| Monday, September 29th, 2014, 00:26 UTC | ||
| [00:26:07] | cesman (cesman!~cesman@pool-173-60-115-40.lsanca.fios.verizon.net) has joined #mythtv | |
| [00:26:07] | cesman (cesman!~cesman@pdpc/supporter/professional/cesman) has joined #mythtv | |
| [00:26:07] | cesman (cesman!~cesman@pool-173-60-115-40.lsanca.fios.verizon.net) has quit (Changing host) | |
| [00:28:37] | knightr_ (knightr_!~Nicolas@69-165-170-178.dsl.teksavvy.com) has joined #mythtv | |
| [00:28:37] | knightr (knightr!~Nicolas@mythtv/developer/knightr) has quit (Read error: Connection reset by peer) | |
| [00:36:06] | SteveGoodey (SteveGoodey!~steve@host109-158-49-145.range109-158.btcentralplus.com) has joined #mythtv | |
| [01:05:58] | sheedy-away is now known as sheedy | |
| [01:09:27] | rich0_ (rich0_!~quassel@gentoo/developer/rich0) has quit (Remote host closed the connection) | |
| [01:09:36] | rich0 (rich0!~quassel@gentoo/developer/rich0) has joined #mythtv | |
| [01:25:52] | tonsofpc1 (tonsofpc1!~mythbuntu@cpe-24-94-56-250.stny.res.rr.com) has quit (Changing host) | |
| [01:25:53] | tonsofpc1 (tonsofpc1!~mythbuntu@rivendell/member/tonsofpcs) has joined #mythtv | |
| [01:25:55] | tonsofpc1 is now known as tonsofpcs | |
| [01:47:02] | SteveGoodey (SteveGoodey!~steve@host109-158-49-145.range109-158.btcentralplus.com) has quit (Quit: Konversation terminated!) | |
| [02:08:44] | sheedy is now known as sheedy-away | |
| [02:28:40] | peper03 (peper03!~peper03@mythtv/developer/peper03) has quit (Ping timeout: 258 seconds) | |
| [02:29:05] | peper03 (peper03!~peper03@mythtv/developer/peper03) has joined #mythtv | |
| [03:10:29] | Warped (Warped!~Warped@108-85-161-113.lightspeed.cicril.sbcglobal.net) has quit (Quit: ChatZilla 0.9.90.1 [Firefox 32.0.3/20140923175406]) | |
| [03:27:10] | fetzerch (fetzerch!~quassel@unaffiliated/fetzerch) has quit (Ping timeout: 260 seconds) | |
| [03:27:55] | Warped (Warped!~Warped@108-85-161-113.lightspeed.cicril.sbcglobal.net) has joined #mythtv | |
| [03:28:24] | fetzerch (fetzerch!~quassel@unaffiliated/fetzerch) has joined #mythtv | |
| [03:50:13] | arescorpio (arescorpio!~arescorpi@223-243-16-190.fibertel.com.ar) has quit (Excess Flood) | |
| [05:56:31] | SteveGoodey (SteveGoodey!~steve@host109-158-49-145.range109-158.btcentralplus.com) has joined #mythtv | |
| [06:06:43] | SteveGoodey (SteveGoodey!~steve@host109-158-49-145.range109-158.btcentralplus.com) has quit (Quit: Konversation terminated!) | |
| [06:12:27] | FabriceMG (FabriceMG!~Thunderbi@217.112.59.207) has joined #mythtv | |
| [06:20:44] | dekarl: | stuarta, the small change helps to not break the IPv6 address anymore. Now it tries to contain the link local address, which doesn't work either. |
| [08:07:26] | disissatya (disissatya!~chatzilla@175.101.4.11) has joined #mythtv | |
| [08:19:07] | Merlin83b (Merlin83b!~Daniel@office.34sp.com) has joined #mythtv | |
| [08:39:36] | FabriceMG (FabriceMG!~Thunderbi@217.112.59.207) has quit (Read error: Connection reset by peer) | |
| [08:41:08] | FabriceMG (FabriceMG!~Thunderbi@LCaen-656-01-100-212.w80-11.abo.wanadoo.fr) has joined #mythtv | |
| [08:48:11] | joki (joki!~joki@p54862FD4.dip0.t-ipconnect.de) has quit (Ping timeout: 272 seconds) | |
| [08:54:43] | joki (joki!~joki@p54860A4F.dip0.t-ipconnect.de) has joined #mythtv | |
| [09:00:24] | dekarl (dekarl!~dekarl@p4FCEF070.dip0.t-ipconnect.de) has quit (Ping timeout: 246 seconds) | |
| [09:22:27] | dekarl (dekarl!~dekarl@p4FE84130.dip0.t-ipconnect.de) has joined #mythtv | |
| [09:42:54] | seld (seld!~seld@h7n7-rny-a12.ias.bredband.telia.com) has quit (Remote host closed the connection) | |
| [09:43:18] | seld (seld!~seld@h7n7-rny-a12.ias.bredband.telia.com) has joined #mythtv | |
| [09:44:04] | seld (seld!~seld@h7n7-rny-a12.ias.bredband.telia.com) has quit (Read error: Connection reset by peer) | |
| [09:45:33] | seld (seld!~seld@h7n7-rny-a12.ias.bredband.telia.com) has joined #mythtv | |
| [10:41:29] | stuartm: | gigem: #12285 – I don't actually remember making that change (over a year ago), and looking at it now I'm confused over the purpose of passing in recordid to LoadFromScheduler(), I tried following it through but the more I see the less I understand |
| [10:41:29] | ** MythLogBot http://code.mythtv.org/trac/ticket/12285 ** | |
| [10:43:11] | stuartm: | e.g. in MainServer::HandleGetPendingRecordings() we use it to create, schedule and then immediately delete a manual search rule? The results of which we don't seem to use ... |
| [12:00:43] | esperegu: | stuartm: this looks also a bit similar: http://qt-project.org/forums/viewthread/22427/P15 maybe somethingin the code anyway? |
| [12:14:26] | rsiebert (rsiebert!~quassel@e178051086.adsl.alicedsl.de) has joined #mythtv | |
| [12:17:27] | rsiebert_ (rsiebert_!~quassel@f052162108.adsl.alicedsl.de) has quit (Ping timeout: 246 seconds) | |
| [12:47:57] | moparsthbest (moparsthbest!~quassel@gateway/tor-sasl/moparisthebest) has quit (Remote host closed the connection) | |
| [12:51:29] | moparisthebest (moparisthebest!~quassel@gateway/tor-sasl/moparisthebest) has joined #mythtv | |
| [13:53:44] | stuartm: | esperegu: let me know if the compile gets any further |
| [14:20:47] | esperegu: | stuartm: yes it does. |
| [14:23:53] | disissatya (disissatya!~chatzilla@175.101.4.11) has quit (Quit: ChatZilla 0.9.90.1 [Firefox 29.0.1/20140506152807]) | |
| [14:35:54] | superm1 (superm1!uid4318@ubuntu/member/superm1) has joined #mythtv | |
| [14:53:06] | gigem: | stuartm: Yeah, that change looks really buggered. The recordid bit goes way back to the bjm days. He handled the schedule diff bit that some user submitted. Basically, manual recordings insert special entries into the program table where manualid = recordid. For most schedule diff previews, we create an alternate record table so as to not affect the real schedule. When the recording rule change being |
| [14:53:08] | gigem: | previewed is a manual rule, however, we also need to do something about the special entries in the program table. It was deemed to expensive and not expected to be used enough to make an alternate copy of the program table for that case. Instead, we simply change the program table as needed, run the preview and then change it back by explicitly rescheduling the real manual rule. |
| [14:55:14] | stuartm: | it all gets very complicated, very quickly |
| [14:57:42] | stuartm: | I could just revert my commit instead of applying the supplied patch, filter the results in the services code instead, although IIRC there was a speed benefit to skipping irrelevant checks on upcoming recordings we're not even interested in |
| [14:58:51] | stuartm: | not the easiest code to follow, and no documentation at all :/ |
| [15:06:20] | gigem: | Yes, that new patch seems like it just adds even more confusion. BTW, I think there is a use after delete bug in the current code. It looks like p gets added to tmpList and then might immediately get deleted and then used again later when tmpList is traversed. |
| [15:09:06] | esperegu: | stuartm: can you fix those files? |
| [15:12:12] | stuartm: | which files? |
| [15:13:10] | esperegu: | recordprofile.h and recordprofile.cpp |
| [15:13:16] | esperegu: | to include those destructors |
| [15:14:16] | stuartm: | gigem: subsequent commits are making it difficult to unpick the changes, suspect I'm just going to end up breaking something else :/ |
| [15:17:42] | gigem: | That's an unfortunate consequence of our code that has 'evolved' so much over time. I'd rather have that and fix the regressions, though, than everyone creating there own version of a function that is almost, but not exactly like, another one. |
| [15:24:52] | stuartm: | gigem: well ironically that's what I was thinking of doing in the near future, but in such a way that it provides a cleaner, consistent and documented API so that it might eventually replace the old stuff which is so hard to comprehend |
| [15:30:08] | gigem: | stuartm: That would be fine. That is, ideally having one, or at most a few, well documented functions to handle all use cases. |
| [15:33:36] | stuartm: | maybe it's just me, but when a method which loads the list of upcoming recordings from the scheduler takes the record schedule id as an argument, logically in the absence of any documentation to the contrary, that's for restricting the results to those resulting from that schedule – not for triggering the creation of new temporary rules and who knows what else |
| [15:39:28] | stuartm: | much prefer some well named methods with proper const correctness e.g. Scheduler::GetUpcomingRecordings() const, of course even then the expectation of 'Upcoming Recordings' is that it returns only those which will actually be recorded, so some thought needs to go into the naming of things |
| [15:53:26] | gigem: | stuartm: It's just you! Just kidding. There is a lot of crap in our code base. |
| [15:58:40] | FabriceMG (FabriceMG!~Thunderbi@LCaen-656-01-100-212.w80-11.abo.wanadoo.fr) has quit (Quit: FabriceMG) | |
| [16:53:02] | stuartm: | . |
| [16:53:59] | stuartm: | !trout MythLogBot |
| [16:53:59] | ** MythLogBot slaps MythLogBot with a trout on behalf of stuartm... ** | |
| [16:54:21] | stuartm: | #12345 |
| [16:54:21] | ** MythLogBot http://code.mythtv.org/trac/ticket/12345 ** | |
| [16:57:40] | gigem: | stuartm: For the record, I take exception to you attributing that regression to "archaic scheduler code." As far as I'm concerned, ViewScheduleDiff() and MainServer::HandleGetPendingRecordings() are no more "scheduler code" than ScheduleEditor(). |
| [17:00:35] | stuartm: | gigem: apologies, 'scheduling related' would have been more accurate and I meant no offence by it, just referencing the fact that it's very old code that is little understood so many years after it was written |
| [17:01:29] | stuartm_ (stuartm_!~gbee@mythtv/developer/stuartm) has joined #mythtv | |
| [17:07:46] | stuartm_: | I should just have left the opinion out of the commit message |
| [17:09:47] | Merlin83b (Merlin83b!~Daniel@office.34sp.com) has quit (Quit: Leaving) | |
| [17:17:50] | gigem: | stuartm_: Accepted. FWIW, I've been known to have too sharp of a tongue myself in commit message before. I still get very frustrated when having to clean up "others' crap", but in my hopefully wiser years, I try not leave a permanent record of it anymore. :) |
| [18:01:31] | esperegu: | mythfrondent running on bananapi... \o/ |
| [18:13:06] | jheizer: | LOL I read that as banana-api. Was googling to see what it was. |
| [18:19:40] | SteveGoodey (SteveGoodey!~steve@host109-158-49-145.range109-158.btcentralplus.com) has joined #mythtv | |
| [18:30:57] | stuartm: | esperegu: I did it hours ago |
| [18:32:28] | stuartm_: | Exit |
| [18:33:01] | stuartm_: | exit |
| [18:33:07] | stuartm: | oh ffs |
| [18:33:45] | stuartm_ (stuartm_!~gbee@mythtv/developer/stuartm) has left #mythtv ("Leaving") | |
| [18:36:36] | esperegu: | stuartm: thx. |
| [19:27:00] | paul-h (paul-h!~Paul@90.204.195.10) has joined #mythtv | |
| [19:41:43] | paul-h: | stuartm: just looking into what was causing tgm4883 theme problems shouldn't a <group> expand to fit all it's children? |
| [19:50:25] | tgm4883: | paul-h: that whole thing was pretty weird. The group was the right size, but way off to the side of the contents |
| [19:50:33] | stuartm: | paul-h: there were points for and against widgets expanding to contain their children, I don't remember all of them unfortunately, but one argument I do recall is whether a widget (group or otherwise) should expand when a themer has given it an explicit size when that may result in confusing behaviour e.g. a shape inside a group with an size of 100%,100% would end up being larger than the themer expects in relation to other widgets on screen |
| [19:51:54] | stuartm: | but whether it's better to break with the QT painter instead ... |
| [19:53:17] | stuartm: | IMHO, a third option, where we somehow warn a themer that there's an issue would be best but I don't know how that would work without cluttering up logs and being generally very annoying to users |
| [19:54:00] | tgm4883: | couldn't the XML verifier figure that out? |
| [19:55:24] | stuartm: | not the XSD one, but we could write a util that parses the xml using libmythui |
| [19:56:44] | paul-h: | Looks like we already attempt to expand parents to fit their children it's just broken because it doesn't take the x and y position of the child into account – https://github.com/MythTV/mythtv/blob/master/ . . . ype.cpp#L857 |
| [19:56:45] | stuartm: | XSD validator only looks at formatting/content, it doesn't check context so can't tell that <area> of child lies outside the <area> of it's parent |
| [19:57:03] | tgm4883: | ah |
| [19:58:19] | stuartm: | paul-h: well there you go then, really couldn't have told you which side I'd come down on, other than to say I'd created a list of pros/cons :) |
| [19:59:57] | stuartm: | paul-h: note that since some themes deliberately take advantage of the ability to place widgets outside their parents, e.g. with negative coordinates, any fix to ExpandArea may actually break themes |
| [20:02:21] | tgm4883: | speaking of the xml validator, did <template> replace <format>? I'm getting an error regarding using <format> and I can't find any documentation on it |
| [20:03:45] | stuartm: | the common example is a button list, especially a grid layout, where the 'selected' button list item appears larger and overlaps it's neighbours – I think Robert M's themes used that, and possibly Aeon |
| [20:03:59] | stuartm: | tgm4883: it did yes |
| [20:04:54] | stuartm: | <format> was only used by the clock widget, <template> was used in <textarea>, for consistency the clock widget was switched to use <template> as well |
| [20:21:21] | paul-h: | Well I've made the change locally and it does fix the problem on #12283 |
| [20:21:21] | ** MythLogBot http://code.mythtv.org/trac/ticket/12283 ** | |
| [20:28:17] | paul-h: | Arclight doesn't work properly on screens using negative coordinates when using the Qt painter with or without the patch |
| [20:30:06] | SteveGoodey (SteveGoodey!~steve@host109-158-49-145.range109-158.btcentralplus.com) has quit (Quit: Konversation terminated!) | |
| [20:30:22] | paul-h: | I'm pretty sure whoever wrote that function intended to expand the parent to contain the child but they just fucked up |
| [20:36:38] | stuartm: | that would be me |
| [20:38:26] | stuartm: | and that was what I originally intended, but even I abused the fact that it didn't actually work in Terra |
| [20:41:04] | paul-h: | Another nail in the coffin for the Qt painter |
| [20:44:13] | stuartm: | aye |
| [20:44:25] | paul-h: | On the other hand if we ever fix the OpenGL painter to clip rather than redraw everything all the time like it should we would wish we fixed this problem sooner :) |
| [20:45:33] | stuartm: | I started adding OpenGL clipping support, it was Isaac who said I shouldn't bother, performance without clipping was already very good and so it wasn't worth the time and effort |
| [20:46:10] | stuartm: | ironically I also fixed clipping in the QT painter which was broken when I started working on mythui |
| [20:47:41] | stuartm: | I don't really see a future for the QT painter, even QT5 uses OpenGL for rendering |
| [20:52:36] | stuartm: | we should re-evaluate which hardware is still unable to do hardware accelerated OpenGL |
| [20:52:55] | paul-h: | Is that not one of the stats we collect for smolt I don't see it |
| [20:54:16] | stuartm: | I don't think it is, we've very few useful mythtv specific tests for smolt, it would help a lot if that were to change |
| [21:22:00] | sraue (sraue!~stephan@xbmc/staff/sraue) has joined #mythtv | |
| [21:22:13] | sraue (sraue!~stephan@xbmc/staff/sraue) has quit (Remote host closed the connection) | |
| [21:26:15] | gigem: | sphery: As always, thanks for taking care of the latest "priority" thread on the -user list. I specifically sat out one of the other recent ones and it was painful to see how much bad information was put out by other, hopefully, well meaning users. I was impressed that in this one, they even got the timezone handling correct in the custom priority. BTW, I should have the first part of the tuner affinity |
| [21:26:17] | gigem: | ready this week, perhaps even tonight. There's more I'm thinking of doing, but I'm going to defer it until after multirec without virtual tuners gets done. |
| [21:33:41] | Warped (Warped!~Warped@108-85-161-113.lightspeed.cicril.sbcglobal.net) has quit (Quit: ChatZilla 0.9.90.1 [Firefox 32.0.3/20140923175406]) | |
| [21:55:50] | paul-h: | All hail the sphery :) |
| [22:06:22] | tgm4883: | thanks stuartm fixed it |
| [22:06:48] | tgm4883: | Do you happen to know about categorycolors? Getting this error "categories.xml:2: element categorycolors: Schemas validity error : Element 'categorycolors': No matching global declaration available for the validation root." but it looks like it should be valid http://www.mythtv.org/wiki/Categories.xml |
| [22:12:50] | paul-h (paul-h!~Paul@90.204.195.10) has quit (Quit: Konversation terminated!) | |
| [22:17:53] | stuartm: | tgm4883: the XSD doesn't currently validate the category colour xml, technically it's not part of the mythui theme stuff |
| [22:18:36] | tgm4883: | stuartm: ok, so it's otherwise valid then |
| [22:19:40] | stuartm: | categories.xml is a remnant from the days before mythui, I'd wanted to ditch it for a set of standardised categories but that never happened, |
| [22:20:18] | stuartm: | having to define colours for every possible language, country and grabber source isn't practical |
| [22:24:24] | stuartm: | one idea was just to let themers set a pallete of ~16–24 colours (COLOR1, COLOR2 ... COLOR16) then assign them to categories randomly |
| [22:25:51] | stuartm: | a lot like IRC clients do with nicks, the same category would always get the same colour, but it would be chosen by the code through a hashing algorithm |
| [22:33:26] | tgm4883: | stuartm: so are you saying that categories.xml isn't needed anymore? |
| [22:34:49] | stuartm: | tgm4883: it's still needed, although it's not necessary for each theme to include it's own copy as it will use the one in the default theme |
| [22:35:08] | stuartm: | but it's a legacy thing that hopefully won't be required forever |
| [23:45:02] | skd5aner: | can someone please re-open #12149, I'm still getting movie artwork being returned for tv shows |
| [23:45:02] | ** MythLogBot http://code.mythtv.org/trac/ticket/12149 ** | |
IRC Logs collected by
BeirdoBot.
Please use the above link to report any bugs.