Back to Home / #mythtv / 2003 / 03 / Prev Day | Next Day
#mythtv IRC Logs for 2003-03-03

00:06<_shad>sleepy time
00:06-!-_shad [] has quit ["[BX] Tony the Tiger uses BitchX. Its Grrrrrrrrreat!"]
00:24-!-rkulagow [~mythtv@] has joined #mythtv
00:38-!-PeteCool [] has quit ["Client Exiting"]
00:53<-- danc_(~danc`` has left #mythtv
01:01* bigguyis [away -={ SLEEP }=- ]
02:05-!-paperclip [] has quit ["grits.. they aren't just for breakfast anymore"]
02:22-!-Soopaman [] has joined #mythtv
02:29-!-bigguy [] has quit ["brb"]
02:29-!-bigguy [] has joined #mythtv
02:29-!-bigguy [] has quit [Remote closed the connection]
02:30-!-bigguy [] has joined #mythtv
04:11-!-FredFunk [] has quit [Read error: 110 (Connection timed out)]
04:19-!-bigguy [] has quit ["A yawn is a silent shout."]
04:30-!-FredFunk [] has joined #mythtv
04:31-!-bigguy [] has joined #mythtv
04:46-!-bigguy [] has quit ["bbiaf"]
04:55-!-choenig [] has joined #mythtv
05:00-!-bigguy [] has joined #mythtv
05:08-!-Soopaman [] has quit [Read error: 60 (Operation timed out)]
06:14-!-justin [] has quit [Read error: 60 (Operation timed out)]
06:23-!-FredFunk [] has quit [Read error: 60 (Operation timed out)]
07:08-!-FredFunk [] has joined #mythtv
07:30-!-the_anti_poptix is now known as divine_poptix
08:34<divine_poptix>holy cow, xmltv is really broken
08:34<divine_poptix>bad regex in 0.5.8
08:53-!-FredFunk [] has quit [Read error: 110 (Connection timed out)]
08:57-!-schwin97 [~schwin97@] has joined #mythtv
09:30-!-FredFunk [] has joined #mythtv
09:52<Chutt>why do people submit patches with lots of totally unnecessary changes?
10:01<divine_poptix>Chutt: are you still running the old xmltv?
10:04<divine_poptix>the new one is broken
10:04<divine_poptix>can't wait for the flood of emails on the lists
10:05<Chutt>0.5.8's been out for awhile
10:05<divine_poptix>it uses some complete ass backwards regex, and can't parse its own config file
10:05<divine_poptix>has it? hmm
10:05<divine_poptix>it's still broken.
10:05<Chutt>couple weeks, iirc
10:05<divine_poptix>had to fix the regex to make it work
10:05<divine_poptix>could be dependant upon an older version of perl
10:47-!-schwin97 [] has quit ["Client exiting"]
11:14-!-schwin97 [~schwin97@] has joined #mythtv
11:36-!-rickter [] has joined #mythtv
11:56-!-paperclip [] has joined #mythtv
12:29-!-paperclip [] has quit [Remote closed the connection]
12:41-!-rickter [] has quit [Read error: 54 (Connection reset by peer)]
13:08-!-toddm [] has joined #mythtv
13:09<-- toddm( has left #mythtv
13:09-!-toddm [] has joined #mythtv
13:10-!-Soopaman [] has joined #mythtv
13:10<Chutt>mdz, could you take a look at that manual recording addition if you get a chance?
13:15<mdz_>Chutt: on the mailing list?
13:15<mdz_>ok, I haven't read it yet today
13:18<-- toddm( has left #mythtv
13:30<hurdel>i'm running cvs from 4 or 5 days ago, with the new volume patch included, and now each time i start watching live tv or a recording, the volume is LOUD
13:30<hurdel>is there a way to save the volume, did I miss a post or readme on how to do this?
13:34-!-paperclip [] has joined #mythtv
13:39<Chutt>VIA wants to call me, now
13:39<Soopaman>new job?
13:39<hurdel>Chutt : VIA the computer company? or the Canadian railroad? LOL
13:40<Chutt>they want mythtv
13:40<Soopaman>don't give it to them
13:40<hurdel>that's cuase it's kicks ass over everything else out there and they know it
13:40<Chutt>it's gpl'd, they can use it however they want
13:40<Soopaman>oh real
13:40<hurdel>tivo will look like sheee-at once myth gets to 1.0
13:40<Soopaman>Chutt, you may find yourself in a paper soon
13:41<Soopaman>chutt vs tivo & microsoft
13:41<hurdel>point chutt
13:41<Soopaman>or should it be chutt^mdz vs tivo vs microsoft
13:41<hurdel>chutt, how much time per week you spend on myth?
13:42<Chutt>i dunno
13:42<Chutt>depends on how much real work i have to get done
13:42-!-justin [] has joined #mythtv
13:42<Chutt>and if i've got other things to do
13:43<Soopaman>Chutt, if i were to join my embedded dist. with mythtv, we would have an environment way more capable than that linuxtv api they are pushing for thousands of dollars
13:44<hurdel>anybody still working on the bootable mythtv cd?
13:45<hurdel>i'm heading out on a trip for 16 days, gotta start selecting my shows for recording now
13:46<hurdel>Chutt : did via say what they actually wanted? closed src ver of myth???
13:47<Chutt>they sent a 2 sentence email
13:47<Chutt>want to call me, which they're not going to get to do =)
13:47<hurdel>sweet :)
13:48<hurdel>i'm assuming you just don't want the headache of multiple versions?
13:48<Chutt>i don't want the headache of dealing with them
13:49<hurdel>do you ask for the copyright to the code when others submit patches, or leave it all publicly open (including copyright)
13:49<Chutt>copyright belongs to the person that wrote the code
13:50<hurdel>that's what I thourgt
13:51<hurdel>i get giddy every time i use myth, can't thank you developers enough
13:51<hurdel>now i just have to recompile my bleeding kernel to get remote working and my wife is hooked too
14:31<hurdel>is Lindows still dicking around with Myth?
14:32<Chutt>haven't heard from them on the mailing list in awhile
14:32<Soopaman>they probably trolling
14:32<Soopaman>wouldn't be surprised if microsoft tried to buy you out
14:33<Chutt>why would they do that?
14:33<Soopaman>because you are making a better version of their MCE
14:33<Soopaman>it would be WAY cheaper for the big 4
14:33<Soopaman>or anyone to convert to linux
14:33<Soopaman>and use myth
14:33<Soopaman>than get licenses for xp/embedded and MC
14:34<Chutt>using xmltv for data wouldn't fly in a commercial product
14:35<Soopaman>aslong as people don't have any other feasible choices, they can charge the premium
14:35<Soopaman>true, but that would be an easy solution
14:35<Soopaman>i currently have semi rights to gemstar programming
14:36<Soopaman>but because of their nda's, i can't give you code for myth
14:36<Soopaman>but if you were to hack my computer....
14:36<Chutt>not really needed
14:36<Chutt>i'm happy with xmltv
14:37<Soopaman>can you download a week at a time with xmltv?
14:37<Soopaman>descriptions and all/
14:37<Soopaman>how is that possible?
14:39<mdz_>solarce: you can't "buy out" a free software project
14:39<Soopaman>how/who do you grab the feeds from?
14:39<mdz_>er, Soopaman
14:39SoopamanMar 03 14:39:47 <jdanner> Soopaman: they come from
14:39<Soopaman>mdz, yeah but you can cut off it's head
14:39SoopamanMar 03 14:40:08 <mdz_> Soopaman: you can only chase it underground
14:40<Soopaman>which would stiffle its progression
14:40<Soopaman>true, but if M$ or someone else would take you and chutt aout of the picture
14:40<Chutt>i'd happily work for ms if they paid me well
14:41<Soopaman>it would take a while for the project to get back to the pace it currently has
14:41<Chutt>they're a good employer, from what i've heard
14:42<Soopaman>hell if i was a ruthless company with deep pockets, i would rape the open source movement
14:42<Soopaman>because they have the most motivated/productive employees
14:42<Soopaman>err would contain
14:44<Soopaman>and change the "rape" with exploit
14:44<Soopaman>'cause that sounds really bad
14:56<Chutt>i wonder if i can make mailman require that only one of the lists be mailed to
14:58<jdanner>to stop people posting to both -dev and -users?
15:20-!-jdanner is now known as moegreen
15:21-!-danc_ [~danc``] has joined #mythtv
15:37-!-schwin97 [] has quit [Remote closed the connection]
15:44-!-schwin97 [~schwin97@] has joined #mythtv
16:00-!-paperclip [] has quit [Remote closed the connection]
16:02-!-paperclip [] has joined #mythtv
16:03<mdz_>Chutt: what is andy davidoff's xinerama patch supposed to do?
16:04<danc_>confuse everyone
16:10<mdz_>it looks like it explicitly disables xinerama handling even when the X server says that xinerama is in use
16:10<mdz_>which seems completely silly
16:11<mdz_>sounds like his window manager isn't xinerama-aware, and so his solution is to change mythtv
16:11<danc_>see? i didn't even look at the patch and i knew what was up
16:12<danc_>i'm surprised he submits patches. he does this "submit first, discuss second" thing then gets upset when people aren't happy with his patches
16:20<Chutt>mdz, well, it is configurable to ignore it
16:24ChuttMar 03 16:24:48 <mdz_> Chutt: yes, but what does it accomplish?
16:25<Chutt>i dunno
16:25<Chutt>i'm just ignoring his stuff now
16:32<Chutt>i dunno
16:32<Chutt>maybe some day he'll contribute something useful
16:33<danc_>maybe. he obviously has the interest and the time. he doesn't seem to be clued in to group development though
16:39-!-choenig [] has quit [Remote closed the connection]
16:44<Chutt>heh, so apparently gcc 3.3 can't translate a long long into a union that's the same size as a long long automatically
16:47<justin>Chutt: if someone uses another program for watching tv, what happens when your watching tv and its time for a recording?
16:48<Chutt>you're screwed
16:48<Chutt>too bad
16:48<justin>thought so:)
16:48<Chutt>there's no way to do the osd in overlay mode, so
16:48<danc_>i wonder if you can cast it as (mmx_t)(long_long)
16:48<Chutt>you can't
16:49<Chutt>error: no matching function for call to mmx_t::._24(long long int)
16:49<Chutt>it's easy enough to fix, i suppose
16:49<Chutt>but it's rather silly
16:49<danc_>yeah. named assignment?
16:51<Chutt>naw, just have to stick braces around it, i think
16:52<Chutt>yeah, that works
16:52<mdz_>Chutt:'re not actually doing that anywhere, are you (long long <-> union)
16:53<Chutt>mdz, the rtjpeg stuff
16:53<Chutt>mmx_t is a union
16:53<Chutt>it's just defined as that for different types of access
16:53<mdz_>is one of the elements of the union a long long?
16:54<Chutt>all of the elements add up to a long long
16:54<Chutt>there's a long long
16:54<Chutt>2 longs
16:54<Chutt>4 shorts, etc
16:54<mdz_>so why not refer to the long long element instead of the union itself?
16:54<Chutt>it's a static variable initialization
16:55<mdz_>c102 initializer?
16:55<Chutt>it was: static mmx_t blah = (long long)0xblahblahblah;
16:55<mdz_>er, c99
16:55<mdz_>couldn't you use that then?
16:56<mdz_>static mmx_t blah = { elementname : value }
16:56<mdz_>I've never used it with a union, but it seems like it should work
16:56<danc_>my thought too. but it's kinda funky if you have to name the members of the union
16:56<Chutt>mdz, that works as well
16:57<mdz_>is it a union because it wants to access the high and low words or something?
16:57<Chutt>so does leaving off the element name
16:57<Chutt>mdz, yeah, there's just multiple ways of accessing the data
16:57<mdz_>I don't think you're supposed to do that
16:58<mdz_>wasn't that the whole big deal with -fstrict-aliasing?
16:58<Chutt>i dunno
16:59<Chutt>i told that dumbass starkeeper to search the mailing list archives
16:59<Chutt>obviously, he can't read
16:59<mdz_>the docs say: "type-punning is allowed, provided the memory is accessed through the union type. "
17:00<mdz_>so it is sort of OK as long as you don't, say, put the address of one of the elements in the union into a pointer of the same type of that element
17:01<mdz_>I think it's still completely nonportable though
17:01<Chutt>nonportability's fine =)
17:01<Chutt>this is all ifdef MMX
17:03<Chutt>probably should go through and make sure everything else compiles with the gcc-snapshot package while i'm at it
17:03ChuttMar 03 17:03:44 <mdz_> Chutt: is this gcc-snapshot you're using?
17:03<mdz_>oh, heh
17:03<Chutt>since that guy complained about it on the mailing list
17:04<mdz_>I'm way behind on the list. lots of traffic lately.
17:04<Chutt>hey, they finally deprecated multi-line string literals
17:04<Chutt>well, removed, they've been deprecated
17:05<Chutt>(ie, goom doesn't compile) =)
17:05-!-choenig [] has joined #mythtv
17:10<mdz_>ah, all caught up
17:10-!-bigguy [] has quit [Read error: 113 (No route to host)]
17:10<mdz_>funny, it was mostly a lot of thread deletion
17:11<mdz_>I must've missed the complaint about gcc
17:11<Chutt>gcc 3.3 and mythtv
17:11<Chutt>Mark Rickard
17:13<mdz_>on mythtv-dev?
17:13<mdz_>ah, I see it
17:14<mdz_>must've gotten mixed in with the 'setup borked in cvs' thread which I deleted
17:19-!-bigguy [] has joined #mythtv
17:19-!-sc00p [] has quit [Read error: 54 (Connection reset by peer)]
17:25-!-Krowi [] has joined #mythtv
18:04<-- Krowi( has left #mythtv
18:22<moegreen>heh, Dreamix
18:28-!-SoopaMech [] has joined #mythtv
18:30-!-Soopaman [] has quit [Read error: 60 (Operation timed out)]
18:31-!-choenig [] has quit [Remote closed the connection]
18:34-!-Soopaman [~soopaman@] has joined #mythtv
18:38-!-SoopaMech [] has quit [Read error: 60 (Operation timed out)]
19:05<Chutt>think i have all the backend<->backend stuff wrapped
19:12<moegreen>what else is left for the release?
19:12<Chutt>debugging this stuff i just wrote
19:13<Chutt>i'm not entirely done with it yet, either
19:14<moegreen>Chutt, should there ever be a 'Changing from None to None' at the frontend?
19:14<Chutt>ah, not really, but it shouldn't hurt anything
19:15<moegreen>it happens when I watch a recording, hit escape from the recording, then I can't move around the list/menu anymore.
19:15<Chutt>sure that's not the same focus bug other people are seeing?
19:16<moegreen>it certainly could be. Is the listbox not taking focus back?
19:16<Chutt>it works fine for me
19:17<moegreen>alright, I wasn't aware of anyone else with that problem, I see if I can mess around with it later
19:26-!-toddm [] has joined #mythtv
19:26<danc_>word up t-dog
19:27-!-rickter [] has joined #mythtv
19:28-!-justin [] has quit [Read error: 110 (Connection timed out)]
19:46-!-sc00p [~oldendic@] has joined #mythtv
19:48<rickter>dan, you certainly gave it to that dreamix guy...
19:49<mdz>Chutt: I get None to None transitions all the time
19:49<danc_>he made it way too easy
19:50<mdz>av@lime:~$ fgrep -c 'None to None' mythfrontend.out
19:50<mdz>doesn't cause any problems though
19:52<Chutt>now that i look at things, i do too
19:52<rickter>danc_: I just hope he responds, i'd love to see it :D
19:53<Chutt>*ahem* -users exists for a reason
19:53<Chutt>not that i'd think he'd be a user, but =)
19:53<danc_>yeah, sorry about that, i knew it would be off topic, but he was just soo stupid
19:56<rickter>he's probably fishing for developers
19:56<danc_>nope, he
19:56<danc_>err. he's trying to see if he can drum up some interest
19:57<Soopaman>interest in what?
19:58<danc_>his product. maybe to raise money, or to lobby some other cause. show that people would buy it
19:58<Soopaman>what is his product?
19:58<Chutt>notice his sig, btw
19:58<Soopaman>nice logo though
19:59<danc_>i liked how "boot splash screen" was one of their top 4 accomplishments
19:59<rickter>yah, too bad that's all he's got so far
19:59<Chutt>the other completed accomplishments were photoshop mockups of uis
19:59<danc_>anyways. it really is a good idea. there is something similar for ps2, which streams audio and video across the broadband adapter
20:00<rickter>I couldnt even find the UI screenshots on that site
20:00<danc_>i just dont think he is telling the whole story. he is telling parts that he wants people to believe. that's why he points out the debian part.
20:00<Chutt>they were hidden last time
20:00<Chutt>i forget the url
20:00<Soopaman>man, i really don't like evangelists...
20:00<rickter>yah, the xbox could certainly handle it, but like you said, no way MS will let that go on
20:01<Chutt>ram would be tight
20:01<Chutt>and you do need an external encoding stuff
20:01<Chutt>usb mpeg2 encoder, say
20:09<rickter>lol, poptix even posted "vaporware. try" on the dreamix forum on
20:10<moegreen>that is just a wierd open source product, "At this time we are accepting resumes from anyone wanting to help out"
20:10<danc_>it's just sillyness
20:12-!-justin [] has joined #mythtv
20:24<moegreen>ok, I think i've fixed that focus problem
20:33<mdz>he can't answer; his window lost keyboard focus
20:36<moegreen>i've added some showFullScreen() and setActiveWindow() commands on startup, after the player is killed and when the player exits normally and it seems to be working for me now. I'm just going through to remove any of them that aren't needed.
20:36<moegreen>this is for the playback window
20:39-!-Soopaman [] has quit [Read error: 54 (Connection reset by peer)]
20:39-!-Soopaman [~soopaman@] has joined #mythtv
20:55<moegreen>Chutt: I can see the windows taking focus after I hit escape from watching the video. The column headers of the playback listbox change in color
20:57<Chutt>so, it's all fixed, or..?
20:57<Chutt>well, the remote backends work for tv viewing
20:58<moegreen>Chutt: yeah, I think it's fixed, I'm just trying a bunch of combinations to make sure I didn't screw up a focus problem somewhere else :)
21:00<Chutt>thanks =)
21:00<Chutt>people had been complaining about that for awhile
21:08<Chutt>this damn thing takes too long to recompile on the p3-550
21:12<moegreen>if it makes you feel better it takes too long to recompile on my athlon @ 1Ghz
21:12<Chutt>i wonder if turning off the excessive opts would do anything
21:12<-- toddm( has left #mythtv
21:15<moegreen>you could always upgrade with this fine specimen from everyone's favorite retailer:
21:19<Chutt>multiple backend playback works seamlessly =)
21:20<Soopaman>moegreen, wow, that is dirt cheap
21:20<Soopaman>Chutt, can you do pip from backends?
21:21<Chutt>any free tuner card
21:21<moegreen>it's also the only place i've seen a 10 gig hard drive in the last 2 years
21:21<Chutt>moegreen, i just started getting a X Error: BadWindow on exiting the playback box
21:21<Chutt>after updating with your fix
21:21<Soopaman>how does the user select the available viewing devices?
21:21<moegreen>Chutt: I'm not seeing that overhere. hmm...
21:22<Chutt>wait, it's not that
21:23<Chutt>doesn't always trigger
21:23<Chutt>not what i thought it was, then
21:25<Chutt>is the showFullScreen/setActiveWindow pair in killPlayer() necessary?
21:25<Chutt>that's just the little preview window, nothing to do with the playback
21:25<moegreen>i tried a couple different combinations and I think it didn't work without it there, let me check that part again
21:26<Chutt>that's what's causing it
21:26<Chutt>it's telling it to show a window that's been killed, when that window is going away
21:27<moegreen>it does indeed work without there.
21:27<Chutt>actually, the other place is semi-incorrect as well
21:27<Chutt>it would be better to do that in play()
21:27<Chutt>before the if (tv->getRequestDelete()) line
21:28<moegreen>should it be before the listview->SetAllowKeypress(true); ?
21:29<Chutt>it could be
21:29<Chutt>shouldn't really matter, i don't think
21:29<Chutt>well, if that works for you, at least
21:29<Chutt>if it doesn't work there, then leave it in startPlayer =)
21:30<moegreen>i'll give it a shot ...
21:31<moegreen>it works in play() as well, I'll recommit the changes :)
21:32<moegreen>ok, should be in there...let me know if there are still issues
21:35-!-Captain_Murdoch [] has joined #mythtv
21:35<divine_poptix>hey chutt
21:36<divine_poptix>found a new logo for mythtv
21:36<moegreen>ha, good stuff!
21:37<danc_>what is google?
21:37<danc_>(i win)
21:37<rickter>i was scared for a second
21:37<Chutt>chris just committed his newest commercial code
21:38* Captain_Murdochis Chris
21:38<Chutt>i'll check it out later
21:38<Captain_Murdoch>now have to see if it works as well for everyone else as it has for me. :)
21:38<Chutt>in the middle of testing all this multiple backend crap
21:39<moegreen>well let's saturate chutt's dsl everybody!
21:40-!-sc00p [] has quit [Read error: 104 (Connection reset by peer)]
21:40<Chutt>ooh, need to make the pixmap preview fall back to the slowass seeking method
21:42<Captain_Murdoch>chutt, did you write the NuppelVideoPlayer::ReencodeFile function? does it work at all? I'm tossing around the idea of starting to work on a function that can be called that will read through a whole file and can apply some/all of the commercial detection stuff to it to come up with a commercial break list which can then be inserted into the database.
21:42<Chutt>it used to work
21:42<Chutt>i doubt it does now
21:42<Chutt>the libavcodec api's changed a bit, and so has the mythtv internals it's referring to
21:44<Captain_Murdoch>ok, I'll just stick with following how tv_play starts up the player. I was thinking of putting this detection function into NuppelVideoPlayer.cpp since it has everything else needed to play through the file. Instead of calling StartPlaying the caller would call the new function to go through and detect commercials. then that function can be called out of the scheduler or an external program.
21:45<Chutt>well, something _like_ reencodefile would probably be best for that
21:46<Captain_Murdoch>once I get something like that working, then I can try to take Liam's patches for the new detection and integrate those as well. I figure run the file through 2-3 or more detection routines parallel or sequential to get a few lists then compare them to get the best shot at the real commercial break list.
21:48<moegreen>Captain_Murdoch: do you have to hit the commercial skip less than 5 secs into the first commercial for it to work best?
21:48<Captain_Murdoch>I'm thinking that to detect the commercials in some methods it will require seeking and didn't want to reinvent the wheel.
21:48<Chutt>moegreen, not with the new stuff =)
21:48<Captain_Murdoch>moegreen, yes with CVS as of 30 minutes ago. :) new cvs should flag them as you record if you have the detection method set.
21:48<Captain_Murdoch>still not catching 100% but getting better
21:49<moegreen>ok, b/c you almost had to have the remote in hand and be ready for it =) Excellent!
21:49<Chutt>moegreen, the new stuff works more like a cutlist, it knows where stuff is in relation to where it's playing
21:49<Captain_Murdoch>the new code stores the blank frame list (and eventually the commercial list) in a table so seeks are pretty much instantaneous.
21:51<moegreen>any thrill seekers with gcc 3.X want to try out some new mythweather stuff for me?
21:51<Chutt>commit it
21:52<Chutt>i'll fix anything that's broke with gcc 3.x
21:52<Captain_Murdoch>is editing in a working state now?
21:52<Chutt>should be
21:53<Captain_Murdoch>ok, must be me then. It tried it a couple days ago but must have been doing something wrong. I was working on getting the editor to load the commercial break list if it existed.
21:54<Chutt>yeah, editing is working
21:54<Chutt>unless you broke it, of course =)
21:56<Captain_Murdoch>displaying video full-screen at 1280x1024 over fast ethernet sure is slow. :)
21:57<Chutt>back in a few
21:59-!-danc_ [] has quit ["Trillian ("]
22:03* Captain_Murdochponders the ability to hit a key while watching a video that will insert a mark at that point which then shows up in the editor. ie, like multiple bookmarks.
22:04<Chutt>i don't want things _too_ complicated
22:06<Captain_Murdoch>:) was only thinking of it when I saw SetBookmark in the code.
22:06<Captain_Murdoch>when I hit 'M' to go into edit mode it just freezes. time to debug.
22:07<Chutt>hit e to go to edit mode
22:07<Chutt>m should be going into the epg
22:08<Captain_Murdoch>keys.txt says e or m while watching a recording. e froze also.
22:10<Chutt>whaddya know, m works too
22:10<Chutt>m doesn't exit edit mode, though
22:11<Captain_Murdoch>ok. it works on your end with current cvs?
22:11<Chutt>i haven't updated since your checkin
22:12<Captain_Murdoch>it's stopping at the getpause in enableedit. I'm looking into it.
22:13-!-Timon [] has joined #mythtv
22:14<Chutt>bah, you didn't rename the markup table
22:15<Captain_Murdoch>sorry, was that in your email?
22:15<Chutt>just thought it was rather generic
22:15<Captain_Murdoch>rename to what? I'll fix it.
22:16<Chutt>recordedmarkup or something more specific
22:16<Chutt>is all i wanted =)
22:16<Chutt>markup could refer to lots of things
22:16<Captain_Murdoch>ahh, I see. I didn't scroll down far enough in your reply.
22:16<Captain_Murdoch>I'll fix it quick and update.
22:17<Chutt>cool, thanks
22:17<Captain_Murdoch>the editing problem is because I'm runing remote and it can't pause audio I think. I'll try at the TV in few minutes.
22:17<Chutt>moegreen, that your mythweather update you wanted tested?
22:18<moegreen>it *should* work :)
22:18<Chutt>it's compiling fine
22:18<Chutt>make that 'compiled', now
22:18<Timon>moegreen, you sound like a true programmer :-)
22:18<moegreen>Chutt: be sure to do a make install from the main directory
22:20<Chutt>your makefile trickery didn't work
22:20<Chutt>it ungziped the file, but didn't install it
22:20<moegreen>did you have an old Makefile in mythweather/ ? I updated and it should have installed it
22:20<Chutt>just did a make distclean to make sure
22:21<Chutt>you don't need to gzip it, though
22:21<Chutt>hopefully people are using the -z option to cvs =)
22:21<moegreen>well it's pretty big and I wasn't sure about that
22:22<Timon>-z? Thats that do?
22:22<Chutt>you can if you want to, but you don't have to
22:22<Chutt>it installed it properly this time
22:23<Chutt>nice setup stuff
22:23<moegreen>cool, to access the setup just hit 'i', just need to make sure the cities are listed in there
22:23<moegreen>the 7 key goes down 25, 8 down 50, 9 down 100 cities
22:23<Chutt>how about adding another level of letters to the city select?
22:24<moegreen>1, 2, 3 are the opposite and 4 to the beginning, 5 to the middle, and 6 to the end
22:24<moegreen>heh, oops.
22:24<Chutt>the unknown transparency doesn't look too hot =)
22:25<moegreen>that's just a debug signal saying you changed something
22:25<moegreen>really? hmm...let me see if I can find an unknown weather type
22:25-!-rickter [] has quit ["Trillian ("]
22:25<Chutt>hit up willoughby, oh
22:25<Timon>Isaac, I'm working on a patch to mythgame to support multiple subdirectories for roms. As such, it will require me to indent large blocks of code to keep consistent formatting. Do you want me to turn on ==ignore-whitespace when I make a diff?
22:27<Chutt>you're reindenting things?
22:27<Chutt>just match what's in the file already
22:27<Chutt>is my preference
22:27<Timon>Not reindenting, just moving stuff in as about 300 lines are now going to be part of a for loop
22:28<moegreen>you're right...hmm...I'll mess around with that some more, did everything else work ok?
22:28<Chutt>yeah, works great
22:28<Chutt>noticed you didn't use the existing settings code :p
22:28<moegreen>getting that list of cities wasn't fun, but its done now!
22:28<Chutt>yeah, very nice =)
22:29<moegreen>i wasn't sure about that, I tried setSetting() but it just changed it temporarily...and then i remembered some discussion on the list about it, so I just took the long way around :)
22:29<Chutt>matt's stuff is pretty easy to use
22:30<Chutt>takes care of saving and loading automatically
22:30<Captain_Murdoch>Chutt, should the InitSound() function set disableaudio if it can't open the audio device?
22:30<Chutt>i thought it was, yes
22:30<Chutt>oh, wait
22:30<moegreen>well there is stuff for mythweather in the frontend's settings right now, I'll take that in a little while
22:30<Chutt>no, it doesn't
22:30<Captain_Murdoch>it just returns now. let me see if that fixes my edit problem to make sure.
22:31<Chutt>should be able to just stick a disableaudio = true before the return in there
22:31<Captain_Murdoch>that's what I did. recompiling now to test.
22:31<Chutt>thanks for changing that table name
22:31<Captain_Murdoch>sorry I missed it the first time. :)
22:32<Captain_Murdoch>makes more sense as recordedmarkup
22:33<Captain_Murdoch>much better, editing works now.
22:33<Captain_Murdoch>do you just want ot commit that disableaudio change on your next commit or should I?
22:33<Chutt>go ahead
22:35<moegreen>Chutt: i've updated the unknown.png, let me know what you think
22:35<moegreen>and now I'll figure out what Unknown #223 is
22:36<Chutt>heh, much nicer =)
22:41<Chutt>now i just need to record something new so i can test out the new commercial skip code
22:41<mdz>Chutt: looked at that manual recording patch
22:41<mdz>it looks pretty good except I think manualrecord should have a lower numeric value so that it takes precedence
22:41<mdz>unfortunately, that means converting the db
22:41<Chutt>does the single record stuff join against the program table?
22:41<mdz>yes it does
22:42<Chutt>does it need to?
22:42<mdz>maybe not
22:42<Chutt>if it doesn't
22:42<Chutt>manual could just be that
22:43<Chutt>the only field not in record is category
22:43<Chutt>and i don't keep that anyway
22:43<mdz>I thought singlerecord used to join with the program table before I changed that, but I guess it didn't
22:43<Chutt>probably should
22:44<mdz>I was planning on changing that stuff anyway
22:44<mdz>to join with oldrecorded
22:45<mdz>I think it does an extra query for that right now
22:45<mdz>for each recording
22:45<mdz>it could just prune it out in the very beginning
22:46<mdz>though it might be nice if it gave some feedback instead
22:46<Chutt>sounds like a good idea
22:46<mdz>I guess that isn't exclusive though
22:46<mdz>it seems to confuse people
22:46<mdz>maybe it should just print a message to the backend log
22:47<Chutt>well, just make sure it doesn't prune single records if you change that =)
22:48<Chutt>let's see
22:48<mdz>changing it to ignore the program table for singlerecords would be pretty easy I think
22:48<mdz>a strategically placed OR in the sql
22:48<Chutt>all i have left to test wrt multiple backends is the master telling a slave when to record
22:48<Chutt>everything else seems to work ok
22:49<Chutt>you are populating the record table with all the info you know at the time, right?
22:49<Chutt>i think so, at least
22:50<Chutt>so single record could just be the manual stuff
22:50<mdz>hmm, not as simple as I thought
22:50<Chutt>that wouldn't work for manually overriding the start time of a timeslot recording, though
22:50<mdz>I'm retrieving program.starttime and program.endtime
22:50<Chutt>but, that's fine by me
22:50<mdz>I am not convinced that the time overriding stuff works at all
22:50<mdz>I have never tried it, but it looks fishy
22:50<Chutt>it probably doesn't
22:51<Chutt>but i was just going by the query there
22:51<Chutt>since timeslot stuff kind of needs to know when to look =)
22:51<mdz>yes, record table is populated with everything it knows
22:51<mdz>but if, for example, the recording type is changed, it doesn't bother to null out the fields
22:52<Chutt>since it doesn't use em, then
22:52<Chutt>but what if it's changed from all record to a single record?
22:52<mdz>currently that can only happen through the epg
22:52<mdz>so it has a full programinfo
22:52<Chutt>ok, good
22:52<mdz>I'll have to remember to deal with that case in the recording editor when I get around to it
22:53<mdz>I guess it should just disallow that particular transition
22:53<mdz>only go from more information -> less
22:53<Chutt>probably should only be able to promote types
22:54<mdz>it could do something clever like search for the first match, but I don't think that'd ever really be what you'd want
22:54<FredFunk>Hey guys -- I'm halfway through 'Fix Conflicts' in mythweb; mind clearning up some confusion for me? Currently if there's a conflict neither programs are recorded? Second it appears I can fix conflicts just by updating the database, am I missing anything?
22:54<moegreen>that might happen in the progfinder as well
22:54<moegreen>it calls the InfoDialog like the epg does
22:54<Chutt>if there's a conflict, it picks by itself
22:55<Chutt>if the user didn't bother to tell it what to record
22:55<mdz>FredFunk: it breaks the tie, I think the lower chanid
22:55<Chutt>well, more important type of recording first
22:55<Chutt>then longer
22:55<Chutt>then lower chanid
22:55<Chutt>i believe
22:55<Chutt>figuring you'd want to record a single record over a timeslot
22:55<FredFunk>Okay; I'd like to show which one will actually be recorded if possible (so if you don't care you don't have to adjust it)
22:55<Chutt>timeslot over an allrecord, etc
22:56<Chutt>fredfunk, can't get that info =)
22:56<mdz>I was thinking it might be nice if, rather than being pruned out completely, duplicates showed up as suppressed, like a conflicting recording that lost
22:56<Chutt>that's true
22:56<mdz>so it would be grayed out in the conflict screen
22:56<Chutt>there wouldn't really be a way to reactivate them, though
22:57<Chutt>at least not with the current scheduler
22:57<mdz>that's another thing
22:57<mdz>I noticed, I think, that it adds the oldrecorded entry early now
22:58<FredFunk>Okay I can resolve the conflicts just by updating the database right? It appears there's backend commands going on to accomplish this
22:58<FredFunk>'no backend' commands
22:58<Chutt>just update the conflict tables in the db and set that one setting to yes
22:58<Chutt>just like the frontend does
22:59<Chutt>i never bothered to write backend commands for those
22:59<FredFunk>right, I didn't know if there was something that triggered the UI to know things were updated or anything along those lines, it looks straight forward
23:02<mdz>Chutt: so if the recording is interrupted or fails somehow, future occurrences of the same program are still suppressed
23:02<mdz>and there is no way for the user to fix that from the UI
23:02<mdz>I thought it used to add the oldrecorded entry at the very end
23:02<justin>i use phpmyadmin a lot:)
23:02<Chutt>it used to
23:03<justin>mdz: i had a recording that got messed up because another app switched the recording device
23:03<Chutt>but then i changed how you stopped in progress recording
23:03<justin>before i started using btaudio:)
23:03<Chutt>and i didn't bother to split the recorded and oldrecorded inserts up
23:03<mdz>Chutt: that doesn't affect oldrecorded, though, right? did it get brought along for the ride?
23:03<Chutt>they were just in the same function
23:04<Chutt>the oldrecorded insert could piggyback on chris's markup table insert, now
23:04<Chutt>if you wanted to move things over easily =)
23:04<mdz>does that happen unconditionally?
23:04<Chutt>i belive so
23:05<mdz>it will always scan for black frames now?
23:05<Chutt>but i'm not certain
23:05<Captain_Murdoch>markups table insert for blank frames only happens if curRecording is set and killPlayer is not.
23:05<Chutt>currently, yes
23:05<Captain_Murdoch>mdz, it scans during record, takes virtually 0 cpu time.
23:06<Captain_Murdoch>doesn't mark commercials just keeps a list of blank frames for later.
23:07<Chutt>i'll have to work on re-encoding for 0.9
23:08<Chutt>i should put up a new survey on the website
23:08<Chutt>i'm _really_ tempted to do the 'most annoying mailing list member' thing
23:08<Captain_Murdoch>chutt, did you delete Kulagowski's FindCommercial() function somewhere along the line?
23:09<Captain_Murdoch>can we vote for that already, I've got my choice already.
23:09<bigguy>well it's a toss up
23:09<Chutt>i should probably just turn off comments and the survey
23:09<bigguy>I'm not a big fan of old Henk
23:09<Chutt>i don't give a damn about what random people visiting the site think =)
23:11<moegreen>I like the comment on HDTV support and the person who mailed the list and made the exact same comment in the survery section.
23:12<Chutt>i'd support hdtv cards if there were drivers
23:12<Chutt>that'd be incentive for me to buy a new tv
23:12<mdz>I'd support hdtv cards if there were drivers AND I had a set
23:12<mdz>and some way to receive hdtv broadcasts
23:13<Chutt>i don't know how good reception is here, either
23:13<mdz>and there were actually any broadcast content in hd
23:13<Chutt>csi =)
23:13<moegreen>a lot of the network's primetime are in hdtv, but other than that Discovery Channel and HBO/Showtime
23:17-!-sc00p18 [~sc00p18@] has joined #mythtv
23:17<Chutt>i'd hate to see people bitching about hdtv being too much for their slow, slow computers to play, though
23:19<moegreen>hmm...i got a commit digest 13 hours early.
23:19<Chutt>might've gotten too large and hit that limit instead of the day
23:19<moegreen>it's 13K, 13 commits
23:20<Chutt>threshold is 30k
23:22<moegreen>they seem to be using the PVR-250
23:22<Chutt>require it, in fact
23:24<justin>that sagetv looks nasty
23:24<justin>and it isnt even free
23:24<Chutt>it's only $75
23:24<Chutt>and $5 a month for data
23:25<justin>who wants to pay for a monthly fee
23:25<Chutt>people that want good data? =)
23:25<mdz>I would pay $5/month for xmltv data
23:25<mdz>for good data
23:26<Chutt>for the full zap2it feed, i would as well
23:26<Chutt>with actor info and better show descriptions
23:27<moegreen>heh, and you thought the TV app looked pretty bad:
23:27<moegreen>looks like a java app
23:28<Chutt>people on avsforum say it works well, though
23:28<Chutt>compared to, what is it, snapstream and showshifter
23:28<Chutt>or whatever they're called
23:30<bigguy>I've used showshifter and its a pile of crap
23:31<mdz>I'm pretty sure that our pal andy is full of shit in his indexing claims
23:31<mdz>a column which may be NULL is not the same as a variable length column
23:32<Chutt>but he's scaled a mysql db up to zillions of records!
23:32<bigguy>ad I've ate a whole xtra large supreme by myself so?
23:32<bigguy>and even
23:33<bigguy>some of his tricks might work when dealing with super complex data or something
23:34<Chutt>have i missed anyone's patches (besides his)?
23:34<mdz>ooh, another helpful list of feature requests
23:34<mdz>ok here is everything I want plz do it ok tx
23:34<Chutt>i just get tons of email and forget about the ones i want to respond to
23:34<Chutt>then i have a hell of a time finding them again
23:34<bigguy>"Can you put in an auto-ScratchMyBack feature?"
23:35<Timon>create a folder and file those ones in a folder to be gotten to later
23:35<Chutt>i do
23:35<Chutt>but i forget to copy em over most of the time
23:37<mdz>that scene change detection patch adds two new configuration options but provides no way to set them
23:37<Chutt>you're supposed to know what to do
23:38<Timon>mdz, the only thing I see in my mysql book regarding nulls and indexing is that isam tables can't index null tables, but myisam can. Myisam will only allow you to put an index on a nullable column if the column isn't a primary key (Kinda of a duh one if you ask me)
23:40<mdz>wow, big commit day
23:40<Captain_Murdoch>mdz, I was planning on looking at integrating those other commercial patches once I get it so we can process offline for commercials.
23:40<Captain_Murdoch>they need a little work before then I think.
23:40<Chutt>0.8's getting close
23:41<Chutt>careful not to break stuff =)
23:41<mdz>Chutt: are you running a multiple-backend setup continuously now?
23:42<Captain_Murdoch>Chutt, I'll probably wait until 0.8 to commit the next commercial stuff.
23:42<Chutt>i just got got that all working a couple hours ago
23:43<Chutt>oh, try to add a new recording profile
23:43<Timon>Chutt, whats your day job if you don't mind my asking
23:43<mdz>0.8 has more than enough now functionality :-)
23:43<Chutt>see if it dies for you like it does for me
23:43<Chutt>timon, i write code
23:43<mdz>adding a new recording profile is a pointless exercise at this point
23:43<mdz>Chutt: this just started recently?
23:43<Chutt>mdz, i wanted to add one for the 2nd encoder box
23:43<Chutt>haven't tried recently
23:43<Timon>I figured that :-) For whom? You work for a multi-nationaly mega corp?
23:43<Chutt>it just refused to
23:43<mdz>it used to work
23:43<Chutt>yeah, i know
23:43<mdz>refused, eh?
23:44<Chutt>buncha errors =)
23:44<mdz>care to share the error messages?
23:44<Chutt>they're long gone
23:44<Chutt>sorry =)
23:44<Chutt>it was awhile ago
23:44<Chutt>i was going to look into it tomorrow
23:45<mdz>what's the verdict on the remotelineedit stuff?
23:45<Chutt>i dunno
23:45<mdz>are you going to change things to use it?
23:45<Chutt>i like it
23:45<mdz>it doesn't seem to break regular keyboard input, so I don't see why not
23:45<Chutt>it's a little odd for entering numbers with a regular keyboard
23:45<Chutt>but, that's ok, really
23:45<mdz>ah, yes. hmm.
23:46<mdz>oh, does it make an enormous edit box for you?
23:46<mdz>it did for me
23:46<mdz>multiple lines
23:46<Chutt>i didn't test it on _everything_ like you did, though
23:46<Soopaman>man, i love nhl94...
23:46<Chutt>and you said that the timeout was too fast?
23:46<Soopaman>too bad i can't get mariokart working on my pocketpc
23:47<mdz>I would press a number, then have to look to see what came up to tell how many more times to press it
23:47<mdz>(or to see what letter I got, since I didn't know the mapping)
23:47<mdz>and while I was looking it would time out and insert the letter
23:47<Chutt>ah, ok
23:47<Chutt>i'll see about lengthing it
23:47<Chutt>or ask thor to
23:48<mdz>hmm, seems to be ok now as far as the hugeness
23:48<mdz>oh, one other thing
23:48<mdz>it doesn't have a textChanged() signal
23:48<mdz>or didn't
23:48<Chutt>maybe it's not ready, then =)
23:49<mdz>without that, it would never have actually modified a setting I don't think
23:50<mdz>hmm, it inherits qtextedit, which does have a textchanged
23:50<mdz>maybe it was my old qt being screwy
23:50<mdz>no, it's even there in 3.0
23:50<mdz>but I quite clearly remember getting an error about it
23:51<mdz>in fact, it's still in my scrollback buffer
23:51<mdz>QObject::connect: No such signal MythRemoteLineEdit::textChanged(const QString&)
23:51<mdz>QObject::connect: (sender name: '-edit')
23:51<mdz>QObject::connect: (receiver name: 'unnamed')
23:51<mdz>oh, the qtextedit textchanged() has no argument
23:51<mdz>what use is that
23:52<mdz>yeah, it's not ready
23:52<Chutt>ah, and he inherits textedit instead of lineedit
23:53<mdz>that would also explain why it's huge
23:53<mdz>maybe he needed qtextedit to get the rich text
23:53<Chutt>check out moegreen's new mythweather config stuff, btw =)
23:53<mdz>yeah, qlineedit doesn't do that
23:54<mdz>I don't even have mythweather checked out; I've never used it
23:54<Chutt>just give it a try
23:54<mdz>I have a window right next to my TV :-)
23:54<Chutt>bah, he did some neat stuff for the config
23:56<Chutt>that focus issue must be with non-qt windows
23:56<Chutt>someone just posted the same fix as moegreen but for mythvideo