| Saturday, January 30th, 2010, 17:24 UTC | ||
| [17:24:52] | Beirdo: | there we go |
| [17:25:05] | Beirdo: | !notice |
| [17:25:05] | MythLogBot: | This channel (#mythtv-bsp) is logged — http://mythtv.beirdo.ca/ircLog/channel/4 |
| [17:25:12] | Beirdo: | I think that's right |
| [17:25:46] | Beirdo: | yup :) |
| [17:26:49] | Beirdo: | I was off taking the car to maintenance... anyways, you now are accumulating logs |
| [17:27:52] | jpabq: | Beirdo, thanks |
| [17:28:12] | Beirdo: | No problemo |
| [17:30:16] | Casper0082 (Casper0082!~Casper@70.20.143.19) has joined #mythtv-bsp | |
| [17:34:35] | skd5aner (skd5aner!~skd5aner@cpe-024-074-043-234.carolina.res.rr.com) has quit () | |
| [17:37:28] | Traveler373 (Traveler373!~traveler3@S010600112515b313.vn.shawcable.net) has quit (Ping timeout: 264 seconds) | |
| [17:45:20] | sphery: | danielk22: re: #7632, kormoc has been telling users in #mythtv-users who have that problem that they're doing things wrong on their Gentoo systems. I think it's also a dup of http://svn.mythtv.org/trac/ticket/7548 . |
| [17:45:30] | sphery: | danielk22: 20091111 18:30:44<+kormoc> it's cause he's using a hardened profile, which is designed for firewalls/webservers/etc |
| [17:48:00] | sphery: | danielk22: Though I should admit I'm not a Gentoo user, so you'd want to ask kormoc for more info. Whole conversation is at http://mythtv.beirdo.ca/ircLog/channel/1/2009-11-11 starting at 19:25:51 (or search for 7548) |
| [17:49:46] | wagnerrp: | you really have a freakishly good memory |
| [17:50:00] | sphery: | s/memory/logs/ :) |
| [17:50:27] | wagnerrp: | but you remembered the comment such that you knew to search the logs for it |
| [17:52:08] | sphery: | Heh, I'll admit to remembering things about tickets I mark as red/orange (wrong/fishy) in my mail client... (Now I'm thinking I want a way to be able to add notes to my e-mails--wonder whether that desire or my desire to stay away from mozilla code will win...) |
| [18:10:10] | bobc (bobc!~bob@64-126-54-49.dyn.everestkc.net) has joined #mythtv-bsp | |
| [18:21:26] | bobc: | is anyone available to help me with an issue I've got with the backend becoming unresponsive to socket requests? |
| [18:23:01] | bobc: | ticket #7608 |
| [18:29:09] | Captain_Murdoch: | pkendall, great, I'll take a look and get it in. I have a reworked version of 7716 as well, but I want to have gbee look it over first since it touches MythUI. |
| [18:37:29] | pkendall: | Could we get http://svn.mythtv.org/trac/ticket/6569 triaged for this release? |
| [18:38:03] | pkendall: | In NZ we have most channels with 2 channel AAc, but one has 5.1 AC3 and is always noticibly quieter than the others. |
| [18:38:27] | pkendall: | I will test this patch and confirm that it fixes the problem. |
| [18:42:21] | Captain_Murdoch: | gbee, could you take a look at the following 2 patches to see if they're OK with you? first is a fix for NetworkControl that sets a flag when a MythScreenType finishes Init(). The second fixes the font scaling for non-square pixels. I have a plasma that's 16:9 but only 1024x768 and this fixes the fonts being too wide. http://www.bc2va.org/chris/tmp/bsp/7716_v1.diff http://www.bc2va.org/chris/tmp/bsp/pixelAspectRatio_v1.di |
| [18:42:21] | Captain_Murdoch: | ff |
| [18:42:26] | Captain_Murdoch: | http://www.bc2va.org/chris/tmp/bsp/pixelAspectRatio_v1.diff |
| [19:17:45] | brfransen (brfransen!~brfransen@adrianDHCP-47.216-254-250.iw.net) has quit (Remote host closed the connection) | |
| [19:18:29] | brfransen (brfransen!~brfransen@adrianDHCP-47.216-254-250.iw.net) has joined #mythtv-bsp | |
| [19:21:25] | eddief (eddief!~root@pool-162-83-144-106.ny5030.east.verizon.net) has joined #mythtv-bsp | |
| [19:21:44] | mrand (mrand!~mrand@cpe-76-184-144-105.tx.res.rr.com) has joined #mythtv-bsp | |
| [19:27:07] | Captain_Murdoch: | pkendall, not ignoring you, just can't answer that question. :| BTW, the patch looks good, committing it now. |
| [19:35:25] | gigem (gigem!~david@cpe-76-184-80-230.tx.res.rr.com) has joined #mythtv-bsp | |
| [19:38:09] | skd5aner (skd5aner!~skd5aner@cpe-024-074-043-234.carolina.res.rr.com) has joined #mythtv-bsp | |
| [19:50:29] | Koch (Koch!~Koch@d40aeb2b.rev.stofanet.dk) has joined #mythtv-bsp | |
| [20:10:24] | abqjp (abqjp!~abqjp@c-98-230-202-207.hsd1.nm.comcast.net) has joined #mythtv-bsp | |
| [20:32:37] | danielk22: | 'class MythUIHelper' has no member named 'IsTopScreenInitialized' <-- build broken? |
| [20:32:52] | gbee: | Captain_Murdoch: sure, I'll take a look in a minute |
| [20:33:39] | gbee: | danielk22: heh, guess Captain_Murdoch didn't wait for me to look at those patches |
| [20:34:51] | Captain_Murdoch: | gbee, ?? which ones? I didn't commit the two I referenced yet. |
| [20:35:01] | Captain_Murdoch: | I didn't think I commited that. |
| [20:35:33] | Captain_Murdoch: | oh, that slipped into the network control patch. :( sorry about that. |
| [20:35:37] | gbee: | Captain_Murdoch: well both are fine to my eyes |
| [20:35:42] | Captain_Murdoch: | I didn't comit the mythui parts. |
| [20:35:49] | Captain_Murdoch: | purposefully. :| |
| [20:36:19] | Captain_Murdoch: | ok. I'll get them in which will fix the error daniel is seeing. thanks. |
| [20:36:42] | gbee: | sorry, I'd left the room just minutes before you asked for the review, only just came back in |
| [20:36:42] | Captain_Murdoch: | my widescreen definitely looks better with the font setStretch(). :) |
| [20:36:58] | Captain_Murdoch: | no problem, I was working on a couple other patches in my view. |
| [20:37:27] | Captain_Murdoch: | I just svn up-ed so I'll commit after I smoke test the compile again. |
| [20:38:36] | abqjp: | I have emacs setup to auto-remove tabs and trailing spaces when I save a c++ file. That results in my "patches" including some whitespace "fixes". Is that OK, or should I turn that off before generating patches? |
| [20:39:09] | Captain_Murdoch: | preferable to turn off so your patches don't touch lines that aren't relevant to the changes you're making. |
| [20:39:32] | stuarta: | re 7991, is there any reason we actually require 5 buffers? |
| [20:39:43] | stuarta: | it seems we could operate with less. |
| [20:39:50] | sphery: | Captain_Murdoch: Any thoughts (off the top of your head) for http://svn.mythtv.org/trac/ticket/7692 ? Seems important to fix, but I don't see how it could happen. Just wondered if you had ideas before I try replicating it (which takes a while to put unimportant recordings in place, fill up disk, then start a recording to autoexpire stuff, so I'm hoping to run as few tests as possible)... |
| [20:45:46] | danielk22: | sphery: I think I may have already fixed this in the last couple months. |
| [20:46:09] | sphery: | danielk22: yeah, I noticed http://svn.mythtv.org/trac/changeset/23054 and wondered if that may have done it |
| [20:48:02] | gbee: | abqjp: it's ok with me, some prefer that whitespace fixes are committed separately but that's generally because it can make commit diffs harder to read and there are ways to exclude whitespace changes from those |
| [20:48:11] | joe___ (joe___!~jmk@64.73.32.135) has joined #mythtv-bsp | |
| [20:48:42] | gbee: | my editor is also configured to strip tabs/whitespaces, so don't worry about it |
| [20:49:49] | eddief (eddief!~root@pool-162-83-144-106.ny5030.east.verizon.net) has quit (Quit: Leaving) | |
| [20:52:50] | abqjp: | gbee: cool. If it results in too much cruft, I will turn if off, but for a couple lines here or there I will leave it. |
| [20:53:27] | stuarta: | abqjp: my editor is configured the same way as gbee's and i tried to do a 4 line patch |
| [20:53:35] | stuarta: | so i've ended up putting in 23383 first |
| [20:54:44] | danielk22: | my emacs used to be configured that way, it caused too much trouble so I disabled it. the key is to do a diff before commit, if there is a lot of cruft in the diff... |
| [20:55:16] | abqjp: | got it. |
| [20:55:30] | stuarta: | example. 23383, 23384 |
| [20:59:54] | gbee: | aside from the odd whitespace added in new commits, the worst has been slowly cleaned up over the last few years |
| [21:00:03] | danielk22: | gbee: you'll be pleased to hear I'm turning my attention to #7987 now... |
| [21:00:33] | danielk22: | gbee: heh, one of my first big commits was a whitespace cleanup... that was years ago... |
| [21:09:01] | gbee: | abqjp: should http://svn.mythtv.org/trac/attachment/ticket/ . . . gative.patch not be w += m_parentArea.width() – m_parentArea.x() + 1; ? |
| [21:11:23] | abqjp: | gbee: yes, I think you are correct. |
| [21:12:04] | brtab4138 (brtab4138!~brtab@c-76-119-217-205.hsd1.ma.comcast.net) has quit (Ping timeout: 258 seconds) | |
| [21:12:16] | abqjp: | I can verify after I am done with the unselect stuff, but what you are saying looks correct. |
| [21:14:23] | gbee: | e.g. if the intent is to give a 10px margin in a 100x100 parent, <area>10,10,-10,-10</area> with the current patch would produce a rect of 90x90 which would meet the right/bottom instead of 80x80 which is what we want |
| [21:15:42] | gbee: | are there any issues that someone wants me to look at it? |
| [21:17:50] | gbee: | looking to offload http://svn.mythtv.org/trac/ticket/7714 |
| [21:25:19] | brtab4138 (brtab4138!~brtab@c-76-119-217-205.hsd1.ma.comcast.net) has joined #mythtv-bsp | |
| [21:25:23] | sphery: | Captain_Murdoch: re: [23382] , are you sure your X is properly configured (and if so, are you sure your Xft/fontconfig is properly configured)? It sounds like you had specified square pixels in your font config and so Qt (and freetype) did things according to config and the setStretch() will basically be changing the font width on properly-configured machines with non-square pixels. |
| [21:29:25] | Koch (Koch!~Koch@d40aeb2b.rev.stofanet.dk) has quit () | |
| [21:31:33] | sphery: | Captain_Murdoch: this should show different dpi for horiz/vertical for non-square pixels: xdpyinfo | grep -B 2 resolution |
| [21:33:15] | sphery: | Captain_Murdoch: also, some distros set Xft.dpi X resource to 96 (which is 96 dpi square pixels) for font sizing, regardless of X DPI or set up fontconfig with a font-specific DPI in an /etc/fonts/* config file |
| [21:37:26] | sphery: | (I'm simply trying to understand why font sizing wasn't working for you, since TTBOMK, it should have. That said, if that is what was breaking it, perhaps (since so many distros are broken), it makes sense to break things, again, in Myth with the changes in [23382] (as I'd guess more systems are misconfigured than properly configured, thanks to distros thinking they should just hard code font sizing to 96dpi square). |
| [21:57:45] | gbee: | are we going to support running mythtv with SELinux? |
| [21:58:36] | stuarta: | maybe, writing selinux policy is a PITA |
| [21:58:39] | sphery: | is this for the mythweather Perl taint mode issue? |
| [21:58:46] | stuarta: | ISTR a ticket about that? |
| [21:59:10] | mlgvl (mlgvl!~4b8b4444@gateway/web/freenode/x-hxxfaeffykgxknvd) has joined #mythtv-bsp | |
| [21:59:19] | gbee: | stuarta: http://svn.mythtv.org/trac/ticket/7503 |
| [21:59:30] | gbee: | may apply to other tickets |
| [21:59:35] | ** stuarta reads 7714 ** | |
| [22:00:27] | sphery: | This one is caused both by SELinux and users running mythfrontend as root: http://svn.mythtv.org/trac/ticket/4954 |
| [22:01:42] | gbee: | I don't see a great need to support it, myth boxes aren't by nature secure anyway |
| [22:02:12] | sphery: | agreed |
| [22:02:12] | stuarta: | i don't mind us fixing the lib if its not too invasive |
| [22:02:26] | mlgvl_ (mlgvl_!~4b8b4444@gateway/web/freenode/x-jybfiuwxfnflmwim) has joined #mythtv-bsp | |
| [22:02:31] | mlgvl_ (mlgvl_!~4b8b4444@gateway/web/freenode/x-jybfiuwxfnflmwim) has quit (Client Quit) | |
| [22:03:08] | mlgvl (mlgvl!~4b8b4444@gateway/web/freenode/x-hxxfaeffykgxknvd) has quit (Client Quit) | |
| [22:03:46] | stuarta: | 4954 should be fixable by doing a bit of sed foo into a placeholder based upon the install dir |
| [22:04:41] | sphery: | stuarta: or even changing the code calling perl as at: http://svn.mythtv.org/trac/ticket/4954#comment:1 |
| [22:05:02] | sphery: | wait, that -I went into the top of the script... |
| [22:05:03] | stuarta: | i'm not a fan of that one |
| [22:05:13] | stuarta: | HACK comes to mind |
| [22:06:05] | stuarta: | by the time we've finished configure we know which dir the files will end up in, so we can explicitly put that in as appropriate, rather than relying on . in the path |
| [22:06:48] | sphery: | yeah, it's more secure to use an absolute path and add a use lib... I just don't care enough about mythweather to want to do the proper fix myself (and don't know qmake, etc. well enough to know how to do it, anyway) |
| [22:07:16] | stuarta: | i'm sure if you take 7714 off gbee he might be nice :-p |
| [22:08:12] | sphery: | heh... yeah, the whole mythtv socket code (and related issues with newer kernels/libc's) scares me, though |
| [22:08:26] | sphery: | though that may just be a case of needing proper locking |
| [22:08:41] | stuarta: | see, mythui scares me, i'm happier with threads |
| [22:09:01] | stuarta: | proper locking or waiting for the other thread to shutdown |
| [22:09:38] | ** wagnerrp is all for not supporting users running mythfrontend as root ** | |
| [22:10:03] | stuarta: | whoever wrote mythfilldatabase's main.cpp doesn't like function calls |
| [22:10:19] | sphery: | heh, that, too--and I think danielk22 was talking about removing support for the realtimepriority threading stuff |
| [22:10:20] | gbee: | I believe locking is all 7714 needs, I'm just not interested enough to do the legwork, I figure there is someone here who already knows the answer even if they haven't yet seen the question |
| [22:10:34] | sphery: | so if the setuid thing some do triggers it |
| [22:10:36] | stuarta: | the whole file is main() |
| [22:11:04] | gbee: | stuarta: waiting for the thread to exit might be the least invasive fix |
| [22:11:29] | stuarta: | i've yet to prove it, but it's the proper fix |
| [22:12:00] | gbee: | almost certainly a one-liner |
| [22:12:28] | ** sphery wonders if Captain_Murdoch might have fixed it for us when he added that: SendMythSystemEvent("MYTHFILLDATABASE_RAN"); ** | |
| [22:12:50] | sphery: | (though the lack of anything between it and delete gContext; implies he didn't) |
| [22:13:02] | sphery: | unless SendMythSystemEvent() is a blocking function |
| [22:13:46] | sphery: | nope, it also uses RemoteSendMessage(), so just a different last message |
| [22:13:49] | stuarta: | where is RemoteSendMessage sending to? |
| [22:13:54] | sphery: | mbe |
| [22:15:23] | gbee: | sphery: won't help if the backend is down, we need to write the last run time to the database and that's stored in the settings cache .... oh, I see |
| [22:15:31] | stuarta: | i thought mfdb was single threaded |
| [22:15:52] | gbee: | yes, if we trigger the settings cache clearance in the backend instead ... |
| [22:16:30] | ** gbee is probably on a different wavelength ** | |
| [22:16:30] | stuarta: | that would work |
| [22:16:44] | stuarta: | no you aren't |
| [22:17:01] | stuarta: | er hang on. |
| [22:17:12] | stuarta: | we are telling mbe to clear the setting cache |
| [22:17:58] | gbee: | yeah, ignore me |
| [22:20:05] | stuarta: | in theory the code is correct. tell backend to clear settings cache, tell it mfdb has run, and shutdown |
| [22:20:22] | ** stuarta wonders if you can deleteLater gContext ** | |
| [22:21:17] | danielk22: | stuarta: no, you never delete gContext until after the event loop has finished... well you could but then you'd have to flush delete events afterward.. |
| [22:21:36] | danielk22: | what is the ticket #? |
| [22:21:50] | stuarta: | 7714 |
| [22:23:10] | danielk22: | So the problem is we're not waiting for RemoteSendMessage to get it's return value? That doesn't sound right. |
| [22:23:20] | danielk22: | It's blocking code. |
| [22:25:57] | jpabq: | gbee: http://svn.mythtv.org/trac/ticket/7781 has been updated |
| [22:26:06] | gbee: | jpabq: thanks |
| [22:27:49] | sphery: | gbee: since it's just requesting a reschedule, it's not a problem if the backend is down, since the backend that does scheduling (master backend) always does a reschedule on startup |
| [22:28:02] | jpabq: | gbee, re http://svn.mythtv.org/trac/ticket/7965 I am sure you are right, but I will verify now. |
| [22:28:08] | stuarta: | danielk22: if it's blocking code, i'm going blind. it boils down to calling MythContext::SendReceiveStringList which defaults to block=true, but block isn't used anywhere |
| [22:28:23] | stuarta: | +that i can see |
| [22:28:32] | danielk22: | hmm, looks like the MythContext dtor needs to join the MythSocket thread. |
| [22:28:35] | sphery: | Not sure about the system event, though--guess that means that you won't get notification of mfdb runs if your backend isn't running (but that seems reasonable as it's the "myth" event system--not system events) |
| [22:28:49] | gbee: | sphery: aye, bit slow tonight, realised that a second after hitting return |
| [22:29:23] | danielk22: | stuarta: I can look at it tmrw if no one beats me to it. |
| [22:29:41] | sphery: | so RemoteSendMessage() is blocking code? |
| [22:29:52] | stuarta: | maybe |
| [22:31:13] | stuarta: | danielk22: should mythcontext dtor be joining the mythsocket thread? |
| [22:31:34] | sphery: | ah, so sounds like it's the processing of the CLEAR_SETTINGS_CACHE that the backend sends back to all connected clients that's the problem |
| [22:31:35] | danielk22: | sphery: Yes, but that isn't where the segfault is occuring, It's occurring on the other socket, the MythEvent socket.. |
| [22:32:37] | danielk22: | stuarta: yeah, I think right now when an event comes in anytime between delete gContext and program exit we'll segfault. |
| [22:33:12] | stuarta: | k, so which event are we receiving? |
| [22:33:21] | stuarta: | is it the clear cache we just sent? |
| [22:33:31] | danielk22: | that used to be pretty darn rare, but not in this case. stuarta: CLEAR_SETTINGS_CACHE, yes |
| [22:33:49] | danielk22: | The backend is reflecting it as an event to all listeners. |
| [22:34:01] | stuarta: | is there some way of shutting down the listener first? |
| [22:35:08] | sphery: | gbee: thanks for #7864 . I was looking at that and came to the same conclusion, but didn't know if we wanted the "Show LiveTV" in the Change Group View to show LiveTVInAllPrograms, like the deprecated setting we handle on line 415 of programs/mythfrontend/playbackbox.cpp . I really don't see a reason to allow users to put LiveTVInAllPrograms |
| [22:35:32] | sphery: | Guess now I don't have to ask what VIEW_LIVETVGRP is meant to do :) |
| [22:38:31] | danielk22: | stuarta: sure, close eventSock, but I think it might be needed for DeleteUPNP() in MythContextPrivate(). |
| [22:39:41] | danielk22: | Someone just needs to make sure we really shut down the eventSock when we intend to in mythcontext.cpp, and no code executed after that point needs the eventSock. |
| [22:40:07] | danielk22: | s/MythContextPrivate()/~MythContextPrivate()/ |
| [22:40:29] | gbee: | sphery: I've no opinion on that one, I don't use livetv and even if I did, I'd have no use for those recordings to appear in Watch Recordings |
| [22:44:02] | sphery: | gbee: is #7913 even worth looking into as TTBOMK, the mythui OSD code will all use QFont and (I /really/ hope) remove all that freetype font rendering code we currently have (which means removing the need to load and cache all the fonts, as it will be handled by the framework) |
| [22:44:22] | sphery: | markk_: is that correct ^^^ |
| [22:45:13] | wagnerrp: | if i do something like 'QString some_var = some_existing_qstring;', is the new one its own copy? or just a pointer to the old one? |
| [22:45:17] | gbee: | sphery: mythui OSD won't land until 0.24 now, if the patch brings the improvement Jim says then I'd say it's definitely worth looking at |
| [22:45:28] | gbee: | wagnerrp: heh, it depends |
| [22:46:15] | gbee: | QString does some smart stuff, it will only create a second copy if you change it (or the original), otherwise it just maintains a pointer |
| [22:46:28] | sphery: | wagnerrp: but if you change the contents of some_var, it won't affect some_existing_qstring (Qt takes care of it for you) |
| [22:46:35] | wagnerrp: | well the existing string is something shared by a number of threads, and im wanting to make it thread safe without having to keep the shared data locked |
| [22:46:37] | sphery: | which is what gbee just said |
| [22:47:14] | wagnerrp: | so if some_existing_string is removed before i touch some_var, does some_var still exist? |
| [22:47:23] | sphery: | if it's multithreaded, you may want to make a deep copy with something like detach() , but danielk22 would know better than me |
| [22:47:28] | stuarta: | wagnerrp: if you want them not to be shared you have to do deepCopy's |
| [22:48:06] | stuarta: | yeah, detach() rings a bell |
| [22:48:09] | gbee: | stuarta: that's changed in QT4, there is now a QString::detach() which does the same |
| [22:48:37] | stuarta: | QString a = QString b; |
| [22:48:40] | stuarta: | <shared> |
| [22:48:44] | stuarta: | a.detach() |
| [22:48:48] | stuarta: | separate |
| [22:48:55] | stuarta: | there you go :) |
| [22:49:05] | wagnerrp: | doesnt seem to be a 'detach' function on the trolltech qstring docs |
| [22:49:29] | stuarta: | check what it inherits from |
| [22:49:32] | gbee: | e.g. QString foo = one; bar = foo; (bar is now just a pointer to foo) – bar + "two"; (bar is detached from foo which remains unmodified) foo == one & bar == onetwo |
| [22:50:04] | gbee: | wagnerrp: it's hidden from the docs because they don't want people using it, and for most purposes there is no need |
| [22:51:14] | gbee: | if you delete some_existing_string, some_var still exists, at the point the original is deleted it calls detach on some_var |
| [22:51:54] | gbee: | QString is fairly idiot proof, which is a good thing since I use it a lot |
| [22:52:08] | sphery: | +1 |
| [22:52:27] | sphery: | (meaning I'm one who needs it idiot proof, not agreeing that you need it :) |
| [22:52:28] | ** stuarta chuckles ** | |
| [22:52:46] | wagnerrp: | sounds good, thanks |
| [22:53:12] | gbee: | for all intents QString makes copies, the fact that in the background it actually does something more complicated to save memory doesn't really matter 99.9% of the time |
| [22:54:42] | gbee: | only time it does matter is when sharing QString across threads, and even then QT reckon it's thread-safe but danielk22's not so sure :) |
| [22:55:40] | Mode for #mythtv-bsp by anthony.freenode.net : +ns | |
| [22:55:41] | gigem (gigem!~david@cpe-76-184-80-230.tx.res.rr.com) has joined #mythtv-bsp | |
| [23:00:37] | Mode for #mythtv-bsp by anthony.freenode.net : -o MythLogBot | |
| [23:00:37] | Mode for #mythtv-bsp by anthony.freenode.net : +ct-s | |
| [23:00:37] | Technophil (Technophil!~David@203-97-234-182.cable.telstraclear.net) has joined #mythtv-bsp | |
| [23:00:37] | abqjp (abqjp!~abqjp@c-98-230-202-207.hsd1.nm.comcast.net) has joined #mythtv-bsp | |
| [23:00:37] | bobc (bobc!~bob@64-126-54-49.dyn.everestkc.net) has joined #mythtv-bsp | |
| [23:00:37] | stuarta (stuarta!~stuarta@unaffiliated/stuarta) has joined #mythtv-bsp | |
| [23:00:37] | danielk22 (danielk22!~danielk@96.57.9.142) has joined #mythtv-bsp | |
| [23:00:37] | mrand (mrand!~mrand@cpe-76-184-144-105.tx.res.rr.com) has joined #mythtv-bsp | |
| [23:00:37] | Casper0082 (Casper0082!~Casper@70.20.143.19) has joined #mythtv-bsp | |
| [23:00:37] | markk_ (markk_!~mark@cm138.delta193.maxonline.com.sg) has joined #mythtv-bsp | |
| [23:00:37] | wagnerrp (wagnerrp!~wagnerrp_@NR-FT1-66-42-240-2.fuse.net) has joined #mythtv-bsp | |
| [23:00:37] | jpabq (jpabq!~jpabq@c-98-230-202-207.hsd1.nm.comcast.net) has joined #mythtv-bsp | |
| [23:00:37] | elmojo_ (elmojo_!~tralph@cpe-075-177-178-210.nc.res.rr.com) has joined #mythtv-bsp | |
| [23:00:37] | foxbuntu (foxbuntu!~foxbuntu@173-29-149-105.client.mchsi.com) has joined #mythtv-bsp | |
| [23:00:37] | joep (joep!~joep@84-105-5-223.cable.quicknet.nl) has joined #mythtv-bsp | |
| [23:00:37] | gbee (gbee!~gbee@cpc1-derb9-0-0-cust213.leic.cable.ntl.com) has joined #mythtv-bsp | |
| [23:00:37] | superm1 (superm1!~superm1@arctangent.net) has joined #mythtv-bsp | |
| [23:00:37] | brtab4138 (brtab4138!~brtab@c-76-119-217-205.hsd1.ma.comcast.net) has joined #mythtv-bsp | |
| [23:00:37] | joe___ (joe___!~jmk@64.73.32.135) has joined #mythtv-bsp | |
| [23:00:37] | doug__ (doug__!~4e69b89b@gateway/web/freenode/x-avjmsueuabxfznpt) has joined #mythtv-bsp | |
| [23:00:37] | GreyFoxx (GreyFoxx!greg@out.of.phaze.org) has joined #mythtv-bsp | |
| [23:00:37] | gnome42 (gnome42!~ngnome42@75-119-228-148.dsl.teksavvy.com) has joined #mythtv-bsp | |
| [23:00:37] | skd5aner (skd5aner!~skd5aner@cpe-024-074-043-234.carolina.res.rr.com) has joined #mythtv-bsp | |
| [23:00:37] | shadash (shadash!~shadn@bbasa1.bakbone.com) has joined #mythtv-bsp | |
| [23:00:37] | gregl (gregl!~greg@cpe-69-204-183-188.nycap.res.rr.com) has joined #mythtv-bsp | |
| [23:00:37] | iamlindoro (iamlindoro!~iamlindor@unaffiliated/iamlindoro) has joined #mythtv-bsp | |
| [23:00:37] | jpharvey (jpharvey!~chatzilla@host217-42-217-211.range217-42.btcentralplus.com) has joined #mythtv-bsp | |
| [23:00:37] | Captain_Murdoch (Captain_Murdoch!~cpinkham@ip72-218-59-20.hr.hr.cox.net) has joined #mythtv-bsp | |
| [23:00:37] | Mode for #mythtv-bsp by anthony.freenode.net : +vvvv stuarta wagnerrp iamlindoro Captain_Murdoch | |
| [23:00:37] | Beirdo (Beirdo!~gjhurlbu@unaffiliated/beirdo) has joined #mythtv-bsp | |
| [23:00:37] | natanojl (natanojl!~jonatan@c83-252-238-84.bredband.comhem.se) has joined #mythtv-bsp | |
| [23:00:37] | yfwork (yfwork!~damo@skynet.skynet.ie) has joined #mythtv-bsp | |
| [23:00:37] | anthony.freenode.net changes topic to MythTV Bug Squashing Party Jan 30th & 31st | |
| [23:00:42] | anykey_ (anykey_!~guedel@84-72-129-72.dclient.hispeed.ch) has joined #mythtv-bsp | |
| [23:00:42] | hads (hads!~hads@argon.nice.net.nz) has joined #mythtv-bsp | |
| [23:02:03] | Captain_Murdoch: | sphery: dimensions: 1024x768 pixels (400x227 millimeters) and resolution: 65x86 dots per inch. does that look correct? |
| [23:02:46] | wagnerrp: | thats a tiny screen |
| [23:03:21] | Captain_Murdoch: | yeah, it's my cell phone. ;) |
| [23:03:40] | wagnerrp: | thats a massive cell phone |
| [23:04:05] | wagnerrp: | an 18" screen... |
| [23:04:44] | sphery (sphery!~mdean@user-0c6sj10.cable.mindspring.com) has joined #mythtv-bsp | |
| [23:04:44] | pkendall (pkendall!~quassel@125-238-108-152.jetstream.xtra.co.nz) has joined #mythtv-bsp | |
| [23:04:44] | ChanServ (ChanServ!ChanServ@services.) has joined #mythtv-bsp | |
| [23:04:44] | Dibblah (Dibblah!~Dibblah@cpc2-dund11-2-0-cust292.sgyl.cable.virginmedia.com) has joined #mythtv-bsp | |
| [23:04:44] | Mode for #mythtv-bsp by anthony.freenode.net : +vov sphery ChanServ Dibblah | |
| [23:04:49] | Captain_Murdoch: | real dimensions are 914x571 |
| [23:04:58] | wagnerrp: | you may want to ask that again |
| [23:05:08] | Captain_Murdoch: | sphery: dimensions: 1024x768 pixels (400x227 millimeters) and resolution: 65x86 dots per inch. does that look correct? |
| [23:05:23] | Mode for #mythtv-bsp by ChanServ!ChanServ@services. : +v Beirdo | |
| [23:05:26] | Captain_Murdoch: | other than the scaled down mm |
| [23:05:49] | Mode for #mythtv-bsp by ChanServ!ChanServ@services. : +v gbee | |
| [23:06:25] | sphery: | Captain_Murdoch: I just got the 65x86dpi--that looks good. Sounds like your Xft/fontconfig may be set to something like 96x96dpi |
| [23:06:43] | sphery: | if you mentioned something about that, I missed it in the netsplit |
| [23:10:29] | mlgvl_ (mlgvl_!~4b8b4444@gateway/web/freenode/x-snevdtakvvgvystv) has quit (Quit: Page closed) | |
| [23:11:42] | Captain_Murdoch: | nah, just that one one and another saying my real dimensions are 914x571. how can I check the Xft/fontconfig setting you're talking about. |
| [23:11:52] | Captain_Murdoch: | I'm happy to revert if this is my misconfig. :) |
| [23:12:16] | sphery: | well, since TTBOMK, the vast majority of distros are misconfigured, even if I'm right, it may be worth keeping the change :) |
| [23:12:29] | sphery: | let me try to find info on setting non-square dpi in fontconfig |
| [23:12:44] | Captain_Murdoch: | I can google as well if you don't know off the top of your head. |
| [23:12:52] | sphery: | It could be something like Xft.dpi X resource set to 96 |
| [23:12:57] | sphery: | Is this GNOME? |
| [23:14:51] | sphery: | In gnome, there's gnome-font-properties that allows changing the DPI used for fonts (and sets the Xft.dpi for you). See http://scanline.ca/dpi/ for some of the info. |
| [23:14:56] | Captain_Murdoch: | no, fvwm2 on fc5 nfsroot ancient |
| [23:16:04] | sphery: | heh, though it's completely misplaced in the wiki, the fact that it's Fedora sounds like it is Xft.dpi: http://www.mythtv.org/wiki/Specifying_DPI_for . . . Fedora_notes |
| [23:16:36] | sphery: | If you comment that (then restart X), X should use the display-specified DPI for fonts |
| [23:17:53] | sphery: | unfortunately, you can set it in so many places that it's hard to guess how someone's system does it (especially when you don't know their distro :) |
| [23:19:06] | danielk22 (danielk22!~danielk@96.57.9.142) has quit (Ping timeout: 240 seconds) | |
| [23:20:25] | sphery: | Captain_Murdoch: you can also run (though I can't guarantee whether it actually returns the info always): xrdb -query | grep dpi |
| [23:20:27] | Captain_Murdoch: | that is set in that file, so that is making it ignore the actual DPI. I'll test that in a little while and revert my other change if this fixes it. |
| [23:20:31] | danielk22 (danielk22!~danielk@96.57.9.142) has joined #mythtv-bsp | |
| [23:20:31] | Mode for #mythtv-bsp by ChanServ!ChanServ@services. : +v danielk22 | |
| [23:20:43] | Captain_Murdoch: | should have put a ? after the "actual DPI?" :) |
| [23:21:10] | sphery: | Captain_Murdoch: do we want to revert it? After all, as confusing as it is to find it/set it, perhaps just "going with the flow" (and assuming a broken config) may work more reliably? |
| [23:21:43] | Captain_Murdoch: | not sure. have you ever seen many people complain about it? it doesn't affect that many I'd think, mainly plasma users probably. |
| [23:21:44] | sphery: | Unfortunately--other than an xrdb query, I don't know of any way of detecting the font DPI used by the underlying fontconfig |
| [23:22:35] | sphery: | Yeah, not many with non-square pixels these days... Some widescreen CRT TV users have complained, and people who use NTSC/PAL resolutions for "better quality output" do |
| [23:22:39] | Captain_Murdoch: | it's bugged me since I got this monitor, but I always thought it was because Qt didn't support it. when I rememberd the setStretch we now support in the themes, I figured that was an easy fix for the scaling. |
| [23:22:44] | Captain_Murdoch: | s/montior/TV/ |
| [23:24:01] | sphery: | Well, I'll leave it up to you to decide what to do with it. Sorry I didn't notice what you were working on before you went to all the trouble. I think--since most people have square pixels--it's not a big deal whether we do it "right" (the way that few systems are actually configured) or just assume that distros are broken. |
| [23:24:37] | Captain_Murdoch: | nah, it wasn't much trouble. the patch is pretty small as you may have noticed. |
| [23:25:12] | sphery: | I will say, though, that it can have a huge effect on themes--when it's set and differs from the display's DPI (even if both are square), it /will/ change font size in Myth |
| [23:25:42] | Captain_Murdoch: | we already allow forcing aspect ratio somehow (I don't know where in the config it is set or if it's a hidden setting), so it wouldn't be that hard to also allow forcing pixel aspect ratio as well and let the user fix it if they needed to. |
| [23:26:00] | sphery: | iamlindoro: which also may be why your font on my Arclight screenshots looked different from yours--I'd bet that your system is configured with a 96dpi font resolution, whereas mine all use the display's DPI |
| [23:26:29] | Captain_Murdoch: | I'll see if this fixes mine and if so I'll revert since we're coming up on a release. rather 'break' things when there's a while to fix instead of right before a release. :) |
| [23:26:30] | sphery: | the aspect ratio forcing is only allowed with Xinerama |
| [23:26:32] | sphery: | right? |
| [23:26:48] | Captain_Murdoch: | not sure, I just saw it in the code somewhere. |
| [23:26:51] | sphery: | if you only have one monitor, you don't get the option (and it doesn't do anything if you change it elsewhere) |
| [23:27:04] | Captain_Murdoch: | ok. makes sense. |
| [23:27:04] | danielk22: | sphery: right. at the time Xinerama didn't allow you to query the aspect ratio for each monitor. |
| [23:27:29] | sphery: | yeah, so if you're talking about XineramaMonitorAspectRatio, that's only used with Xinerama |
| [23:27:49] | sphery: | don't know what variable it gets stuffed into in code, though (probably doesn't say Xinerama) |
| [23:27:49] | danielk22: | We always assume that Xorg has the correct display info in video playback, except in this one case where we know it doesn't. |
| [23:29:22] | sphery: | Captain_Murdoch: but a hidden setting (to go into your Property Editor setting page :) probably makes sense as a "I don't care to learn how X works" solution for users with non-square pixels. |
| [23:29:51] | sphery: | (as we really shouldn't expect them to learn all about X and dpi's :) |
| [23:30:44] | Captain_Murdoch: | ok, I may just wrap that setStretch call with a if GetNumSetting("blah") == 1) |
| [23:30:55] | Captain_Murdoch: | gotta run again for a few, be back in a little while. |
| [23:31:09] | sphery: | thanks for checking--I was really curious why it was happening for you :) |
| [23:31:19] | brfransen (brfransen!~brfransen@adrianDHCP-47.216-254-250.iw.net) has joined #mythtv-bsp | |
| [23:32:29] | Captain_Murdoch: | thanks for the insight. :) |
| [23:32:40] | sphery: | better late than never? |
| [23:34:47] | natanojl: | sphery, Captain_Murdoch : fwiw, I also have a 1024x768 plasma. dimensions: 1024x768 pixels (1130x629 millimeters), resolution: 23x31 dots per inch |
| [23:35:30] | natanojl: | xrdb -q | grep dpi doesn't find any Xft.dpi |
| [23:35:39] | jpabq: | gbee, http://svn.mythtv.org/trac/ticket/7965 updated |
| [23:37:21] | skd5aner: | danielk: your "very wide button" changsets, could they be related (indirectly) to some of the issues raised in 7983? |
| [23:37:33] | sphery: | natanojl: what distro? I've found that xrdb doesn't usually mention it, even when it a font DPI is specified (but never looked into why). Also, if you're using a desktop environment, it's very likely setting a font DPI for you. |
| [23:39:41] | natanojl: | sphery: Debian testing. I'm running Xfce and have "System default" as DPI setting. It was set to 100 DPI before and then it turned up in xrdb |
| [23:40:13] | sphery: | ah, yeah, so you've reconfigured yours properly! :) Nice. |
| [23:42:03] | natanojl: | Yeah, finally after 4 years or so :). But setStretch seems necessary to get the fonts to look correct in myth |
| [23:42:37] | sphery: | natanojl: Are you running trunk? If so, you may see fonts get too narrow after r23382 (until Capt M puts in the setting to disable the stretching). In the interim, you could just set the fonts to a specified square DPI. |
| [23:43:27] | sphery: | does that mean that fonts don't look right before that change? |
| [23:45:35] | natanojl: | No I'm not but I added a setStretch(75) call in mythuitext.cpp. Now it looks much better |
| [23:46:20] | natanojl: | s/mythuitext.cpp/mythfontproperties.cpp/ |
| [23:49:49] | sphery: | natanojl: in /etc/fonts/*.conf or /etc/fonts/conf.d/* , do you have any mention of dpi ? It sounds like yours is still using the wrong font DPI. |
| [23:52:46] | natanojl: | no, grep -ir dpi doesn't find anything at least |
| [23:53:41] | sphery: | brfransen: are you still here? I have some questions for you on #7996 (assuming that's yours) |
| [23:54:19] | doug__ (doug__!~4e69b89b@gateway/web/freenode/x-avjmsueuabxfznpt) has quit (Quit: Page closed) | |
| [23:56:59] | stuarta: | hmmm livetv isn't getting a lock |
| [23:59:15] | natanojl: | sphery: I've also removed '-dpi 100' from the command in gdm |
| [23:59:34] | sphery: | yeah, that would be the X DPI, which is right from your numbers above |
IRC Logs collected by
BeirdoBot.
Please use the above link to report any bugs.