| Wednesday, February 15th, 2012, 00:02 UTC | ||
| [00:02:15] | sphery: | stuartm: Well, IMHO, we should be fixing our broken class designs that allow invalid data into objects (nulls in not-null fields), but that's a whole other argument. (And, really, doing a: query.bindValue("STRING", value.isEmpty() ? "" : value); would be the same as an isNULL() check--whether empty or null, we want to use empty string, but as I originally said, "where possible". :) |
| [00:03:01] | sphery: | superm1: re: rsyslog permissions problem... I just need someone who knows what ubuntu's logging is set at to change the owner/group values for http://www.mythtv.org/wiki/Rsyslog_Configuration to something appropriate |
| [00:03:50] | sphery: | you can use sys:admin or you can modify sys or admin to have access to mythtv group and change dir to 0775 and file to 0664 or ... |
| [00:05:03] | Jester86_Mobile (Jester86_Mobile!~Jesse@cpe-174-097-199-056.nc.res.rr.com) has joined #mythtv | |
| [00:05:07] | sphery: | whatever works best for the distro is what we should have on that page--since it will probably be mainly ubuntu users who need that config page |
| [00:05:13] | Jester86_Mobile: | hello all |
| [00:05:35] | superm1: | sphery: yeah what we're shipping is what was on that page. i'm not familiar with rsyslog enough yet myself |
| [00:05:40] | superm1: | it looks to me that it runs as the syslog user |
| [00:06:03] | superm1: | so it might just be a matter of setting the directory to have the right permissions for the syslog user rather than mythtv group |
| [00:06:22] | Jester86_Mobile: | I'm not sure if this is the location to post an idea for future releases or not but.. would it be possible to add the option (at the frontend level) to exit and reboot after current recording |
| [00:06:58] | Jester86_Mobile: | have it instantly reboot if there are no recordings happening.. if there are recordings in process then wait until they complete to reboot |
| [00:07:30] | Jester86_Mobile: | may need to reword how it appears on the screen as not to sound like it relies on a recording to be ongoing to do anything but you get the point |
| [00:07:54] | Jester86_Mobile: | obviously this would be useful only to FE/BE boxes |
| [00:08:01] | sphery: | each user I've seen who asked about it not working was first told enough by someone else to let them fix it in such a way I couldn't see what state things were in when not working before they read my request for directory listing information, so I don't know, but it seems that there's a sys user and admin group involved in file ownership of log files on your systems... |
| [00:09:10] | superm1: | i'm going to upgrade one of my boxes soon to 0.25 so i'll try to make sense of it |
| [00:09:48] | sphery: | chances are that by the time it creates the /etc/syslog/mythtv directory, syslog has already dropped permissions to the point that it can't create a directorory owned by mythtv:mythtv... meaning you probably need to switch it to the default owner/group (meaning you can just delete that top section and the bottom section and all will work properly) |
| [00:10:04] | superm1: | ah |
| [00:10:16] | superm1: | which would probably be a fine solution then |
| [00:10:17] | ** jya still trying to decipher sphery earlier message ** | |
| [00:10:23] | sphery: | there may be something more elegant you can do that marks them as more mythtv-like, but the simplest fix is likely to just get rid of those things |
| [00:10:31] | superm1: | no need to restrict who can read logs to mythtv group imo |
| [00:11:02] | sphery: | yeah, well we have world exec on /etc/syslog/mythtv dir and world read on log files, anyway, so we're not really restricting |
| [00:11:04] | jya: | I can set the /var/log/mythtv directory to 777 and see if it starts logging again |
| [00:11:11] | jya: | then we'll know if it's a permission issue |
| [00:11:15] | Jester86_Mobile (Jester86_Mobile!~Jesse@cpe-174-097-199-056.nc.res.rr.com) has left #mythtv ("Leaving") | |
| [00:11:58] | superm1: | jya: actually if you've got a busted system right now, try doign what sphery said about just taking the top section out of that rsyslog config |
| [00:12:05] | superm1: | if that sorts it out, then probably just go with that solution |
| [00:12:20] | jya: | where is it configured ? |
| [00:12:35] | superm1: | /etc/rsyslog.d/40-mythtv.conf |
| [00:13:06] | jya: | well, /var/log/mythtv is in 777 mode, and still no log being created |
| [00:13:27] | sphery: | jya: better test... just delete /var/log/mythtv dir and its contents (or move them out of the way if you want to preserve old log files) and then use http://pastebin.com/7PA64gTN for your /etc/rsyslog.d/40-mythtv.conf , then restart/reload rsyslog, then restart mythtv apps and ensure dirs/files are properly created |
| [00:14:24] | sphery: | if that works, I'll change the wiki config to remove the owner/group stuff and it will just use syslog defaults |
| [00:15:14] | superm1: | and i'll update the packaging to do that too |
| [00:16:16] | sphery: | superm1: you might also need to have the package (at least on upgrade?) remove any existing /var/log/mythtv dir, since that will almost certainly already be set up to use mythtv:mythtv since 0.24 and below logs were written directly by the user running mythtv apps |
| [00:16:35] | superm1: | sphery: ah yeah, that should be do-able on the postinst script |
| [00:16:55] | sphery: | "at least on upgrade" meaning from 0.24, but might make sense to do it for all upgrades to clear out any broken stuff for those who have been on master for a while |
| [00:17:14] | superm1: | right |
| [00:17:27] | sphery: | or at least have the package chown the relevant files--which might be nice to prevent losing old logging? |
| [00:17:33] | sphery: | files and dirs |
| [00:17:50] | jya: | sphery: restarted mythbackend |
| [00:18:08] | sphery: | *fingers crossed* |
| [00:18:13] | jya: | I had moved the /var/log/mythtv into /var/logs/mythtv.old |
| [00:18:22] | jya: | no mythtv dir created in logs |
| [00:18:43] | sphery: | and your mythbackend was started with --syslog local7 |
| [00:18:54] | jya: | ps ax | grep mythbackend |
| [00:18:59] | jya: | /usr/bin/mythbackend --syslog local7 --user mythtv |
| [00:19:15] | superm1: | did you restart rsyslog too though? |
| [00:19:30] | jya: | did anyone tell me to restart syslog ? :P |
| [00:19:37] | superm1: | haha |
| [00:19:43] | sphery: | hehe, it was buried in my novel :) |
| [00:20:18] | jya: | how do you restart it ? doesn't seem to have a service called rsyslog |
| [00:20:30] | jya: | never mind |
| [00:20:41] | jya: | for some reason autocompletion didn't work when typing service |
| [00:21:03] | sphery: | you'll also need to re-restart mythbackend |
| [00:21:08] | jya: | # service rsyslog restart |
| [00:21:08] | jya: | rsyslog start/running, process 3223 |
| [00:21:14] | jya: | service mythtv-backend start |
| [00:21:14] | jya: | mythtv-backend start/running, process 3273 |
| [00:21:27] | jya: | # ls -l /var/log/mythtv |
| [00:21:27] | jya: | ls: cannot access /var/log/mythtv: No such file or directory |
| [00:21:31] | jya: | so no go |
| [00:21:40] | jya: | even after restarting rsyslog |
| [00:22:38] | sphery: | can you pastebin: ls -ld /var/log{,/*,/mythtv.old{,/*}} |
| [00:22:56] | jya: | looking at the rsyslog conf , where is mythtvFile defined ? |
| [00:23:04] | sphery: | $template mythtvFile,"/var/log/mythtv/%programname%.log" |
| [00:23:10] | sphery: | line 8 |
| [00:23:27] | jya: | it's commented out no? |
| [00:23:39] | jya: | ahhh |
| [00:23:46] | jya: | hence when the ensure one is uncommented |
| [00:23:49] | sphery: | (looks like it's commented, because the $ at the beginning is a non-alphanumeric character, but it's actually for defining the $template variable) |
| [00:24:29] | jya: | never mind what I wrote… something isn't right with me today |
| [00:25:18] | jya: | http://pastebin.com/YyM83CGm |
| [00:26:09] | sphery: | and which user is rsyslog running as? |
| [00:27:01] | jya: | root@htpc:/etc/rsyslog.d# ps aux | grep rsyslog |
| [00:27:01] | jya: | syslog 3223 0.0 0.0 123820 1464 ? Sl 11:20 0:00 rsyslogd -c4 |
| [00:27:10] | jya: | so syslog |
| [00:28:00] | jya: | this is on 10.04 LTS if that makes a difference |
| [00:29:05] | sphery: | so looks like the problem is that /var/log is root:root at drwxr-xr-x , and rsyslog is running as syslog user, who doesn't have permission to write to /var/log |
| [00:30:29] | sphery: | so it means that the package needs to create a /var/log/mythtv directory with appropriate ownership/permissions (I'd suggest syslog:adm drwxr-xr-x) and then just get rid of the owner/group sections in rsyslog.conf |
| [00:31:00] | sphery: | jya: if you do: sudo install -d -m755 -o syslog -g adm /var/log/mythtv |
| [00:31:14] | jya: | I had already done so.. |
| [00:31:21] | jya: | nothing created in /var/log/mythtv |
| [00:31:35] | sphery: | then restart rsyslog (just in case) and mythbackend, does it work with the current rsyslog config (the one without the owner/group sections) |
| [00:31:38] | jya: | oh never mind |
| [00:31:47] | jya: | it has created a mythbackend.log now |
| [00:31:51] | pla (pla!~htplainf@99-95-202-230.lightspeed.chtnsc.sbcglobal.net) has joined #mythtv | |
| [00:31:54] | sphery: | yay... |
| [00:31:55] | superm1: | great! |
| [00:32:05] | superm1: | okay so end solution will be updated config and set permissions to syslog:adm then |
| [00:32:07] | jya: | does mythfrontend use the same rsyslog mechanism ? |
| [00:32:28] | jya: | file created is owned by rsyslog:rsyslog |
| [00:32:41] | jya: | in 640 mode |
| [00:33:03] | superm1: | ls |
| [00:33:05] | sphery: | do other /etc/rsyslog.d/*.conf have owner/group sections? |
| [00:33:05] | superm1: | oops |
| [00:33:06] | jya: | i think log should be readable by all for troubleshooting perspective |
| [00:33:43] | jya: | what do they look like ? |
| [00:33:55] | sphery: | that implies that your /etc/syslog.conf has default set to rsyslog:rsyslog @ 0640, but the list of log directory info says other logs aren't using those defaults |
| [00:34:24] | sphery: | the $FileOwner syslog\n$FileGroup adm type stuff (the top and bottom sections of http://www.mythtv.org/wiki/Rsyslog_Configuration |
| [00:34:37] | jya: | I don't see anything in the conf in regards to group / user ownership or permission |
| [00:35:01] | superm1: | the top level /etc/rsyslog.conf (before the ones in /etc/rsyslog.conf.d are sourced) sets the default |
| [00:35:09] | superm1: | which has $FileCreateMode 0640 |
| [00:35:21] | jya: | $FileOwner syslog |
| [00:35:22] | jya: | $FileGroup adm |
| [00:35:22] | jya: | $FileCreateMode 0640 |
| [00:35:31] | jya: | dir createmode is 755 |
| [00:35:56] | sphery: | jya: that's in syslog.conf? |
| [00:35:59] | jya: | yes |
| [00:36:03] | jya: | rsyslog.conf |
| [00:36:26] | gregL (gregL!~greg@cpe-74-76-125-87.nycap.res.rr.com) has quit (Read error: Connection reset by peer) | |
| [00:36:31] | sphery: | weird... wonder why we didn't inherit that... I'm guessing there's a /etc/rsyslog.d/*.conf that comes before 40 that sets it to rsyslog:rsyslog |
| [00:38:02] | superm1: | other weird thing is that there is no rsyslog user (at least on oneiric). maybe that was only on lucid |
| [00:38:25] | sphery: | jya: can you try http://www.mythtv.org/wiki/Rsyslog_Configuration configuration, then delete files in /var/log/mythtv (but leave the directory at syslog:adm @ 0755) |
| [00:38:37] | sphery: | then restart rsyslog and mythbackend and see if it looks better |
| [00:47:46] | sphery: | superm1: so, if that config works (with syslog:adm), looks like the package should use http://www.mythtv.org/wiki/Rsyslog_Configuration configuration file and should: chown -R syslog:adm /var/log/mythtv && chmod 0755 /var/log/mythtv && chmod 0644 /var/log/mythtv/* |
| [00:48:16] | sphery: | jya / superm1 : thanks for the help figuring this out. I really want logging to work for Ubuntu users, so really appreciate your time. |
| [00:52:40] | jya: | sorry, was trying a bug reported to me in Lion |
| [00:53:47] | gregL (gregL!~greg@cpe-74-76-125-87.nycap.res.rr.com) has joined #mythtv | |
| [00:54:19] | jya: | Lion has introduced an oddities when playing passthrough. You can play passthrough once, then you have to exit the app and come back as otherwise the audio device is hogged for digital forever and isn't usable anymore |
| [00:55:08] | jya: | same with vlc |
| [01:05:07] | stichnot (stichnot!chatzilla@nat/intel/x-mqidxysegdamcuwo) has quit (Ping timeout: 276 seconds) | |
| [01:05:54] | jpabq_ (jpabq_!~jpabq@mythtv/developer/jpabq) has quit (Quit: jpabq_) | |
| [02:04:44] | davide (davide!~david@host70.16.intrusion.com) has quit (Remote host closed the connection) | |
| [02:05:07] | davide (davide!~david@host70.16.intrusion.com) has joined #mythtv | |
| [02:25:16] | jya: | sphery: actually, I just looked into mythbackend.log |
| [02:25:35] | jya: | only looking if it had been created |
| [02:25:41] | jya: | it's actually empty |
| [02:28:31] | sphery: | sounds like it's able to create it, but not able to write to it... this is with syslog:adm@0644 and the http://www.mythtv.org/wiki/Rsyslog_Configuration that sets fileowner/group to syslog/adm? |
| [02:28:41] | jya: | I'm restarting from 0 |
| [02:28:48] | jya: | let see if it works any better |
| [02:29:36] | jya: | i used the config from the wiki as-is |
| [02:29:59] | jya: | the file mythbackend.log is created , owned by syslog:syslog, in 644 more |
| [02:30:01] | jya: | mode |
| [02:30:11] | jya: | byt it stays with a size of 0 |
| [02:30:24] | sphery: | syslog:syslog with the new config in http://www.mythtv.org/wiki/Rsyslog_Configuration ? |
| [02:30:30] | sphery: | the one that says to use syslog:adm? |
| [02:30:40] | jya: | directory /var/log/mythtv was created by syslog:adm chmod of 755 |
| [02:30:45] | jya: | that's right |
| [02:30:48] | jya: | I restarted rsyslog |
| [02:31:14] | jya: | oh.. my fault |
| [02:31:25] | jya: | i didn't uncomment any of the line |
| [02:31:36] | jya: | ooops |
| [02:31:39] | jya: | it is uncommented |
| [02:31:47] | jya: | so now, it's exactly as in the wiki |
| [02:31:56] | jya: | specifying the syslog user |
| [02:32:31] | jya: | actually, user running syslog is "syslog" |
| [02:32:47] | jya: | and is a member of syslog only |
| [02:32:53] | jya: | it's not a member of adm |
| [02:33:03] | jya: | so it can't create a file that is of group adm |
| [02:33:27] | jya: | that is if rsyslog is owned as user syslog |
| [02:34:17] | sphery: | well, does syslog:syslog work? I was just trying to make it match the other syslog-owned files in your /var/log |
| [02:34:46] | jya: | looking on the wiki |
| [02:35:01] | jya: | the bottom part of the script needs to be copied in rsyslog.conf right? |
| [02:35:20] | sphery: | would just mean changing $FileGroup adm and $DirGroup adm to syslog |
| [02:35:24] | jya: | otherwise you set FileOwner twice in that file |
| [02:35:37] | sphery: | no, the bottom part of the script should be identical to what's already in rsyslog.conf |
| [02:35:45] | jya: | haven't looked in details on how rsyslog works |
| [02:36:02] | sphery: | i.e. it sets things back to the system defaults after setting up our application's settings |
| [02:36:03] | jya: | it isn't |
| [02:36:43] | jya: | this is what rsyslog.conf look on my system |
| [02:36:44] | jya: | http://pastebin.com/QHGXMsiz |
| [02:38:37] | jya: | ok.. so I've changed it to match exactly what was in rsyslog.conf |
| [02:39:07] | sphery: | hehe... https://bugs.launchpad.net/ubuntu/+source/rsyslog/+bug/484336 |
| [02:39:28] | TheAsp (TheAsp!~asp@blk-222-139-8.eastlink.ca) has quit (Read error: Connection reset by peer) | |
| [02:40:05] | jya: | jeez, how long can they keep a bup open for? |
| [02:41:40] | jya: | still, you define the DirOwner and DirGroup in the mythtv rsyslog conf file |
| [02:41:51] | jya: | but those are not defined in the rsyslog.conf one |
| [02:42:18] | jya: | so it's hard to make them "agree" with the main rsyslog.conf file as specified in the wiki |
| [02:42:40] | sphery: | yeah, looks like we need to use syslog:syslog for file and dir, though... the bottom section can be whatever--because rsyslog default configuration in Ubuntu is broken, anyway, so setting back to default probably isn't that important :) |
| [02:44:04] | jya: | i'm applying the patch in that ticket |
| [02:44:05] | jya: | will see |
| [02:44:09] | jya: | what a pain... |
| [02:45:31] | jya: | i can't believe a bug can stay like this for 2.5 years and not being actioned on.. I thought only Apple did things that way |
| [02:46:09] | zombor (zombor!~zombor_@kohana/developer/zombor) has quit (Remote host closed the connection) | |
| [02:47:30] | jya: | not sure what I changed, but now the file isn't even created in in the mythtv log directory |
| [02:48:40] | sphery: | I'm thinking http://www.mythtv.org/wiki/Rsyslog_Configuration should work, even with broken rsyslog configuration on Ubuntu, if we pre-create the /var/log/mythtv directory with syslog:syslog @ 0755 |
| [02:49:53] | jya: | i believe so too |
| [02:49:57] | jya: | anyhow.. this works now |
| [02:50:17] | sphery: | hehe, well glad you got it working for you |
| [02:50:28] | sphery: | we've at least made some progress for the generic case, too |
| [02:50:42] | jya: | well halfway there |
| [02:50:48] | jya: | mythfrontend doesn't create a log in there |
| [02:51:15] | jya: | ah |
| [02:51:20] | jya: | looking at how it is started |
| [02:51:32] | jya: | the log file is now in /tmp/mythfrontend.PID.log |
| [02:51:41] | jya: | superm1: that's one for you I think |
| [02:52:17] | sphery: | mythfrontend should only be creating log files if you use --logpath or --logfile to start mythfrontend (rather than using --syslog user7 ) |
| [02:55:17] | stichnot (stichnot!~chatzilla@192.55.54.40) has joined #mythtv | |
| [02:55:20] | jya: | that's what it does |
| [02:55:28] | jya: | it is started with /usr/bin/mythfrontend.real --logfile /tmp/mythfrontend.24394.log |
| [02:55:50] | jya: | 23494 being the pid of the mythfrontend wrapper that started it |
| [02:56:26] | sphery: | so, another option for logging with rsyslog is to do what ubuntu must be doing for the rest of the apps... rather than use a template to dynamically choose the log file name, we could hard code in a file name for each app, then have the package pre-create the files with permissions that rsyslog can use |
| [02:56:48] | zombor (zombor!~zombor_@kohana/developer/zombor) has joined #mythtv | |
| [02:56:51] | sphery: | superm1: if you want me to create such a config, please let me know... this is due to https://bugs.launchpad.net/ubuntu/+source/rsyslog/+bug/484336 and https://bugs.launchpad.net/ubuntu/+source/rsyslog/+bug/407862 |
| [02:57:12] | sphery: | (which basically mean that templated file/directory creation is broken in Ubuntu) |
| [03:03:15] | sphery: | er, above I meant --syslog local7 (for some reason I always want to say user 7) |
| [03:31:12] | plink212 (plink212!~jircii@S0106c0c1c01b51cf.vc.shawcable.net) has joined #mythtv | |
| [03:35:29] | plink212: | the commit earlier for getserverbackendip seems to have stopped me watching new recordings |
| [03:35:39] | plink212: | old ones are fine |
| [03:40:19] | wagnerrp: | plink212: could you look through your logs and see what address it is trying to connect to to stream the recordings? |
| [03:42:40] | plink212: | I get mythcorecontext.cpp:1007 (GetBackendServerIP) – No address defined for host: hybrid |
| [03:43:12] | plink212: | in mythtv setup i have an ipv4 address set |
| [03:45:01] | wagnerrp: | what do you get from... "select data from settings where value like 'BackendServerIP%' and hostname='hybrid';"? |
| [03:46:27] | plink212: | 172.16.1.78 |
| [03:47:17] | stichnot: | Is anyone able to reproduce issues 1 and 3 from #10333? (The test file is the 650MB http://ncamftp.wgbh.org/DTV/CEA%20test%20mate . . . v1.2zero.trp .) |
| [03:49:41] | wagnerrp: | plink212: so BackendServerIP6 is ::1? |
| [03:50:07] | plink212: | should line 1000 in mythcorecontext be isempty rather than just empty |
| [03:51:14] | wagnerrp: | same thing |
| [03:51:37] | plink212: | backendserverip6 = ::1 |
| [03:54:31] | sphery: | superm1: I have a new simpler approach to rsyslog logging at http://www.mythtv.org/wiki/Rsyslog_Configuration that I recommend using for Ubuntu because of the above-mentioned bugs. You'll need to ensure that the configuration specified in the comment at top is taken care of by the packages (either using the install commands I gave--which will truncate any existing log files with the given names--or using touch/chown/chmod type approach). (I ... |
| [03:54:37] | sphery: | ... haven't actually tested it, but am pretty sure it's correct. I'll probably give it an initial sanity check tomorrow.) |
| [04:13:49] | kurre2 (kurre2!~tomimo@xdsl-83-150-88-111.nebulazone.fi) has quit (Quit: leaving) | |
| [04:14:00] | plink212: | so that recording has now finished, and it plays fine. it seems to only be in progress recordings that have the issue |
| [04:15:04] | tomimo_ (tomimo_!~kurre@xdsl-83-150-88-111.nebulazone.fi) has quit (Read error: Connection reset by peer) | |
| [04:18:34] | zombor (zombor!~zombor_@kohana/developer/zombor) has quit (Remote host closed the connection) | |
| [04:22:14] | kurre2 (kurre2!~tomimo@xdsl-83-150-88-111.nebulazone.fi) has joined #mythtv | |
| [04:31:02] | Beirdo: | stichnot: you got a moment? |
| [04:37:47] | Computer_Czar (Computer_Czar!~Unknown@69.4.155.83) has joined #mythtv | |
| [04:42:18] | jpabq: | I am getting random crashes in the channel icon downloader. Looks like it is trying to deal with a redirect... |
| [04:44:59] | Beirdo: | did that code get changed lately? I think it has |
| [04:50:00] | stichnot (stichnot!~chatzilla@192.55.54.40) has quit (Ping timeout: 252 seconds) | |
| [04:55:10] | jpabq: | I think so. I want to say it was stuartm, but I don't know why. Captain_Murdoch makes more sense. |
| [04:57:45] | jpabq: | Version is from several days ago. I will try updating later, and reproduce. |
| [04:57:45] | superm1: | sphery: okay i'll get things adapted to that instead. unfortunate that the dynamic creating files stuff isn't working |
| [04:59:41] | superm1: | jya: alsa lib v 1.0.25 is already in precise. do you know what part of ALSA 1.0.25 is what breaks things? it sounds like it's one of the kernel driver pieces. the kernel in precise has mostly ALSA 1.0.24. it might be best that you load up a precise install and see how things look, potentially with the ubuntu audio dev PPA that has all the rest of the pieces that haven't yet landed for 1.0.25 i |
| [04:59:41] | superm1: | n precise |
| [05:00:05] | superm1: | if things are at all broke, now would be the right time to share with those commits that fix the problems |
| [05:00:14] | jya: | superm1: it's alsa-drivers |
| [05:00:18] | jya: | the actual kernel modules |
| [05:00:27] | superm1: | eg the pieces that are synced into various kernel versions then |
| [05:00:49] | sphery: | superm1: well, the fixed file approach has the benefit that it will work regardless of chosen syslog facility and it prevents non-mythtv messages from getting into mythtv logs when other apps use the same facility |
| [05:01:02] | superm1: | sphery: ah that makes sense then |
| [05:01:13] | ** sphery always looks for the silver lining :) ** | |
| [05:02:43] | Computer_Czar (Computer_Czar!~Unknown@69.4.155.83) has quit (Quit: Leaving for a long time ... see you later!) | |
| [05:06:58] | stichnot (stichnot!~chatzilla@192.55.54.40) has joined #mythtv | |
| [05:08:19] | stichnot: | Beirdo: back now. |
| [05:08:32] | Beirdo: | K, you got my email? |
| [05:08:55] | stichnot: | ah yeah, I see it now |
| [05:09:06] | Beirdo: | Cool. I think you are pretty much set :) |
| [05:09:20] | stichnot: | cool, thanks! |
| [05:09:26] | plink212 (plink212!~jircii@S0106c0c1c01b51cf.vc.shawcable.net) has quit (Quit: jIRCii - http://www.oldschoolirc.com) | |
| [05:10:40] | Beirdo: | any time, man. |
| [05:11:18] | superm1: | jya: okay so here's that commit that broke things for you: http://kernel.ubuntu.com/git?p=ubuntu/ubuntu- . . . b0ab7070ad0e |
| [05:11:23] | superm1: | it's in the precise kernel right now |
| [05:13:27] | superm1: | and i suspect this is the fix you were looking for: http://kernel.ubuntu.com/git?p=ubuntu/ubuntu- . . . 4b491fd678fd |
| [05:13:32] | superm1: | which is already in the kernel as well |
| [05:13:36] | jya: | superm1: that's the one |
| [05:14:36] | jya: | it will be in the kernel, but will it be in the kernel shipping with myth? |
| [05:14:55] | superm1: | that's in the kernel that's going to be shipping with precise |
| [05:15:22] | superm1: | so yeah shouldn't be any concerns. other distros will be picking up that patch too with kernel 3.2 (i think is when it was tagged) |
| [05:26:38] | MaverickTech (MaverickTech!~MaverickT@111.86.233.220.static.exetel.com.au) has joined #mythtv | |
| [05:29:21] | wahrhaft_ (wahrhaft_!~quassel@cpe-24-210-71-26.columbus.res.rr.com) has quit (Read error: Operation timed out) | |
| [05:30:35] | wahrhaft (wahrhaft!~quassel@cpe-24-210-71-26.columbus.res.rr.com) has joined #mythtv | |
| [05:30:42] | xris (xris!~xris@mythtv/developer/xris) has quit (Ping timeout: 252 seconds) | |
| [05:31:03] | xris (xris!~xris@xris.forevermore.net) has joined #mythtv | |
| [05:34:03] | xris (xris!~xris@xris.forevermore.net) has quit (Changing host) | |
| [05:34:03] | xris (xris!~xris@mythtv/developer/xris) has joined #mythtv | |
| [05:36:15] | superm1: | sphery: i think this will need a little bit more experimentation. i just looked at the ufw rule in rsyslog.d, and all it has is ":msg,contains,"[UFW " /var/log/ufw.log" |
| [05:36:26] | superm1: | and i can't find anything that actually creates that log at package creation time |
| [05:36:58] | superm1: | it's created with these permissions '-rw-r----- 1 syslog adm' and seems to be doing the right thing |
| [05:42:03] | sphery: | the pre-create may not be necessary with fixed file names since they'll likely be created before rsyslog drops privileges, but if nothing else, the package will have to change permissions on the existing /var/log/mythtv and files within |
| [05:42:31] | sphery: | (since they'll be mythtv:mythtv from 0.24 and below) |
| [05:43:11] | superm1: | hopefully they won't have to be pre-created, i'll see though |
| [05:43:16] | superm1: | i'm upgrading my 0.24 to 0.25 right now |
| [05:44:24] | sphery: | yeah, will probably create them no problem with fixed file names, but if any exist, already, they could prevent it from writing to them |
| [05:44:59] | superm1: | if that's all that is the problem i'll just set up a postinst rule to chown everything in the directory ahead of time |
| [05:45:08] | sphery: | oh, and just remembered that ubuntu's default FileCreateMode is 0640, so we'd need to add an override in our file |
| [05:47:45] | sphery: | added FileCreateMode back in to http://www.mythtv.org/wiki/Rsyslog_Configurat . . . onfiguration |
| [05:48:08] | sphery: | superm1: please try yours with ^^^ (it should make sure the files are world readable) |
| [05:49:37] | superm1: | it looks like just restarting rsyslog with that file went and created all the files for me |
| [05:49:43] | superm1: | and the ones that existed already it chowned |
| [05:49:54] | sphery: | and note that rsyslog won't change owner/group/mode for any existing files, so you'll need to ensure they're all reasonable |
| [05:50:02] | sphery: | is it using syslog:adm for the files? |
| [05:50:13] | superm1: | yup |
| [05:51:07] | sphery: | ok, I'll update the "pre-create" instructions to use syslog:adm, but I'm leaving them in the wiki page as a simple way to ensure files are the right ownership/permissions. You can leave those comments out of the in-package config if you like |
| [05:51:17] | superm1: | ok |
| [05:51:37] | superm1: | i'll certainly be doing more experimentation on other systems too |
| [05:52:15] | superm1: | i didn't have to change directory permissions either though, rsyslog seems to have handled that too |
| [05:52:20] | jya: | superm1: can you put back the mythfontend log in /var/log/mythtv too ? |
| [05:52:26] | superm1: | jya: yeah i'm gonna fix that too |
| [05:52:30] | jya: | rather than in /tmp |
| [05:53:09] | jya: | sphery: do you have to create all the files by hand? |
| [05:53:24] | jya: | cause if that's the case, when it rolls over and backup them, it's not going to work |
| [05:53:36] | superm1: | jya: it seemed that rsyslog created them for me |
| [05:53:41] | jya: | how is it going to create .log.1 etc.. if it can't create .log to start with |
| [05:53:47] | superm1: | i think that the logrotate job needs to reinvoke rsyslog |
| [05:53:52] | superm1: | i saw it doing that for another app |
| [05:54:09] | jya: | superm1: i saw the same thing, I didn't manually create any of those files |
| [05:54:55] | sphery: | jya: with fixed filenames it will create non-existent before dropping privileges, so it will do it right, but you do need to change ownership/permissions on any existing files with the same name |
| [05:55:14] | sphery: | the commands in that section will do that for you: http://www.mythtv.org/wiki/Rsyslog_Configurat . . . onfiguration (just needs to be run once) |
| [06:03:50] | sphery: | ok, description in the page is much better, now. thanks for the testing, both of you |
| [06:33:13] | stoffel (stoffel!~quassel@pD9E4265D.dip.t-dialin.net) has joined #mythtv | |
| [06:37:48] | danielk22 (danielk22!~danielk@96.57.9.142) has joined #mythtv | |
| [06:38:23] | danielk22: | stuartm: FYI The delete problem has re-occurred.. |
| [06:42:26] | danielk22: | Beirdo: Was the abort in 7c1b73704 intentionally committed or debugging left in? |
| [06:42:34] | Beirdo: | I just noticed |
| [06:42:37] | Beirdo: | heh, it's gone |
| [06:42:55] | Beirdo: | I noticed just about exactly when you said it :) |
| [06:43:28] | danielk22: | :) |
| [06:49:05] | rsiebert_ (rsiebert_!~quassel@e179131036.adsl.alicedsl.de) has quit (Remote host closed the connection) | |
| [06:50:53] | rsiebert (rsiebert!~quassel@e179131036.adsl.alicedsl.de) has joined #mythtv | |
| [07:12:14] | pheld (pheld!~heldal@cl-5.osl-01.no.sixxs.net) has quit (Quit: Leaving.) | |
| [07:46:15] | stoffel (stoffel!~quassel@pD9E4265D.dip.t-dialin.net) has quit (Read error: Connection reset by peer) | |
| [08:19:16] | netw1z (netw1z!~overseas_@cpe-72-225-175-91.nyc.res.rr.com) has quit () | |
| [08:19:37] | Chutt (Chutt!~ijr@cpe-24-29-225-175.neo.res.rr.com) has quit (Read error: Connection reset by peer) | |
| [08:27:36] | Robert_S (Robert_S!14852855@gateway/web/freenode/ip.20.133.40.85) has joined #mythtv | |
| [09:02:56] | Eruphus (Eruphus!~quassel@89.245.155.154) has joined #mythtv | |
| [09:17:36] | Memphis (Memphis!~Memphisau@124-171-212-254.dyn.iinet.net.au) has quit (Ping timeout: 245 seconds) | |
| [09:23:59] | Memphis (Memphis!Memphisau@203-206-255-191.dyn.iinet.net.au) has joined #mythtv | |
| [09:36:21] | xavierh (xavierh!~xavier@cpc1-swin3-0-0-cust274.3-1.cable.virginmedia.com) has joined #mythtv | |
| [10:01:28] | mrand (mrand!~mrand@ubuntu/member/mrand) has quit (Read error: Connection reset by peer) | |
| [10:02:18] | Eruphus (Eruphus!~quassel@89.245.155.154) has quit (Read error: Connection reset by peer) | |
| [10:11:46] | jeff999 (jeff999!~jeff@203-173-13-174.dyn.iinet.net.au) has joined #mythtv | |
| [10:32:38] | jeff999 (jeff999!~jeff@203-173-13-174.dyn.iinet.net.au) has quit (Quit: Leaving) | |
| [11:05:02] | mike|3 (mike|3!~mike@c-76-115-119-121.hsd1.or.comcast.net) has quit (Remote host closed the connection) | |
| [11:05:53] | mike (mike!~mike@c-76-115-119-121.hsd1.or.comcast.net) has joined #mythtv | |
| [11:06:18] | mike is now known as Guest51960 | |
| [11:33:31] | joki (joki!~joki@p5486456D.dip.t-dialin.net) has quit (Ping timeout: 252 seconds) | |
| [11:33:40] | joki- (joki-!~joki@p54864EEE.dip.t-dialin.net) has joined #mythtv | |
| [11:33:51] | joki- is now known as joki | |
| [11:35:43] | pheld (pheld!~heldal@cl-5.osl-01.no.sixxs.net) has joined #mythtv | |
| [11:38:16] | Seeker`: | has markk hidden his mythtv git? |
| [11:50:24] | stuarta: | eh? |
| [11:52:21] | stuarta: | he's committing to his own fork |
| [11:52:29] | stuarta: | https://github.com/Torc/torc |
| [11:53:30] | stuarta: | that's pretty selfish |
| [11:55:37] | stuarta: | anyone want to volunteer to merge back his useful changes |
| [11:55:46] | Seeker`: | ah, didn't realise he'd changed it from a mythtv named fork to Torc |
| [11:55:57] | Seeker`: | I think iamlindoro has been cherrypicking stuff |
| [11:56:35] | stuartm: | yes and no, I can undestand him not wanting to fight over every substantial change, e.g. removing the QT painter, XVideo support etc |
| [11:57:05] | stuartm: | all things we've agreed to do, but can't seem to agree when they should be done |
| [11:58:43] | stuartm: | MythTV has too many people invested in the past, Mark now has the freedom to move forward and create something that remains competitive, I envy him a little |
| [11:59:19] | Seeker`: | perhaps this is a 'good' thing then. It looks like markk is ripping out/rebuilding significant sections of functionality. If we can pull in the stuff when he is done it means that the tree doesn't get in to a broken state in the middle. |
| [11:59:42] | Seeker`: | It also means that it would be possible to advertise the features as 'going away' for a release before they actually do |
| [12:01:07] | stuartm: | it won't work out that way, big changes mean conflicts, there's no way we'll be able to merge in his changes after a few weeks/months have passed |
| [12:01:50] | Seeker`: | well, he seems to be pulling our changes in to his tree |
| [12:02:05] | stuarta: | stuartm: it's always doable, it just gets more painful |
| [12:02:28] | stuarta: | which is why configuration management for source is an artform |
| [12:03:32] | Seeker`: | if markk carries on pulling in all of our commits, then his tree should remain in a compatiable state |
| [12:03:57] | stuarta: | indeed |
| [12:04:06] | stuarta: | it's just a variation on a branch |
| [12:04:27] | Seeker`: | it only becomes problematic if he rips out features that we need to keep on his branch |
| [12:05:01] | Seeker`: | but if he is only ripping out things that were agreed that should be then it shouldn't be too much of a problem |
| [12:05:56] | stuarta: | this is where git really comes into it's own, merging multiple branches together and keeping track of it |
| [12:06:12] | ** stuarta waits for stuartm to mock git... ** | |
| [12:14:56] | stuarta: | https://github.com/Torc/torc/commit/0ba772d6d . . . 001b622f898c <- that's one thing we've talked about |
| [12:24:05] | Seeker`: | is there any reason it couldn't be pulled in after 0.25 release? |
| [12:24:30] | stuarta: | no technical reason |
| [12:25:09] | stuarta: | whether the project "wants to", is a completely different question |
| [12:36:22] | wagnerrp: | Beirdo: heh... thanks got that catch |
| [12:36:30] | wagnerrp: | s/got/for/ |
| [12:44:13] | Robert_S (Robert_S!14852855@gateway/web/freenode/ip.20.133.40.85) has quit (Quit: Page closed) | |
| [13:03:35] | stuartm: | danielk22: remind me, these are recordings deleted from the menu/keybinding in pbb? Versus deleted during playback using DELETE? |
| [13:05:05] | davide (davide!~david@host70.16.intrusion.com) has quit (Remote host closed the connection) | |
| [13:05:29] | davide (davide!~david@host70.16.intrusion.com) has joined #mythtv | |
| [13:30:54] | xavierh: | In Playback Setting, Vertical Scaling is from -100% to 100%, should it not be from 0% to 200% ? (same for horizontal) |
| [13:34:50] | Jordack (Jordack!~jordack@h69-131-44-221.plmomi.dedicated.static.tds.net) has joined #mythtv | |
| [13:42:35] | stuartm: | kProgramInfoTypeVideoStreamingHTML << Is that HTTP streaming, or something more specific? |
| [13:47:11] | stuartm: | answered my own question |
| [13:57:05] | zombor (zombor!~zombor_@kohana/developer/zombor) has joined #mythtv | |
| [13:59:42] | joki (joki!~joki@p54864EEE.dip.t-dialin.net) has quit (Quit: c ya!) | |
| [14:01:45] | xavierh: | stuartm: I would have prefered if you did answer mine ;) |
| [14:05:15] | stuartm: | xavierh: I don't know the answer sorry |
| [14:06:21] | stuartm: | I suspect you'd be right though |
| [14:06:33] | Seeker`: | xavierh: what do the settings do? half size to doublesize? |
| [14:06:39] | j-rod|afk is now known as j-rod | |
| [14:13:49] | xavierh: | Seeker`: I did not test it yet |
| [14:25:56] | Cougar (Cougar!~cougar@2001:ad0:2:5:a0b9:cc52:173f:a894) has quit (Ping timeout: 260 seconds) | |
| [14:36:53] | Chutt (Chutt!~ijr@cpe-24-29-225-175.neo.res.rr.com) has joined #mythtv | |
| [14:36:58] | Chutt: | guh, how annoying |
| [14:37:12] | Chutt: | at&t has my zip code wrong on my landline account |
| [14:37:24] | Chutt: | so now i can't port my number to time warner's phone service |
| [14:37:37] | Cougar (Cougar!~cougar@2001:ad0:2:5:fc39:9074:447b:8249) has joined #mythtv | |
| [14:37:52] | stuarta: | no doubt that takes several weeks to fix |
| [14:37:57] | Chutt: | mythtv related question being – how well does the hd prime work? |
| [14:38:11] | Chutt: | (since that's what i'm changing all around) |
| [14:44:00] | wagnerrp: | Chutt: works great... as long as you dont have time warner |
| [14:44:04] | wagnerrp: | :) |
| [14:44:08] | Chutt: | hah |
| [14:44:18] | Chutt: | why, they mark everything as copy protected or something? |
| [14:44:23] | wagnerrp: | aye |
| [14:44:25] | Chutt: | damn |
| [14:44:36] | wagnerrp: | comcast and verizon work great |
| [14:44:42] | Chutt: | not an option here |
| [14:44:42] | ** stuarta senses a flaw in the otherwise cunning plan ** | |
| [14:44:52] | wagnerrp: | some other companies are hit or miss, or dont have enough users reporting |
| [14:44:53] | Chutt: | yeah, you're ruining my plans of saving money :p |
| [14:45:03] | wagnerrp: | time warner is almost certainly a waste due to copy protection |
| [14:46:38] | Chutt: | ah well |
| [14:46:50] | Chutt: | i'm finally tired of not being able to record non-local channel hd |
| [14:47:11] | wagnerrp: | well at least those are required to be transmitted clear |
| [14:47:37] | Chutt: | yeah, i can already get them with my old hdhr |
| [14:49:47] | wagnerrp: | ron frazier set up a page tabulating CCI data from a handful of users... http://www.ronfrazier.net/mythtv/cci/index.php |
| [14:50:14] | wagnerrp: | and robert kulagowski is doing something similar through schedules direct, but i dont think any of that has been made available yet |
| [14:55:53] | mrand (mrand!~mrand@ubuntu/member/mrand) has joined #mythtv | |
| [14:56:05] | Chutt: | ah well, thanks for the info |
| [15:03:57] | danielk22: | stuartm: hmm, don't know.. I usually delete from within the video though. |
| [15:04:53] | stichnot: | To simplify and bug-fix the text subtitle code, I would like to call the equivalent of SubtitleScreen::Initialise708Fonts() at the start of playback instead of delaying until the first subtitle rendering. On my ION frontend, logs indicate this will add 5ms to the start of playback. Are there any objections? |
| [15:05:40] | danielk22: | stuartm: right now both are working.. :| |
| [15:07:01] | danielk22: | stichnot: Not a good idea. At one point we were spending 5 seconds at playback startup on lots of tiny 5–10ms init's. I was a nightmare to fix. |
| [15:07:22] | danielk22: | s/I was/It was/ |
| [15:09:10] | stichnot: | danielk22: OK. I'll keep initialization deferred (though not as a static method...). |
| [15:17:54] | rsiebert_ (rsiebert_!~quassel@e179132207.adsl.alicedsl.de) has joined #mythtv | |
| [15:18:01] | rsiebert (rsiebert!~quassel@e179131036.adsl.alicedsl.de) has quit (Read error: Connection reset by peer) | |
| [15:23:40] | zombor (zombor!~zombor_@kohana/developer/zombor) has quit (Remote host closed the connection) | |
| [15:26:26] | Seeker`: | would it be bad/wrong of . e to send a mail to -users suggesting that they dont upgrade to 0.25 just yet, seeing hiw there will likely still be bugs from the feature freeze ? |
| [15:27:26] | jya: | well, that would only give us wider test bed for debugging before release |
| [15:27:39] | jya: | so i do hope people are upgrading before it's official |
| [15:27:42] | stuarta: | it would be wrong, because one of the points of the ff is to get users who are in the mood to install it and test it |
| [15:28:08] | stuarta: | we don't have to ask them, some will do it, and some will wait for a package |
| [15:29:11] | Seeker`: | isnt that what rc is for? :P |
| [15:31:38] | danielk22: | Seeker`: We really try to keep a fairly stable trunk at all times and more testing is good; we just ask that users not get abusive of the devs and ticket triage folks when a bug in trunk can't be fixed immediately. |
| [15:32:39] | Seeker`: | fair enough. was just wondering whether the thread on 0.25/ipv6 was going to start a lot of people upgrading |
| [15:40:13] | xavierh: | Found a little in te setting for the database, the default wolRetry is 0, when in the mysql.txt a comment say that the default is 5 if not specified. furthermore the ui only allow to set between 1 and 10, so the "real" default value fall out of range |
| [15:40:50] | xavierh: | Not sure it is worth fixing thought |
| [15:43:50] | stuarta: | we've fixed stuff like that before |
| [15:47:00] | xavierh: | reading mythdb.cpp, the default should be 5, don't know why I get 0 |
| [15:47:32] | stuarta: | check your db, and see if 0 has been stuffed in there |
| [15:49:31] | xavierh: | those setting are also in the db ? |
| [15:54:54] | zombor (zombor!~zombor_@kohana/developer/zombor) has joined #mythtv | |
| [15:59:29] | xavierh: | stuarta: It is not in the DB. I replace the line 894 of mythdb.cpp by: |
| [15:59:32] | xavierh: | params.wolRetry = settings.GetNumSetting("WOLsqlConnectRetry",5); |
| [15:59:58] | xavierh: | to fix the problem, should I create a patch and open an issue with it ? |
| [16:00:01] | jrwren (jrwren!~evarlast@about/csharp/regular/jrwren) has left #mythtv () | |
| [16:01:48] | stichnot (stichnot!~chatzilla@192.55.54.40) has quit (Ping timeout: 245 seconds) | |
| [16:04:14] | stuartm: | Seeker`: if you like you can quote the following – "users upgrading to master now do so at their own risk, we won't discourage them from doing so but they should expect an unfinished product which is a little rough around the edges." As a reminder the first beta is still more than two weeks away, in it's current form 0.25 is still technically at the alpha stage. |
| [16:05:44] | stuartm: | we do want people to test, but if you upgrade now and find you've broken your production setup then it's your own fault, not ours |
| [16:07:15] | RDV_Linux (RDV_Linux!~doug@CPE1caff7df6774-CM00252eac6f40.cpe.net.cable.rogers.com) has joined #mythtv | |
| [16:07:35] | RDV_Linux (RDV_Linux!~doug@CPE1caff7df6774-CM00252eac6f40.cpe.net.cable.rogers.com) has left #mythtv ("Ex-Chat") | |
| [16:07:48] | xavierh: | stuartm: is that ok to open bug issue and set its milestone to 0.25? the fix is trivial... |
| [16:08:55] | stuartm: | stuarta: ^^ |
| [16:09:20] | ** stuarta shrugs ** | |
| [16:09:21] | stuartm: | dekarl: is there a ticket open for that generic matching bug? |
| [16:09:38] | stuarta: | just don't assign it to me. i have zero time to work on mythtv atm |
| [16:09:50] | stuartm: | xavierh: open a ticket |
| [16:15:18] | xavierh: | stuartm: ok. For the Setting UI, I use the old UI method name such as setText instead of SetText. should I convert the name to the later for the new UI ? |
| [16:15:35] | stuartm: | stuarta: the eit subtitle fixup still causes the odd unintended issue, the other night I noticed a programme with the title 'The Forty year' and subtitle 'old virgin' ;) |
| [16:16:43] | stuartm: | xavierh: that would be good since the coding standards call for that case and it would be consistent with the rest of the code |
| [16:17:47] | xavierh: | stuartm: I did not do it at first because I wanted to keep the change at a minimum, but the change are quite big now anyway |
| [16:17:55] | stuarta: | stuartm: was that one of mine? |
| [16:18:25] | stuarta: | i think i know the bit of code you are talking about |
| [16:18:31] | stuarta: | i'm not happy with it |
| [16:18:38] | jpabq_ (jpabq_!~jpabq@mythtv/developer/jpabq) has joined #mythtv | |
| [16:19:02] | stuarta: | was somebody else's code iirc |
| [16:19:22] | stuartm: | http://www.mythtv.org/wiki/Coding_Standards#C . . . riable_names |
| [16:20:09] | stuartm: | stuarta: I can't remember who wrote it, I thought you had written all of the UK fixups, but obviously I have that wrong |
| [16:20:21] | stuarta: | nah, just many of the recent ones |
| [16:20:29] | stuarta: | it's not a specific fixup iirc |
| [16:20:59] | stuarta: | there's a new function somebody wrote to ponder what needed moving around an it's that i believe which gets it around |
| [16:21:13] | stuarta: | s/around/wrong |
| [16:22:10] | stuarta: | i runs after many of the fixups have already been done |
| [16:25:02] | stuartm: | I don't use EIT enough for it to be high on my list of priorities unfortunately :( Since any fix would likely need extensive testing with EIT enabled on all channels that's another reason why I'm unlikely to look at it any time soon |
| [16:26:00] | stuarta: | well once i've finished the DIY i have a dvb-s2 card to install in my backend, and i should be able to run parallel prod/dev rigs on that box |
| [16:26:35] | stuarta: | and play mix-n-match eit data across dvb-t and dvb-s |
| [16:26:45] | stuarta: | that'll iron out a few issues |
| [16:41:06] | Goga777 (Goga777!~Goga777@2.95.15.81) has joined #mythtv | |
| [16:42:11] | xavierh: | done #10341 |
| [16:43:12] | joki (joki!~joki@p54865969.dip.t-dialin.net) has joined #mythtv | |
| [16:49:01] | xavierh: | stuartm: I created a MythSpinBoxDialog class so I can use it in the settings to select value in a range, I am going to submit a patch for it now. May I set the Milestone to 0.26 ? |
| [16:52:21] | xavierh: | stuartm: btw, the milestone for #10092 is wrongly set to 0.25 |
| [17:03:14] | zombor (zombor!~zombor_@kohana/developer/zombor) has quit (Remote host closed the connection) | |
| [17:03:46] | gregL (gregL!~greg@cpe-74-76-125-87.nycap.res.rr.com) has quit (Quit: Leaving) | |
| [17:08:09] | stichnot (stichnot!~chatzilla@192.55.54.42) has joined #mythtv | |
| [17:12:04] | davide (davide!~david@host70.16.intrusion.com) has quit (Remote host closed the connection) | |
| [17:12:26] | davide (davide!~david@host70.16.intrusion.com) has joined #mythtv | |
| [17:14:20] | zombor (zombor!~zombor_@kohana/developer/zombor) has joined #mythtv | |
| [17:28:06] | Jordack (Jordack!~jordack@h69-131-44-221.plmomi.dedicated.static.tds.net) has quit () | |
| [17:28:50] | stuartm: | Beirdo: would it be difficult to add id="L{linenum}" to each line of the irc logs so that we can produce a url which jumps to the exact line? |
| [17:29:58] | pheld (pheld!~heldal@cl-5.osl-01.no.sixxs.net) has quit (Quit: Leaving.) | |
| [17:33:31] | Beirdo: | stuartm: ummm, I guess, but it's in the millions, and the id is not per-channel, so it might be confusing ;) |
| [17:33:55] | Beirdo: | you can jump to it by time anyways |
| [17:35:01] | stuartm: | Beirdo: not per line of the entire history, but per line of the page |
| [17:35:09] | stuartm: | Beirdo: you can jump by time? |
| [17:36:05] | stuartm: | how would I jump to [23:24:46] on the following page? http://irc.mythtv.org/ircLog/channel/4/2012-02-12 |
| [17:37:27] | Beirdo: | one sec |
| [17:38:04] | stuartm: | knightr: I just rediscovered your question about the date/time formats – yes, that would be fine but I wouldn't want translators to think that's an alternative to a proper locale config |
| [17:38:33] | Beirdo: | http://irc.mythtv.org/ircLog/channel/4/2012-02-12:23:24:46 i think |
| [17:38:53] | Beirdo: | yup |
| [17:39:07] | Beirdo: | and if you want to put an end on the range: |
| [17:39:43] | Beirdo: | http://irc.mythtv.org/ircLog/channel/4/2012-0 . . . -12:23:29:57 |
| [17:39:46] | stuartm: | Beirdo: ok, cool, didn't know about that |
| [17:40:28] | Beirdo: | IIRC, you don't need to put the seconds on the end either |
| [17:41:38] | stuartm: | Beirdo: pushing my luck but my follow up question would then be, can clicking on the time in the log link to that? So you don't have to remember the format and add in the time manually? |
| [17:42:44] | stuartm: | e.g. <a href="{path}">{timestring}</a> |
| [17:44:20] | Beirdo: | Oooh, I guess that can be done. :) |
| [17:44:48] | Beirdo: | I'll take a look today whenever I get a chance between tasks and all |
| [17:45:00] | jeff999 (jeff999!~jeff@203-173-13-174.dyn.iinet.net.au) has joined #mythtv | |
| [17:45:07] | stuartm: | thank you :D |
| [17:50:34] | Beirdo: | no prob. :) |
| [17:52:23] | eharris_ (eharris_!~eharris@99-179-7-82.lightspeed.austtx.sbcglobal.net) has quit (Ping timeout: 265 seconds) | |
| [18:00:22] | stichnot: | Beirdo: re your mythtv-dev comment on my commit — is that because it violates the coding conventions (oops), or because of something cppcheck related? |
| [18:01:28] | danielk22: | coding conventions |
| [18:03:05] | stichnot: | is it acceptable to fix that and the preexisting instances during the freeze? |
| [18:05:38] | danielk22: | 0->NULL? Sure. |
| [18:06:18] | davide (davide!~david@host70.16.intrusion.com) has quit (Remote host closed the connection) | |
| [18:06:21] | Goga777 (Goga777!~Goga777@2.95.15.81) has quit (Remote host closed the connection) | |
| [18:06:35] | stichnot: | ok. (though now that I look, I'm having a hard time finding the other instances I thought were there...) |
| [18:06:39] | davide (davide!~david@host70.16.intrusion.com) has joined #mythtv | |
| [18:13:46] | MavT (MavT!~MaverickT@111.86.233.220.static.exetel.com.au) has joined #mythtv | |
| [18:17:20] | MaverickTech (MaverickTech!~MaverickT@111.86.233.220.static.exetel.com.au) has quit (Ping timeout: 276 seconds) | |
| [18:32:39] | kormoc (kormoc!~kormoc@mythtv/developer/kormoc) has joined #mythtv | |
| [18:33:56] | stuartm: | stichnot: we prefer NULL because it is unambiguous, if you see that you know you're dealing with a pointer without having to see it's declaration, 0 could be an int or a pointer |
| [18:37:28] | stichnot: | stuartm: makes perfect sense. |
| [18:38:05] | stuartm: | especially helpful since we shun Hungarian notation |
| [18:44:52] | stuartm: | I just wish the coding standards were stricter on variable naming, so we had less instances of strings called 's' or other meaningless short names, it's significantly easier to jump into the middle of code and know what's going on when everything is named clearly and descriptively |
| [18:51:25] | Beirdo: | yeah, I'd think for a string called "s", for instance, I'd expect that with a very small limited scope maybe, but not all over the place |
| [18:52:32] | Beirdo: | if the temporary variable is in a 5-line scope of an if statement or something, it's pretty easy to decipher |
| [19:48:19] | xavierh (xavierh!~xavier@cpc1-swin3-0-0-cust274.3-1.cable.virginmedia.com) has quit (Quit: leaving) | |
| [19:49:03] | xavierh (xavierh!~chatzilla@cpc1-swin3-0-0-cust274.3-1.cable.virginmedia.com) has joined #mythtv | |
| [19:49:43] | taylorr: | j-rod: how do you know when something has been committed to the kernel? is there something similar to our github commits? |
| [19:50:21] | j-rod: | taylorr: depends on the maintainer |
| [19:50:59] | j-rod: | with the media tree, mauro generates notices based on s-o-b, cc, reported-by, etc. lists and sends mail when the patch gets pushed for certain branches |
| [19:52:22] | xavierh: | danielk22: not trying to argue but what are the reasons to move #10341 from 0.25 to 0.26 when it is a bug fix? |
| [20:04:41] | pheld (pheld!~heldal@cl-5.osl-01.no.sixxs.net) has joined #mythtv | |
| [20:18:51] | sphery: | So, I really wish markk had explained to us whether people should be using opengl es on desktop systems or not. Seems that's the problem for users on opensuse, and since we have no way to disable it--it's always enabled if you (wrongly?) installed the es headers when installing Mesa. http://www.gossamer-threads.com/lists/mythtv/dev/502791#502791 and http://www.gossamer-threads.com/lists/mythtv/users/504193#504193 and now ... |
| [20:18:58] | sphery: | ... http://code.mythtv.org/trac/ticket/10336 |
| [20:20:21] | sphery: | does anyone have any preference for how to "fix" the issue (without requiring us to either fix opengles implementation or fix the nvidia drivers to provide better es support on desktop systems or whatever is wrong)... I can make opengl es be disabled by default at compile time and require an --enable-opengl-es argument to configure, or I can disable the code at runtime (but have no idea what that will do when it's compiled with gl es support) ... |
| [20:20:27] | sphery: | ... or ... |
| [20:21:10] | stuartm: | sphery: ask Mark? |
| [20:21:49] | sphery: | yeah, that's probably best approach |
| [20:25:48] | kormoc (kormoc!~kormoc@mythtv/developer/kormoc) has quit (Ping timeout: 252 seconds) | |
| [20:27:10] | stuartm: | danielk22: I've managed to reproduce the delete bug, although not when I had any additional logging or debugging in place to pinpoint the issue |
| [20:27:14] | danielk22: | sphery: I'd go with disabling opengl-es by default if it is causing problems, but it doesn't hurt to shoot mark an email. |
| [20:27:50] | sphery: | ok, will plan to do both--that will clear up the issue for now and give us a better idea of the intent of the ES code |
| [20:29:00] | danielk22: | xavierh: It's just lower priority and I want to take care of that when I push #10305, I still have ten 0.25 bugs left after that bit of triage this morning. |
| [20:30:42] | Seeker`: | why would you use es on a desktop? |
| [20:31:08] | xavierh: | danielk22: make sense, thx for the reply |
| [20:35:47] | danielk22: | Seeker`: For development.. |
| [20:40:22] | jeff999 (jeff999!~jeff@203-173-13-174.dyn.iinet.net.au) has quit (Remote host closed the connection) | |
| [20:44:43] | stuartm: | danielk22: ok, I think I know what is causing the issue, without the logs it's just an educated guess but I believe we're receiving an UPDATE event for the programme after the DELETE event, the UPDATE event is changing the available status back from asDeleted |
| [20:45:48] | natanojl (natanojl!~jonatan@c83-252-237-63.bredband.comhem.se) has joined #mythtv | |
| [20:46:10] | stuartm: | that explains why when I fixed the UPDATE events it broke the DELETE and fits with my experience when debugging the earlier issue, I was seeing multiple UPDATE events flying around at key moments such as exiting playback |
| [20:46:43] | stuartm: | solution is easy, don't handle UPDATE events where the recording is marked as deleted/pending delete |
| [20:54:33] | taylorr: | j-rod: thanks – I have no idea how the attachment got the mime-type set to octet – I just use gmail and attach it the same way every time |
| [20:55:26] | j-rod: | gmail could stand to be a lot less crap wrt patches... |
| [20:55:32] | stuartm: | although ... maybe safer to update but ensure the available status isn't changed from deleted |
| [20:55:48] | j-rod: | I've given up on using anything but git send-email or mutt for patches |
| [20:57:16] | sphery: | stuartm / danielk22: fwiw, I've been thinking we should change mythtv to always delete to the deleted recgroup (which can be done instantaneously, and regardless of in-use status), then have the auto-expirer delete based on "Deleted max age (days)" (which should be changed to allow "As needed", also) when the recording is not in use. We could then simplify code with a single path for deletion, work around the "can't delete the show because of ... |
| [20:57:18] | stuartm: | danielk22: do you know if it's possible for a recording marked as pendingDelete to return to asAvailable again? Or once marked as pendingDelete is it certain it will be deleted? I'm uncertain how to handle that |
| [20:57:23] | sphery: | ... previewgen" issue, ger rid of "Auto-Expire instead of delete recording" setting, and won't have people thinking they need to mark shows as watched before deleting to Deleted recgroup |
| [20:57:35] | sphery: | granted, not possible until after 0.25, but, just thought I'd mention the idea |
| [20:58:53] | xavierh: | how to do a "svn revert" equivaent with git on a set of file ? |
| [20:59:08] | xavierh: | s/equivaent/equivalent |
| [20:59:20] | stichnot (stichnot!~chatzilla@192.55.54.42) has quit (Ping timeout: 260 seconds) | |
| [20:59:32] | sphery: | git checkout path/to/file path/to/otherfile will get you back to last-committed version |
| [20:59:52] | sphery: | if you've committed it on your tree, you'll have to do more work |
| [20:59:59] | xavierh: | and what about commited files? |
| [21:00:04] | xavierh: | ok |
| [21:00:24] | sphery: | not sure how to handle ones you've committed when you want to undo those commits |
| [21:00:25] | danielk22: | sphery: I |
| [21:00:29] | xavierh: | thanks, the files I like to revert are not commited anyway |
| [21:00:52] | stichnot (stichnot!~chatzilla@192.55.55.41) has joined #mythtv | |
| [21:01:08] | xavierh: | sphery: I have done a reset, thinking it was the equivalent to reset, is it bad doctor ? |
| [21:01:38] | danielk22: | m not sure about the deleted recgroup idea, but lets worry about that after 0.25 is out.. as for the pending Delete, it is possible for a delete to fail, in which case I assume it probably returns to asAvailable. |
| [21:02:04] | sphery: | I don't know reset--it's never done good things for me... |
| [21:03:06] | kormoc (kormoc!~kormoc@mythtv/developer/kormoc) has joined #mythtv | |
| [21:03:18] | danielk22: | I think it's probably fine to ignore update events for deleted programs.. |
| [21:04:52] | taylorr: | j-rod: cool, I didn't know about git send-email |
| [21:05:29] | j-rod: | yeah, get your git tree, create your patches, git format-patch, then git send-email *.patch |
| [21:05:31] | j-rod: | and it'll dtrt |
| [21:05:43] | j-rod: | though you have to be careful about not having excess patches around |
| [21:06:45] | j-rod: | I'll work in a branch, rebase it on master, rm -f *.patch, then git format-patch --cover-letter master.., then git send-email *.patch |
| [21:07:04] | j-rod: | (--cover-letter for a 0/x stub for a >1 patch series) |
| [21:07:31] | stuartm: | danielk22: http://pastebin.com/NRbTVcep |
| [21:11:39] | stuartm: | danielk22: I'm reasonably certain that will fix it, but it would be nice to have confirmation before I push |
| [21:13:24] | danielk22: | stuartm: It will take a few days for me to be confident it's fixed due to the random nature of the bug so you may as well commit and wait. |
| [21:15:58] | stuartm: | heh, no it won't work, at least PI::Clone() won't overwrite the available status when called with the args we're using in ProgramInfoCache::Update() ... |
| [21:16:42] | stuartm: | crap, I thought it was going to be a nice and simple bug |
| [21:17:57] | danielk22: | doh |
| [21:34:57] | wagnerrp (wagnerrp!~Wagner@mythtv/developer/wagnerrp) has quit (Remote host closed the connection) | |
| [21:35:44] | stuartm: | ok, I'm going to run with some additional logging and debugging for as long as it takes to reproduce and get something useful |
| [21:46:08] | ilya-- (ilya--!ilya@gateway/shell/trekweb.org/x-xrrgjkpacpxnvets) has joined #mythtv | |
| [21:46:17] | kenni: | stuartm: Is Metallurgy going to be available for the themedownloader for 0.25? Currently it is available, but with it being dropped from git, I'm wondering if we should continue to make its strings translatable? If you plan to keep it for 0.25, I'll leave the strings, otherwise I'll remove them. |
| [21:48:32] | stuartm: | kenni: it's being removed, I thought Captain_Murdoch had already removed it but I guess not |
| [21:48:33] | Guest51960 (Guest51960!~mike@c-76-115-119-121.hsd1.or.comcast.net) has quit (Read error: Connection reset by peer) | |
| [21:49:47] | Guest51960 (Guest51960!~mike@c-76-115-119-121.hsd1.or.comcast.net) has joined #mythtv | |
| [21:55:17] | stuartm: | kenni: in case it wasn't clear, don't bother translating it |
| [21:58:36] | mike|2 (mike|2!~mike@c-76-115-119-121.hsd1.or.comcast.net) has joined #mythtv | |
| [21:58:57] | sphery: | wow, devinheitmueller just linked this in #mythtv-users: http://www.phoronix.com/scan.php?page=news_item&px=MTA1Nzg |
| [21:59:02] | kenni: | stuartm: Thanks, I'll remove them. |
| [21:59:08] | sphery: | Torc made phoronix |
| [22:00:15] | ilya--: | :) |
| [22:01:24] | Guest51960 (Guest51960!~mike@c-76-115-119-121.hsd1.or.comcast.net) has quit (Ping timeout: 245 seconds) | |
| [22:03:13] | davide: | stuartm: as danielk22 already noted, deletes can fail. it happens to me when i use my dev system as a part-time slave and run the backend as my own user instead of mythtv. i use the nfs directories from the master on the slave. when the master later tries to delete the files it gets a permission error. i think it's best to update the PI's, but keep them as delete pending until the deletion actually fails. |
| [22:08:35] | davide: | sphery: i've considered, but never used the Deleted recgroup stuff. i really like your suggestion to make it the only way. it should be [reasonably] acceptable to everyone, eliminates extra code paths and fixes bugs/misfeatures like the "can't delete because of inuse" issue. i might even consider it enough of a bug fix to go in now if the reading of the setting is hard coded. removal of the extra code paths could be done in 0.26. |
| [22:09:29] | wagnerrp (wagnerrp!~wagnerrp_@2001:470:1f11:12f::a27) has joined #mythtv | |
| [22:09:29] | wagnerrp (wagnerrp!~wagnerrp_@2001:470:1f11:12f::a27) has quit (Changing host) | |
| [22:09:30] | wagnerrp (wagnerrp!~wagnerrp_@mythtv/developer/wagnerrp) has joined #mythtv | |
| [22:09:48] | xris: | love that.. removing tons of legacy code like it was his idea. rather than that he just couldn't wait the 2 months so we could release a stable .25 |
| [22:10:22] | Seeker`: | xris: but will it get removed from our codebase after 0.25 is released? stuarta didn't seem confident it would be |
| [22:10:23] | danielk22: | If we just hard code the reading of the setting without a lot of code changes that would remove my top worry about the change. |
| [22:10:45] | xris: | Seeker`: depends on the specifics. I know a lot of stuff is scheduled for removal |
| [22:11:06] | xris: | wondering when they're going to approach SD about becoming an approved app |
| [22:11:38] | danielk22: | I do worry a bit about people trying to delete to free up space. It's not how we use deletion, but I can image some perplexed users.. |
| [22:11:59] | sphery: | getting rid of the setting/adding the "immediate" value to the "Deleted max age (days)" (seems currently 0 = "as required", not immediate) would be relatively easy, but I don't think we'd be able to change code paths/fix the "allow delete even during preview gen" stuff without some major changes to simplify down to one code path |
| [22:12:01] | Seeker`: | xris: stuarta's take was that a lot of it has been discussed and agreed that it should go, but there wasn't a consensus on when, other than 'not now' |
| [22:12:15] | xris: | Seeker`: yeah, something like that. |
| [22:12:23] | xris: | I thought some parts were more in agreement than others |
| [22:13:45] | Seeker`: | xris: but if the consensus is perpetually 'not now', then it will never get done |
| [22:14:26] | xris: | except that "not now" was specifically in relation to 0.25 |
| [22:14:36] | xris: | granted, I think it was supposed to be released a lot sooner that it was. |
| [22:14:56] | xris: | but mark was as much a part of that delay as anyone else. |
| [22:16:38] | Seeker`: | xris: well, if once 0.25 is released we can pull in all of markk's work in ripping stuff out then great |
| [22:16:51] | Seeker`: | perhaps we would be able to convince him to come back at that point |
| [22:16:56] | davide: | what other parts besides xv did mark want to rip out? that's the only big one i can remember. with the current state of vdpau, opengl, va..., i doubt there'd be much resistance to removing it 0.26. |
| [22:17:10] | xris: | davide: I heard rumors of qtpainter, too |
| [22:17:23] | kormoc: | One painter to rule them all |
| [22:17:53] | Seeker`: | erm |
| [22:18:01] | Seeker`: | seen the last 2 commits to the torc repo? |
| [22:18:01] | davide: | oh, yes. that's right. |
| [22:18:35] | danielk22: | Heh, I can't see doing that while backend setup still requires a GUI. |
| [22:18:55] | xris: | Seeker`: would be nice. I know that there have been tensions between mark+robert and other devs, though. so some people may resist bringing them back. |
| [22:19:33] | xris: | Seeker`: looks like robert's just getting a jump on yet another thing planned for immediately-after-.25 |
| [22:20:01] | kormoc: | xris, huh? External player support wasn't planned for removal |
| [22:20:11] | kormoc: | I thought that was the consensus from the mailing list thread |
| [22:20:16] | j-rod is now known as j-rod|afk | |
| [22:20:16] | xris: | right. |
| [22:20:23] | davide: | so has robert officially jumped ship, then? |
| [22:20:33] | kormoc: | so it's not a jump on post 0.25 changes |
| [22:20:37] | xris: | well, the consensus I got was that we'd be making significant changes… obviously not far enough for robert |
| [22:20:59] | Seeker`: | I haven't seen anything official anywhere |
| [22:21:01] | kormoc: | xris, no, the only change was we would unify how external player support is offered, not remove it |
| [22:21:29] | wagnerrp: | xris: markk actually wanted to remove all that stuff prior to 0.24 |
| [22:21:49] | wagnerrp: | but was forced to proceed with the OSD rewrite in a manner he didnt want in order to maintain Xv support |
| [22:22:41] | xris: | kormoc: maybe I misunderstood the discussion. thought there was talk about simplifying it down to a single possible external command, which would contain any logic the user wanted. rather than a ton of extra db fields for stuff like specific players for specific video files. |
| [22:23:11] | davide: | xris: that's what i remember. |
| [22:23:30] | xris: | which is in effect stripping out 99% of the external player code. |
| [22:23:55] | Seeker`: | that isn't what the commit does though, according to the message |
| [22:24:03] | Seeker`: | 'begin removing external player support' |
| [22:24:05] | xris: | Seeker`: yeah. robert wanted to remove it completely |
| [22:24:16] | xris: | on his own project, now he can. |
| [22:24:34] | kormoc: | xris, that wasn't entirely decided |
| [22:24:38] | ** xris should go back to working on the brewery web page, and waiting for recruiters to call. ** | |
| [22:24:44] | kormoc: | xris, other devs wanted per file extension player support |
| [22:24:48] | xris: | kormoc: point taken. |
| [22:25:05] | danielk22: | kormoc: really? |
| [22:25:08] | xris: | though the "move the logic into the user script" solves that, too. |
| [22:25:30] | xris: | and that was sort of the end of the conversation. it did sort of peter out rather than coming to an all-approved conclusion |
| [22:25:58] | gregL (gregL!~greg@cpe-74-76-125-87.nycap.res.rr.com) has joined #mythtv | |
| [22:26:38] | danielk22: | xris: I thought move the logic into the script was the final & best idea. |
| [22:27:05] | kormoc: | I thought someone objected to that |
| [22:27:10] | danielk22: | I didn't speak up since I hardly ever use MythVideo except to play the occasional potty video. |
| [22:27:10] | kormoc: | but I can't go look right now, so perhaps I'm wrong |
| [22:30:22] | stuartm: | Chutt: hey, I should probably not drag you into this, but there's a user in #mythtv-users who is insisting (quite forcefully) that back when you were still running things that mythfrontend never crashed, it would be amusing if not for his militant attitude |
| [22:30:39] | danielk22: | I'm pretty sure Robert isn't leaving MythTV just yet. Committing to Torc just means he can commit to a non-frozen repo that someone else is keeping up with the required merging. |
| [22:31:55] | xavierh (xavierh!~chatzilla@cpc1-swin3-0-0-cust274.3-1.cable.virginmedia.com) has quit (Ping timeout: 240 seconds) | |
| [22:31:59] | sphery: | Nigel preferred keeping the per-video player field so that users didn't have to learn to script (or find a script that did what they needed) |
| [22:33:56] | Seeker`: | danielk22: it would be nice to hear him say that :) |
| [22:39:01] | kormoc: | To be fair, until he says he's leaving, he's just taking a break |
| [22:39:01] | stuartm: | Chutt: on second thoughts, stay well clear, the level of stress this guy induces can't be good for anyone |
| [22:40:07] | kormoc: | Yeah. That guy was on my ignore list. After checking the logs, I can't say I'm surprised |
| [22:45:59] | skd5aner: | re – deleting recordings... since the feature went in, I've used the setting that puts deleted recordings in a recording group and deletes them after 3 days. I LOVE it, because it basically provides a "recycle bin" feature where I can undelete if I want, but at the same time moves it to the top of the autoexpire list if I actually need the space immediatenly (or before 3 days) |
| [22:46:07] | skd5aner: | it works very well |
| [22:46:25] | skd5aner: | only had 1 bug with it last year, and someone fixed it pretty quickly |
| [22:47:32] | xavierh (xavierh!~chatzilla@cpc1-swin3-0-0-cust274.3-1.cable.virginmedia.com) has joined #mythtv | |
| [22:54:18] | natanojl (natanojl!~jonatan@c83-252-237-63.bredband.comhem.se) has quit (Ping timeout: 245 seconds) | |
| [22:54:59] | xris: | kormoc: there was some attempt at furthering the logic-in-a-script thing but I don't remember it going very far. I still think it's an idea worth hashing out but I'm personally of the opinion that we need to simplify mythtv's config stuff a LOT more than we currently have. |
| [23:00:23] | skd5aner: | hmmm – wonder why mark got rid of mythmessage in his fork? |
| [23:00:53] | ** stuartm wonders about adapting the GPL license with a term which reads along the lines of "We retain the right to revoke this license if you aggravate us in any way" ** | |
| [23:01:57] | kormoc: | Sadly, that'd be a violation of the gpl |
| [23:05:25] | stuartm: | I don't know why idiots like that wind me up so much, nor why I try to reason with them instead of ignoring them but as a result I'm no longer in the mood to go bug hunting tonight |
| [23:05:40] | mike|2 (mike|2!~mike@c-76-115-119-121.hsd1.or.comcast.net) has quit (Remote host closed the connection) | |
| [23:06:02] | skd5aner: | stuartm: just to irk him, go change some things that would cause segfaults – or better yet, something that would cause driver instability! |
| [23:06:03] | Seeker`: | stuartm: just kicking them from the channel works wonders |
| [23:06:10] | stuartm: | I'd love to know how danielk22 manages to remain so calm and detached :) |
| [23:06:36] | mike|2 (mike|2!~mike@c-76-115-119-121.hsd1.or.comcast.net) has joined #mythtv | |
| [23:06:46] | danielk22: | heh, I just joined the channel to observe the fireworks. |
| [23:07:33] | stuartm: | danielk22: I left a while back, I've no idea whether he's still at it |
| [23:08:47] | mike|2 (mike|2!~mike@c-76-115-119-121.hsd1.or.comcast.net) has quit (Remote host closed the connection) | |
| [23:09:07] | Beirdo: | danielk22: if someone wants to commit to a non-frozen repo, git has this concept of a branch |
| [23:09:11] | Beirdo: | it should be used :) |
| [23:09:40] | kormoc: | We have 49 other forks. it's not like forking isn't a common thing with git |
| [23:09:50] | Beirdo: | true |
| [23:09:50] | kormoc: | and we can cherry-pick to our hearts content |
| [23:09:54] | Beirdo: | yep |
| [23:10:04] | Beirdo: | I have no issues with a fork |
| [23:10:20] | Beirdo: | now, I'd like to know what Robert's intentions are though. |
| [23:10:29] | sphery: | though this is the first fork I know of that made phoronix :) |
| [23:11:38] | Beirdo: | intentional announcement 1 day after the fork was made |
| [23:11:43] | stuartm: | xris: the silly thing about all this is that we do have a majority agreement on the need to simplify setup, configuration and even day to day usage, but for one reason or another it's not happening as fast as most people would like – several years on we're still using a standalone database despite acknowledging that an embedded database would be so much simpler for first time users, we're still not auto-detecting capture cards, we're still |
| [23:11:44] | stuartm: | pretty much where we were 3–4 years ago on all those issues |
| [23:11:57] | mike|2 (mike|2!~mike@c-76-115-119-121.hsd1.or.comcast.net) has joined #mythtv | |
| [23:12:10] | skd5aner: | design by committee |
| [23:12:22] | xris: | stuartm: yeah. I think those of us with more experience in enterprise environment are just used to slower progress.. and yes, design by committee. |
| [23:12:29] | Beirdo: | stuartm: granted. But part of that is due to the people with the ideas not being able/willing to complete them |
| [23:12:51] | xris: | especially when people beg for organization and then refuse to take even basic suggestions. |
| [23:13:19] | stuartm: | that and feature creep, things like the embedded db have become more and more complicated because of the other changes that were hooked onto it |
| [23:13:20] | kormoc: | design by committee tends to also mean 'overtaken by the new, lean hawtness' |
| [23:13:36] | xris: | yeah, mythtv would be an entirely different entity if we really organized like a product.. like, say, all of the huge majorly successful open source projects like firefox. |
| [23:13:51] | skd5aner: | yea, but... some of the changes are not that complex, and could be done in a day if the devs wanting to do felt they weren't restricted or always had to get a majority vote |
| [23:14:18] | purserj_ is now known as purserj | |
| [23:14:21] | kormoc: | skd5aner, it's not always majority vote, sometimes it's purely absolute consensus |
| [23:14:21] | stuartm: | xris: when we get paid like those working on firefox is when I think a formalised hierarchy might work |
| [23:14:30] | xris: | I think part of the issue with mythtv is that we try to please everyone. one tiny dissent among devs and we don't do something. |
| [23:14:31] | skd5aner: | kormoc: yea, very true |
| [23:14:39] | Beirdo: | stuartm: that's one reason I like to work on large chunks at once rather than fiddle-faddle with little parts incrementally if possible :) |
| [23:14:42] | xris: | stuartm: that'd be nice. |
| [23:14:54] | skd5aner: | I mean, markk's committed a ton of big changes to his fork in 10 days... it wasn't the "difficulty" stopping him |
| [23:15:37] | skd5aner: | and since dinner is ready – I'm going to hand over the soapbox :) |
| [23:15:41] | xris: | skd5aner: no. it was not wanting to wait until .25. :) |
| [23:16:17] | Beirdo: | he didn't have to |
| [23:16:28] | Beirdo: | all he had to do was make a branch for that work, and carry on |
| [23:16:31] | skd5aner: | well, you mean the one where he sent out the freeze and release dates for since it seemed like 12+ months was long enough? |
| [23:16:40] | stuartm: | it might have been the fact that he was told, 'wait for 0.25' and then months later he's the one who has to insist on setting a release date for 0.25 |
| [23:16:42] | stuartm: | just a thought |
| [23:16:54] | Beirdo: | Umm |
| [23:17:07] | Beirdo: | he was also the one that objected to the potential freeze in October |
| [23:17:42] | Beirdo: | so... SHRUG |
| [23:17:59] | stuartm: | Beirdo: shows how bad my memory is, I don't remember that :) |
| [23:18:04] | Beirdo: | just like my GPU commflag stuff. It will not be put in until after 0.25 |
| [23:18:09] | Beirdo: | but it's on a branch |
| [23:18:20] | Beirdo: | I didn't need to fork the entire project and rename it to do that |
| [23:18:56] | Beirdo: | there's nothing stopping anyone from creating a branch right now to work on something that's going in after 0.25 |
| [23:19:07] | stuartm: | yeah, but let's not try to vilify either Mark or Robert here, I think we'd still like to work with them in future |
| [23:19:14] | Beirdo: | other than the desire to focus on bug-squashing for 0.25 :) |
| [23:19:27] | Beirdo: | True enough |
| [23:19:49] | Beirdo: | I guess what I'm saying is, it's more than just a workflow issue that caused this |
| [23:20:08] | Beirdo: | There's plenty that we just don't know |
| [23:21:32] | Beirdo: | We do know that Mark at least has officially quit MythTV (for now at least) |
| [23:21:36] | Beirdo: | that's about it |
| [23:27:04] | stichnot: | Do we have an official policy on "long long"? |
| [23:27:23] | Beirdo: | what about it? |
| [23:27:42] | stichnot: | like getting rid of it in favor of int64? |
| [23:27:46] | Beirdo: | I've seen both "long long" and "int64_t" in our code |
| [23:27:55] | Beirdo: | I personally prefer int64_t |
| [23:27:58] | stichnot: | sorry, int64_t |
| [23:28:03] | danielk22: | stichnot: we usually use int64_t or uint64_t |
| [23:28:03] | Beirdo: | it's more... obvious :) |
| [23:28:14] | Beirdo: | and platform independent |
| [23:28:29] | stichnot: | right, I thought there might be platform variations with long long |
| [23:28:38] | danielk22: | stichnot: janne was working on removing the long long but never finished that project. |
| [23:29:10] | stichnot: | so if I happen across long long, go ahead and convert it? |
| [23:29:34] | danielk22: | stichnot: right, but you often need to follow a long trail.. |
| [23:29:55] | stichnot: | ok |
| [23:30:32] | Beirdo: | there may be some system calls that use long long, but I don't recall |
| [23:32:41] | sphery: | stichnot: I think the plan was to use int64_t or uint64_t in player-related code, since that's what the libav* code uses |
| [23:33:04] | sphery: | but might be better to use qint64 in non-player code for better platform independence |
| [23:33:20] | Beirdo: | how is that more platform independent? |
| [23:33:33] | Beirdo: | it's a 64-bit integer :) |
| [23:33:44] | xavierh: | Is Robert quiting to ? :( |
| [23:33:53] | sphery: | (meaning when we use int64_t/uint64_t we usually end up having to include compat.h and such in windows code) |
| [23:33:56] | Beirdo: | xavierh: you'd have to ask him |
| [23:34:02] | sphery: | or something like that |
| [23:34:25] | Beirdo: | yeah, that could be. |
| [23:34:26] | Beirdo: | hmm |
| [23:35:13] | danielk22: | I think you just need to include #include <inttypes.h> |
| [23:35:48] | stichnot: | tons of long long in the decoder code — is that considered player-related? |
| [23:36:47] | danielk22: | stichnot: yeah.. but you don't really want to be making lots of those changes in trunk right now.. it's easy to slip in a bug when changing types. |
| [23:37:29] | stichnot: | nope, no intentions of making any changes like that now :) |
| [23:38:15] | stichnot: | it just helps to understand the history, future intentions, etc. |
| [23:40:38] | kormoc: | xavierh, robert committed to mythtv an hour ago. I don't think he left |
| [23:40:46] | sphery: | yeah, just saying that Qt already has qint64 defined as long long int on *nix and __int64 on Windows and we don't have to remember any includes of <stdint.h> or anything... it just works |
| [23:40:52] | kormoc: | I think people should just relax and let things work out |
| [23:46:32] | stichnot: | danielk22: re #7525. I have already abandoned my patch for issue 3 (too much complexity for too little gain). Do you think the first two issues are still worth pursuing? |
| [23:48:48] | stichnot: | I think this ticket was created in the days when so many people were complaining about 7% or 14% CPU usage from an "idle" frontend, before we knew why. |
IRC Logs collected by
BeirdoBot.
Please use the above link to report any bugs.