Back to Home / #mythtv / 2008 / 02 / Prev Day | Next Day
#mythtv IRC Logs for 2008-02-06

---Logopened Wed Feb 06 00:00:29 2008
00:01-!-xris [] has joined #mythtv
00:06-!-ColForbin [] has joined #mythtv
00:06-!-ColForbin [] has left #mythtv []
00:12-!-skamithi [] has quit ["WeeChat 0.2.3"]
00:30-!-Johan-- [n=chatzill@] has quit ["ChatZilla 0.9.80 [Firefox]"]
00:33-!-xris [] has quit [Remote closed the connection]
00:33-!-xris [] has joined #mythtv
00:50-!-Gokee2_Laptop [] has quit [Remote closed the connection]
00:55-!-grokky [n=grokky@] has joined #mythtv
00:57-!-grokky [n=grokky@] has quit [Client Quit]
00:57-!-fuhgawz [] has quit [Read error: 110 (Connection timed out)]
01:01<hads>xris: ping
01:02<hads>Just wondering about your question on #4600 (python bindings).
01:02<hads>I don't quite get what you are asking.
01:03<xris>someone said the bindings should not go under plugins... they don't
01:04<hads>Yeah, was me. I weas just referring to moving them since they are now in mythplugins/mythvideo/
01:04<hads>But I intended them to be in mythtv/bindings/ eventually.
01:05<xris>oh. this guy's bindings are all new, I thought
01:05<hads>Na, it's a split up of the stuff that I worked with Anduin and someone else to get in.
01:05-!-loops [] has quit [Read error: 110 (Connection timed out)]
01:06<xris>he might be the someone else, then?
01:06<xris>dunno. I didn't know that there were any official python bindings yet
01:06<hads>Nope, don't think so. THe someone else was the guy who wrote for mythvideo. I did most of the actual bindings bit.
01:06-!-loops [] has joined #mythtv
01:07<hads>Just making a comment on the ticket now, I've prepared a diff to put them in mythtv/bindings with a to install them.
01:07-!-fuhgawz [n=beexwax@] has joined #mythtv
01:07-!-MrGandalf [] has quit [Read error: 104 (Connection reset by peer)]
01:08<hads>Here's where the original stuff is;
01:19<xris>probably not going to do anything with it tonight.
01:19<xris>but if his patch is just a re-thinking of your stuff, we might as well just wait for anduin to show up and ask him what he thinks about porting stuff over to mythtv/
01:20<hads>Yeah for sure.
01:20<xris>adding them to configure should be a pretty easy task... not sure how to deal with creating a .pro file for them, though.
01:20<xris>someone else had to do that for me for the perl stuff
01:21<xris>I need to dig into, though... I want nuvexport to be able to export compatible meta files
01:23<hads>I'll whip up some patches for and other stuff to reflect the move to mythtv/ if needed.
01:23<xris>theoretically the libs should just install wherever python wants them to be put
01:24<hads>Yeah, the will do that, just needs to be called from the Makefile somehow like the perl one is.
01:30-!-beata-- [] has joined #mythtv
01:30-!-beata [] has quit [Read error: 104 (Connection reset by peer)]
01:31-!-onyxsoft [] has joined #mythtv
01:33-!-kormoc [n=kormoc@unaffiliated/kormoc] has joined #mythtv
01:36<superm1>so the bindings are going to stick as parts of the plugins for 0.21 though right hads ?
01:36<superm1>rather than their own "bindings" package
01:37<hads>superm1: I'm not sure, it's not up to me where they go :) There's isn't that much in-tree using them, I don't know how many people have out of tree stuff that uses them though.
01:38<hads>It would be nice to have them in the bindings package so they install system wide.
01:39<superm1>well if something is decided before feb 14th that they will go in the bindings package, can you let me know?
01:39<hads>Sure thing. Are you using them?
01:39<superm1>we'll have some stuff to work around on our end for that, but if they are in the plugin packages, it wont be a big deal
01:39<superm1>we won't be using them for our stuff, but ubuntu has a freeze at the 14th, so introducing a new package is really hard past that
01:39<superm1>they would sit in their own libmyth-python similar to libmyth-perl has its own
01:40<superm1>if its after the 14th, we'll probably just include them in mythtv-common or similar
01:40<hads>Ah yep, well hopefully anduin pops in and gives his opinion either way and we can get it sorted soonish.
01:44<superm1>well i idle in here, but dont always catch backtrace. just throw my name in the conversation at some point and i'll see x-chat light up and i'll read the backtrace if i'm not around
01:45<superm1>thanks. okay off to bed for me then
01:45<superm1>night :)
01:52-!-Cougar [] has joined #mythtv
01:54-!-mzb [] has quit [Read error: 113 (No route to host)]
01:56-!-mzb_d800 [] has quit [Read error: 113 (No route to host)]
01:57<hads>Ah that's right, it was visit0r that did
01:57-!-jhulst [n=jhulst@unaffiliated/jhulst] has joined #mythtv
02:02-!-ahbritto [] has joined #mythtv
02:03-!-Cougar [] has quit []
02:05-!-Cougar [] has joined #mythtv
02:06-!-fuhgawz [n=beexwax@] has quit [Read error: 110 (Connection timed out)]
02:10-!-mzb_d800 [] has joined #mythtv
02:21-!-PointyPumper [i=Pintlezz@] has quit [Read error: 110 (Connection timed out)]
02:27-!-gnome42 [] has quit [Remote closed the connection]
02:34-!-onixian [] has joined #mythtv
02:41-!-jhulst [n=jhulst@unaffiliated/jhulst] has quit [Read error: 113 (No route to host)]
02:41-!-onixian [] has quit ["This computer has gone to sleep"]
02:49-!-briand [] has quit [Read error: 110 (Connection timed out)]
02:55-!-grokky [] has joined #mythtv
03:06-!-_spike [] has quit [Remote closed the connection]
03:07-!-_spike [] has joined #mythtv
03:30-!-onixian [] has joined #mythtv
03:41-!-siXy [i=siXy@] has joined #mythtv
03:42<xris>hads: yes... I could only remember that he had a 0 in his nick. heh.
04:04-!-PointyPumper [i=Pintlezz@] has joined #mythtv
04:06<xris>libmythappearance went away?
04:10<kormoc>merged into the frontend iirc
04:10-!-trisooma [] has joined #mythtv
04:12<hads>trisooma: You opened #4600 didn't you.
04:12-!-trisooma [] has quit [Client Quit]
04:12-!-trisooma [] has joined #mythtv
04:13<trisooma>yes i did
04:14<trisooma>hasds: it was just a start to create the bindings of off
04:14<trisooma>hads: it was just a start to create the bindings of off
04:15<hads>Yeah, as I mentioned in the comments it was my intention to get them into the bindings dir eventually. We're just waiting to see what Anduin thinks.
04:16<trisooma>hads: great, i read that he might have some more stuff to add to it
04:16<hads>Yeah, so do I. I've only added methods that I am using in my little scripts here so far.
04:20<trisooma>i'll keep checking progress
04:21<trisooma>hads: out of curiousity - let us say that these files are 'accepted' - will they be added to the trunk at some point?
04:22<hads>trisooma: They are in trunk at the moment, just in mythplugins. They will probably be moved to mythtv/bindings at some stage but that's not up to me. I'm not a dev I just contributed the bindings.
04:22-!-xris [] has quit []
04:25<trisooma>hads: k, thnx
04:40-!-MaverickTech [] has joined #mythtv
04:59-!-MavT [] has quit [Read error: 113 (No route to host)]
05:01-!-briand [] has joined #mythtv
05:19-!-mattwire [i=d43ae981@gateway/web/ajax/] has joined #mythtv
05:21-!-gardz [] has joined #mythtv
05:52<justinh>just had an email informing me the uk icon grabber script I wrote is out of date. can that be removed now since mythtv-setup is meant to be doing all the downloading?
06:00-!-MGisbers [n=mgisbers@] has joined #mythtv
06:12<gbee>justinh: yup
06:13-!-xian__ [] has joined #mythtv
06:16-!-MGisbers [n=mgisbers@] has quit [Remote closed the connection]
06:19-!-onixian [] has quit [Read error: 110 (Connection timed out)]
06:23<justinh>mmm I want one of these nice camera modules for home. no, infact four of them
06:23<justinh>dynamic range is amazing
06:24<stuarta>morning ppl
06:24<stuarta>how are we all today?
06:25<justinh>tired. you?
06:25<justinh>anybody object to me pulling uk_icons then? if not, they go byebyes :)
06:26<stuarta>fine. stuck in the data centre pulling cables
06:29-!-grokky [] has quit []
06:56*justinh chuckles at the "It's hard to describe the quality of the Virgin operation in MK but try to imagine watching a Baird televisor through an old sock." on el reg
07:06-!-sc00p [] has quit [Success]
07:11-!-TelnetManta [] has quit ["Ex-Chat"]
07:32<gbee>justinh: ouch
07:33<gbee>dunno if I mentioned watching a Sky HD presentation at the shopping centre around Christmas? Quality was piss poor and yet this was what they were using for a promotion (i.e. their best stuff)
07:33<gbee>honestly BBC SD content looks better
07:34<justinh>"but it's aitch dee !"
07:38-!-sc00p [] has joined #mythtv
07:39<stuarta>gullible comes to mind
07:39<justinh>the Emperor's new telly :)
07:41-!-mattwire [i=d43ae981@gateway/web/ajax/] has quit [" ajax IRC Client"]
07:41<justinh>really needed to test those font changes I'm working on today but I forgot to switch the dev box on
08:10-!-anykey_ [] has quit [Remote closed the connection]
08:11<justinh> - for anyone interested
08:15-!-anykey_ [] has joined #mythtv
08:16-!-rooau1 [] has joined #mythtv
08:28-!-briand [] has quit [Read error: 110 (Connection timed out)]
08:28-!-briand [] has joined #mythtv
08:29-!-Gringo_ [] has joined #mythtv
08:29-!-Gringo_ [] has left #mythtv ["Leaving."]
08:39-!-rooau1 [] has quit ["Leaving."]
09:03*GreyFoxx sees prices on 1TB drives and starts to get the upgrade itch
09:03-!-beata-- is now known as beata
09:04-!-xian__ [] has quit ["Leaving"]
09:42-!-mattwire [i=d43ae981@gateway/web/ajax/] has joined #mythtv
09:42-!-skamithi [] has joined #mythtv
09:43-!-Cardoe [n=Cardoe@gentoo/developer/Cardoe] has joined #mythtv
09:47-!-jmk_ [] has joined #mythtv
09:49-!-MrGandalf [] has joined #mythtv
09:58*gbee sees the prices of 1TB drives and wishes he wasn't so tight fisted
09:58<MrGandalf>you'd have to get two and mirror them
10:01<MrGandalf>would not do to have a gig of recordings suddenly go away
10:02<stuarta>no, that would be cleansing :)
10:02<stuarta>1g of recordings = too much tv
10:03<MrGandalf>lemme correct myself, 1TB
10:03<MrGandalf>so you're saying my 2.1TB store is too big them?
10:04<MrGandalf>and my 200gig LiveTV store as well?
10:09<stuarta>i haven't the time to watch my 320gb
10:09<MrGandalf>well, you may have a point I suppose
10:09<MrGandalf>I end up recording a lot of stuff I "have to have" but never watch again.
10:15<GreyFoxx>_mre|666: Nah, I'd buy 5 of them and raid them.
10:15<GreyFoxx>MrGandalf: Nah I'd buy 5 of them and raid them.
10:15<GreyFoxx>stupid nick complete and lazy me
10:15<MrGandalf>I thought of that.. I'm be worried about speed though
10:16-!-cattelan [] has quit [Remote closed the connection]
10:18<stuarta>MrGandalf: speed? who cares when you have about 4Tb of storage
10:19<MrGandalf>when you have 2 or 3 recordings going on at once and a couple are HD, speed matters. :)
10:20<stuarta>well most raid controllers are gunna have cache anyway
10:21<MrGandalf>I'm definately going to look into it.. I need to start getting away from my 1000 watt EMC array.
10:21<stuarta>nice toy
10:21<MrGandalf>expensive electric bill
10:22-!-cndougla_ [] has left #mythtv ["Konversation terminated!"]
10:22<MrGandalf>with what I spend on electric for it, I could buy a single 1TB drive a month.
10:24<GreyFoxx>MrGandalf: right now I have 3 TB spread amongst 2 machines. I'd LOVE to trim it to everything in a single box
10:25<MrGandalf>greyfoxx: scst
10:25<GreyFoxx>scsi ?
10:25<GreyFoxx>Maybe if scsi prices have taken a 180 in the last while :)
10:25<MrGandalf>no, scst
10:26<GreyFoxx> That what you are referrng to ?
10:26<MrGandalf>I used to run it and likely will again
10:27<MrGandalf>I wrote the configuration script for it
10:31-!-skamithi [] has quit ["WeeChat 0.2.3"]
10:36<GreyFoxx>ugh, time to replace the UPS at home
10:37<GreyFoxx>looks like there was a power bump and everything shut off improperly :)
10:38<MrGandalf>that sucks
10:40<GreyFoxx>yeah, I have a APC 1400 rackmount sitting here waiting to take home for the last 2 weeks hehe
10:41<MrGandalf>I used to run a tripplite APS unit hooked up to a couple 12v deep cycle marine batteries. I could get 4 hours in a single charge
10:42<MrGandalf>you can get them on ebay for a few hundred bucks
10:42<GreyFoxx>I've looked at using a couple car batteries connected to an old lawnmower engine as a basic generator
10:42<MrGandalf>and the batteries are dirt cheap
10:42<MrGandalf>that's interesting :)
10:43<GreyFoxx>I'd love to setup something to power the machines long enough to power down well, and 1 or 2 lights + TV/cablemodem
10:43<GreyFoxx>MrGandalf: YEah
10:43<MrGandalf>well, if power is out likely cablemodem is out as well since the line amps require power from the poles.
10:43<GreyFoxx>the plans are simple enough really
10:44<GreyFoxx>MrGandalf: Our cableco does phones over the lines. Their stuff is powered and has 12 hours of battery at the poles
10:44<MrGandalf>wow, slick
10:44<GreyFoxx>My phone and internet and cabletv come in on 1 wire
10:44<justinh>1 coax
10:44<GreyFoxx>into a box on the side of the house which splits them out
10:44<justinh>that's how mine comes in
10:45<justinh>coax into the street cabinet then fibre all the way back apparently
10:45<GreyFoxx>They have massive batteries every couple poles heh
10:48<MrGandalf>well, if that's how your land line comes in then they are required by law to maintain connectivity for a certain amount of time during a power outage.
10:48-!-gnome42 [] has joined #mythtv
10:49<GreyFoxx>I think legally it's 6 or 8 hours. The local phone company shoots for 10 as a standard
10:50<GreyFoxx> that the sort of tripplite you are using?
10:50<GreyFoxx><<-- Likes to follow instructions when it comes to electricity :)
10:56-!-fuhgawz [] has joined #mythtv
10:58<MrGandalf>WOW, they've really come down in price
11:04-!-skamithi [] has joined #mythtv
11:04<MrGandalf>there's a 1250 for $250 as well, 120/12
11:04<MrGandalf>they used to go for >$800
11:07<janneg>argh, subversion's branches are almost useless. svn merge is a miserable failure
11:08*GreyFoxx needs to find something to make it easier to track series of patches in 1 source tree
11:08<GreyFoxx>I'm often working on several different things at once and it sucks having to carefully trim out stuff that isn;'t meant for a diff
11:09<janneg>I'm trying to merge a branch and svn merge is just stupid, not mythtv related
11:09<janneg>GreyFoxx: have you tried quilt?
11:10<GreyFoxx>janneg: nope, not yet
11:11<janneg>another possibility would be a distributed scm with fast branching like git
11:11*janneg uses git and quilt to track patches
11:12<janneg>for mythtv development
11:15-!-jdhall [] has joined #mythtv
11:15-!-jhulst [n=jhulst@unaffiliated/jhulst] has joined #mythtv
11:19<justinh>omg. just found a c++ book on my desk under some junk. wonder how that got there
11:21<MrGandalf>ancient Chinese secret
11:22<justinh>wonder if it's any good
11:22<justinh>then again I dunno if I can get any worse at it
11:28<MrGandalf>If I have to work in C++, I prefer Qt.
11:28<MrGandalf>but then again I've never called myself a developer
11:30<justinh>I call myself a mangler of code & a developer of themes
11:40-!-trisooma [] has quit []
11:43-!-fuhgawz [] has quit [Read error: 110 (Connection timed out)]
11:45-!-siXy [i=siXy@] has quit ["bye!"]
11:47-!-jhulst [n=jhulst@unaffiliated/jhulst] has quit ["Konversation terminated!"]
11:49-!-mattwire [i=d43ae981@gateway/web/ajax/] has quit [" ajax IRC Client"]
11:55-!-mattwire [i=d43ae981@gateway/web/ajax/] has joined #mythtv
12:14<MrGandalf>yes, but rather nice themes
12:19-!-nordenm [] has quit []
12:22-!-jgarvey [] has joined #mythtv
12:27-!-jfg [] has joined #mythtv
12:40-!-jfg [] has quit ["Leaving."]
12:50<stuarta>dammit. the osx build is failing again on something that it creates itself
12:50-!-Chutt [] has joined #mythtv
12:53-!-reynaldo [] has quit ["leaving"]
12:53-!-reynaldo [] has joined #mythtv
12:57*stuarta trouts osx-packager
12:59<stuarta>wtf is it putting a make unistall cleanup target in the mythplugins
13:01<stuarta>feck it.
13:02*stuarta goes shopping
13:02-!-JoeBorn [] has joined #mythtv
13:12-!-mattwire [i=d43ae981@gateway/web/ajax/] has quit [" ajax IRC Client"]
13:24-!-xris [] has joined #mythtv
13:41-!-ahbritto [] has quit [Client Quit]
13:41-!-dekar1 [] has left #mythtv []
13:50-!-xris [] has quit []
13:53<okolsi>gbee: ping
13:54<okolsi>one thing in MythControls.. if you have unsaved edits and try to exit there is a dialog asking what to do
13:55<okolsi>if you press Esc, it doesn't just close the dialog/popup but exits back to menu
13:56<okolsi>it could be like this.. but I think elsewhere Esc just closes the popup/dialog
13:57<okolsi>btw, nice to see mythui in action :)
14:00<gbee>okolsi: ok, I'll fix it
14:01<gbee>okolsi: mythcontrols is a pretty poor example of mythui, I kept the old theme/layout etc so it barely scratches the surface of what is possible with mythui
14:02-!-Puhi [] has quit [Remote closed the connection]
14:04<okolsi>gbee: can't wait to see the new stuff.. it'll become more visible when theme developers start to use it
14:04-!-xris [] has joined #mythtv
14:15-!-nordenm [] has joined #mythtv
14:33-!-jdhall [] has left #mythtv []
14:34-!-\S2 [] has joined #mythtv
14:34-!-\S2 is now known as S2
14:35<xris>hads: so that python stuff is ready to plug into bindings/ ?
14:35-!-xris [] has left #mythtv []
14:36-!-xris [] has joined #mythtv
14:36<xris>this irc client really annoys me sometimes
14:37<Chutt>HOME=/blah mythbackend
14:38<Chutt>'way to pass in env var on commandline' :p
14:41-!-czth__ [n=dbrobins@nat/microsoft/x-86f95fe6898cefbe] has joined #mythtv
14:44<xris>Chutt: true
14:45<xris>wasn't sure if it would work that way in inittab
14:58-!-czth_ [n=dbrobins@nat/microsoft/x-ea4c2a0a41316ce2] has quit [Read error: 110 (Connection timed out)]
14:59<justinh>okolsi: there aren't enough theme developers around to try out the mythui stuff. and I'm way too busy :)
14:59<justinh>gonna be a chicken & egg situation for a while I think
15:04<hads>xris: Yep, that patch should be ready to go.
15:05-!-ahbritto [] has joined #mythtv
15:06<hads>The only thing is someone should look over the file as I just copied it over from perl.po and adjusted it without really understanding it too much.
15:11-!-onixian [n=xian@] has joined #mythtv
15:11-!-onixian [n=xian@] has quit [Remote closed the connection]
15:15-!-reynaldo [] has quit [Read error: 113 (No route to host)]
15:21-!-foxhunt [] has joined #mythtv
15:28-!-reynaldo [] has joined #mythtv
15:30-!-trisooma [n=remko@] has joined #mythtv
15:34<skamithi>hello justinh. would like ur opinion on a theme change for #4568 (embed upcoming recordings). just uploaded what i'm using for the titivillus theme into the ticket. pls let me know by mail or update the ticket with whether i'm complying with the current way of creating new containers in ui.xml.
15:37<justinh>skamithi: heh been waiting for that patch to see what it's all about
15:38<justinh>only thing I'll say is - can't _wait_ for more area inheritance
15:40<justinh>skamithi: is there no way to re-use the existing upcoming recordings container?
15:41<skamithi>i dunno..u tell me :)
15:41<justinh>maybe just put contexts in to deal with displaying one of 2 areas & to make the video preview optional
15:42<justinh>ui.xml is big enough already :)
15:42<xris>so what happened to libmythappearance?
15:43<justinh>xris: merged
15:43-!-kormoc [n=kormoc@unaffiliated/kormoc] has quit []
15:43<xris>into libmyth?
15:43<xris>threw me off when I was compiling my mythplugins rpm
15:44<justinh>built into mythfrontend now
15:48<justinh>it was never intended to stay as a plugin - or even start life as a plugin - just the way it happened
15:50-!-mattwire [] has joined #mythtv
15:50-!-foxhunt [] has quit [Read error: 110 (Connection timed out)]
15:51<okolsi>justinh: yeah, too bad (small number of theme devs)
15:52-!-MrGandalf [] has quit ["home"]
15:54<skamithi>justinh, is there a good example of the use of contexts in a theme ? i've never tried that before.
16:12-!-kormoc [n=kormoc@unaffiliated/kormoc] has joined #mythtv
16:13-!-skamithi [] has quit ["WeeChat 0.2.3"]
16:15<GreyFoxx>Sweet, someone just threatened to sue me (my company really) because I disabled his access to webspace he should not have had access to for 8 months and he has been storing content for websites we don't host in it and no longer has any copies ofit himself :)
16:16<GreyFoxx>it's because of a typo that we was able to login at all
16:16<GreyFoxx>when I fixed the type it killed his access
16:17<GreyFoxx>and then he send me a reaming email that I saw him using it and disabled it and now he's gonna sue if I don't give him the data
16:17<GreyFoxx>and he'll get the people who'se content it is to sue as well :)
16:17<GreyFoxx>heh silly users :)
16:22<gnome42>kormoc: program detail dates are good now. Thanks!
16:22<kormoc>I would be temped to email him back threatening to sue him for unauthorized access and seek damages for the extra bandwidth and disk space/io/etc he took up.
16:23<kormoc>gnome42, you're quite welcome :)
16:23<GreyFoxx>kormoc: I'm sure my boss will say "pay for the account for the last 8 months and I'll happily reenable access" :)
16:24<kormoc>heh, fair 'nuff
16:25<gnome42>kormoc: It kinda makes apache a little cpu hungry too. Makes a nice pattern though. :)
16:26<kormoc>hah, whoops
16:30<gnome42>kormoc: I was doing some fiddling with trying to make cpu use go down. So, I removed the "Try to skip ahead in the timeout queue...", but no great success.
16:32<kormoc>gnome42, what I'm planning to do is to actually make the auto-load a option that defaults to off, that way it's only the few requested ones that get loaded
16:32<kormoc>that should lower cpu usage a lot
16:33<kormoc>I'm also going to look into batching the info for the autoload into larget chunks (20,50,etc) rather then one at a time, that way less requests
16:33<gnome42>yeah, I was playing with that option too. Sounds good.
16:33<kormoc>also looking at streamlining the php loading, so hopefully it'll be a lot lighter, as right now it does a full load and even connects to the BE
16:34<gnome42>yeah, batching ... same idea :)
16:36<gnome42>kormoc: I think the caching you just put in is working fairly well too. Was wondering if that could be leveraged in more places.
16:36-!-dekarl [] has joined #mythtv
16:38<kormoc>gnome42, Well, really the webserver should be tuned to do that caching, I'm sorta being a bad dev by forcing it (it'll be an option defaulted on soon, so it can be swapped)
16:39<kormoc>gnome42, if you're running apache, later I can help you setup a htaccess file to do more aggressive caching and the like.
16:39<kormoc>Or perhaps I should just write up a wiki page on it...
16:40<gnome42>ah, okie. I didn't realize that it should be done on the server.
16:42<kormoc>Well, it's the sad mix of certain users will have setup the server with how they want it to be cached and I'm overriding that (no cookie for me :/), but also a lot of users won't have, and they'll love it (yay cookie!), so it's one of those lose-lose things
16:43<gbee>guess I should try out these mythweb changes, just a pain to install updates
16:43<xris>mythweb is easy... `svn up`
16:44<xris>kormoc even added a mythproto override setting so it ignores mythproto version
16:44<kormoc>For those who love to live on the edge!
16:45<gbee>xris: if your mythweb install is an svn checkout, mine isn't (for a couple of reason), so I need to svn export then resync the files across to the webserver
16:45-!-S2 [] has quit [Remote closed the connection]
16:46<xris>no need to export.. rsync --cvs-exclude
16:46<xris>it'll ignore .svn and other revision control files
16:47<gbee>saves me a step I guess ;)
16:47<gnome42>xris: yeah, me too
16:48<gbee>another problem I had originally was that I kept overwriting the htacess file every time I updated, I switched to using a proper config instead but for a while that used to drive me crazy
16:54<gbee>first problem seems to be that I need to mouseover the title twice before the description will appear, does appear the first time although it seems to make the request
17:01-!-foxhunt [] has joined #mythtv
17:04<kormoc>Yeah, I need to fix that too, wiggling the mouse should do it as well
17:06-!-MrGandalf [] has joined #mythtv
17:14-!-onixian [] has joined #mythtv
17:14-!-onixian [] has quit [Remote closed the connection]
17:18<gbee>another bug, when I clicked on a title to go to the "program detail" page, I get a bunch of description popups appear - looks like they are the ones belonging to the "possible conflicts"
17:20<kormoc>can you take a screen shot?
17:32-!-xris [] has quit []
17:36<sphery>Anyone know why mfdb uses UPDATE settings SET data = ... for mythfilldatabaseLastRunStart/End rather than using MythContext::SaveSettingOnHost() with an empty host?
17:36<gbee>kormoc: sure
17:38<janneg>sphery: probably no good reason
17:39-!-foxhunt [] has quit [Remote closed the connection]
17:39<sphery>Cool. I'll use the latter for my code since it doesn't require a dbcheck update to insert my BackupDBLastRunStart/End.
17:44<kormoc>Hrm. funky
17:45<gbee>popups are for the conflicts
17:49-!-mzb [] has joined #mythtv
17:54<gbee>janneg: I assume decoding would be slower on some systems if --enable-ffmpeg-pthreads was used by default?
17:59<janneg>I'm not sure. It's definitely slower with two threads on a single cpu system. But it isn't enabled by default in ffmpeg
18:00-!-Cardoe [n=Cardoe@gentoo/developer/Cardoe] has quit ["Leaving"]
18:00<MrGandalf>really only usefull for h264 anyway and it seems that's still broken for most DVB providers anyway
18:01<MrGandalf>even with the latest spacial direct commit
18:01<gbee>I'm just thinking about ease of installation/setup for users, if there isn't a significant slowdown with threading enabled then it would be easier if it was made the default
18:03<gbee>does compiling with --enable-ffmpeg-pthreads explicitly turn on threading? or does that only happen when the number of processors is increased in the video profiles?
18:03<janneg>It's probably safe as long nobody uses avcodec_thread_init
18:04<janneg>multithreaded decoding is only enabled if max_cpus is greater than 1
18:13-!-jgarvey [] has quit ["Leaving"]
18:16<MrGandalf>does this looks wrong to anyone: DELETE FROM program WHERE manualid = -1 OR (manualid <> 0 AND -1 = -1);
18:16<MrGandalf>from the scheduler
18:16<trisooma>can anybody tell me how generate a patch that is accepted as such (ie. parameters to patch command (-ur?))
18:16-!-_spike [] has quit [Remote closed the connection]
18:17-!-CDev [] has left #mythtv []
18:17-!-CDev [] has joined #mythtv
18:17<trisooma> for posting on trac
18:17<hads>trisooma: svn diff
18:18-!-_spike [] has joined #mythtv
18:18<trisooma>hads: hi again... these files are not in svn (it's a new patch for python bindings)
18:18<hads>Why a new patch? What's it for?
18:19<MrGandalf>diff -rupN <old tree> <new tree>
18:19-!-_spike [] has quit [Remote closed the connection]
18:19<hads>Anyway, svn add them and then a svn diff will work.
18:19<kormoc>(as long as they're not binary files)
18:19<trisooma>hads: this is a patch to make the package more usable (and some comments and an example file)
18:19<MrGandalf>that bit of sql in the scheduler takes typically 14 seconds on my system. The whole scheduler run takes 16-17.
18:20<hads>trisooma: Cool, is it based on the patch that I posted?
18:20-!-_spike [] has joined #mythtv
18:20<trisooma>hads: yup
18:21<hads>Cool. I'll keep an eye out for it and have a look.
18:21<trisooma>hads: i created two directory structures to create a diff
18:21<trisooma>hads: i'll try mrgandalf's options
18:21<sphery>the svn add is better...
18:22<sphery>it will allow the code to show up formatted in trac
18:22<trisooma>kk, i'll try svn add first, when it fails ill try plain diff
18:22<sphery>also, make sure you use a .patch or .diff extension
18:23<kormoc>when it fails? you have such a positive outlook on things don't ya?
18:23<trisooma>kormoc: lol, these files aren't actually added to svn (yet)
18:24<kormoc>trisooma, that's fine, svn add will add them to your local tree and svn diff will handle them fine
18:24*kormoc uses that method daily
18:25<trisooma>kormoc: but it still diffs against files that aren't in the trunk
18:25<kormoc>Aye, if they're added, yes
18:27<trisooma>kormoc: srry, i have two sets of files 'previous' (not in svn) and 'current' (new situation, diff should be against previous)
18:27<trisooma>it's in proposal stage
18:28<trisooma>kind of virtual diff 8-)
18:28<kormoc>whatever, just seems strange to not use svn for that
18:28<janneg>MrGandalf: the rest runs after that query? then it's probably a caching effect
18:29<trisooma>kormoc: i'll give it a try anyhow
18:29-!-mattwire [] has quit ["Leaving"]
18:29<MrGandalf>janneg: I'm looking at the -1.. doesn't look valid. I'm putting together an email to the dev list now. By removing that sql, I've cut my scheduler runs down froim 17 seconds to a 5-6 seconds.
18:31<janneg>the query is equivalent to DELETE FROM program WHERE manualid <> 0;
18:31<MrGandalf>what is manualid?
18:32-!-timofonic [] has quit ["WeeChat 0.2.6"]
18:32<MrGandalf>there's no index associated so it's doing a full table scan
18:32<MrGandalf>this why it takes 11 seconds.
18:33<janneg>I think it is used for manual schedules
18:36<sphery>MrGandalf: is the -1 part put there by code (i.e. a placeholder that's used sometimes and not used other times)
18:38<sphery>looks like: DELETE FROM program WHERE manualid = :RECORDID OR (manualid <> 0 AND :RECORDID = -1)
18:38<MrGandalf>sphery: yes, it's a bind variable
18:39<MrGandalf>seems there is an index.. hmm
18:42<sphery>lol... I think you're the reason why:
18:42-!-JoeBorn [] has quit ["running out"]
18:43<sphery>Should have given you this one: --you submitted the patch to put that index in because of the same query
18:44<sphery>MrGandalf: When was the last time you ran (Or at least an ANALYZE TABLE on all your tables.)
18:44<MrGandalf>I optimized about a week ago
18:45<janneg>according to EXPLAIN the index isn't used
18:45<sphery>Since the data withing the program table is nearly replaced within a week, it may help to do it again. The OPTIMIZE will compress it and the ANALYZE should prep the indices
18:45<MrGandalf>ah, I didn't try that
18:47<sphery>EXPLAIN SELECT count(*) FROM program WHERE manualid = -1 OR (manualid <> 0 AND -1 = -1); shows Using where; Using index on mine
18:48<MrGandalf>as does mine
18:48<sphery>only with the count, though... with SELECT * it says Using where
18:49<MrGandalf>and how does either of those equate to a delete I wonder..
18:49<sphery>no idea
18:51<MrGandalf>I'll try dropping the index and see how that differs
18:52<sphery>I'd be interested to know if performance changes once you re-run
18:52<MrGandalf>my scheduler runs always take that long..
18:52<MrGandalf>even after an optimize
18:54<MrGandalf>in a sense, re-creating the index will be nearly the same as an optimize on that table
18:54<sphery>makes sense
18:55<gbee>stuarta: any idea if #2394 is still true after the multirec merge? I'd can check right now as I've got recordings in progress
18:55<MrGandalf>that's if the db ever gets done dropping the index..
18:55<gbee>trisooma: svn add *new file* ... then svn diff
18:56-!-xris [] has joined #mythtv
18:56*gbee notices the question was asked 40 minutes ago
18:59<MrGandalf>heh, went up to 26.5 seconds without the index, so it must be using it
18:59<trisooma>gbee, thx.... i got it working now
19:01<MrGandalf>or maybe not.. oddly, in the mysql-slow log it shows as still taking 11 seconds
19:01-!-trisooma [n=remko@] has quit ["Enuff for today"]
19:03<sphery>guess he didn't want anyone else telling him to use svn add :)
19:03<MrGandalf>when that breaks, he can use diff.... ;)
19:05<janneg>gbee: yes, #2394 is still valid. I think the problem is the playback. it seems to cache the last tuned channel and doesn't notice the card change
19:08<gbee>janneg: ok, that's a shame :)
19:09<janneg>I'm looking into it now
19:10<gbee>#3773 seems familiar, wasn't there a recording bug causing similar problems?
19:11<gbee>there has been too much traffic on the -commits list for me to find the relevant discussion
19:12<janneg>trac doesn't respond
19:12-!-mntmst [] has joined #mythtv
19:12<sphery>#3773 sounds a lot like a corrupt seektable (especially the "starting to spread to other channels" part), except for the not playing properly in mplayer. Signal issues /and/ a corrupt seektable?
19:14<sphery>gbee: re: #3971 (MythWeather scripts), I think I covered all the possibilities for running with libs not in @INC at . I do agree, though, that MythWeather scripts should not be in @INC (though I was making an argument for why Perl bindings should be in @INC in that post).
19:16<MrGandalf>sphery: down to 7.9 (single run only).. I hate to have to optimize the database every day though - that seems overkill for one sql statement.
19:16<sphery>gbee: Also with #3773, video playing too fast sounds like he is (but shouldn't be) using "Use video as timebase."
19:17<sphery>MrGandalf: Cool. On the bright side, "soon" you won't have to optimize the DB ever--Myth will automatically do it every day for you. :)
19:18*gbee cracks the whip
19:18<MrGandalf>sphery: hmm, dunno about that :) that kills recordings for me while it's optimizing. That may be good for small DBs though.
19:18<janneg>I'm looking into it now~.
19:18<sphery>Yeah. I'm working the "only when there's nothing else happening" part of it.
19:19<MrGandalf>sphery: my program table has near 1,000,000 entries.
19:19<gbee>sphery: thanks for that post, I'll look at the options - I think I already implemented the "use lib " solution now that I've been reminded of it, so ...
19:20<janneg>hmm, my connection was still alive
19:20<sphery>I'm getting the basics of a backup in place now. Once there, I'll mod the housekeeper stuff to allow configurable JobConstraints (including leadtime, no playback, ...)
19:21<MrGandalf>sphery: (another) stupid question, sphery == ?? (on dev list)
19:21<MrGandalf>don't tell me another Stuart.
19:21<sphery>backup will only be done on DB schema upgrade at first. once the housekeeper/jobconstraints stuff is done, then it will be automatic along with optimize/analyze/etc.
19:21<sphery>"Stuart" Mike Dean
19:22<MrGandalf>too many Stuarts :)
19:22<janneg>gbee: re #2394 we are checking if the channum matches the last channum on the prevChan list and SwitchCards/Switch
19:23<janneg>Inputs doesn't updates prevChan
19:23<sphery>Yeah. No Stuart in my name, though. (Hope I didn't offend any Stuarts by posing..."
19:24<gbee>I could make it easier and go by my second name
19:25<MrGandalf>janneg: "option 3..." were you serious about the patch? I can provide one pretty easily which I think does the correct thing.
19:27-!-nsaspook [] has quit [Read error: 110 (Connection timed out)]
19:27<janneg>if the patch is simple and has no side effects, sure. but mythtv isn't designed for identical callsigns on the same source
19:28<MrGandalf>janneg: a lot of discussion around that.. I believe it's callsign + channum
19:29<sphery>Other than for the class (or other top-level stuff), should any doxygen comments be in a header file? Or should they just be in the cpp file? I've seen what looked like the content of doxygen comments copied from .cpp to .h files, but that looks like a maintenance issue...
19:35<sphery>MrGandalf: BTW, you could easily do something like the following to get unique callsigns (change falsechannel to channel): SET @row = 0; UPDATE falsechannel SET callsign = CONCAT(callsign, ' (', (SELECT @row := @row + 1), ')') WHERE callsign = 'somevalue';
19:35<sphery>gives 'somevalue (1)', etc.
19:37<sphery>I wonder if the scanner should do it's own suffixing when it finds a multiple channels with the same callsign reported...
19:37<MrGandalf>there is a lot of discussion around this. Janneg just stated that having the same callsigns on the same source is not supported. Other devs argue that having the same callsign on any source isn't supported. Some say (and I agree with) that callsign + channum should be the constraint same or different source.
19:38<sphery>I say that same callsign with different content isn't supported (though I know that Bruce says, "same callsign with mostly identical content" is supported)
19:39<sphery>I don't like his approach because of the ambiguity of "mostly identical"
19:40<MrGandalf>sphery: ok, that's fine, but I would argue that if the user has the same callsign on multiple channels, the guide should always attempt to tune the channel the user selects, and not a random channel with the same callsign on the same source.
19:40<gbee>for me, same callsign means same channel, whatever the source - but I've had this discussion already and we didn't get very far
19:41<gbee>MrGandalf: in that case, for tuning purposes, I'd agree
19:41<MrGandalf>well, same callsign is easily gotten around for me. My bigger issue is with same channum
19:41<MrGandalf>and different callsign
19:41<MrGandalf>and different source
19:42<janneg>MrGandalf: your mixing different things. it's callsign + channum for ommitting channels in the guide, it's only the callsign for the scheduler and having the same channel twice on the same source doesn't make much sense
19:42<sphery>janneg: weren't you working on a patch to fix the default channum being used by the scanner?
19:43<MrGandalf>janneg: I'm only talking about the guide and in trunk it's callsign OR channum. Callsign + channum is what I'm working towards. :)
19:43<sphery>per the conversation with jarle_ in early Jan (Jan 10, 2008 at about 8:00pm in my timezone_
19:44<sphery>< janneg> we should use something more reasonable than the serviceid in the dvb channel scanner serviceid modulo 1000 for starters
19:44<janneg>MrGandalf: true, and I said I would accept a patch but I don't start to fix a problem caused by an unsupported configuration
19:44<MrGandalf>janneg: if you read that bit of code I sent a patch for, you see that it first matches on channum and adds them to the list, then matches on callsign and adds them as well. My patch combines those to be callsign + channum
19:45<MrGandalf>janneg: if callsign + channum is supported, and that's what the patch gives, how is that fixing an unsupported configuration?
19:48<xris>janneg: are you talking about switching the group-by stuff for channels?
19:48<janneg>sphery: that avoids only the big channums (iirc up to 2^16) we would still need to avoid duplicates
19:49<janneg>MrGandalf: having multiple distinct channels with the same callsign on a single source is not supported
19:50<sphery>Oh. I see.
19:50<janneg>xris: no, it's just related
19:50<MrGandalf>janneg: understood. What is the standard for multiple channums with different callsigns?
19:54<janneg>don't expect livetv tuning to work, i.e. being able to tune to every channel. so essentialy the same
19:55<janneg>bed time for me, good night
19:56<gbee>reckon we should start closing PVR-350 tickets as feature requests without patches?
19:56-!-j-rod [n=jarod@nat/redhat/x-3ad737eb3b7d5ec1] has quit [Read error: 110 (Connection timed out)]
19:57<MrGandalf>janneg: Then I'll say this (an no more), whether or not these are supported configuration, users will identify these two items as bugs and report them as such since previous Myth versions did at least appear to support them. Also, if these configurations are unsupported, users will ask that work be done to both mythfilldatabase and the scanner to support multiple providers which happen to use the same channums.
19:57<MrGandalf>Because at it stands, Myth only supports one provider at a time unless you manually tweek channums in the database.
19:58<gbee>#3822 - any difference between a PVR150 and a 250? I can't reproduce this problem with my 150
20:19-!-gardz [] has quit [Remote closed the connection]
20:35-!-briand [] has quit [Connection timed out]
20:36-!-briand [] has joined #mythtv
20:43-!-hemul [] has joined #mythtv
20:43<hemul>hi people
20:45-!-reynaldo_ [] has joined #mythtv
20:46-!-xris [] has quit []
20:52-!-reynaldo1 [] has joined #mythtv
21:01-!-gardz [] has joined #mythtv
21:03-!-reynaldo_ [] has quit [Read error: 110 (Connection timed out)]
21:04-!-reynaldo [] has quit [Read error: 110 (Connection timed out)]
21:18-!-grim[GameOp]_ [] has joined #mythtv
21:28-!-gardz [] has quit [Read error: 101 (Network is unreachable)]
21:30-!-turbo [] has joined #mythtv
21:31-!-briand [] has quit [Connection timed out]
21:33<sphery>Do we have any functionality for choosing a temp directory? Bonus points if it allows specifying a preference for local filesystems.
21:34<Chutt>uh, /tmp?
21:35<sphery>Guess I could use that and let the Windows guys sort it out...
21:36<MrGandalf>should use $TMPDIR or /tmp
21:36<Chutt>sphery, qt doesn't have one?
21:37<sphery>Good question... It looks like Daniel used /tmp in DataDirectProcessor.
21:38<sphery>QT4 has QTemporaryFile
21:38<hads>Yeah, can't see anything in 3 though
21:39<sphery>Yeah. I'll look more into what we're doing with DataDirectProcessor. If it hasn't caused issues for the Windows port there, it won't with my code. :)
21:39-!-xris [] has joined #mythtv
21:43<MrGandalf>sphery: btw, my shorter schedule runs were short lived.. now they're up to just over 20 seconds.
21:45<sphery>Hmmm. Looks like createTempFile() (in libs/libmyth/util.cpp) handles it for Windows, but it takes a template argument that includes the dir... Windows ignores the whole template. Whatever. Guess I can use createTempFile()...
21:46<sphery>MrGandalf: That prety much disproves my theory about the amount of changes to program causing it to be unoptimized/needing analysis for indices.
21:47<MrGandalf>sphery: maybe I should switch to Oracle :)
21:47<MrGandalf>oh wait, then I'd need a 2ghz+ system with 4 CPUs and 8gigs of memory
21:47<sphery>I think you should try Apache's Java-based derby... :)
21:48<MrGandalf>yeah, there ya go :)
21:49<MrGandalf>I'm not familiar with manual recordings.. if a manual recording is setup, it should also exist in record, correct?
21:49<MrGandalf>maybe a simple check on that table before running the sql would work
21:49<MrGandalf>I'll have to experminent this weekend.
21:50-!-j-rod [n=njarod@nat/redhat/x-98338348ee0b9469] has joined #mythtv
22:04<sphery>Anyone know why DataDirectProcessor::GetPostFilename() (libs/libmythtv/datadirect.cpp +1751) would make a QDeepCopy of the QString returned by createTempFile()? The same is done for every other usage of createTempFile() in DataDirectProcessor. Wonder why createTempFile() doesn't just do that...
22:05<sphery>(no other usage of createTempFile() in mythtv)
22:05-!-turbo [] has quit [Connection timed out]
22:05-!-turbo [] has joined #mythtv
22:30-!-briand [] has joined #mythtv
22:32-!-turbo [] has quit [Read error: 110 (Connection timed out)]
22:36-!-mikew [] has joined #mythtv
22:41-!-reynaldo1 [] has quit ["leaving"]
22:41-!-reynaldo [] has joined #mythtv
22:43-!-jmk_ [] has quit ["Leaving"]
22:43-!-jhulst [n=jhulst@unaffiliated/jhulst] has joined #mythtv
22:58-!-turbo [] has joined #mythtv
22:58-!-briand [] has quit [Read error: 110 (Connection timed out)]
23:15-!-mikew [] has quit [Read error: 110 (Connection timed out)]
23:23-!-jk1joel [] has quit [Remote closed the connection]
23:23-!-jk1joel [] has joined #mythtv
23:43-!-jmk [] has quit ["Leaving"]
23:46-!-jmk [] has joined #mythtv
23:51-!-kormoc [n=kormoc@unaffiliated/kormoc] has quit []
23:53-!-ahbritto [] has quit [Client Quit]
---Logclosed Thu Feb 07 00:00:04 2008