Back to Home / #mythtv / 2003 / 03 / Prev Day | Next Day
#mythtv IRC Logs for 2003-03-12

00:00<thor_>But I would not be suprised if Qt resolves to the same thing
00:00<thor_>err ... mysql
00:00<mdz>if it did, its date type wouldn't be very suitable for SQL dates
00:00<mdz>I'm pretty sure SQL dates have a wider range than (32-bit) unix timestamps
00:01<inman>dates alone are 4 bytes in mysql and datetimes are 8.
00:01<thor_>yeah, you're probably right ... been sitting in front of the computer too long .. not thinking straight
00:01<mdz>"The supported range is '1000-01-01 00:00:00' to '9999-12-31 23:59:59'"
00:01<mdz>according to the mysql docs
00:01<mdz>that is a very long time in the future
00:01<inman>y10k bug
00:06<rkulagow>hey guys, i'm getting this when running mythprogfind: (this is on a remote frontend running QT3.1)
00:06<rkulagow>QTime::setHMS Invalid time 28:00:00.000
00:06<rkulagow>QDate::setYMD: Invalid date 2935-93-02
00:06<rkulagow>i saw in scrollback that you guys were talking about this.
00:06<rkulagow>just wanted to add a datapoint if this was a qt3.0 vs. 3.1 potential issue.
00:08<mdz>Chutt: I'd like to fix that viewscheduled warning by finishing handleDuplicate so that it displays the show info. should it work ok to cut and paste the code from the constructor?
00:08<mdz>rkulagow: please try the stuff I just added to cvs.sql and see if they go away
00:09<thor_>rkulagow, do you have any mechanism in place for images (gif's, jpeg's) in the doc system?
00:09<mdz>the gridlayout is the top level of that part right?
00:09<Chutt>mdz, it might be better to cut and paste from the infodialog stuff
00:12<rkulagow>mdz: i'm syncing now and i'll let you know.
00:12<rkulagow>thor: i suppose i could add graphics. what do you have in mind?
00:12<mdz>Chutt: hmm...does that have the channel in it?
00:12<Chutt>i don't know
00:12<mdz>doesn't seem to
00:12<mdz>rkulagow: thanks
00:13<thor_>rkulagow: I keep meaning to write up some docs for LCD devices ... would be nice to have a picture as example and motivation
00:13<mdz>Chutt: it would be nice to make a routine to create one of these out of a programinfo
00:14<mdz>dunno where you would want it though
00:14<rkulagow>thor: if you give me the picture and docs, i guess that will give me the motivation to figure out how to get it in there. :)
00:15<thor_>sounds good ... I will send them along at some point
00:16<Captain_Murdoch>just FYI, I just made the cvs.sql date/time not null mods and tested adding a timeslot recording via mythweb and still worked fine. filled in 0000-00-00 for the dates same as all the other entries got when I did the alter table.
00:18<inman>the rdbms is your friend. :-)
00:24<rkulagow>mdz: ran the cvs.sql on the master db; the remote mythprogfind is still spitting out the same error...
00:24<rkulagow>is this only going to take effect on new records?
00:26<inman>no, will affect everything.
00:26<inman>can you check your db? issue a `show fields from record like '%time';`
00:27-!-Justin_ [] has quit [Read error: 104 (Connection reset by peer)]
00:28<rkulagow>mysql> show fields from record like '%time';
00:28<rkulagow>| Field | Type | Null | Key | Default | Extra |
00:28<rkulagow>| starttime | time | | | 00:00:00 | |
00:28<rkulagow>| endtime | time | | | 00:00:00 | |
00:28<inman>looks fine...
00:29<mdz>rkulagow: select starttime, startdate, endtime, enddate from record
00:29<mdz>rkulagow: when you say the same error, it still says exactly the same numbers?
00:30<rkulagow>mdz: yes, looks like the same thing:
00:30<rkulagow>[mythtv@backend2 mythtv-0.8]$ mythprogfind
00:30<rkulagow>QTime::setHMS Invalid time 28:00:00.000
00:30<rkulagow>QDate::setYMD: Invalid date 2935-93-02
00:30<rkulagow>i've got 5 pairs of these
00:30<inman>are you planning on recording anything in the year 2935?
00:30<inman>that might explain it...
00:31<rkulagow>mdz: pasting...
00:31<rkulagow>mysql> select starttime, startdate, endtime, enddate from record;
00:31<rkulagow>| starttime | startdate | endtime | enddate |
00:31<rkulagow>| 19:00:00 | 2003-02-06 | 19:30:00 | 2003-02-06 |
00:31<rkulagow>| 19:30:00 | 2003-02-06 | 20:00:00 | 2003-02-06 |
00:31<rkulagow>| 00:00:00 | 0000-00-00 | 00:00:00 | 0000-00-00 |
00:31<rkulagow>| 00:00:00 | 0000-00-00 | 00:00:00 | 0000-00-00 |
00:31<rkulagow>| 00:00:00 | 0000-00-00 | 00:00:00 | 0000-00-00 |
00:31<rkulagow>| 22:00:00 | 0000-00-00 | 22:30:00 | 0000-00-00 |
00:31<rkulagow>| 00:00:00 | 0000-00-00 | 00:00:00 | 0000-00-00 |
00:31<rkulagow>7 rows in set (0.00 sec)
00:33<Captain_Murdoch>you are getting a pair of those for each row in record that has a date of 0000-00-00. I just ran mythprogfind and it came back with 20 pairs cause I have 20 programs setup in record.
00:34<rkulagow>weird. startdate = 2003-02-06? where'd that come from?
00:34<inman>well, 0000-00-00 isn't a valid QDate.
00:35<mdz>ok, so this doesn't fix it
00:35<Captain_Murdoch>maybe set a default date to 1970-01-01
00:35<inman>hmm, is 0 == 0000? if so, then it should assume it's 1900.
00:35<mdz>it did fix it when I set all of my NULLs to a valid date/time
00:38<mdz>bah, I knew it was at least half-assed, shouldn't have done it
00:40<Captain_Murdoch>is there a way to set default == curdate() or something like that.
00:41<inman>if(somecolumn, somecolumn, curdate())
00:41<inman>if the columns are NULLable, it's easier
00:41<Captain_Murdoch>I mean when you create the table.
00:41<inman>ifnull(somecol, curdate())
00:41<inman>Captain_Murdoch: nope.
00:47-!-Captain_Murdoch [] has quit ["X-Chat!"]
01:03-!-sc00p [~oldendic@] has joined #mythtv
01:11-!-nathanziarek [] has quit [Read error: 60 (Operation timed out)]
01:18-!-mirk_dt [] has quit ["Lost terminal"]
02:01* bigguyheads to bed
03:52-!-david_ [] has joined #mythtv
03:52-!-hurdel [] has quit [Read error: 104 (Connection reset by peer)]
03:52-!-Viddy [] has quit [Read error: 104 (Connection reset by peer)]
03:54-!-david_ is now known as Viddy
04:13-!-hurdel [] has joined #mythtv
08:28-!-nathanziarek [] has joined #MythTV
09:49-!-PeteCool [] has joined #mythtv
09:58-!-nathanziarek [] has quit [Read error: 60 (Operation timed out)]
12:43* paperclipis away: I'm busy
12:44<rkulagow>slow today, or am i broken?
12:44<nziarek>slow today
12:46<rkulagow>i've realized that between work, family and contributing where i can on mythtv i don't even have time to _watch_ tv anymore!
12:48<nziarek>good thing you can tape it all then
12:53<rkulagow>tape? how 80's! my wife still says "tape", even though we've had a tivo for 2 years.
12:54<nziarek>hmmm...i am not any older than the 80s...what do you call it? "Hon, can you Myth Friends tonight?"
12:56<nziarek>I have a question for someone...I wanted to work on MythSports (like MythWeather) and wrote a script in ASP (don't hate me) that does NCAA basketball. If I transferred that to PHP, would that get me any closer to having a working program?
12:57<nziarek>"trasnferred that to PHP" --> "rewrote the code in PHP"
13:02<thor_>There are ways to make PHP run as shell scripts (outside of Apache)
13:03<nziarek>my current script, when done, just come up with a string: dsd,sdsd,sdsd,sdsds,d|sds,sdssd,,sdsd,, that can be split at the "|" and "," into an array.
13:03<nziarek>could this string be passed somehow to the MythSports GUI then?
13:04<nziarek>I don't need to know how yet, just that I am angling down the right path. Step one is learning PHP
13:05<thor_>It's possible ... but not the "right" way to do this (right way would be to learn C++, which is no harder than PHP)
13:06<nziarek>that what y'all keep saying, but I know ASP, PHP is not too far off, and C++ for dummies starts off about 30 steps above my understanding. I'll keep reading though :)
13:06<thor_>You have Qt installed, correct?
13:07<thor_>Go into /usr/local/src/qt/examples/ and start screwing around...
13:08<nziarek>yeah, i did. I didn't understand anything. Like, little things would make sense, but no big picture stuff.
13:09<nziarek>I don't think it helps that I know VB badly; my code is never elegant nor especially correct. It just does what it should.
13:09* paperclipis back (gone 00:26:11)
13:10<thor_>Don't worry about elegance ... just try and make it do things (can I say hello world, can I create a window, can I create a button)
13:10<nziarek>yeah, i know. sometimes I am too impatient for baby steps =)
13:11<thor_>Your biggest problem is probably that VisualBasic has taught you that programming means pointing and clicking in a nice GUI
13:12<nziarek>sorta - I just use notepad for VBScript
13:12<thor_>Ah, well then you're half way there
13:12<nziarek>I only installed VisualC++ because I thought the color coding would help make sense of it all
13:12<nziarek>ok, you inspried me: Hello WOrld, here I come!
13:13<thor_>You'd be surprised how quickly Hello World can become a rudimentary "MythSports"
13:13<nziarek>here's to taht coming true
13:13<Chutt>steal code from mythweather
13:14<Chutt>it's probably doing something similar
13:15<nziarek>i tried :) i think that might be too ambitious for a beginning.
13:18<thor_>Get hello world figured out (.h file go here, .cpp goes there, Make does this, I have to #include things, etc.). Then try a Qt hello world. Then you're about 30% done
13:28<nziarek>all right, I'm game...Hello World, here I come! (I think that was a pun =)
13:38-!-paperclip [] has quit [Read error: 104 (Connection reset by peer)]
13:38-!-paperclip [] has joined #mythtv
13:46-!-MrWilby [] has joined #mythtv
13:50-!-MrWilby [] has quit [Client Quit]
13:59<rkulagow>chutt: are you going to make a general purpose settings screen for 0.9? i'm looking at mythgallery here - not too many variables, quick change to code, and something like mythweather "force setup, then use a hotkey" vs. the standard "seperate menu item" might be good. what do you think?
13:59<rkulagow>general purpose meaning it's going to incorporate module setup, vs. each having their own.
14:00<PeteCool>I think in the root menu there should be a setup section
14:00<PeteCool>and you have everything there
14:00<PeteCool>just moving the actual setup screen up on level
14:01<PeteCool>and add module-specific items
14:50<thor_>Chutt, AAAAGGGGGGHHHHHHH your TreeCheckItem's destructors were deleting without new;ing metadata !!!
14:56<Chutt>of course they were
14:57<thor_>They get passed a pointer, they don't allocate memory for it.
14:57<thor_>so whoever allocated the memory should be deleteing it
14:57<Chutt>in the original databasebox.cpp stuff, they were passed in a pointer that they then were owning.
14:58<Chutt>i didn't really want to keep track of the metadata pointers in the databasebox object, so it seemed reasonable to have the listview own its data
14:58<thor_>Yeah, I know ... it's my fault ... but the owners are now way up in main() (hit DB once), and I've been trying to find this segfault for 2 hours
14:58<thor_>debugging destructors sucks
15:00<thor_>Ah .. works like a dream now ... Select Music comes up in well under a second
15:00<Chutt>anything left over is qt being slow, then =)
15:00<thor_>Well, I would like to presort right at the beginning
15:01<thor_>but let me get the popups cleaned up and try and get you a copy of this before I die of old age
15:21<inman>nice work thor
16:01<mdz_>Chutt: does the setup device selector work OK for you?
16:02<Chutt>sorry, i haven't tried it yet
16:06<mdz_>works fine for me, but people complained about segfaults on the list
16:07<mdz_>I got that when there were no devices there, but I added a check for that before committing it
16:07<mdz_>instead of giving back an iterator with no items in it, it gives a NULL pointer.
16:14<Chutt>tried it with bad permissions?
16:21<mdz_>but I've never seen a mainstream system where /dev isn't o+rx
16:26<thor_>mdz, how about badly setup devfs
16:34<mdz_>thor_: it would have to be intentionally broken
16:34<mdz_>nothing that two random people on the mailing list would do
16:43<thor_>mdz, I noted the use of "two"
16:58-!-coffmant [] has joined #mythtv
17:00<coffmant>Does anybody have any CPU number for dual MPEG4 encoding/single decode
17:02<PeteCool>>xp 1800+
17:03<PeteCool>Chutt says his xp 1800+ can almost do it
17:14-!-eli [] has joined #mythtv
17:24-!-m0tion [] has joined #mythtv
18:13<-- m0tion( has left #mythtv
18:19<-- coffmant( has left #mythtv
19:11-!-Justin_ [~justin@H233-225.STATE.RESNET.ALBANY.EDU] has joined #mythtv
19:12<Justin_>anyone else have the mythtv window not go fullscreen/decorationless?
19:14-!-Chutt [] has quit [Read error: 54 (Connection reset by peer)]
19:15-!-Chutt [] has joined #mythtv
19:15<Chutt>stupid laptop was all locked up.
19:17* Justin_presses up
19:17<Justin_>anyone else have the mythtv window not go fullscreen/decorationless?
19:17<Justin_>i use pwm, 0.70 worked fine, cvs has never worked...
19:17<Justin_>the window is the right size, if i toggle decore and move it, its good
19:18<Justin_>i also tried running mythtv remotely when I was at work, and it did the same thing, and then crashed, shitty windows X server
19:19<Chutt>i don't recall cvs doing anything differently than 0.7 wrt displaying the windows
19:21<Chutt>mdz, the setup device stuff works for me as a regular user
19:30-!-PeteCool [] has quit [Read error: 110 (Connection timed out)]
19:39<Chutt>rkulagow, you around at all?
19:48<mdz>Chutt: yep, me too
19:49<mdz>Chutt: which qt were you on when you tried it?
19:54<mdz>me too
19:55<mdz>qt is almost like having a real standard library
20:16<poptix>i slipped and hurt my back
20:25-!-PeteCool [] has joined #mythtv
20:27<PeteCool>Chutt: do you think you could put a check somewhere to see if the myth files directory is full? when it keeps writing nowhere the box hangs (after many many minutes though)
20:31<mdz>it will only cause problems with programs other than mythtv if you don't have a dedicated filesystem for it
20:31<mdz>otherwise it is harmless when it runs out of space (except to mythtv of course)
20:32<mdz>besides, the system reserves 5% of the space for root, so that things don't break too horribly if it fills up
20:32<mdz>and you shouldn't run mythtv as root
20:34-!-bigguy [] has quit [Read error: 113 (No route to host)]
20:38<yebyen>47 days of uptime on my mythtv box, w000t
20:38-!-PeteCool [] has quit [Read error: 104 (Connection reset by peer)]
20:41<mdz>yebyen: the box, or mythtv?
20:44-!-PeteCool [] has joined #mythtv
20:44<PeteCool>Chutt: had you answered my previous question? XChat crashed :(
20:44<Chutt>mdz did
20:45<PeteCool>can you paste it again please?
20:45<Chutt><mdz> it will only cause problems with programs other than mythtv if you don't have a dedicated filesystem for it
20:45<Chutt><mdz> otherwise it is harmless when it runs out of space (except to mythtv of course)
20:45<Chutt><mdz> besides, the system reserves 5% of the space for root, so that things don't break too horribly if it fills up
20:45<Chutt><mdz> and you shouldn't run mythtv as root
20:46<Chutt>i'll handle full diskspace eventually someday
20:46<Chutt>but, just not yet
20:46<PeteCool>it's already on its dedicated 20 gigs hard disk
20:46<PeteCool>but it often gets full
20:47<mdz>what should it do? delete some of it for you?
20:47<Chutt>going through 4 months worth of commits mail to extract a changelog is going to suck
20:47<Chutt>mdz, yeah, automatic deletion i think would be best
20:47<mdz>aborting upcoming recordings seems like the only sensible action
20:48<Chutt>of stuff you've watched and not told it to keep
20:48<PeteCool>mdz: yep, just stop recording so it doesn't hang/crash
20:49<PeteCool>deleting could be nice too, as long as it was either watched once, or gets removed from the oldrecorded table so it can be recorded again
20:49<Chutt>i'd like to be able to tell it on a per program title and per show basis what not to delete automatically
20:50<Chutt>so, reading over the docs, everything looks good for 0.8 except for the initial setup stuff that still references 0.7 stuff
20:50<Chutt>and my name spelled wrong at the beginning =)
20:52-!-bigguy [] has joined #mythtv
20:52<PeteCool>misspelled wrong? you mean they failed misspelling?
20:53<Chutt>spelled wrong is what i said :p
20:56<Chutt>someone should port mythfrontend to a zaurus
20:56<Chutt>i wonder if it'd have enough power for low-res playback
20:57<mdz>it does
20:57<mdz>not scaling though
20:57<mdz>so you'd have to record at 320x240
20:57<Chutt>might be able to reencode on the fly in the backend
20:57<PeteCool>Chutt: the incomplete recorings made when the disk was full mess up mythfrontend
20:58<PeteCool>as if it tried to test the end of the file, which isn't there
20:58<Chutt>the preview window doesn't hit the end of the file
20:58<Chutt>it plays from the beginning
20:58<PeteCool>I know
20:58<mdz>that means you should have no problem deleting it
20:59<yebyen>mdz: the box
21:00<mdz>yebyen: bah, that's nothing special then
21:00<yebyen>mdz: i've been upgrading mythtv much more regularly than 47 days
21:00<mdz>I wonder how long folks have mythtv running continuously for
21:00<yebyen>i've had it going since march 7
21:06<yebyen>any changes gone into commercial detection lately?
21:07<yebyen>since... 2003-03-06
21:07<Chutt>there won't be any more until after 0.8
21:08<PeteCool>oh, the ads around here are often black backgrounds with a little white text... the ad skipper fallen into it each time I tried it (not that many times, but a few times)
21:11<PeteCool>the commercial detector can be fooled with amostly-black advertisement
21:11<PeteCool>missing a space there
21:15-!-nathanziarek [] has joined #MythTV
21:15<PeteCool>nathanziarek: so did you figure out hello world? :)
21:16<mdz>Chutt: do you have the beginnings of some release notes?
21:17<mdz>oh, you were just dreading it then :-)
21:17<Chutt>well, i've got the UPGRADING file updated
21:17<Chutt>and README has all the old instructions removed, it just references the docs now
21:17<nathanziarek>PeteCool: no!
21:18<nathanziarek>i can't get cout to work :(
21:18-!-nyquiljer [] has joined #mythtv
21:24<nyquiljer>so whats the oldest hardware anyone successfully runs myth on?
21:24<nyquiljer>i'm looking to go the extremely cheap route
21:25<PeteCool>nyquiljer: celeron 1100A with i815 mobo and 512 megs of ram (256 might be fine)
21:26<PeteCool>nyquiljer: that's not very expensive stuff
21:26<nyquiljer>i dont really know intel prices
21:26<nyquiljer>rough ballpark figure?
21:27<PeteCool>I think the mobo + cpu + case cost ~400$ CDN
21:27<PeteCool>it was a gift
21:28<PeteCool>100$ tuner
21:28<PeteCool>65$ tv-out vid card
21:28<nyquiljer>i have a scan converter, i may just use that
21:28<PeteCool>+ remote and receiver I'll be getting soon
21:28<PeteCool>you could get a used card, or a mono one, they'd be cheaper
22:12-!-m0tion [] has joined #mythtv
22:12-!-PeteCool [] has quit [Read error: 54 (Connection reset by peer)]
22:13-!-nyquiljer [] has quit [Read error: 54 (Connection reset by peer)]
22:31-!-PeteCool [] has joined #mythtv
22:35-!-nathanziarek [] has quit [Read error: 60 (Operation timed out)]
22:42-!-PeteCool [] has quit ["Client exiting"]
22:44<Chutt>i can't reproduce that remote frontend hang at all
22:51-!-m0tion [] has quit ["using sirc version 2.211+KSIRC/1.2.4"]
23:03<yebyen>is cvs.sql up to date with current changes?
23:05<Chutt>should be
23:05<yebyen>i just pasted the whole thing into mysql a few lines at a time
23:05<yebyen>didn't seem to be any errors beyond "Err, you already did that last time you upgraded"
23:27<Chutt>heh, the qpixmap->qimage->qpixmap stuff has some issues at 16bpp
23:30<Chutt>at least in 3.0.5
23:56-!-PeteCool [] has joined #mythtv