| Monday, May 7th, 2012, 00:30 UTC | ||
| [00:30:34] | gigem: | stichnot: Oops! I was in a hurry and skipped right over the "xml-mode" in your question. |
| [00:42:57] | knightr: | wagnerrp, 10613 is you mean handle both "Series" and "Season" as equivalent, I am looking into it... I could be wrong but I think the problem can partially be fixed partially by modifying the translation of one of the "Season" by "Series|Season" since one of them is directly inserted into the regex... |
| [00:45:04] | wagnerrp: | that could do it |
| [00:45:11] | wagnerrp: | feel free to claim the ticket if you want to do so |
| [00:45:24] | wagnerrp: | i have no intention of changing it on my own |
| [00:45:30] | knightr: | assuming that work there's another place where I think there might be a problem but I am not entirely sure so I'll have to test... |
| [00:45:57] | knightr: | wagnerrp, thanks, I'll assign it to myself... |
| [00:49:14] | knightr: | argh, I should have reopened before assigning it to myself... that switch took more steps than necessary... |
| [00:49:21] | wagnerrp: | heh |
| [00:53:24] | Captain_Murdoch: | jya, the HLS server only has supports encoding 2 streams now. the first is A/V at the bandwidths specified and the seocnd is autio only at 32 kbit I think. |
| [00:54:22] | Captain_Murdoch: | wouldn't that hard to make more streams if we wanted to. the only reason I added the audio only stream originally was because that's required for any HLS app in the App store. |
| [00:55:07] | Captain_Murdoch: | dont' want to just encode a bunch of streams because then we're wasting CPU making smaller or larger streams than the user can support. best to let them specify I think. |
| [00:55:25] | wagnerrp: | Captain_Murdoch: i had something i wanted to ask you |
| [00:55:28] | Captain_Murdoch: | right now it uses the threaded slice encoding in libx264 |
| [00:55:30] | wagnerrp: | ... but i simply cannot remember what it was |
| [00:56:04] | wagnerrp: | oh, right! |
| [00:56:31] | wagnerrp: | does the System Event stuff handle applications with a manually configured LocalHostName correctly? |
| [00:57:05] | wagnerrp: | seems like it would mis-detect that as two separate systems connecting |
| [00:58:13] | Captain_Murdoch: | not sure. I assume it would mis-detect. only way to prevent that would be to check unique IPs instead of 'hostname's. |
| [00:59:28] | jya_: | Captain_Murdoch: you don't have to encode them before-hand. Aren't you doing the encoding on the fly? |
| [00:59:45] | Captain_Murdoch: | jya, yes, but you have to encode all streams at the same time. |
| [01:00:07] | Captain_Murdoch: | the player can switch at any time and the server doens't know which stream the player is playing at any given time. |
| [01:00:24] | jya_: | Captain_Murdoch: that's not how i'm understanding the system of the m3u8 playlist… The client request the stream it wants, only when it wants them |
| [01:00:39] | jya_: | Captain_Murdoch: the server would know, according to which URL is fetched |
| [01:00:47] | Captain_Murdoch: | so why waste time encoding 3 different bitrates when the user can request the one they want. |
| [01:00:55] | jya_: | you don't have to. |
| [01:01:01] | Captain_Murdoch: | jya_, but you have to encode it before the player can fetch it |
| [01:01:36] | jya_: | i see here, i have the apple playlist, it has 3 different bitrates available, each of them with a different relative URI, the server would encode according to which URI is fetched from the client |
| [01:01:38] | Captain_Murdoch: | the player requests the main playlist. that contains links to other playlsits for each stream. the player can switch between streams at any point during playback |
| [01:02:14] | jya_: | sure, but with the buffering required , it will always pre-fetch segments |
| [01:02:25] | Captain_Murdoch: | the player can be playing along fine and then hit a network issue and switch to an audio only stream, only to swtich back to the highest bitrate A/V stream 2 segments later. |
| [01:02:38] | jya_: | that time delay should be sufficient to only encode on the fly according to the URI used |
| [01:05:08] | Captain_Murdoch: | the player can switch at any time though, you don't want hte player to have to wait while you encode the stream if it all the sudden decided to try a lower bandwidth stream. especially if it can seek backwards. |
| [01:05:40] | jya_: | Currently, I'm fetching with 1 segment in advance |
| [01:05:45] | jya_: | that's 10s ahead |
| [01:06:12] | jya_: | anyhow, soon we'll be able to do more tests |
| [01:06:41] | jya_: | just fetching a http://blah.m3u8 will make it use the new httplivestream ringbuffer |
| [01:12:03] | Captain_Murdoch: | but the player can fetch multiple copies of the same segment if it decides to switch streams. |
| [01:13:03] | Captain_Murdoch: | trying to outsmart the player is only asking for heartache IMHO. I figured it was easier to let the player (via the API) request the bitrate it wanted. |
| [01:14:32] | Captain_Murdoch: | wagnerrp, looks like we store the IP in the PlaybackSock, so instead of checking pbs->getHostname(), I could check pbs->getIP() possibly to guard against that situation. |
| [01:15:29] | wagnerrp: | not urgent, just more curious than anything else |
| [01:15:49] | wagnerrp: | since ive been trying to push away from LocalHostName, and use of it as a "profile" instead |
| [01:16:07] | Captain_Murdoch: | I made a note, but can't say when I'll have time to test since I'm in the middle of a move. |
| [01:29:41] | mgeneral (mgeneral!~Spyder@pool-71-186-143-24.bflony.east.verizon.net) has joined #mythtv | |
| [02:07:46] | stichnot (stichnot!~chatzilla@mythtv/developer/stichnot) has quit (Ping timeout: 272 seconds) | |
| [02:08:07] | stichnot (stichnot!~chatzilla@192.55.55.41) has joined #mythtv | |
| [02:08:18] | stichnot (stichnot!~chatzilla@192.55.55.41) has quit (Changing host) | |
| [02:08:18] | stichnot (stichnot!~chatzilla@mythtv/developer/stichnot) has joined #mythtv | |
| [02:12:35] | mgeneral (mgeneral!~Spyder@pool-71-186-143-24.bflony.east.verizon.net) has quit (Remote host closed the connection) | |
| [02:25:05] | jya: | is there a class to provide basic http fetch ? |
| [02:26:11] | jya: | hum.. mythdownloadmanager would do the trick i think |
| [02:51:54] | skd5aner: | jya: yup, you've got it... I think Beirdo even used it to remove the need for wget within mythfilldatabase shortly before 0.25 went it, so it should do the trick for you |
| [02:52:57] | jya: | Only thing I would have liked is a "peek" function, where it stops reading after receiving xx bytes, so it doesn't need to download the whole url |
| [02:54:00] | wagnerrp: | you could probably let it get a small bit in and terminate |
| [02:54:12] | wagnerrp: | how big is this file you want to fetch? |
| [02:54:48] | jya: | wagnerrp: I have no idea on how big the file could be |
| [02:55:16] | skd5aner: | jya: you thinking of throttling a stream or something? |
| [02:55:32] | wagnerrp: | the download manager will fire off update events every few seconds |
| [02:55:50] | jya: | skd5aner: not really, I want to check if a stream is of a specific type before downloading the whole lot, and for this I only need about 50 bytes worth |
| [02:55:55] | wagnerrp: | you can write some loop to listen on those and wait until it hits a certain size, or otherwise completes |
| [02:56:09] | wagnerrp: | heh... well not the best solution for mere bytes |
| [02:56:30] | jya: | wagnerrp: the download method, seems to be blocking and only returns data once complete, it downloads 10s worth |
| [02:56:30] | wagnerrp: | better to get Captain_Murdoch to add that feature internally |
| [02:58:21] | jya: | hum.. I wonder is this why I've seen hang in download from time to time, a lock isn't released there |
| [03:41:38] | Seeker` (Seeker`!~cjo20@unaffiliated/seeker) has quit (Ping timeout: 240 seconds) | |
| [03:43:02] | Seeker` (Seeker`!~cjo20@host86-178-252-146.range86-178.btcentralplus.com) has joined #mythtv | |
| [03:43:20] | Seeker` (Seeker`!~cjo20@host86-178-252-146.range86-178.btcentralplus.com) has quit (Changing host) | |
| [03:43:21] | Seeker` (Seeker`!~cjo20@unaffiliated/seeker) has joined #mythtv | |
| [03:49:09] | stichnot (stichnot!~chatzilla@mythtv/developer/stichnot) has quit (Ping timeout: 248 seconds) | |
| [04:06:34] | stuartm (stuartm!~stuartm@mythtv/developer/stuartm) has quit (Remote host closed the connection) | |
| [04:22:26] | stichnot (stichnot!~chatzilla@mythtv/developer/stichnot) has joined #mythtv | |
| [05:14:41] | stoffel (stoffel!~quassel@pD9E42742.dip.t-dialin.net) has joined #mythtv | |
| [05:48:51] | Yancho_ (Yancho_!~mpulis@unaffiliated/yancho) has quit (Remote host closed the connection) | |
| [05:48:51] | Yancho (Yancho!~mpulis@unaffiliated/yancho) has quit (Remote host closed the connection) | |
| [05:58:10] | dekarl (dekarl!~dekarl@p4FE84706.dip.t-dialin.net) has joined #mythtv | |
| [06:00:21] | hashbang (hashbang!~alex@213-152-35-50.dsl.eclipse.net.uk) has joined #mythtv | |
| [06:00:42] | dekarl1 (dekarl1!~dekarl@p4FCEFA63.dip.t-dialin.net) has quit (Ping timeout: 265 seconds) | |
| [06:58:50] | natanojl (natanojl!~jonatan@c83-252-237-63.bredband.comhem.se) has joined #mythtv | |
| [07:07:57] | solars (solars!~solars@clnet-kmu02-090.ikbnet.co.at) has joined #mythtv | |
| [07:31:24] | natanojl (natanojl!~jonatan@c83-252-237-63.bredband.comhem.se) has quit (Ping timeout: 272 seconds) | |
| [07:50:09] | rsiebert_ (rsiebert_!~quassel@g226063156.adsl.alicedsl.de) has joined #mythtv | |
| [07:53:37] | rsiebert (rsiebert!~quassel@g226060015.adsl.alicedsl.de) has quit (Ping timeout: 276 seconds) | |
| [08:47:18] | jya_ (jya_!~jyavenard@mythtv/developer/jya) has quit (Read error: Connection reset by peer) | |
| [08:47:39] | jya_ (jya_!~jyavenard@mythtv/developer/jya) has joined #mythtv | |
| [09:30:20] | stuartm (stuartm!~stuartm@cpc1-derb9-0-0-cust441.8-3.cable.virginmedia.com) has joined #mythtv | |
| [09:30:20] | stuartm (stuartm!~stuartm@cpc1-derb9-0-0-cust441.8-3.cable.virginmedia.com) has quit (Changing host) | |
| [09:30:21] | stuartm (stuartm!~stuartm@mythtv/developer/stuartm) has joined #mythtv | |
| [09:33:51] | Steve_Goodey (Steve_Goodey!~steve@host86-182-4-91.range86-182.btcentralplus.com) has joined #mythtv | |
| [10:05:02] | Guest56746 (Guest56746!~mike@c-98-232-220-158.hsd1.or.comcast.net) has quit (Remote host closed the connection) | |
| [10:05:50] | mike|3 (mike|3!~mike@c-98-232-220-158.hsd1.or.comcast.net) has joined #mythtv | |
| [10:07:53] | Steve_Goodey (Steve_Goodey!~steve@host86-182-4-91.range86-182.btcentralplus.com) has quit (Remote host closed the connection) | |
| [10:29:30] | Steve_Goodey (Steve_Goodey!~steve@host86-182-4-91.range86-182.btcentralplus.com) has joined #mythtv | |
| [10:47:17] | zombor (zombor!~zombor_@kohana/developer/zombor) has joined #mythtv | |
| [10:48:22] | zombor (zombor!~zombor_@kohana/developer/zombor) has quit (Remote host closed the connection) | |
| [11:10:15] | XDS2010_ (XDS2010_!u1218@gateway/web/irccloud.com/x-akoircfojsdpijsw) has quit (Excess Flood) | |
| [11:10:43] | XDS2010_ (XDS2010_!u1218@gateway/web/irccloud.com/x-xhcwfevgcovqpxwj) has joined #mythtv | |
| [11:34:20] | Steve_Goodey (Steve_Goodey!~steve@host86-182-4-91.range86-182.btcentralplus.com) has quit (Remote host closed the connection) | |
| [12:04:44] | highzeth (highzeth!~hz@205-20-11.connect.netcom.no) has joined #mythtv | |
| [12:12:41] | Anssi (Anssi!hannulaa@mandriva/developer/anssi) has quit (Remote host closed the connection) | |
| [12:16:32] | Anssi (Anssi!hannulaa@mandriva/developer/anssi) has joined #mythtv | |
| [12:40:49] | dmfrey (dmfrey!~dmfrey@webdefence.cluster-h.websense.net) has joined #mythtv | |
| [12:42:55] | aqwwertty (aqwwertty!~na@87.254.70.6) has joined #mythtv | |
| [12:58:28] | Captain_Murdoch: | jya, for async, you'd want MDM::queueDownload(), not ::download() |
| [12:59:10] | Captain_Murdoch: | might be useful to add a 'peek by downloading the first X bytes' option to MDM. |
| [13:01:41] | Captain_Murdoch: | no guarantee whether the underlying Qt code would terminate without downloading the whole thing though. we do issue an abort() call to stop the download when MDM::cancel() is called on an active download. |
| [13:21:23] | zombor (zombor!~zombor_@kohana/developer/zombor) has joined #mythtv | |
| [13:30:37] | danielk22: | Captain_Murdoch: jya: For peeking we'd probably want to use HTTP range support when available. |
| [13:36:43] | stoffel (stoffel!~quassel@pD9E42742.dip.t-dialin.net) has quit (Ping timeout: 260 seconds) | |
| [13:43:25] | highzeth (highzeth!~hz@205-20-11.connect.netcom.no) has quit (Quit: Leaving) | |
| [13:46:00] | jya_: | Captain_Murdoch: synchronous download is fine, I'm doing the reading in a secondary thread, so I don't care if it's blocking |
| [13:46:14] | Jordack (Jordack!~jordack@h69-131-44-221.plmomi.dedicated.static.tds.net) has joined #mythtv | |
| [13:56:56] | gregL (gregL!~greg@cpe-74-76-125-87.nycap.res.rr.com) has quit (Read error: Connection reset by peer) | |
| [14:12:56] | stoffel (stoffel!~quassel@pD9E42742.dip.t-dialin.net) has joined #mythtv | |
| [14:13:11] | gregL (gregL!~greg@cpe-74-76-125-87.nycap.res.rr.com) has joined #mythtv | |
| [14:55:12] | skd5aner: | stuartm: I saw this coverity scan thread on -dev... Eric mentions he got through, but suggest leveraging -fixes. I would think it'd likely be better to set up master instead at this point, don't you think? |
| [14:58:25] | stuartm: | definitely master, checking fixes would be a case of putting the cart after the horse, we want the next release to be bug free not fixing the bugs after it's already out there |
| [14:59:51] | Captain_Murdoch: | danielk22, yeah, thanks. I'd think we should be able to hook that header up fairly easily. |
| [15:00:07] | skd5aner: | stuartm: I was just checking the gossamer archive, as I typically don't log in to that email account that often... was sure if you had already sent him a reply or not stating as such but if not you might want to do so (unless you /really/ want me to do it) |
| [15:00:16] | skd5aner: | ") |
| [15:00:20] | skd5aner: | er, :) |
| [15:00:50] | skd5aner: | damnit, s/was/wasn't |
| [15:01:13] | ** skd5aner gives up on typing – needs neural interface ** | |
| [15:01:14] | Captain_Murdoch: | danielk22, and if hooking up in MDM, we may as well go ahead and support requesting bytes X-Y instead of just the first Z bytes. |
| [15:10:39] | danielk22: | wagnerrp: In the UTC branch I dropped support for timezone queries this weekend, but I wasn't sure what the Python bindings were using that info for. I've just use the fallback "Unknown" w/ 0 offset. |
| [15:12:08] | danielk22: | sphery: I retained the time variance checking when connecting to backends, replacing QUERY_TIMEZONE with QUERY_TIME. But I didn't retained the time variance checking for the frontends. Do they really misbehave if their clocks are wrong? |
| [15:20:13] | cocoa117 (cocoa117!~cocoa117@188-222-31-239.zone13.bethere.co.uk) has joined #mythtv | |
| [15:39:14] | stichnot: | Question about the ffmpeg code. I'm trying to get the cutlist editor to show thumbnails for various frames around the current frame. It uses a slightly modified version of MythPlayer::GetScreenGrabAtFrame() to get the target frame. That method calls avpicture_fill(), avpicture_deinterlace(), etc. My problem is that it basically works if I'm using the Slim profile, but segfaults with... |
| [15:39:16] | stichnot: | ...VDPAU Normal. Does anyone know how to make this work with VDPAU? |
| [15:39:58] | stichnot: | (this may be a question for Beirdo) |
| [15:40:45] | wagnerrp: | danielk22: python bindings dont touch time zone... likely the perl bindings dont either |
| [15:40:52] | wagnerrp: | it may cause problems with mythweb |
| [15:41:25] | wagnerrp: | well the UTC conversion in general is going to cause problems with mythweb, but i mean i believe it performs a timezone check |
| [15:44:26] | Steve_Goodey (Steve_Goodey!~steve@host86-182-4-91.range86-182.btcentralplus.com) has joined #mythtv | |
| [15:56:14] | gregL (gregL!~greg@cpe-74-76-125-87.nycap.res.rr.com) has quit (Read error: Connection reset by peer) | |
| [16:00:48] | joki (joki!~joki@p54862DC6.dip.t-dialin.net) has quit (Ping timeout: 260 seconds) | |
| [16:00:58] | joki- (joki-!~joki@p54863142.dip.t-dialin.net) has joined #mythtv | |
| [16:01:06] | joki- is now known as joki | |
| [16:04:51] | danielk22: | wagnerrp: Yeah, I already talked a little bit with Chris Petersen about mythweb some months back. |
| [16:06:56] | wagnerrp: | although with a move to UTC, it would probably be good to add some form of time zone support into the bindings |
| [16:07:14] | wagnerrp: | detecting the users' local time zone so it can translate accordingly |
| [16:08:12] | danielk22: | wagnerrp: The timezone stuff didn't really work very well since there is no cross-platform standard for the info. |
| [16:08:25] | wagnerrp: | right now, the only timezone support is for content coming from the services API, since it is in ISO format with the +- values |
| [16:08:56] | wagnerrp: | to be fair, the bindings really arent cross platform anyway |
| [16:09:32] | wagnerrp: | they do linux and freebsd, and would probably work on OSX if someone went through the effort of getting them installed |
| [16:10:11] | wagnerrp: | ive never even tried using them on windows, but i know right now a lot of stuff wouldnt work right |
| [16:19:35] | gregL (gregL!~greg@cpe-74-76-125-87.nycap.res.rr.com) has joined #mythtv | |
| [16:26:07] | highzeth (highzeth!~hz@205-20-11.connect.netcom.no) has joined #mythtv | |
| [16:39:07] | sphery: | danielk22: time variance on frontends seems to cause Live TV failures in some circumstances (never explored details, though) |
| [16:39:24] | danielk22: | sphery: Thanks, I'll test that. |
| [16:40:05] | sphery: | danielk22: also, FWIW, mythweb was requesting the time zone and using that to set the php time zone... not sure if we still want that, but I think the Services API capability was much better for it than myth proto |
| [16:40:31] | sphery: | I'm all for getting rid of the myth proto check--that came before we'd properly connected everything--for time/time zone |
| [16:41:01] | sphery: | I had planned to switch it to just use Services API, but dblain was working on a "services client" implementation that would allow us to use services from within the app |
| [16:41:52] | danielk22: | sphery: I ripped out all the timezone stuff in both the Myth Proto and Services API. But I'm wondering if we still need it.. |
| [16:42:08] | danielk22: | Ideally we can drop it, but if we can't so be it. |
| [16:42:15] | sphery: | we could just remove it and make desired time zone a setting... |
| [16:42:59] | sphery: | in mythweb, makes sense to have per-session time zone--and would allow different time zones depending on where you're connecting from when traveling |
| [16:43:00] | danielk22: | For PHP I can just add a strtotime and then use the normal time formatting stuff which uses up the local time zone. |
| [16:43:34] | sphery: | yeah, that's probably fine, too... then users see their "home" times that they're used to |
| [16:44:15] | danielk22: | sphery: That was my thinking.. |
| [16:45:19] | sphery: | getting rid of that code is a good thing from a maintenance standpoint--especially as it's already chock full of conditions for various different distro's approaches for configuring time zone, so if distros change things around, it would likely be more work than it's worth to fix |
| [16:51:24] | solars (solars!~solars@clnet-kmu02-090.ikbnet.co.at) has quit (Ping timeout: 245 seconds) | |
| [16:54:26] | danielk22: | sphery: Yeah, that was my hope. I started off just removing the timezone check for the MythProto connections. But I saw there was a whole lot of code there that we're maintaining. |
| [16:59:28] | sphery: | yeah, I'd say go ahead and leave it gone, and if any issues crop up, we can handle them later. |
| [17:03:29] | sphery: | danielk22: also, I'd say to get rid of the offset stuff in QUERY_TIME, too (sounds like you may have left that in?)... just change it to return only the current time on mbe (ideally just return UTC time or epoch time or whatever) |
| [17:03:42] | NightMonkey (NightMonkey!~NightrMon@pdpc/supporter/professional/nightmonkey) has joined #mythtv | |
| [17:07:38] | zombor (zombor!~zombor_@kohana/developer/zombor) has quit (Ping timeout: 240 seconds) | |
| [17:09:19] | highzeth (highzeth!~hz@205-20-11.connect.netcom.no) has quit (Quit: Leaving) | |
| [17:11:29] | kormoc (kormoc!~kormoc@mythtv/developer/kormoc) has joined #mythtv | |
| [17:12:29] | zombor (zombor!~zombor_@kohana/developer/zombor) has joined #mythtv | |
| [17:13:55] | cocoa117 (cocoa117!~cocoa117@188-222-31-239.zone13.bethere.co.uk) has quit (Remote host closed the connection) | |
| [17:17:39] | cattelan_away is now known as cattelan | |
| [17:33:19] | solars (solars!~solars@089144206027.atnat0015.highway.a1.net) has joined #mythtv | |
| [17:38:06] | Beirdo: | stichnot: it's always worked for me with VDPAU Normal (IIRC) |
| [17:38:18] | Beirdo: | we'll have to dig into that |
| [17:44:14] | gigem: | stuartm: Are buttonlists and spinboxes (and perhaps others) able to show the "disabled" state? If so, what does the theme need to provide to accomplish it? |
| [18:01:11] | aqwwertty (aqwwertty!~na@87.254.70.6) has quit (Quit: Because F/LOSSing is so passe ...... :)) | |
| [18:07:59] | stichnot: | Beirdo: Further digging into ffmpeg/vdpau. I found that mythpreviewgen does actually work (and it uses MythPlayer::GetScreenGrabAtFrame) if I change the code to allow VDPAU. I think the right approach may involve creating a "new MythPlayer((PlayerFlags)(kAudioMuted | kVideoIsNull | kDecodeAllowGPU))" to handle the frame grabs. |
| [18:08:51] | stichnot: | like what is done in PreviewGenerator::GetScreenGrab(), with kDecodeAllowGPU added |
| [18:10:06] | stichnot: | my next concern would be whether the code does the right thing if for some reason there aren't enough VDPAU resources to create the null video output |
| [18:15:19] | danielk22: | sphery: I dropped the offset too, all it does is return the UTC time. |
| [18:15:35] | zombor_ (zombor_!~zombor_@kohana/developer/zombor) has joined #mythtv | |
| [18:16:39] | danielk22: | stichnot: if you allow use of the GPU how will resource allocation work? i.e. how do you make sure the VDPAU usage by the preview generator or mythtranscode, mythcommflag, etc. doesn't cause problems for video playback? |
| [18:17:25] | danielk22: | Ideally the OS would handle this of course, but we know it doesn't.. :) |
| [18:18:59] | Beirdo: | danielk22: libvdpau should take care of that |
| [18:19:15] | zombor (zombor!~zombor_@kohana/developer/zombor) has quit (Ping timeout: 265 seconds) | |
| [18:19:17] | Beirdo: | *should* :) |
| [18:19:25] | Beirdo: | obviously, we'd need to test |
| [18:19:41] | stichnot: | danielk22: I have the same questions. In this case, I would only intend to create a new player during cutlist editing during normal playback, not turn it on for all applications. |
| [18:20:41] | stichnot: | I would hope that it would fall back to software decoding if there aren't enough HW decoding resources, and it would be nice to know how to stress-test that... |
| [18:25:22] | danielk22: | I'm not too familiar with the API, but from my looking at the doxygen docs I don't see any support for prioritizing different decode or render streams so I would think we'd need to do the resource allocation. |
| [18:26:01] | danielk22: | Chutt: Maybe you can give some insight? |
| [18:26:43] | Chutt: | hmm? |
| [18:27:44] | danielk22: | Can we label one VDPAU decode, deint & render stream as high priority and add additional ones to do as resources are available for commflag, transcode, etc. |
| [18:27:46] | Chutt: | oh, no idea |
| [18:28:22] | Chutt: | generally, no |
| [18:29:16] | Chutt: | but i think there aren't many _resources_ other than the amount of time everything takes |
| [18:29:20] | Chutt: | and memory |
| [18:29:34] | Chutt: | should be possible to manage things from the client side, though |
| [18:29:46] | Chutt: | kill transcode when something else comes along |
| [18:29:48] | Chutt: | etc |
| [18:29:51] | danielk22: | Yeah, we'd just need to do it. |
| [18:32:41] | danielk22: | stichnot: I think for the preview case in editing it is only the memory usage that matters. |
| [18:33:43] | stichnot: | danielk22: What are the memory usage issues, apart from being careful not to cache too many frame grabs? |
| [18:34:20] | stichnot: | i.e., vdpau-specific memory issues? |
| [18:34:54] | Beirdo: | I would presume you mean GPU memory... |
| [18:35:35] | danielk22: | stichnot: Well if you are using the same vdpau decoder as is being used for playback then I wouldn't expect any issues. If you start a new vdpau decoder and you are using 300 MB for decoding for playback then you don't want to allocate another 300 MB for preview on a 512 MB card. |
| [18:35:57] | danielk22: | Yep, GPU memory. |
| [18:37:00] | Beirdo: | I seriously doubt you want to use the current vdpau decoder as it ties directly to the screen via presentation queues |
| [18:37:23] | stichnot: | yeah, I think that is what led to segfaults |
| [18:37:25] | Beirdo: | you might want to look at how I did it in the gpu-commflag branch for gpucommflagging |
| [18:38:06] | Beirdo: | it's a different way of hooking into it, although you likely need even less as you just need it decoded, then pulled back to CPU memory |
| [18:38:35] | Beirdo: | I have it decoding, then hand control over to OpenGL, then to OpenCL so I can use it in the GPU memory directly |
| [18:38:54] | Beirdo: | and yeah, debugging that stuff is rife with segfaults |
| [18:38:55] | Beirdo: | heh |
| [18:39:16] | Beirdo: | and much cursing... until you have it right, and then it's lovely |
| [18:39:41] | Beirdo: | any reason not to use mythpreviewgen for your screenshots? (just curious) |
| [18:40:12] | Beirdo: | I may be misunderstanding your use case of course :) |
| [18:40:46] | wagnerrp: | is there any chance of hooking into the VDPAU deinterlacers if we extract the video back through OpenGL? |
| [18:40:47] | danielk22: | mythpreviewgen takes a long time to startup.. if we had mythpreviewd implemented I imagine it would be more attractive :) |
| [18:41:04] | Beirdo: | danielk22: it won't in a day or two |
| [18:41:05] | wagnerrp: | can those be applied before hand? |
| [18:41:06] | Beirdo: | :) |
| [18:41:17] | sphery: | FWIW, he's working on the new editing interface that shows frames around current – #9556 – to get something like http://code.mythtv.org/trac/attachment/ticket/9556/bball.jpg |
| [18:41:17] | ** MythLogBot http://code.mythtv.org/trac/ticket/9556 ** | |
| [18:41:49] | dmfrey (dmfrey!~dmfrey@webdefence.cluster-h.websense.net) has quit (Remote host closed the connection) | |
| [18:41:54] | sphery: | so we don't want these previews stored to disk as time-based previews (as all will be the same second of video) |
| [18:42:00] | Beirdo: | danielk22: much of that is the slowness of bringing up logging, and I have that solved in a totally different way |
| [18:42:23] | Beirdo: | sphery: Ahhh, makes sense |
| [18:42:27] | dmfrey (dmfrey!~dmfrey@webdefence.cluster-h.websense.net) has joined #mythtv | |
| [18:42:51] | sphery: | might be possible to tell mythpreviewgen to not store the preview, but we do need to be able to create previews of specific frames |
| [18:42:54] | Beirdo: | was just wondering what the differentiation was (if that makse sense) |
| [18:43:33] | ben1066_ (ben1066_!~quassel@host109-152-58-224.range109-152.btcentralplus.com) has quit (Quit: http://quassel-irc.org - Chat comfortably. Anywhere.) | |
| [18:43:53] | stichnot: | well, one could change mythpreviewgen to take a frame number, and perhaps also an expiration time that the housekeeper would honor... |
| [18:43:54] | sphery: | not sure if mythpreviewgen allows you to specify frame number and if it only grabs key frame previews? |
| [18:44:25] | ben1066 (ben1066!~quassel@host109-152-58-224.range109-152.btcentralplus.com) has joined #mythtv | |
| [18:45:00] | Beirdo: | it will allow you to put a time offset, no reason we can't make that a frame number |
| [18:45:06] | gregorcy (gregorcy!~gregorcy@strongbad.chemeng.utah.edu) has joined #mythtv | |
| [18:45:18] | Beirdo: | and it might need a bit of tweaking, but I'm sure that can be handled |
| [18:45:37] | Beirdo: | it just seems a bit of a shame to duplicate the functionality yet again, that's all. |
| [18:46:35] | danielk22: | SetPreviewTimeAsFrameNumber() |
| [18:46:37] | stichnot: | fwiw, my intention on 9556 is to allow the <osd_program_editor> window to add <imagetype> objects that get filled in with frame grabs. The number of frames, size, position, alpha, reflection, etc. would all be under themer control. |
| [18:46:45] | stuartm: | gigem: only standalone buttons are currently able to reflect that state, it wouldn't take much to add it for the others ... I think I simply overlooked it when many widgets like spinboxes were created |
| [18:49:05] | danielk22: | I'm not sure if the mythpreviewgen app supports everything needed, I'm pretty sure the PreviewGenerator class does. However it would not be the most efficient if you want say 30 frames in a row.. |
| [18:51:41] | ben1066_ (ben1066_!~quassel@host109-152-85-91.range109-152.btcentralplus.com) has joined #mythtv | |
| [18:51:56] | ben1066_ (ben1066_!~quassel@host109-152-85-91.range109-152.btcentralplus.com) has quit (Client Quit) | |
| [18:52:31] | ben1066_ (ben1066_!~quassel@host109-152-85-91.range109-152.btcentralplus.com) has joined #mythtv | |
| [18:52:58] | danielk22: | Beirdo: it doesn't sound like this duplicates much functionality. Jim would be using the same GetScreenGrabAtFrame() function that the preview generator does. |
| [18:53:20] | Beirdo: | K, cool :) |
| [18:53:58] | Beirdo: | it's the "do not repeat yourself" part that concerned me, but yeah, using a common function is all good. |
| [18:54:10] | ben1066 (ben1066!~quassel@host109-152-58-224.range109-152.btcentralplus.com) has quit (Ping timeout: 260 seconds) | |
| [18:54:57] | stichnot: | danielk22: basically the same GetScreenGrabAtFrame, but perhaps parameterized to not necessarily call OpenFile() and InitVideo(), and maybe other slight changes |
| [18:55:28] | ben1066_ (ben1066_!~quassel@host109-152-85-91.range109-152.btcentralplus.com) has quit (Client Quit) | |
| [18:55:30] | Beirdo: | stichnot: if we can avoid cut/paste, that would be best for future code maintenance. |
| [18:55:39] | stichnot: | yep |
| [18:55:57] | Beirdo: | i.e. add parameters and make the function work for both (unless it gets messy), and so on |
| [18:56:00] | ben1066 (ben1066!~quassel@host109-152-85-91.range109-152.btcentralplus.com) has joined #mythtv | |
| [18:56:20] | stichnot: | I meant that the existing GetScreenGrabAtFrame would be further parameterized if necessary |
| [18:56:41] | stichnot: | but I should get it to actually work first :) |
| [18:56:48] | Beirdo: | :) |
| [18:59:21] | Beirdo: | yeah, functionality is always a good thing :) |
| [19:00:19] | mrand (mrand!~mrand@ubuntu/member/mrand) has quit (Ping timeout: 265 seconds) | |
| [19:06:58] | mrand (mrand!~mrand@ubuntu/member/mrand) has joined #mythtv | |
| [19:07:41] | mzanetti_ (mzanetti_!~mzanetti@server1.muehlhaeuser.de) has quit (Excess Flood) | |
| [19:09:01] | mzanetti_ (mzanetti_!~mzanetti@server1.muehlhaeuser.de) has joined #mythtv | |
| [19:14:29] | NightMonkey (NightMonkey!~NightrMon@pdpc/supporter/professional/nightmonkey) has quit (Ping timeout: 244 seconds) | |
| [19:15:10] | zombor_ (zombor_!~zombor_@kohana/developer/zombor) has quit (Ping timeout: 260 seconds) | |
| [19:21:27] | kth (kth!~kth@unaffiliated/kth) has joined #mythtv | |
| [19:25:43] | zombor (zombor!~zombor_@kohana/developer/zombor) has joined #mythtv | |
| [19:30:34] | mrec (mrec!~markus@sundtek.de) has joined #mythtv | |
| [19:30:59] | mrec: | /usr/bin/mythfrontend.real: symbol lookup error: /usr/lib/x86_64-linux-gnu/libva-x11.so.1: undefined symbol: fool_postp |
| [19:31:07] | mrec: | does anyone know how to solve that issue? |
| [19:31:19] | mrec: | libva-x11 seems to be the issue that it doesn't contain those symbols or something like that |
| [19:31:33] | mrec: | everytime installing mythtv on a new ubuntu system new issues are coming up .. |
| [19:38:43] | dmfrey (dmfrey!~dmfrey@webdefence.cluster-h.websense.net) has quit (Remote host closed the connection) | |
| [19:38:52] | danielk22: | mrec: You need to ask for help in the ubuntu user support channel (or at least #mythtv-users, but I think this is probably an ubuntu specific query). |
| [19:39:20] | dmfrey (dmfrey!~dmfrey@webdefence.cluster-h.websense.net) has joined #mythtv | |
| [19:43:20] | stuarta: | well thats the debian build slave migrated to a new home |
| [19:45:07] | ** stuarta hugs kvm virt ** | |
| [19:46:50] | stuarta: | now why is my prod frontend crashing in the audio startup.... |
| [19:52:07] | natanojl (natanojl!~jonatan@c83-252-237-63.bredband.comhem.se) has joined #mythtv | |
| [20:08:05] | stuartm: | stuarta: what's not to like about easy to fix segfaults? :) |
| [20:13:32] | mrand (mrand!~mrand@ubuntu/member/mrand) has quit (Ping timeout: 256 seconds) | |
| [20:14:36] | stuarta: | if it were easy, i'd agree with you |
| [20:14:48] | stuarta: | damn thing is crashing inside libasound |
| [20:14:59] | stuartm: | eugh |
| [20:15:14] | stuarta: | and it's hit many different apps from my quick google |
| [20:16:54] | ** stuarta tries the old make clean; make; make install trick ** | |
| [20:17:08] | mrand (mrand!~mrand@ubuntu/member/mrand) has joined #mythtv | |
| [20:34:24] | stuarta: | why is the audio test pattern white noise? |
| [20:36:19] | stuartm: | Beirdo: cppcheck 1.54 was released a couple of weeks back, I think we can settle on that version/tag for the duration of the 0.26 cycle :) |
| [20:37:34] | Beirdo: | OK, I'll get that updated shortly :) |
| [20:37:34] | stuartm: | stuarta: I think it was simply because white noise can be generated but anything more sophisticated would mean shipping a wav/mp3/ogg file and playing it back, which is a little more involved |
| [20:37:47] | stuarta: | \o/ the old make clean did the trick |
| [20:38:03] | stuartm: | Beirdo: no rush, and thanks for handling it :) |
| [20:39:12] | Beirdo: | no problemo :) |
| [20:39:13] | stuartm: | all the talk about coverity reminded me that it had been a while since I looked at the cppcheck output – I'm just about to investigate one new feature in particular that might be interesting, an option to only check files changed since the last run and therefore speed up checks |
| [20:39:29] | mrand (mrand!~mrand@ubuntu/member/mrand) has quit (Read error: Operation timed out) | |
| [20:39:30] | Jordack (Jordack!~jordack@h69-131-44-221.plmomi.dedicated.static.tds.net) has quit () | |
| [20:39:44] | stuarta: | Beirdo: you need to lart your windows build slave |
| [20:40:14] | Beirdo: | stuarta: will do :) It needs a serious smack far too often |
| [20:40:37] | Beirdo: | stuartm: just go with git master for cppcheck then? |
| [20:40:40] | stuarta: | aye |
| [20:41:11] | zombor (zombor!~zombor_@kohana/developer/zombor) has quit (Ping timeout: 252 seconds) | |
| [20:41:40] | stuartm: | Beirdo: not sure if it's better to stick with the last release tag to avoid new and unfixed false positives |
| [20:42:00] | stuartm: | ah, the 'recheck' option only applies to their GUI client, shame |
| [20:42:32] | Beirdo: | hmm, good point |
| [20:43:17] | Beirdo: | well, it only takes a couple minutes to update it later. I'll put it to the tag for now |
| [20:43:32] | stuartm: | that's not to say that there won't be false positives in the release version, but the hope is that there will be fewer of them |
| [20:47:55] | hashbang (hashbang!~alex@213-152-35-50.dsl.eclipse.net.uk) has quit (Quit: Leaving.) | |
| [20:49:00] | stoffel (stoffel!~quassel@pD9E42742.dip.t-dialin.net) has quit (Remote host closed the connection) | |
| [20:52:49] | mrand (mrand!~mrand@ubuntu/member/mrand) has joined #mythtv | |
| [20:53:25] | NightMonkey (NightMonkey!~NightrMon@pdpc/supporter/professional/nightmonkey) has joined #mythtv | |
| [20:54:04] | Beirdo: | MythBuild: force build cppcheck-master now |
| [20:54:04] | MythBuild: | build forced [ETA 14m14s] |
| [20:54:04] | MythBuild: | I'll give a shout when the build finishes |
| [20:56:55] | MythBuild: | Hey! build cppcheck-master #1055 is complete: Success [3build successful] |
| [20:56:55] | ** MythLogBot http://code.mythtv.org/trac/ticket/1055 ** | |
| [20:56:55] | MythBuild: | Build details are at http://code.mythtv.org/buildbot/builders/cppc . . . /builds/1055 |
| [20:57:18] | Captain_Murdoch: | hmm, now the bots are talking to each other?? |
| [20:57:20] | danielk22: | MythLogBot fail! :P |
| [20:57:20] | Beirdo: | looks like it's working :) |
| [20:57:41] | Beirdo: | bloody hell, another thing to fix in the regexp |
| [20:57:55] | Beirdo: | !trout MythLogBot |
| [20:57:55] | ** MythLogBot slaps MythLogBot with a trout on behalf of Beirdo... ** | |
| [20:58:00] | dmfrey (dmfrey!~dmfrey@webdefence.cluster-h.websense.net) has quit (Quit: Ex-Chat) | |
| [20:58:00] | stuarta: | :-p always the way |
| [20:58:53] | Beirdo: | pretty much |
| [20:59:26] | Beirdo: | got the windows VM rebooting |
| [21:02:47] | obo_ (obo_!~obo@rockbox/developer/obo) has quit (Ping timeout: 246 seconds) | |
| [21:02:55] | stuarta: | i'm experimenting with running the debian build slave as a VM on my main backend :) |
| [21:03:19] | Beirdo: | hehe. my 64-bit slave IS my main backend :) |
| [21:03:48] | stuarta: | same now with the debian one |
| [21:03:49] | kurre2 (kurre2!~tomimo@xdsl-83-150-88-111.nebulazone.fi) has quit (Quit: leaving) | |
| [21:04:05] | obo (obo!~obo@70-140-190-120.lightspeed.hstntx.sbcglobal.net) has joined #mythtv | |
| [21:04:05] | obo (obo!~obo@70-140-190-120.lightspeed.hstntx.sbcglobal.net) has quit (Changing host) | |
| [21:04:05] | obo (obo!~obo@rockbox/developer/obo) has joined #mythtv | |
| [21:05:28] | mrand (mrand!~mrand@ubuntu/member/mrand) has quit (Ping timeout: 245 seconds) | |
| [21:06:22] | tomimo (tomimo!~kurre@xdsl-83-150-88-111.nebulazone.fi) has quit (Quit: Ex-Chat) | |
| [21:06:29] | Steve_Goodey (Steve_Goodey!~steve@host86-182-4-91.range86-182.btcentralplus.com) has quit (Remote host closed the connection) | |
| [21:35:19] | stuartm: | Captain_Murdoch: it's when they start a complete conversation that we need to worry ;) |
| [21:41:52] | cattelan is now known as cattelan_away | |
| [21:47:25] | cattelan_away is now known as cattelan | |
| [21:48:03] | Beirdo: | windows build is borked. |
| [21:48:16] | Beirdo: | configuration.cpp:128:24: error: 'fsync' was not declared in this scope |
| [21:49:23] | Beirdo: | holy crap. my work inbox is 2 emails short of 20k messages |
| [21:49:31] | Beirdo: | I should clean that out sometime |
| [21:49:46] | ** stuarta recommends napalm ** | |
| [21:49:52] | Beirdo: | hehe |
| [21:58:32] | stuartm: | https://github.com/MythTV/mythtv/blob/master/ . . . ext.cpp#L561 << These are a no-brainer to fix but ... |
| [21:59:09] | Beirdo: | let me guess.. m_display is NULL? |
| [21:59:11] | stuartm: | Beirdo: hmm, wonder why we're calling fsync there |
| [21:59:32] | Beirdo: | not sure, I think it's the new config.xml stuff |
| [21:59:36] | stuartm: | Beirdo: nah, the allocations following that are wrong |
| [22:00:15] | Beirdo: | ooh, the memsets :) |
| [22:00:16] | stuartm: | Beirdo: ah right, that makes sense then, we need to ensure config.xml is written immediately to avoid the overwrite issues we had before |
| [22:00:16] | Beirdo: | heheh |
| [22:00:37] | stuartm: | http://code.mythtv.org/cppcheck/index.html – 6 & 7 |
| [22:00:56] | Beirdo: | yup |
| [22:01:01] | stuartm: | I can push a fix, but I can't compile test, which doesn't bother me too much ... |
| [22:02:04] | Beirdo: | yeah, probably worth it anyways |
| [22:02:32] | Beirdo: | as long as it passes visual inspection, of course. |
| [22:06:06] | gigem: | stuartm: re disabled UI types, it's not a high priority, but it'd be nice to have. Please put it on your TODO list to get to whenever. |
| [22:22:04] | danielk22: | Beirdo: stuartm: Yep, the fsyncs are to ensure that if the plug is pulled at just the wrong moment we have either the old or new config.xml, but not something partially written out (or more likely completely blank.) |
| [22:25:02] | natanojl (natanojl!~jonatan@c83-252-237-63.bredband.comhem.se) has quit (Ping timeout: 272 seconds) | |
| [22:25:05] | Beirdo: | danielk22: OK, that makes sense. We just need to make Winblows builds happy :) |
| [22:25:55] | ** stuarta recommends napalm again ** | |
| [22:26:55] | danielk22: | Beirdo: windows doesn't have fsync? sync will work as well, just a bit slow.. |
| [22:27:09] | Beirdo: | hehe, I tend to agree, napalm would be good. |
| [22:27:31] | Beirdo: | it might just need another #include, not sure, but the buildbot slave barfed on it |
| [22:28:05] | Beirdo: | I'm sure it's not a huge rush to fix :) |
| [22:31:13] | yb0t (yb0t!quassel@nat/nokia/x-zucmsjfnabbicilb) has quit (Remote host closed the connection) | |
| [22:31:45] | yb0t (yb0t!quassel@nat/nokia/x-dbkpfjxdqbylsvwq) has joined #mythtv | |
| [22:33:39] | pheld (pheld!xrhloq@109-109-76-195.bb.cust.telefiber.no) has quit (Quit: Leaving.) | |
| [22:35:29] | Beirdo: | stuartm, I think you aren't zeroing enough therer |
| [22:35:51] | Beirdo: | those are arrays of VASurfaceID and vaapi_surface |
| [22:36:27] | Beirdo: | you need m_numSurfaces * sizeof(blah) |
| [22:36:57] | stuartm: | ah crap, yeah, wasn't paying enough attention |
| [22:37:04] | Beirdo: | it happens :) |
| [22:37:32] | Beirdo: | better than just the first 4 or 8 bytes as it was before |
| [22:46:09] | stuartm: | anyway ... plenty of low hanging fruit in http://code.mythtv.org/cppcheck/index.html |
| [22:51:11] | solars (solars!~solars@089144206027.atnat0015.highway.a1.net) has quit (Ping timeout: 255 seconds) | |
| [22:52:11] | zombor (zombor!~zombor_@kohana/developer/zombor) has joined #mythtv | |
| [22:52:34] | Beirdo: | yup, plenty of it |
| [23:10:39] | Beirdo: | MythBuild: force build cppcheck-master now |
| [23:10:39] | MythBuild: | build forced [ETA 6m02s] |
| [23:10:40] | MythBuild: | I'll give a shout when the build finishes |
| [23:11:10] | Beirdo: | I changed the URLs it generates to point to cgit (as it's incredibly faster, especially on large files) |
| [23:12:18] | ** wagnerrp would still rather use the trac browser... if it can be made faster with mysql and repo caching ** | |
| [23:13:07] | wagnerrp: | as it is, there is no comparison |
| [23:13:19] | wagnerrp: | trac is even slower than github |
| [23:13:21] | Beirdo: | well, *if* we can get that working, it's worth considering for sure |
| [23:13:29] | wagnerrp: | although without the nasty javascript issues |
| [23:13:35] | Beirdo: | yeah, and it's server-side slow to boot |
| [23:13:59] | MythBuild: | Hey! build cppcheck-master #1057 is complete: Success [3build successful] |
| [23:13:59] | ** MythLogBot http://code.mythtv.org/trac/ticket/1057 ** | |
| [23:13:59] | MythBuild: | Build details are at http://code.mythtv.org/buildbot/builders/cppc . . . /builds/1057 |
| [23:14:21] | Beirdo: | there |
| [23:14:23] | wagnerrp: | in theory, all that operating in RAM cache in a nice innodb table should work a lot better |
| [23:14:31] | Beirdo: | should, yeah |
| [23:15:05] | wagnerrp: | ill have to feed my test install a copy of the repo and see what happens |
| [23:15:16] | Beirdo: | OK, that works anyways. MUCH faster than github |
| [23:15:32] | Beirdo: | not quite as pretty though, but I think there's something that can be done for that too |
| [23:15:42] | wagnerrp: | cgit didnt appear to have any sort of 'blame'? |
| [23:15:50] | Beirdo: | nope |
| [23:16:13] | Beirdo: | not that I've seen yet |
| [23:18:35] | danielk22: | We're a moderately sized project with a moderate # of developers, it surprises me how poorly github and trac scale for us. I would think we'd be exactly the type of project they would optimize for. |
| [23:18:57] | Beirdo: | yeah, agreed |
| [23:19:10] | wagnerrp: | were a pretty massive project |
| [23:19:57] | wagnerrp: | most open source projects dont exceed a couple hundred thousand lines |
| [23:20:14] | wagnerrp: | including ffmpeg and the themes, were over 2M lines |
| [23:20:35] | danielk22: | wagnerrp: Just a couple million lines of code, about 10% the size of some code bases I've worked on. |
| [23:20:56] | Beirdo: | for opensource, it's fairly large though. |
| [23:23:24] | stuartm: | Beirdo: re appearance, googling suggests there are syntax highlighting plugins for cgit and I'd guess it can be themed to some extent too |
| [23:23:59] | Beirdo: | yeah, I'm looking at putting in a highlighting plugin to make it look nicer :) |
| [23:27:06] | stuartm: | btw what happened to the irc commit hook? |
| [23:27:32] | Beirdo: | the one I tried is very unresiliant... |
| [23:27:57] | Beirdo: | freenode disconnected it for a flood once, and it never even tried to reconnect |
| [23:28:02] | Beirdo: | not pleased :) |
| [23:28:20] | Beirdo: | I'll try to get that fixed too... thanks, I forgot about it |
| [23:28:58] | Beirdo: | it shouldn't have flooded in the the first place. |
| [23:29:08] | stuartm: | ah, ok, not really essential but it was nice to have, easier at times to follow the flow of commits than my email client |
| [23:29:15] | Beirdo: | yeah |
| [23:29:35] | Beirdo: | I liked having it too. Just need to get it behaving. |
| [23:30:08] | Beirdo: | I might try warming up the RSS-feed plugin in beirdobot, and point it at a feed from cgit or trac or... |
| [23:30:16] | danielk22: | wagnerrp: I don't think we're small, just smaller than KDE, Gnome, Linux, etc. We're in that middle ground were Cinelarra, gimp, openoffice, and many other useful projects live. There are of course hundreds of thousands of smaller projects in the long tail.. |
| [23:30:23] | stuartm: | I need to get the theme validation stuff finished so we might get that operational before the 0.26 release |
| [23:30:36] | Beirdo: | that would be cool |
| [23:31:32] | stuartm: | it would be simpler if xmllint had an xml output mode, seems ironic that it doesn't |
| [23:31:43] | Beirdo: | hehe |
| [23:31:52] | Beirdo: | s/ironic/idiotic/ |
| [23:33:19] | danielk22: | found this for fsync on windows... http://lists.gnu.org/archive/html/bug-gnulib/ . . . sg00004.html |
| [23:33:20] | stuartm: | it's not difficult to parse, but something like XML::Simple trumps a bunch of regexps |
| [23:34:53] | Beirdo: | danielk22: cool, looks like something we could add to compat.h and whatever .cpp |
| [23:42:43] | danielk22: | Beirdo: I think we may just need to add "#include unistd.h" I think this was added to MinGW 4 years ago according to this e-mail exchange. |
| [23:42:53] | Beirdo: | oh, even better |
| [23:43:48] | danielk22: | Hmm, unistd.h is already included in configuration.cpp |
| [23:45:10] | kth (kth!~kth@unaffiliated/kth) has quit (Read error: Connection reset by peer) | |
| [23:52:03] | amejia_ (amejia_!~andres@129.174.97.130) has joined #mythtv | |
| [23:52:03] | amejia_ (amejia_!~andres@xbmc/staff/amejia) has joined #mythtv | |
| [23:52:03] | amejia_ (amejia_!~andres@129.174.97.130) has quit (Changing host) | |
IRC Logs collected by
BeirdoBot.
Please use the above link to report any bugs.
