:39] | stuarta: | interesting. http://news.netcraft.com/archives/2014/02/03/ . . . -survey.html |
| [09:31:40] | jya: | i guess I’m just venting my frustration that it’s been a full week, and still no announcement. Sure I may be responsible, but that no one else seem to care enough to take over the task is very underwhelmong |
| [09:31:56] | stuarta: | "Microsoft gained a staggering 48 million sites this month, increasing its total by 19% — most of this growth is attributable to new sites hosted by Nobis Technology Group. Along with Microsoft, nginx also made a large gain of 14 million sites, whereas Apache fell by 7 million." |
| [09:32:42] | jya: | stuarta: probably all web servers running at microsoft head office to twist the stats :) |
| [09:32:50] | stuarta: | :) |
| [09:33:17] | jya: | simply look at the active sites share... |
| [09:33:20] | stuartm: | jya: no-one likes writing news announcements, you're not the only one |
| [09:33:28] | jya: | how could there be such discrepency? |
| [09:33:32] | stuarta: | jya: i promise to look at wordpress again |
| [09:34:00] | stuartm: | fwiw I've never felt that there is a particular rush to publish a news item following a point release, it's not going anywhere after all :) |
| [09:34:08] | jya: | what I like with wordpress is how easy it is to skin. and you can get a custom skin made for like $50 |
| [09:34:56] | jya: | stuartm: I guess I feel differently because I’ve put so many hours in that code |
| [09:35:38] | ** stuarta bribes stuartm with coffee to get the release announcement out to keep jya happy ** | |
| [09:35:58] | jya: | too late. I’m done , just need to publish :) |
| [09:36:59] | jya: | what’s the difference between a “news†and an announcement? |
| [09:37:13] | jya: | just the $topic value? |
| [09:37:35] | jya: | and just when the announcement of 0.27.1 is made, here comes 0.27.2 :) |
| [09:38:25] | stuartm: | jya: I could have written something, but it probably wouldn't have sold the changes that you consider to be most important simply because I've been too busy doing other things to keep track :) I hate writing those announcements, I hate writing in general, it's not my thing |
| [09:38:53] | jya: | stuartm: I’m surprised you had a crash due to the ffmpeg upgrade. in theory, the binaries should have been backward compatible. I’ve taken great care to not change any of the APIs in libmyth*. I did add a new functions here and there, but the ABI should have remained compatible |
| [09:39:27] | stuarta: | our website sucks |
| [09:39:31] | stuartm: | jya: plugins used the ffmpeg API so they needed a binary change |
| [09:39:52] | jya: | stuartm: they sure do, but ffmpeg add new functions, they don’t change the existing ones |
| [09:40:04] | jya: | and all the functions I added were private |
| [09:40:25] | stuartm: | jya: difficult to say, releases are definitely announcements, news for general stuff e.g. "MythTV is used by NASA!" |
| [09:40:44] | jya: | let’s make news... |
| [09:40:55] | stuarta: | it appears to change the side label |
| [09:40:56] | jya: | after all, today is a new day |
| [09:41:02] | stuartm: | stuarta: it does |
| [09:42:45] | stuartm: | btw, if we do redesign the look of the website, it would be nice to keep the general colour scheme otherwise the colour coordinated Forum and WebFrontend won't match (lot of work went into theming both of those recently) |
| [09:43:03] | jya: | stuarta: can I please ask a favour? can we make forum.mythtv.org plural? |
| [09:43:08] | jya: | it’s driving me nuts |
| [09:43:44] | jya: | i can’t think of a single web site, using forum vs forums |
| [09:44:21] | stuarta: | i'll get the process started |
| [09:44:40] | stuarta: | but we'll keep forum, and just make forums -> forum |
| [09:45:44] | stuartm: | jya: Any idea how difficult this would be? https://forum.mythtv.org/viewtopic.php?f=9&t=177 |
| [09:46:50] | jya: | stuartm: I’ve been thinking of having our own audio server running, similar to what xbmc does know |
| [09:46:58] | jya: | e.g. your audio card is always open. |
| [09:47:11] | stuartm: | makes sense |
| [09:47:26] | jya: | that allows to have sound effect, and prevent the cut in audio with amp that takes a while to get a carrier |
| [09:47:29] | stuartm: | I've long wanted the ability to dial down the volume before starting playback |
| [09:47:39] | jya: | my SP32 amp can take up to 7s to get a lock |
| [09:47:45] | jya: | very annoying |
| [09:48:14] | jya: | the volume itself is controlled completely outside the audio class |
| [09:48:27] | jya: | there’s no reason why you can’t change the volume without playback going |
| [09:49:06] | jya: | just a matter of mapping the feature to keys… it’s a database setting which is read when audio is opened |
| [09:49:43] | stuarta: | jya: do you just normally type forums.mythtv.org ? and hit it via http not https? |
| [09:50:23] | jya: | I type forums.mythtv.org |
| [09:50:26] | jya: | no http or https |
| [09:50:39] | jya: | and I let it do its thing |
| [09:50:44] | stuartm: | so http |
| [09:50:52] | stuarta: | okay cool. that'll hit http which will hit the redirect i just put in. now to request dns update |
| [09:51:00] | jya: | and as I’ve attempted to open forums.mythtv.org in the past |
| [09:51:26] | jya: | when I type forum it automatically gets auto-completed to forums.mythtv.org which leads to an error and point me to my ISP search page |
| [09:51:49] | jya: | so I can’t simply edit the URL, as it has now become something like: bigpond.com/blah |
| [09:52:57] | stuartm: | ouch, typo hijacking sucks |
| [09:54:18] | stuarta: | dns update request sent |
| [09:56:14] | jya: | like on www.mythtv.org, you hover on Support, it says “Forums†and it takes you to: ….. forum |
| [09:56:27] | jya: | my OCD is going into overdrive there |
| [09:58:54] | jya: | you got to be kidding me |
| [09:58:58] | jya: | git push -> ! [remote rejected] master -> master (n/a (unpacker error)) |
| [10:03:14] | stuartm: | wtf does that even mean? |
| [10:03:52] | stuartm: | http://blog.shamess.info/2011/05/06/remote-re . . . acker-error/ |
| [10:03:57] | stuartm: | stuarta: ^^ |
| [10:04:08] | jya: | I could use publish |
| [10:04:18] | jya: | but I couldn’t push from my home computer back to code |
| [10:04:21] | ** stuarta fiddles ** | |
| [10:05:04] | jya: | http://www.mythtv.org/ yeah |
| [10:05:11] | jya: | I’m “Jean Yves" |
| [10:06:41] | ** jya feel so happy ** | |
| [10:07:08] | jya: | allright.. let’s see why using vdpau old api with 2.2 now causes a segfault after a couple of playback |
| [10:08:00] | stuartm: | jya: you can edit the name that's displayed in includes/conf.php |
| [10:08:06] | jya: | yeah, just did it |
| [10:08:27] | stuarta: | jya: try again? |
| [10:08:39] | jya: | pushing ? I have nothing to push now :) |
| [10:08:54] | jya: | or I could do a 0.27.2 for the sake of it |
| [10:09:16] | jya: | I;m keen on getting mythbuntu having a bug-free release, or as close as possible |
| [10:10:46] | stuartm: | hmm, I'm not getting any of your changes when I pulled just now |
| [10:13:13] | stuartm: | stuarta: is it possible those changes have disappeared into the ether? |
| [10:15:31] | andreaz (andreaz!~Andreaz@tmo-106-161.customers.d1-online.com) has joined #mythtv | |
| [10:16:35] | stuarta: | erm shouldn't have. they should still be in jya's repo, as the push failed |
| [10:16:58] | jya: | stuarta: the push succeeded when I used sudo and the publish script |
| [10:17:08] | jya: | I manually copied my changes I had made locally to alcor |
| [10:17:24] | jya: | i may not have pushed. |
| [10:17:29] | jya: | but I certainly used publish |
| [10:17:49] | jya: | ah ok.. I hadn’t pushed |
| [10:17:53] | jya: | so there it is |
| [10:20:37] | stuarta: | hmmm weird, i don't het them either when i pull |
| [10:21:12] | stuartm: | I do now |
| [10:21:19] | jya: | I did |
| [10:21:28] | stuartm: | and I've just pushed a change to robots.txt |
| [10:21:28] | jya: | just pulled on my home machine |
| [10:21:37] | jya: | what did that git error mean? |
| [10:22:12] | stuartm: | jya: perms on the actual directory were wrong |
| [10:22:23] | stuartm: | http://blog.shamess.info/2011/05/06/remote-re . . . acker-error/ |
| [10:23:34] | stuartm: | gigem: are you working on allowing the use of recording id in services, e.g. as an argument to Dvr/GetRecorded() etc? |
| [10:23:35] | stuarta: | there must have been 1 thing which didn't allow group write, which i fixed |
| [10:26:28] | ** jya love obvious error messages ** | |
| [10:34:17] | jya_: | myth:// is always on the 6543 port? never 6544 ? |
| [10:42:28] | Steve-Goodey (Steve-Goodey!~steve@host86-134-63-27.range86-134.btcentralplus.com) has joined #mythtv | |
| [10:44:49] | jheizer (jheizer!~jheizer@73.51.93.177) has quit (Read error: Connection reset by peer) | |
| [10:45:13] | jheizer (jheizer!~jheizer@73.51.93.177) has joined #mythtv | |
| [10:50:28] | ghoti (ghoti!~paul@scratch.it.ca) has joined #mythtv | |
| [10:56:11] | ghoti (ghoti!~paul@scratch.it.ca) has quit (Ping timeout: 255 seconds) | |
| [10:56:51] | ghoti (ghoti!~paul@scratch.it.ca) has joined #mythtv | |
| [10:58:28] | Steve-Goodey (Steve-Goodey!~steve@host86-134-63-27.range86-134.btcentralplus.com) has quit (Quit: Konversation terminated!) | |
| [11:01:38] | jya_ (jya_!~jyavenard@mythtv/developer/jya) has quit (Quit: jya_) | |
| [11:02:07] | ghoti (ghoti!~paul@scratch.it.ca) has quit (Ping timeout: 260 seconds) | |
| [11:02:48] | ghoti (ghoti!~paul@scratch.it.ca) has joined #mythtv | |
| [11:07:46] | ghoti (ghoti!~paul@scratch.it.ca) has quit (Ping timeout: 240 seconds) | |
| [11:08:49] | ghoti (ghoti!~paul@scratch.it.ca) has joined #mythtv | |
| [11:13:38] | ghoti (ghoti!~paul@scratch.it.ca) has quit (Ping timeout: 240 seconds) | |
| [11:15:04] | ghoti (ghoti!~paul@scratch.it.ca) has joined #mythtv | |
| [11:20:10] | ghoti (ghoti!~paul@scratch.it.ca) has quit (Ping timeout: 240 seconds) | |
| [11:21:03] | ghoti (ghoti!~paul@scratch.it.ca) has joined #mythtv | |
| [11:25:58] | ghoti (ghoti!~paul@scratch.it.ca) has quit (Ping timeout: 240 seconds) | |
| [11:27:19] | ghoti (ghoti!~paul@scratch.it.ca) has joined #mythtv | |
| [11:27:23] | stuartm: | yes |
| [11:32:29] | ghoti (ghoti!~paul@scratch.it.ca) has quit (Ping timeout: 252 seconds) | |
| [11:33:45] | ghoti (ghoti!~paul@scratch.it.ca) has joined #mythtv | |
| [11:35:36] | jya_ (jya_!~jyavenard@mythtv/developer/jya) has joined #mythtv | |
| [11:38:38] | ghoti (ghoti!~paul@scratch.it.ca) has quit (Ping timeout: 245 seconds) | |
| [11:40:09] | ghoti (ghoti!~paul@scratch.it.ca) has joined #mythtv | |
| [11:45:05] | ghoti (ghoti!~paul@scratch.it.ca) has quit (Ping timeout: 265 seconds) | |
| [11:46:36] | ghoti (ghoti!~paul@scratch.it.ca) has joined #mythtv | |
| [11:51:59] | ghoti (ghoti!~paul@scratch.it.ca) has quit (Ping timeout: 255 seconds) | |
| [11:52:38] | ghoti (ghoti!~paul@scratch.it.ca) has joined #mythtv | |
| [11:57:46] | ghoti (ghoti!~paul@scratch.it.ca) has quit (Ping timeout: 240 seconds) | |
| [11:58:42] | ghoti (ghoti!~paul@scratch.it.ca) has joined #mythtv | |
| [12:03:29] | skd5aner (skd5aner!~skd5aner@50-90-5-146.res.bhn.net) has quit (Quit: Leaving) | |
| [12:04:11] | ghoti (ghoti!~paul@scratch.it.ca) has quit (Ping timeout: 260 seconds) | |
| [12:05:11] | ghoti (ghoti!~paul@scratch.it.ca) has joined #mythtv | |
| [12:10:42] | ghoti (ghoti!~paul@scratch.it.ca) has quit (Ping timeout: 265 seconds) | |
| [12:11:05] | ghoti (ghoti!~paul@scratch.it.ca) has joined #mythtv | |
| [12:16:14] | ghoti (ghoti!~paul@scratch.it.ca) has quit (Ping timeout: 252 seconds) | |
| [12:17:11] | ghoti (ghoti!~paul@scratch.it.ca) has joined #mythtv | |
| [12:21:58] | ghoti (ghoti!~paul@scratch.it.ca) has quit (Ping timeout: 240 seconds) | |
| [12:23:16] | ghoti (ghoti!~paul@scratch.it.ca) has joined #mythtv | |
| [12:28:10] | ghoti (ghoti!~paul@scratch.it.ca) has quit (Ping timeout: 240 seconds) | |
| [12:29:10] | ghoti (ghoti!~paul@scratch.it.ca) has joined #mythtv | |
| [12:34:27] | ghoti (ghoti!~paul@scratch.it.ca) has quit (Ping timeout: 252 seconds) | |
| [12:35:04] | ghoti (ghoti!~paul@scratch.it.ca) has joined #mythtv | |
| [12:40:04] | ghoti (ghoti!~paul@scratch.it.ca) has quit (Ping timeout: 252 seconds) | |
| [12:41:10] | ghoti (ghoti!~paul@scratch.it.ca) has joined #mythtv | |
| [12:45:59] | ghoti (ghoti!~paul@scratch.it.ca) has quit (Ping timeout: 255 seconds) | |
| [12:47:06] | ghoti (ghoti!~paul@scratch.it.ca) has joined #mythtv | |
| [12:52:03] | ghoti (ghoti!~paul@scratch.it.ca) has quit (Ping timeout: 252 seconds) | |
| [12:53:13] | ghoti (ghoti!~paul@scratch.it.ca) has joined #mythtv | |
| [12:59:05] | ghoti (ghoti!~paul@scratch.it.ca) has quit (Remote host closed the connection) | |
| [12:59:11] | ghoti (ghoti!~paul@scratch.it.ca) has joined #mythtv | |
| [13:05:00] | ghoti (ghoti!~paul@scratch.it.ca) has quit (Remote host closed the connection) | |
| [13:05:07] | ghoti (ghoti!~paul@scratch.it.ca) has joined #mythtv | |
| [13:10:17] | ghoti (ghoti!~paul@scratch.it.ca) has quit (Ping timeout: 255 seconds) | |
| [13:11:05] | ghoti (ghoti!~paul@scratch.it.ca) has joined #mythtv | |
| [13:16:10] | ghoti (ghoti!~paul@scratch.it.ca) has quit (Ping timeout: 240 seconds) | |
| [13:17:02] | ghoti (ghoti!~paul@scratch.it.ca) has joined #mythtv | |
| [13:22:07] | ghoti (ghoti!~paul@scratch.it.ca) has quit (Ping timeout: 252 seconds) | |
| [13:23:00] | ghoti (ghoti!~paul@scratch.it.ca) has joined #mythtv | |
| [13:28:00] | andreaz1 (andreaz1!~Andreaz@tmo-106-161.customers.d1-online.com) has joined #mythtv | |
| [13:28:13] | ghoti (ghoti!~paul@scratch.it.ca) has quit (Ping timeout: 245 seconds) | |
| [13:29:00] | ghoti (ghoti!~paul@scratch.it.ca) has joined #mythtv | |
| [13:31:44] | andreaz (andreaz!~Andreaz@tmo-106-161.customers.d1-online.com) has quit (Ping timeout: 252 seconds) | |
| [13:33:58] | ghoti (ghoti!~paul@scratch.it.ca) has quit (Ping timeout: 240 seconds) | |
| [13:34:56] | ghoti (ghoti!~paul@scratch.it.ca) has joined #mythtv | |
| [13:35:18] | joki (joki!~joki@p54860639.dip0.t-ipconnect.de) has quit (Ping timeout: 240 seconds) | |
| [13:40:29] | ghoti (ghoti!~paul@scratch.it.ca) has quit (Ping timeout: 264 seconds) | |
| [13:40:58] | ghoti (ghoti!~paul@scratch.it.ca) has joined #mythtv | |
| [13:41:33] | stichnot (stichnot!~chatzilla@adsl-68-125-53-82.dsl.pltn13.pacbell.net) has quit (Ping timeout: 252 seconds) | |
| [13:42:22] | joki- (joki-!~joki@p5486223F.dip0.t-ipconnect.de) has joined #mythtv | |
| [13:46:04] | ghoti (ghoti!~paul@scratch.it.ca) has quit (Ping timeout: 252 seconds) | |
| [13:46:57] | ghoti (ghoti!~paul@scratch.it.ca) has joined #mythtv | |
| [13:52:08] | ghoti (ghoti!~paul@scratch.it.ca) has quit (Ping timeout: 255 seconds) | |
| [13:53:00] | ghoti (ghoti!~paul@scratch.it.ca) has joined #mythtv | |
| [13:55:49] | jya_: | i was looking at what was being sent over the network, and it seems that everytime a QUERY_SG_GETFILELIST is sent, the first later of the path is dropped. |
| [13:55:58] | jya_: | e.g. my SG location is /data/videos/ |
| [13:56:05] | jya_: | the path is /movies |
| [13:56:31] | jya_: | QUERY_SG_GETFILELIST first called for /data/videos//ovies/ |
| [13:57:01] | jya_: | somehow that still works bcause it default in searching differently if nothing is found the first time |
| [13:57:27] | jya_: | culprit is this line: QString subdir = folder_path.right(folder_path.length() – (prefix.length() + 2)); |
| [13:58:18] | jya_: | that’s for the cover image… maybe that’s why i don’t have all the cover images being displayed from time to time |
| [13:58:29] | ghoti (ghoti!~paul@scratch.it.ca) has quit (Ping timeout: 265 seconds) | |
| [13:59:00] | ghoti (ghoti!~paul@scratch.it.ca) has joined #mythtv | |
| [13:59:13] | jya_: | I love the comment there: “// TODO: This can likely get a little cleaner†|
| [14:03:48] | ghoti (ghoti!~paul@scratch.it.ca) has quit (Ping timeout: 240 seconds) | |
| [14:05:01] | ghoti (ghoti!~paul@scratch.it.ca) has joined #mythtv | |
| [14:05:08] | stuarta: | crap that's horrible |
| [14:09:48] | ghoti (ghoti!~paul@scratch.it.ca) has quit (Ping timeout: 240 seconds) | |
| [14:11:02] | ghoti (ghoti!~paul@scratch.it.ca) has joined #mythtv | |
| [14:14:54] | jya_: | that line means that it will never find any images found in a folder to be used a cover image |
| [14:15:48] | ghoti (ghoti!~paul@scratch.it.ca) has quit (Ping timeout: 240 seconds) | |
| [14:17:03] | ghoti (ghoti!~paul@scratch.it.ca) has joined #mythtv | |
| [14:20:20] | ** jya_ wonders how often one can do copy/paste of a multi-pages code ** | |
| [14:21:58] | ghoti (ghoti!~paul@scratch.it.ca) has quit (Ping timeout: 245 seconds) | |
| [14:22:46] | jya_: | what decides when a folder will contain the cover image of a video located inside of it? |
| [14:22:56] | ghoti (ghoti!~paul@scratch.it.ca) has joined #mythtv | |
| [14:22:59] | jya_: | I have two folders here. both with videos in it, with fanart. |
| [14:23:12] | ** stuarta offers jya the broom of code cleaning ** | |
| [14:23:24] | jya_: | being at the upper level, one shows the image of the first video in it |
| [14:23:29] | jya_: | the other one absolutely nothing |
| [14:23:51] | jya_: | stuarta: i only stumbled upon it as I was testing if my changes to enable IPv6 damaged something |
| [14:23:59] | jya_: | and those wrong paths caught my attention |
| [14:24:42] | jya_: | the funn thing, you can tell the entire code is a copy/paste of the other (including the comment about being a dirty code), yet one has the right string manipulation, and the other doesn't |
| [14:25:09] | jya_: | stuarta: code cleaning is pretty much all I’ve been doing for months… its getting tiring |
| [14:27:38] | ghoti (ghoti!~paul@scratch.it.ca) has quit (Ping timeout: 240 seconds) | |
| [14:28:53] | jya_: | why do we have to tell the backend whenever a popup is being created or dismissed ? |
| [14:28:58] | ghoti (ghoti!~paul@scratch.it.ca) has joined #mythtv | |
| [14:29:12] | jya_: | like whenever I press info on a video, a message is sent to the backend |
| [14:29:40] | stuarta: | doesn't the popup need data about the video which is held by the backend? |
| [14:30:20] | gigem: | stuarta: Yes, that's the type of chanid/recstartts => recordedid changes I intend to make. I haven't decided, though, how aggressively I want to attack them all at the expense of other things I'd like/need to to. If there's a particular change you'd like earlier rather than later, let me know. |
| [14:31:58] | jya_: | stuarta: that it requests the image to be displayed is one thing |
| [14:32:17] | stuarta: | stuartm: ^^^ |
| [14:32:50] | jya_: | but there’s an extra message: BACKEND_MESSAGE[]:[]SYSTEM_EVENT SCREEN_TYPE CREATED itemdetailpopup SENDER jyamba |
| [14:33:07] | jya_: | same when the screen is closed |
| [14:34:06] | ghoti (ghoti!~paul@scratch.it.ca) has quit (Ping timeout: 252 seconds) | |
| [14:34:51] | ghoti (ghoti!~paul@scratch.it.ca) has joined #mythtv | |
| [14:40:03] | ghoti (ghoti!~paul@scratch.it.ca) has quit (Ping timeout: 265 seconds) | |
| [14:40:20] | jya_: | surely there have got to better ways to find an image for a particular folder thant seinding a multitudes of QUERY_SG_GETFILELIST in a row, each with a different extension |
| [14:40:53] | ghoti (ghoti!~paul@scratch.it.ca) has joined #mythtv | |
| [14:41:59] | jya_: | being able to send a single request the backend to search for a file name with a regex ought to be faster |
| [14:45:43] | ghoti (ghoti!~paul@scratch.it.ca) has quit (Ping timeout: 245 seconds) | |
| [14:46:44] | ghoti (ghoti!~paul@scratch.it.ca) has joined #mythtv | |
| [14:49:22] | bas-t (bas-t!~tycho@52484E89.cm-4-1b.dynamic.ziggo.nl) has joined #mythtv | |
| [14:49:42] | stuarta: | especially given that the backend has the SG list for a given frontend available |
| [14:49:58] | ** stuarta hits his mac mini ** | |
| [14:50:53] | bas-t (bas-t!~tycho@52484E89.cm-4-1b.dynamic.ziggo.nl) has quit (Client Quit) | |
| [14:52:08] | ghoti (ghoti!~paul@scratch.it.ca) has quit (Ping timeout: 265 seconds) | |
| [14:52:46] | ghoti (ghoti!~paul@scratch.it.ca) has joined #mythtv | |
| [14:53:00] | jya_: | stuarta: the funny thing is that it will query the backend several times to retrieve a single file, each time only changing the extension |
| [14:53:17] | jya_: | and then it retrieves the entire files list to search with a regex in it |
| [14:53:42] | jya_: | and it retrieves the entire content each time for the different search type. |
| [14:54:07] | jya_: | so three different methods: three times retrieve the entire list |
| [14:54:17] | ** jya_ shaking head ** | |
| [14:55:40] | stuartm: | jya_: those messages are there to support remote control of the frontend and also to trigger system events defined by the user |
| [14:55:59] | stuartm: | and, no they weren't added by me, although I understand why they exist |
| [14:56:58] | stuartm: | gigem: I asked because if it wasn't something you were working on I'd be happy to start |
| [14:57:13] | jya_: | stuartm: i can understand why you would want to receive such information from the frontend (using like a subscribe event mechanism), sending it every time to the backend, no matter what seems highly inefficient to me |
| [14:57:51] | mythfan (mythfan!~mythfan@24.226.229.100) has joined #mythtv | |
| [14:58:07] | ghoti (ghoti!~paul@scratch.it.ca) has quit (Ping timeout: 276 seconds) | |
| [14:58:51] | ghoti (ghoti!~paul@scratch.it.ca) has joined #mythtv | |
| [14:59:23] | stuartm: | jya_: we'll I can't speak for the window open/close stuff, but I'm currently working on sending the frontend status info (see http://{frontendip}:6547/Frontend/GetStatus) to the backend whenever there is a change so that the backend can then include that information in the WebFrontend |
| [14:59:49] | stuartm: | note that with GetStatus it's really only supporting playback of videos/dvds atm |
| [15:00:47] | stuartm: | compared to the multitudes of data that get sent during playback of videos etc, those SYSTEM_EVENT messages are a drop in the ocean |
| [15:02:38] | stuartm: | not even as large as a tweet :) |
| [15:03:41] | ghoti (ghoti!~paul@scratch.it.ca) has quit (Ping timeout: 255 seconds) | |
| [15:04:30] | ghoti (ghoti!~paul@scratch.it.ca) has joined #mythtv | |
| [15:05:19] | dekarl (dekarl!~dekarl@p4FE84186.dip0.t-ipconnect.de) has joined #mythtv | |
| [15:06:06] | jya_: | I guess, when looking when how the artwork is retrieved for folders, that extra useless events was just the famous drop of water |
| [15:08:11] | mythfan (mythfan!~mythfan@24.226.229.100) has quit (Quit: Colloquy for iPad - http://colloquy.mobi) | |
| [15:09:53] | ghoti (ghoti!~paul@scratch.it.ca) has quit (Ping timeout: 264 seconds) | |
| [15:10:30] | ghoti (ghoti!~paul@scratch.it.ca) has joined #mythtv | |
| [15:15:47] | ghoti (ghoti!~paul@scratch.it.ca) has quit (Ping timeout: 252 seconds) | |
| [15:16:06] | stuartm: | yes, well obviously searching for a file in that manner is bad design, I was specifically referring to SYSTEM_EVENT messages and not any other bits of the protocol |
| [15:16:31] | ghoti (ghoti!~paul@scratch.it.ca) has joined #mythtv | |
| [15:21:53] | ghoti (ghoti!~paul@scratch.it.ca) has quit (Ping timeout: 264 seconds) | |
| [15:22:30] | ghoti (ghoti!~paul@scratch.it.ca) has joined #mythtv | |
| [15:27:31] | ghoti (ghoti!~paul@scratch.it.ca) has quit (Ping timeout: 252 seconds) | |
| [15:28:24] | stuartm: | not reassuring that the Sans Institute have a white paper titled "Cross-Sight Scripting Vulnerabilities" |
| [15:28:27] | ghoti (ghoti!~paul@scratch.it.ca) has joined #mythtv | |
| [15:32:00] | gigem: | stuartm: Give me a short while to add an initial PI constructor using just recordedid, and then feel free to go ahead. |
| [15:33:18] | ghoti (ghoti!~paul@scratch.it.ca) has quit (Ping timeout: 240 seconds) | |
| [15:34:28] | ghoti (ghoti!~paul@scratch.it.ca) has joined #mythtv | |
| [15:37:08] | jheizer: | jya_, dod I remember right that you had some kind of testing/stress program you were using while making the livetv changes? I was the one with the log on -users a little bit ago. Trying to see if I can track down any storage system issues. |
| [15:39:15] | jya_: | jheizer: i do have that somewhere... |
| [15:39:27] | jheizer: | I recently switched the drives from a mdadm raid5 to a zfs raidz1. I just started 5 recordings and was able to play back those all 5 in vlc at once, thought windows/vlc seemed to not like it after a minute or so. |
| [15:39:30] | ghoti (ghoti!~paul@scratch.it.ca) has quit (Ping timeout: 265 seconds) | |
| [15:39:34] | jya_: | the liveTV patches only creates new recordings every x seconds |
| [15:39:40] | jya_: | that was to test the transition |
| [15:39:54] | jheizer: | those 4 seem to continue playing ok |
| [15:39:56] | jheizer: | though |
| [15:40:28] | ghoti (ghoti!~paul@scratch.it.ca) has joined #mythtv | |
| [15:40:35] | jya_: | what you would probably more interested with however is that I wrote unit tests that simulate simultaneous recordings in both write and read and measure the throughput |
| [15:41:01] | jheizer: | ah ok, that is what I was thinking of. |
| [15:41:04] | jya_: | the only issue, is that I have no ideas where those units tests are. |
| [15:41:05] | jheizer: | Missed they were tests. |
| [15:41:36] | jya_: | because in the end, what caused stalling while reading the disks, was the attempt to read more than how much data was available |
| [15:41:46] | jya_: | on NFS that can hang for up to 30s |
| [15:42:21] | jya_: | that’s one of the changes I did to FileRingBuffer, which is to read how much data is available before attempting to read, and never read more than that amount |
| [15:42:47] | jya_: | before it would attempt to read a 64kB block, and expect it to returns early with a lesser amount of EOF was encountered |
| [15:42:58] | jya_: | which is what occurs when reading from a local disk |
| [15:43:03] | jheizer: | Got ya. I tried to keep up on it all here as I found it all interesting. |
| [15:43:37] | jya_: | but with nfs or my mdadm array, if you were exactly at the end, and attempted to read X bytes, it would take 40–50s for the call to read to return |
| [15:44:11] | jheizer: | Last night and once last week were just was the first time I had in progress issues in ages. But hard when multiple things have changed. |
| [15:44:26] | jya_: | problem is that I wrote those units tests many moons ago, and because they took several minutes to run, i wasn’t that keen in committing them |
| [15:44:51] | jya_: | I had stashed them, but since I’ve rsync several times between machines, likely that stash got destroyed |
| [15:45:19] | FabriceMG (FabriceMG!~Thunderbi@217.112.59.207) has quit (Quit: FabriceMG) | |
| [15:45:43] | ghoti (ghoti!~paul@scratch.it.ca) has quit (Ping timeout: 245 seconds) | |
| [15:46:07] | jheizer: | Ah I know how that goes. |
| [15:47:00] | ghoti (ghoti!~paul@scratch.it.ca) has joined #mythtv | |
| [15:47:04] | jheizer: | Hope I didn't shoot myself in the foot with this ZFS switch as the same drives under mdadm had been working fine for a few years. |
| [15:47:18] | jya_: | I have daily backups , but it may take me ages to find them, as I first have to remember when I did write those |
| [15:47:42] | jya_: | jheizer: I use ZFS myself, and I’ve set up a friend with ZFS RAIDZ2 array |
| [15:47:51] | jya_: | he had many issues, but none related to zfs |
| [15:48:02] | jheizer: | I have used it with freebsd for a while on a SAN for work |
| [15:48:07] | jheizer: | but this is my first time on linux. |
| [15:50:43] | jya_: | yeah, likely gone |
| [15:51:04] | jheizer: | Ok, no problem. Don't worry about digging too much just for me. |
| [15:51:15] | jya_: | if you run the backend with -v file —loglevel=debug |
| [15:51:25] | jya_: | you will see where about it took so long to read |
| [15:51:34] | jheizer: | Was hoping I could easily prove I had an issue or maybe give you a piece of data. |
| [15:51:42] | jheizer: | ok, will do that. |
| [15:51:43] | ghoti (ghoti!~paul@scratch.it.ca) has quit (Ping timeout: 252 seconds) | |
| [15:51:51] | jya_: | either in TFW, where it takes too long to write, or in ringbuffer |
| [15:51:58] | jya_: | is this a combo unit? |
| [15:52:03] | jya_: | frontend+backend ? |
| [15:52:11] | jheizer: | standalone BE |
| [15:52:25] | jya_: | I mean where was the frontend you used? |
| [15:52:28] | jheizer: | 5x2TB ZFS gbit to atom ion FE |
| [15:52:36] | jheizer: | standalone FE |
| [15:52:48] | jya_: | ok… and the atom doesn’t mount the same directory via nfs or what else? |
| [15:52:59] | ghoti (ghoti!~paul@scratch.it.ca) has joined #mythtv | |
| [15:53:10] | jheizer: | it does via samba and I just removed that. |
| [15:53:24] | jheizer: | I should have the always stream from BE on, but dropped it just to be sure. |
| [15:54:31] | jya_: | yes, stream via the myth protocol, *much* easier to troubleshoot |
| [15:54:41] | jheizer: | that FE also has a slave BE with a single tuner but is almost never used. |
| [15:54:42] | jya_: | and far less overhead than the single threaded samba |
| [15:54:51] | jheizer: | Only needed for livetv which we rarely use |
| [15:55:07] | jya_: | and accessing via myth is fully threaded on the backend. |
| [15:56:01] | jya_: | I wouldn’t be surprised if samba was the cause… and you can’t track the problem by only looking at the backend, you would need to run the frontend with -v file —loglevel=debug too |
| [15:56:20] | jheizer: | ah I never realized it was single threaded. |
| [15:56:42] | jya_: | if you stream from the backend, in a single place, you can see where it wrote , how long it takes to write, and at the same time the read |
| [15:57:38] | jya_: | run this patch though |
| [15:57:38] | jya_: | http://pastebin.com/9UD0TwUE |
| [15:57:57] | ghoti (ghoti!~paul@scratch.it.ca) has quit (Ping timeout: 252 seconds) | |
| [15:58:05] | jya_: | otherwise TFW is way too verbose.. if you record a mpeg-ts stream, it would show every time 188 bytes is being written |
| [15:58:17] | jya_: | makes it impossible to identify anything |
| [15:58:44] | jheizer: | ok, I'll have to compile from source. Lazy since mythbuntu is updated to regularly. |
| [15:58:51] | jheizer: | Cause yeah, 90% ATSC here. |
| [15:58:57] | ghoti (ghoti!~paul@scratch.it.ca) has joined #mythtv | |
| [16:01:26] | jheizer: | Thanks for the help. I'll report back if I find anything interesting. |
| [16:02:54] | andreaz1 (andreaz1!~Andreaz@tmo-106-161.customers.d1-online.com) has quit (Read error: Connection reset by peer) | |
| [16:04:29] | ghoti (ghoti!~paul@scratch.it.ca) has quit (Ping timeout: 264 seconds) | |
| [16:04:56] | ghoti (ghoti!~paul@scratch.it.ca) has joined #mythtv | |
| [16:09:46] | ghoti (ghoti!~paul@scratch.it.ca) has quit (Ping timeout: 240 seconds) | |
| [16:10:56] | ghoti (ghoti!~paul@scratch.it.ca) has joined #mythtv | |
| [16:15:46] | ghoti (ghoti!~paul@scratch.it.ca) has quit (Ping timeout: 240 seconds) | |
| [16:16:55] | ghoti (ghoti!~paul@scratch.it.ca) has joined #mythtv | |
| [16:21:46] | ghoti (ghoti!~paul@scratch.it.ca) has quit (Ping timeout: 240 seconds) | |
| [16:22:56] | ghoti (ghoti!~paul@scratch.it.ca) has joined #mythtv | |
| [16:28:17] | ghoti (ghoti!~paul@scratch.it.ca) has quit (Ping timeout: 255 seconds) | |
| [16:28:58] | ghoti (ghoti!~paul@scratch.it.ca) has joined #mythtv | |
| [16:34:19] | ghoti (ghoti!~paul@scratch.it.ca) has quit (Ping timeout: 276 seconds) | |
| [16:41:56] | caelor: | stuartm: do you have plans for event/notification subscription for the webfrontend? e.g. a longpoll or websocket approach |
| [16:46:32] | stuartm: | websocket |
| [16:47:13] | stuartm: | as noted here – https://code.mythtv.org/trac/wiki/WebFrontendWishlist |
| [16:48:48] | NightMonkey_ (NightMonkey_!~NightrMon@173-164-139-193-SFBA.hfc.comcastbusiness.net) has joined #mythtv | |
| [16:49:42] | NightMonkey_ (NightMonkey_!~NightrMon@173-164-139-193-SFBA.hfc.comcastbusiness.net) has quit (Remote host closed the connection) | |
| [16:49:44] | caelor: | ah, I completely missed that! thanks. I'm considering replacing my long serving info screen (hacked picture frame displaying https://github.com/caelor/mythtv-infoscreen) with a cheap android tablet running a webpage |
| [16:50:13] | caelor: | I'll set it up using a polling mechanism initially, and switch to websocket when available |
| [17:00:28] | knightr (knightr!~Nicolas@mythtv/developer/knightr) has quit (Read error: Connection reset by peer) | |
| [17:00:41] | knightr (knightr!~Nicolas@69-165-170-178.dsl.teksavvy.com) has joined #mythtv | |
| [17:00:41] | knightr (knightr!~Nicolas@69-165-170-178.dsl.teksavvy.com) has quit (Changing host) | |
| [17:00:41] | knightr (knightr!~Nicolas@mythtv/developer/knightr) has joined #mythtv | |
| [17:02:27] | dekarl: | jya, mythbuntu users should simply enable the correct fixes repository and be offered daily updates. its all in mythbuntu control centre. No need for a release |
| [17:03:30] | Merlin83b (Merlin83b!~Daniel@office.34sp.com) has quit (Quit: Leaving) | |
| [17:08:08] | Weaselweb (Weaselweb!~quassel@77-64-221-251.dynamic.primacom.net) has joined #mythtv | |
| [17:22:05] | stuartm: | dekarl: I think he means the package ubuntu ships with rather than the private repo users need to explicitly enable (and so many don't) |
| [17:23:14] | stuartm: | it's one of the things that's disappointing about Ubuntu, they never ship bug fix packages for stuff like MythTV in their main repos |
| [17:23:28] | stuartm: | or at least they make it so difficult to do that it never seems to happen |
| [17:24:56] | dekarl: | I still prefer the official Mythbuntu PPA with daily fixes/0.27 over other distributions' "we can only update to point releases, updating to fixes/0.27 with good bugfixes is not our thing" |
| [17:28:20] | stuartm: | dekarl: I've shifted the recording title/subtitle editing wishlist item to 'big features' since it seems there's no current endpoint for altering recording metadata/options and that will need to be added first |
| [17:28:51] | dekarl: | stuartm, that's sad. But thanks for looking into it |
| [17:29:09] | stuartm: | will be added, just not a five minute job |
| [17:30:40] | stuartm: | can anyone think of a good reason to keep the Content/DeleteFile and Content/RenameFile endpoints? They seem broken by design |
| [18:05:48] | brfransen (brfransen!~brfransen@64.179.169.226) has quit (Ping timeout: 240 seconds) | |
| [18:06:30] | brfransen (brfransen!~brfransen@64.179.169.226) has joined #mythtv | |
| [18:14:20] | dekarl1 (dekarl1!~dekarl@p4FCEFA2A.dip0.t-ipconnect.de) has joined #mythtv | |
| [18:17:17] | dekarl (dekarl!~dekarl@p4FE84186.dip0.t-ipconnect.de) has quit (Ping timeout: 252 seconds) | |
| [18:32:34] | natanojl (natanojl!~jonatan@mythtv/developer/natanojl) has joined #mythtv | |
| [18:33:19] | stuartm: | dblain: I'm trying to handle HEAD/OPTIONS correctly, they shouldn't trigger any endpoints to be executed but atm they do |
| [18:35:04] | stuartm: | well OPTIONS doesn't but only because we don't recognise it as a valid method (it is part of HTTP 1.1 though) |
| [18:35:56] | dblain: | stuartm: HEAD must return the same header as GET but without message content. The problem is we need to perform the actual action in order to get what the content-length should be in the returned header. |
| [18:37:39] | stuartm: | dblain: true, I'm looking at RFC 2616 which states "In particular, the convention has been established that the GET and HEAD methods SHOULD NOT have the significance of taking an action other than retrieval. These methods ought to be considered "safe"" |
| [18:38:44] | dblain: | Which all our GET methods adhear to. POST calls are the only methods that have the ability to modify. |
| [18:39:36] | stuartm: | not entirely true, I noticed that Frontend/PlayRecording() triggered actions even though they are GET methods |
| [18:39:40] | dekarl1 is now known as dekarl | |
| [18:39:51] | dblain: | that should be changed. |
| [18:40:27] | stuartm: | that makes sense |
| [18:40:29] | dblain: | We don't want a web crawler to be able to take an action |
| [18:41:58] | dblain: | not that I recomment exposing this directly to the web :) |
| [18:42:05] | dblain: | recommend |
| [18:42:13] | stuartm: | exactly, nor do we want XSS vulnerabilities which is what I was looking into, the origin policy requires the browser to look at the Access-Control-Allow-Origin header which it can only do by actually making a request |
| [18:43:21] | stuartm: | dblain: no, though with XSS it wouldn't need to be exposed to the web for issues like this to still be exploited – already created a proof of concept where simply visiting a page (hosted on mythtv.co.uk) triggers my frontend to start playing a recording :) |
| [18:43:43] | dblain: | oops! |
| [18:44:17] | dblain: | obviously, I'm all for anything that will harden our implementation. |
| [18:44:33] | stuartm: | I've a second line of defence in mind, one which seems to go beyond what most servers do, and that's to check the Origin header of the request and if it doesn't match our hostname/ip then we reject it |
| [18:45:46] | dblain: | I'd have to think about that one... there may be legitimite reasons for them not to match |
| [18:46:34] | dblain: | I was thinking of reversing our GET/POST default... everything defaults to POST unless marked as GET. Would at least help with developers forgetting to make it a POST. |
| [18:47:12] | stuartm: | technically, if they don't match, the browse should reject the request anyway – but since that still means the we have to build the response first it's still sub-optimal |
| [18:47:25] | stuartm: | browser |
| [18:47:52] | stuartm: | http://www.w3.org/TR/cors/#introduction |
| [18:48:13] | dblain: | When I read the spec, the only thing I figured HEAD would be helpful form a performance perspective would be static content. |
| [18:49:55] | dblain: | btw: sorry I haven't been able to work on any of the changes we discussed. Just too many things happening these past 6–8 months. |
| [18:57:23] | tris (tris!tristan@2001:1868:a00a::4) has quit (Quit: Leaving) | |
| [19:05:29] | stuartm: | dblain: np, I understand :) |
| [19:33:02] | dekarl: | jya, looks like this variable isn't assigned before testing for NULL, http://pastebin.com/jJ8MULdp |
| [20:06:25] | gigem: | Has anyone looked at the patch for #11954? I've been meaning to bring that up, but kept forgetting to do so. I looked at George's patch several weeks ago as I'd promised him. The thing I didn't like about it was it added yet another field to ProgramInfo. In this case, one that probably only needs to be displayed on the program details screen. I worked up an alternative approach that didn't add anything |
| [20:06:26] | gigem: | to PI, but instead queried the recorded table on demand. The problem is that our ToMap() way of doing things where everything gets shoved into a map whether it's needed or not makes my on demand function got called a lot. Does anyone have a better idea that won't further bloat PI? There are things I'd actually like to remove from PI before adding any more to it. |
| [20:16:33] | stuartm: | #11954 |
| [20:16:33] | ** MythLogBot http://code.mythtv.org/trac/ticket/11954 ** | |
| [20:22:13] | stuartm: | gigem: well it ought to go in RecordingInfo rather than ProgramInfo for a start, but if we want to only display it in ProgramDetails then I'd just add the query there – not everything has to go into the map, the maps are just a convenience for making large amounts of data available to be displayed (optionally) without any code burden in each screen |
| [20:25:05] | Weaselweb (Weaselweb!~quassel@77-64-221-251.dynamic.primacom.net) has quit (Remote host closed the connection) | |
| [20:47:29] | dekarl: | stuartm, does vertical scrolling still work in the webfrontend's program guide for you? I'm missing the scroll bar and manual scrolling doesn't work either. But zooming out shows more channels, so the data is there and is being loaded. |
| [20:48:20] | dekarl: | when entering the guide I can see the scrollbar while the page is build but then it disappears |
| [20:52:47] | stuartm: | works here, which browser? |
| [20:53:17] | dekarl: | Firefox 26.0 on Mythbuntu 12.04lts |
| [20:54:51] | stuartm: | I tested with 24.5.0, though it's fairly standard css so it shouldn't have stopped working in a later version |
| [20:55:23] | stuartm: | dekarl: when it first loads do you see a full screen of programs? |
| [20:56:53] | dekarl: | Well. I do see the scrollbar that suggests there is a full screen but before I can scroll the bar disappears. |
| [20:57:11] | dekarl: | If I zoom out I can see the full screen, still without scroll bar |
| [20:57:31] | dekarl: | So its only the scrolling that is broken, the page appears to be present |
| [20:57:42] | stuartm: | should look like this – http://mythtv.co.uk/imagebin/WebFrontend_guide_example.png |
| [20:58:59] | dekarl: | looks almost the same over here, only the scroll bar is missing |
| [20:59:22] | stuartm: | are you familiar with using the firefox debugger (right click Inspect Element)? |
| [20:59:50] | dekarl: | http://picpaste.de/Bildschirmfoto-zoom-out-lots-lbqR7APV.png |
| [21:00:23] | dekarl: | no yet |
| [21:00:25] | dekarl: | not yet |
| [21:00:42] | stuartm: | trippy |
| [21:02:22] | stuartm: | curious to know what the height of the "channels" div is, and what the overflow-y value is under the 'Computed' tab |
| [21:02:33] | dekarl: | hmm, fixed vertical height, overflow is hidden, but lots of channels are present |
| [21:03:05] | dekarl: | that was from the rules tab |
| [21:03:06] | stuartm: | that after selecting "<div class="channels" ..." in the source browser |
| [21:03:24] | dekarl: | in the computed tab its overflow:hidden, overflow-x:hidden, overflow-x:hidden |
| [21:03:32] | dekarl: | aye, after selecting the channels div |
| [21:03:49] | dekarl: | height is 632.667px |
| [21:04:05] | dekarl: | appears to be about what I see with the debugger openb |
| [21:04:23] | stuartm: | strange, maybe you've got an old cached css file there? Have you tried a Shift-reload? |
| [21:04:29] | dekarl: | ahzh |
| [21:04:55] | stuartm: | gesundheit |
| [21:06:38] | dekarl: | hmm, its still the same :( |
| [21:07:37] | dekarl: | ahh, manually clearing the cache helped, not the scroll bar is back |
| [21:08:13] | dekarl: | no I get overflow-y:scroll :) |
| [21:09:38] | SteveGoodey (SteveGoodey!~steve@host86-134-63-27.range86-134.btcentralplus.com) has quit (Quit: Konversation terminated!) | |
| [21:09:46] | dekarl: | s/no/now/ time to hit sack judging from the typos |
| [21:13:21] | stuartm: | heh |
| [21:13:37] | stuartm: | overflow-y: scroll; is the expected result |
| [21:14:07] | stuartm: | think we need some cache control improvements |
| [21:15:13] | dekarl (dekarl!~dekarl@p4FCEFA2A.dip0.t-ipconnect.de) has quit (Remote host closed the connection) | |
| [21:20:25] | stuarta: | jya_: forums now works and redirects to forum |
| [21:28:34] | gigem: | stuarta: Okay, I have no problem with only making it available on the details screen. Is it something you want in the services API? |
| [22:12:33] | natanojl (natanojl!~jonatan@mythtv/developer/natanojl) has quit (Ping timeout: 240 seconds) | |
| [22:46:12] | stuartm: | gigem: maybe, I'll need to think about it when I'm not so tired :) |
| [22:53:43] | gigem: | stuartm: No problem. I won't be doing anything on tonight anyway. |
| [22:57:50] | gregL (gregL!~greg@cpe-74-76-121-109.nycap.res.rr.com) has quit (Remote host closed the connection) | |
| [23:13:49] | gregL (gregL!~greg@cpe-74-76-121-109.nycap.res.rr.com) has joined #mythtv | |
| [23:20:00] | xris- (xris-!~xris@xris.forevermore.net) has joined #mythtv | |
| [23:20:22] | jya_: | dekarl: actually that variable shouldn’t even be there. |
| [23:20:31] | xris (xris!~xris@mythtv/developer/xris) has quit (Quit: Goodbye.) | |
| [23:20:43] | cecil (cecil!~cesman@pool-173-60-115-40.lsanca.fios.verizon.net) has joined #mythtv | |
| [23:21:54] | jya_: | i’ve made it a class member now… which can be re-used for the entire duration of the encoder |
| [23:24:40] | cesman (cesman!~cesman@pdpc/supporter/professional/cesman) has quit (Ping timeout: 252 seconds) | |
| [23:25:30] | jya_: | stuarta: thanks |
IRC Logs collected by
BeirdoBot.
Please use the above link to report any bugs.
