| Wednesday, December 8th, 2010, 00:00 UTC | ||
| [00:00:00] | markk_: | stuartm: has anything changed in the terra osd theme recently? I seem to be getting font size issues again with xvideo. |
| [00:03:36] | ** iamlindoro has a fun way to solve that ;) ** | |
| [00:15:38] | jannau: | git rm -r themes/terra? |
| [00:16:32] | iamlindoro: | heh, no, I imagine it looks fine under a worthwhile render :) |
| [00:16:37] | jannau: | markk_: make sure you rerun configure |
| [00:16:55] | jpharvey__ (jpharvey__!~jpharvey@host86-185-255-217.range86-185.btcentralplus.com) has quit (Ping timeout: 245 seconds) | |
| [00:16:56] | iamlindoro: | so something like git rm libs/libmythtv/videoout_xv.* |
| [00:18:31] | jannau: | allcodeccs looks for CONFIG_XVMC_DECODER |
| [01:00:41] | gkeen316 (gkeen316!~gkeen316@d75-156-147-11.abhsia.telus.net) has quit (Quit: HydraIRC -> http://www.hydrairc.com <-) | |
| [01:01:06] | gkeen316 (gkeen316!~gkeen316@d75-156-147-11.abhsia.telus.net) has joined #mythtv | |
| [01:04:19] | [1]gkeen316 ([1]gkeen316!~gkeen316@d75-156-147-11.abhsia.telus.net) has joined #mythtv | |
| [01:04:39] | sphery: | Hmmm... I wonder why git describe --dirty --always is showing dirty on everything except mythtv. Is it the lack of tags/names? |
| [01:05:25] | Beirdo: | -dirty means you have uncommitted changes |
| [01:05:46] | gkeen316 (gkeen316!~gkeen316@d75-156-147-11.abhsia.telus.net) has quit (Ping timeout: 255 seconds) | |
| [01:05:47] | sphery: | but I haven't changed anything |
| [01:06:25] | Beirdo: | hmm, I dunno, but that's what it means :) |
| [01:06:37] | Beirdo: | what does git status say? |
| [01:07:55] | sphery: | hmmm.. maybe it's just my repo--seems ok on a new clone of extras |
| [01:08:02] | sphery: | nothing to commit (working directory clean) from status, though |
| [01:08:27] | sphery: | heh, and now the describe works on that repo |
| [01:08:35] | sphery: | guess I just needed to status first |
| [01:11:07] | Beirdo: | how odd |
| [01:11:21] | Beirdo: | I think I'm going home now. |
| [01:11:37] | Beirdo: | Fighting with asterisk is not fun. |
| [01:16:30] | iamlindoro: | Heh, fun, got Dailymotion working with remote now too |
| [01:21:44] | jannau (jannau!~janne@chybek.jannau.net) has quit (Ping timeout: 260 seconds) | |
| [01:23:22] | sphery: | iamlindoro: nice. |
| [01:23:50] | Captain_Murdoch: | stuartm, I didn't add the feature, but my wife and I use that feature to jump to another program during playback. we do it sometimes when one of us has to leave the room for a few minutes. Then,we use the keybinding to takeus back to the previous program when the other person comes back in the room. |
| [01:25:39] | jannau (jannau!~janne@chybek.jannau.net) has joined #mythtv | |
| [01:36:46] | Captain_Murdoch: | sphery, is your mythtv.org email registered on github? |
| [01:36:58] | Captain_Murdoch: | the commit email said "no github username" for your commit |
| [01:37:31] | sphery: | I typod the address |
| [01:37:51] | sphery: | sorry... in my git config, I had an extra t in there--should have been mdean@mythtv.org |
| [01:37:55] | Captain_Murdoch: | ah, see your comment in -users now. :) |
| [01:38:26] | sphery: | I've fixed my config, now, though. But of all the things I could mess up on my first git commit, that's probably not too bad of one, right? |
| [01:39:05] | Captain_Murdoch: | nah, trivial, and your position in the commit-count race won't be affected much if it's just one commit. :) |
| [01:39:43] | sphery: | heh, yeah, I'll just make sure to commit something, then wait for someone to push, then I'll push to get an extra merge commit |
| [01:44:27] | iamlindoro: | This needing to git pull before git pushing thing is obnoxious as hell |
| [01:45:06] | iamlindoro: | Guess I just get some bonus commits courtesy of the nice people at git :P |
| [01:46:18] | jannau (jannau!~janne@chybek.jannau.net) has quit (Ping timeout: 260 seconds) | |
| [01:46:24] | jannau (jannau!~janne@chybek.jannau.net) has joined #mythtv | |
| [02:19:40] | pheld (pheld!~heldal@cl-5.osl-01.no.sixxs.net) has quit (Quit: Leaving.) | |
| [02:42:57] | ceros (ceros!~andres@xbmc/staff/ceros) has quit (Remote host closed the connection) | |
| [02:43:12] | ceros (ceros!~andres@c-68-49-39-112.hsd1.dc.comcast.net) has joined #mythtv | |
| [02:43:12] | ceros (ceros!~andres@xbmc/staff/ceros) has joined #mythtv | |
| [02:43:12] | ceros (ceros!~andres@c-68-49-39-112.hsd1.dc.comcast.net) has quit (Changing host) | |
| [03:06:51] | danielk22: | markk_: It is quite possible it is no longer effective. The threading was quite a bit different when it was added. IIRC it was about a 200 ms difference so it should be noticable even without looking at the timing. |
| [03:09:22] | markk_: | danielk22: I did a little timing this morning and startup times on 2 different machines improved slightly and exit times were much improved. it also fixes an issue whereby playback failure took far too long to timeout – though the root cause was different. |
| [03:10:01] | danielk22: | well good riddence then :) |
| [03:12:58] | markk_: | anything to simplify tv.cpp ! |
| [03:34:46] | jpharvey (jpharvey!~jpharvey@host86-135-83-20.range86-135.btcentralplus.com) has joined #mythtv | |
| [03:52:10] | kenni (kenni!~kenni@pfsense.dhcp.pop.k-net.dk) has quit (Ping timeout: 245 seconds) | |
| [04:00:07] | [1]gkeen316 ([1]gkeen316!~gkeen316@d75-156-147-11.abhsia.telus.net) has quit (Ping timeout: 264 seconds) | |
| [04:00:12] | m4xx (m4xx!~spooky@c-76-19-95-158.hsd1.ct.comcast.net) has joined #mythtv | |
| [04:00:57] | m4xx: | when i have live tv running and i try to play a small sound byte from the command line using mplayer it freezes |
| [04:01:14] | m4xx: | is there any way i can successfully do so? |
| [04:02:31] | m4xx: | using mplayer or any other app |
| [04:08:12] | ** wagnerrp points at the topic ** | |
| [04:08:32] | m4xx: | oh crap, sorry i knew that |
| [04:08:35] | m4xx: | sorry for intruding |
| [04:08:53] | m4xx (m4xx!~spooky@c-76-19-95-158.hsd1.ct.comcast.net) has left #mythtv () | |
| [04:13:41] | AriX (AriX!~Ari@c-76-99-118-183.hsd1.pa.comcast.net) has quit (Quit: Leaving...) | |
| [04:14:43] | AriX (AriX!~Ari@c-76-99-118-183.hsd1.pa.comcast.net) has joined #mythtv | |
| [04:15:28] | AriX (AriX!~Ari@c-76-99-118-183.hsd1.pa.comcast.net) has quit (Client Quit) | |
| [04:16:35] | kenni (kenni!~kenni@pfsense.dhcp.pop.k-net.dk) has joined #mythtv | |
| [04:46:51] | jya (jya!~avenardj@60-242-40-141.static.tpgi.com.au) has quit (Quit: jya) | |
| [04:49:47] | holomntn (holomntn!~Joe@71-93-231-1.dhcp.mdfd.or.charter.com) has quit (Ping timeout: 255 seconds) | |
| [04:52:26] | holomntn (holomntn!~Joe@71-93-231-1.dhcp.mdfd.or.charter.com) has joined #mythtv | |
| [04:53:06] | jya (jya!~avenardj@60-242-40-141.static.tpgi.com.au) has joined #mythtv | |
| [04:59:42] | gkeen316 (gkeen316!~gkeen316@d75-156-147-11.abhsia.telus.net) has joined #mythtv | |
| [05:12:49] | iamlindoro: | jesus |
| [05:12:55] | iamlindoro: | what is happening with commit messages |
| [05:12:59] | iamlindoro: | it's resending EVERYTHING |
| [05:13:30] | ** iamlindoro waits for the storm to die down ** | |
| [05:22:10] | kenni (kenni!~kenni@pfsense.dhcp.pop.k-net.dk) has quit (Ping timeout: 245 seconds) | |
| [05:22:51] | xris: | iamlindoro: could be someone did a merge and then a push |
| [05:23:29] | wagnerrp: | it was exactly that |
| [05:23:44] | xris: | so not dupes. |
| [05:23:45] | wagnerrp: | jya merged the last few days of master commits to his audio branch |
| [05:23:51] | xris: | and then pushed |
| [05:23:54] | xris: | that makes sense |
| [05:24:01] | kenni (kenni!~kenni@pfsense.dhcp.pop.k-net.dk) has joined #mythtv | |
| [05:24:09] | xris: | iamlindoro: check the branch name in the commit message |
| [05:24:10] | wagnerrp: | so theyre dupes |
| [05:24:10] | iamlindoro: | So... we want this to happen? |
| [05:24:16] | iamlindoro: | Because that's insane |
| [05:24:36] | jya: | just do like me: ignore mythtv-commit :) |
| [05:24:39] | xris: | they're duplicate diffs, but different commits |
| [05:24:45] | iamlindoro: | I'm about to start |
| [05:24:57] | xris: | but yeah.. git log is WAY more useful than commit emails |
| [05:24:59] | jya: | I use the github interface, far less confusing |
| [05:25:37] | iamlindoro: | xris, Most of us still rely on e-mails to keep up to date-- should we abandon that practice? Another casualty of this migration? |
| [05:25:54] | xris: | we should filter based on branches.... |
| [05:26:12] | jya: | I push my branch to master, because my main dev machine is a VM... And to try on another PC, I can't easily accept the git repo on the VM from the local network |
| [05:26:20] | xris: | maybe only send messages out for commits directly to master |
| [05:26:56] | xris: | Beirdo and I were discussing git with a coworker at lunch today... the basic bit of understanding that separates git from svn is to realize that svn revisions are slices in time across the whole repo. |
| [05:27:10] | xris: | e.g. you can check r12345 out from any branch and you will get something.. |
| [05:27:22] | xris: | but with git, each revision is a node on a tree, tied to one or more parents... |
| [05:27:43] | wagnerrp: | perhaps change the email to indicate branch in the topic |
| [05:27:54] | xris: | so when you merge in svn it creates a new patch against that branch — probably a single commit for the entire merge (even if you're merging 100 bits) |
| [05:27:56] | wagnerrp: | so they can be filtered/organized as such |
| [05:28:11] | xris: | when you merge in git, that original node just gets a second (third, etc) parent node. |
| [05:28:24] | wagnerrp: | instead of 'mythtv commit', do 'master commit', or 'fixes/0.24 commit' |
| [05:28:25] | xris: | and a new sha to go with it |
| [05:28:36] | xris: | wagnerrp: well, mythtv is there to separate extras, mythweb, etc. |
| [05:28:47] | xris: | but yeah.. I think getting it into the subject would be handy |
| [05:28:51] | wagnerrp: | mythtv/master, mythtv/fixes/0.24 |
| [05:29:05] | Beirdo: | that can be done |
| [05:29:07] | xris: | and only sending out notifications for master and fixes/* |
| [05:29:24] | Beirdo: | that I'm not so keen on |
| [05:30:00] | xris: | meh. I'm with JYA on the "ignore commits" comment... |
| [05:30:10] | xris: | what we really need is a way for devs to mark the commit as "must read this" |
| [05:30:25] | iamlindoro: | all commits to trunk are must read |
| [05:30:27] | xris: | actually... now that we're generating the emails, we could parse commit messages for a tag like that, and bump up msg priority |
| [05:30:34] | Beirdo: | but if that's what is desired, sure, we can look at filtering |
| [05:30:40] | iamlindoro: | That's just a responsibility of being a dev and/or running trunk |
| [05:30:45] | xris: | iamlindoro: there are too many for everyone to read all the time. |
| [05:30:56] | xris: | I haven't read the commits list regularly in about 4 years |
| [05:30:57] | wagnerrp: | 'trunk' being 'master' |
| [05:31:18] | xris: | as a mythweb dev only about 1 in 1000 commits affected me |
| [05:31:44] | iamlindoro: | xris, And as a steering committee member, you still don't feel an obligation to keep track? |
| [05:32:05] | xris: | of projects and directions, not individual commits |
| [05:32:20] | iamlindoro: | I just think that's unfortunate |
| [05:32:34] | xris: | besides. I have maybe an hour a day of personal time at a computer |
| [05:32:37] | iamlindoro: | Because then in addition to trying to get you on board with something, it falls on the rest of us to bring you up to speed |
| [05:32:48] | iamlindoro: | And FWIW, I had noticed you don't read commits |
| [05:32:56] | davide: | markk_: do you intend to apply your tv thread change to fixes/0.24? before you decided to commit it to master, i applied it to my 0.24 to test with this evening. the ui seemed to be slightly more responsive and i didn't notice any ill effects. |
| [05:33:38] | jya: | it's not realistic to expect everyone to read all commits... and don't see the link of being in the steering comitee and reviewing all the comits. Just like you don't expect your project manager to review in details everything you do on a day to day basis. |
| [05:33:51] | xris: | iamlindoro: like I said. a way to mark commits as important would do the job better. the only commits I regret not reading are the ones where people use the commit message as documentation, instead of writing actual documentation. |
| [05:34:18] | jya: | xris: that's a good point there.. too often we have those. |
| [05:34:39] | jya: | And I'm one of the first to do it |
| [05:34:41] | xris: | iamlindoro: jya's right, though. steering committee isn't here to babysit things... we're here to offer advice on direction when it's needed, and resolve disputes that you can't amongst yourselves. |
| [05:34:48] | xris: | mythtv is an open source project, not a company... |
| [05:35:00] | iamlindoro: | Well, we'll all just have to agree to disagree then-- In my opinion accepting commit rights means accepting the responsibility of keeping yourself informed of what's going on-- expecting others to spend their time recounting commits to you because you choose not to read them is just disrespectful |
| [05:35:23] | xris: | believe me, over half of our team would have been fired several times over the years if we operated like a real company |
| [05:35:36] | xris: | iamlindoro: I don't expect people to recount commits. |
| [05:35:45] | iamlindoro: | xris, And yet I've had to do so a number of times to you |
| [05:36:00] | davide: | fwiw, i try to read all of the commit messages. all of these merge dupes are going to make that tough. isn't there any way we can filter out just those on the mailing list? |
| [05:36:00] | xris: | there's a difference between recounting commits and answering questions about commits. |
| [05:36:19] | xris: | remember, I'm not a c++ dev... I can read the code but rarely understand much of it |
| [05:36:24] | Beirdo: | davide: I'm trying to concoct a plausible way of doing so |
| [05:36:30] | iamlindoro: | xris, Is it still asking questions when the question is specifically and pointedly addressed in the commit message? |
| [05:36:52] | xris: | was it information that should have been in documentation? |
| [05:36:54] | iamlindoro: | If it is, then it's just giving it a new name to paint a rosier picture |
| [05:37:12] | iamlindoro: | No, it was not |
| [05:37:13] | jya: | that falls back to the case of not proper documentation written and instead put in the commit log :) |
| [05:37:30] | iamlindoro: | it was specific information about protocol, metadata, and mythvideo internals |
| [05:37:33] | xris: | I think most of my questions are generally of the "yes", "no", and "look here:..." variety. |
| [05:37:57] | iamlindoro: | I happen to write more documentation that anyone besides perhaps wagnerrp-- so accusing me of not writing it is a straw man |
| [05:38:04] | xris: | even if I read commit messages I wouldn't necessarily be able to remember when something specific happened. |
| [05:38:17] | iamlindoro: | Anyway, as with every discussion, this is a waste of my time |
| [05:38:18] | jya: | we should be having a protocol and architecture document... commit logs are really not designed for that |
| [05:38:34] | xris: | anyway, I think it's wrong that we use commit messages to communicate changes in behavior. |
| [05:38:45] | iamlindoro: | Go on asking questions, but if you're not reading the commits, I'm just not going to answer them any more |
| [05:38:53] | wagnerrp: | the doxygen stuff is supposed to be that, but its not used much |
| [05:38:58] | iamlindoro: | then we can both be selfish with our time |
| [05:39:37] | xris: | sorry. but as much as you are *almost* as whiny as my 1-year-old, he still takes priority. call it selfish is you wish. |
| [05:39:51] | jya: | can we take a middle ground approach and try to have a centralised info where you can get all at a glanc e when it comes to architecture, protocol and API ? |
| [05:40:08] | xris: | afk |
| [05:40:11] | iamlindoro: | And as much as you're *almost* a functional member of this team, I'll apportion my energy to those who act like one |
| [05:40:14] | jya: | as wagnerrp mentioned: doxygen is a start |
| [05:40:33] | jya: | but that's often too detailed and you need to know where to look at |
| [05:41:15] | jya: | commit logs are worse, because it changes in time all the time. What you read in a commit log, may not be relevant any longer. it's not reasonable to have to replay all changes to get to the current info |
| [05:41:33] | wagnerrp: | Beirdo: is that something buildbot could do? nightly refreshes of doxygen which could be stuffed onto the server? |
| [05:41:47] | Beirdo: | yeah, we could do that |
| [05:42:08] | Beirdo: | I hope to start on that soon too, once all the current kerffuffle calms some |
| [05:42:10] | jya: | it's a pity that the only place to get a bit of info about our API and classes is using the stuartm site |
| [05:42:38] | Beirdo: | there's no reason we can't keep a copy of the autogenerated doxygen on the server |
| [05:42:44] | jya: | (doxygen on miffteevee ) |
| [05:42:55] | Beirdo: | which would hopefully inspire us to doxygenize the code more |
| [05:43:10] | jya: | unfortunately, the current doxygen info in our code is close to being useless |
| [05:43:25] | jya: | it's often just showing what arguments it takes |
| [05:43:32] | Beirdo: | mostly because we haven't been trying to do anything with it |
| [05:44:14] | jya: | I've tried adding comments in the code that are suitable for doxygen, but I'm the first to admin I don't spent anywhere as much time as required |
| [05:44:33] | Beirdo: | me neither, I always forget it |
| [05:45:09] | Beirdo: | wagnerrp: you were suggesting "repository/branch"? |
| [05:45:22] | jya: | i usually would quickly wrap something in the commit, and when it comes to remember how it works, absolute pain to retrieve that info |
| [05:45:25] | wagnerrp: | for the commit messages? makes sense |
| [05:45:36] | Beirdo: | OK, I'm tweaking at the moment |
| [05:46:16] | jya: | what's the format in a commit log to put a reference to another commit ? |
| [05:46:23] | jya: | I tried [xxxx] rxxxx |
| [05:46:30] | jya: | nothing appears as a link in gitgub |
| [05:46:36] | Beirdo: | SHA: xxxxxxxx |
| [05:46:40] | jya: | or is this not a git way of doing thing ? |
| [05:46:42] | Beirdo: | that's what github wants |
| [05:46:46] | jya: | ok thanks |
| [05:46:59] | Beirdo: | I think there was some other stuff too |
| [05:47:44] | markk_: | davide: I'm not entirely sure – it potentially has a large impact – so maybe I'll give it a few days and see if anyone's been screaming. |
| [05:48:55] | jya (jya!~avenardj@60-242-40-141.static.tpgi.com.au) has quit (Quit: jya) | |
| [05:48:57] | Beirdo: | I have an idea for suppressing identical commits |
| [05:49:29] | Beirdo: | I'll have to think about how well it will work though |
| [05:49:40] | Beirdo: | meantime, I'll put the branch in the subject |
| [05:49:51] | jya (jya!~avenardj@60-242-40-141.static.tpgi.com.au) has joined #mythtv | |
| [05:50:04] | iamlindoro: | Beirdo, That would be great (especially the former, but the latter works for now) |
| [05:52:08] | Beirdo: | OK. Should be in place as of now |
| [05:52:32] | Beirdo: | for suppression, I'm thinking of doing SHA1 on the commit message + file details |
| [05:52:54] | Beirdo: | + author, perhaps |
| [05:53:56] | Beirdo: | I think if I implement it, I'll have it just mark the duplicates at first, and then if it is marking correctly, THEN have it suppress |
| [05:54:00] | carmony (carmony!~justin@c-24-10-194-172.hsd1.ut.comcast.net) has quit (Quit: carmony) | |
| [05:54:44] | Beirdo: | I'm worried about false positives though |
| [05:57:55] | xris: | Beirdo: shoud be careful with that suppression... you could hide things that get merged into master |
| [05:59:00] | iamlindoro: | Beirdo, Is there no concept of "directionality" (for lack of a better term) of the merges? Could just suppress any merge from master to any branch not-fixes? |
| [06:00:10] | Beirdo: | xris: good point, could suppress only NOT on master, but then it won't do the "trivial merge" suppression either |
| [06:00:25] | xris: | yeah, that's the problem |
| [06:00:39] | Beirdo: | iamlindoro: not really, unless we can match it to the original commit, which is info we aren't given |
| [06:00:45] | xris: | which is why I was thinking about just not showing branches... |
| [06:01:09] | xris: | but that's only because I assume that people working on other branches together are already communicating a lot about the fixes going in |
| [06:02:17] | Beirdo: | well, I still see value to having the emails for the branches, but they are more filterable now anyways |
| [06:02:42] | Beirdo: | but yeah, we could consider only showing master and fixes/* |
| [06:04:42] | Beirdo: | you'd think that we can GET the merge directionality information from the Network API |
| [06:22:10] | superm1: | Beirdo, after playing with it a bit, i'm not convinced submodules are the proper solution for a meta type branch. it seems that in the meta branch you have to constantly be updating the reference commit for submodules, and they're normally in detached states |
| [06:23:03] | superm1: | so the only way that could really work is if the bootstrap script that would live in the meta branch would check out everything to an attached state and then you just stop using the git features of the meta project at that point |
| [06:23:11] | Beirdo: | could well be |
| [06:23:52] | superm1: | which that would beg the question ,why even bother with complexities of submodules |
| [06:31:04] | CppIsWeird (CppIsWeird!~user@unaffiliated/cppisweird) has joined #mythtv | |
| [07:04:05] | gkeen316 (gkeen316!~gkeen316@d75-156-147-11.abhsia.telus.net) has quit (Ping timeout: 265 seconds) | |
| [07:05:08] | xris: | time for whiskey (not to be confused with whisky) |
| [07:19:28] | Beirdo: | my scotch is at work, unfortunately |
| [07:19:45] | Beirdo: | coulda used some today, good thing I didn't have a tumbler at my desk |
| [07:33:56] | XChatMav (XChatMav!~MaverickT@111.86.233.220.static.exetel.com.au) has quit (Ping timeout: 265 seconds) | |
| [07:45:40] | natanojl (natanojl!~jonatan@c83-252-237-63.bredband.comhem.se) has joined #mythtv | |
| [07:46:36] | MaverickTech (MaverickTech!~MaverickT@111.86.233.220.static.exetel.com.au) has joined #mythtv | |
| [07:48:11] | CppIsWeird: | mythtv isnt just something you can install ontop of an existing ubuntu system? it seems to me from my reading that the backend part of mythtv that gets input from tv tuners and so forth have to be mythtv ran systems where as clients will run on other systems. |
| [07:48:38] | xris: | Captain_Murdoch: wrong channel — check the topic |
| [07:48:54] | CppIsWeird: | right, sorry. |
| [07:57:19] | MaverickTech (MaverickTech!~MaverickT@111.86.233.220.static.exetel.com.au) has quit (Ping timeout: 276 seconds) | |
| [07:58:06] | MaverickTech (MaverickTech!~MaverickT@123.200.222.98) has joined #mythtv | |
| [08:01:40] | natanojl (natanojl!~jonatan@c83-252-237-63.bredband.comhem.se) has quit (Ping timeout: 240 seconds) | |
| [08:17:04] | pheld (pheld!~heldal@cl-5.osl-01.no.sixxs.net) has joined #mythtv | |
| [08:23:03] | xris: | ok, svn.mythtv.org now perma-redirects to code.mythtv.org |
| [08:23:27] | xris: | or rather... will once I re-enable it later. |
| [08:33:42] | simonckenyon (simonckenyon!~simoncken@195.7.61.12) has joined #mythtv | |
| [08:40:40] | jpharvey (jpharvey!~jpharvey@host86-135-83-20.range86-135.btcentralplus.com) has quit (Ping timeout: 245 seconds) | |
| [08:46:17] | kormoc is now known as kormoc_afk | |
| [08:52:51] | jpharvey (jpharvey!~jpharvey@host86-135-81-74.range86-135.btcentralplus.com) has joined #mythtv | |
| [09:06:04] | xris: | wow these things run faster when you turn off nested layers of gzip... |
| [09:07:26] | pheld (pheld!~heldal@cl-5.osl-01.no.sixxs.net) has quit (Quit: Leaving.) | |
| [09:14:40] | Beirdo: | hehe |
| [09:14:55] | Beirdo: | gzip-gunzip-gzip-gunzip-gzip? |
| [09:15:48] | xris: | probably |
| [09:15:58] | xris: | doesn't seem to be a way to un-proxy jira's static content, though. |
| [09:16:24] | MaverickTech (MaverickTech!~MaverickT@123.200.222.98) has quit (Ping timeout: 265 seconds) | |
| [09:19:43] | xris: | though it looks like jira clips along pretty quickly once it can cache things.. first load of an issue is slow, but refreshes are pretty fast |
| [09:20:22] | MaverickTech (MaverickTech!~MaverickT@111.86.233.220.static.exetel.com.au) has joined #mythtv | |
| [09:23:50] | xris: | yeah, crazy HUGE cpu usage by apache is definitely trac |
| [09:32:35] | xris: | and... googlebot + trac timeline. ok, that'll stop as soon as it reloads robots.txt |
| [09:34:38] | Beirdo: | no timeline for you, Google! |
| [09:35:07] | stuarta: | hah |
| [09:38:42] | xris: | looks like things have calmed down a bit |
| [09:38:53] | xris: | lol. or not |
| [09:39:01] | xris: | regardless, I'm going to sleep |
| [09:39:25] | Beirdo: | yeah, sounds like a plan |
| [09:39:31] | xris: | made good progress on this stuff today |
| [09:40:55] | xris: | stuarta: if you want to poke around in the mail server, I think getting stuff like basic sending confirmed working, make sure we're not an open relay, turn on magic like greylisting.... |
| [09:41:17] | Beirdo: | we have basic sending |
| [09:41:18] | xris: | would like to find some way to confirm that mailman will work without turning on the whole mailing list |
| [09:41:39] | Beirdo: | we can't be a relay at the moment as we are only listening on loopback :) |
| [09:41:41] | xris: | Beirdo: we should configure jira, fisheye, crowd, wiki for sending, then |
| [09:41:45] | xris: | ah |
| [09:41:48] | Beirdo: | and that's about the extent of it |
| [09:42:20] | Beirdo: | yeah, sure. trac emails are working, the commit emails are working |
| [09:42:22] | xris: | would like to find some way for us to get authenticated smtp running... but I'm just not sure how that'd work |
| [09:42:26] | xris: | ah, good point. :) |
| [09:42:31] | Beirdo: | no reason jira and friends shouldn't |
| [09:43:26] | Beirdo: | hopefully tomorrow I won't be QUITE as wiped out from the day job and can tackle the mail out from jira, etc |
| [09:43:46] | Beirdo: | but if stuarta wants to get it up and running to recieve mail too, even better. |
| [09:43:50] | Beirdo: | :) |
| [09:43:52] | MaverickTech (MaverickTech!~MaverickT@111.86.233.220.static.exetel.com.au) has quit (Ping timeout: 240 seconds) | |
| [09:44:17] | Beirdo: | gonna have to set an MX to point there though to finish testing. Maybe for new.mythtv.org for now? |
| [09:49:04] | xris: | the atlassian apps haven't been configured yet |
| [09:49:35] | Beirdo: | yeah, not tonight :) |
| [09:49:47] | xris: | or a new lists.mythtv.org |
| [09:57:05] | MaverickTech (MaverickTech!~MaverickT@111.86.233.220.static.exetel.com.au) has joined #mythtv | |
| [10:00:21] | stuartm: | ok, commits list is now officially useless |
| [10:09:31] | jamesba (jamesba!~jamesba@gateb.kw.bbc.co.uk) has joined #mythtv | |
| [10:12:18] | wagnerrp: | stuartm: its been partially fixed |
| [10:12:34] | wagnerrp: | commits emails now give the repository and branch |
| [10:12:39] | wagnerrp: | so you can filter them as needed |
| [10:16:14] | stuartm: | I could have filtered them before based on that information being in the body of the message, but I'm supposed to create new filters every time someone creates a new branch? |
| [10:16:35] | wagnerrp: | no, just dump everything but the branches you want |
| [10:17:01] | jamesba (jamesba!~jamesba@gateb.kw.bbc.co.uk) has quit (Remote host closed the connection) | |
| [10:18:40] | jamesba (jamesba!~jamesba@gateb.kw.bbc.co.uk) has joined #mythtv | |
| [10:18:55] | wagnerrp: | like i said... partially fixed |
| [10:19:12] | wagnerrp: | see backlog several hours ago for initial complaints |
| [10:29:37] | simonckenyon (simonckenyon!~simoncken@195.7.61.12) has quit (Quit: KVIrc 4.0.0 Insomnia http://www.kvirc.net/) | |
| [10:38:54] | jya: | `something has been changed very recently in trunk, trying to play some files result in Player(0): Waited 100ms for video buffers |
| [10:40:08] | jya: | like this file http://filebin.ca/wwajzc/nachtschicht.asf |
| [10:53:25] | stuartm: | Beirdo: think we can stick the branch detail into the headers? it's much easier to construct !match rules that way, there is already an X-Repository header, an X-Branch would complete it |
| [11:01:05] | jya: | downgraded to 55baf4532f6836c3681a96441cbbf626cb968eaf (December 1st, and it's working.. |
| [11:03:24] | stuartm: | xris: do we think it's time to split the trac stuff out of -commits? I'd guess most people are already filtering that into a separate folder anyway, it might as well be a separate list |
| [11:23:00] | jya: | ah.. it's a fix of mine .. 98a196941ac35f9a4601bcb93314cda00aa0b7ec |
| [11:23:26] | jya: | must have revealed another bug somewhere |
| [11:28:50] | stuarta: | stuartm: i must be different -commits goes to my -dev inbox |
| [11:36:05] | stuartm: | there's no reason that as two separate lists that it couldn't go to the same inbox, although maybe I'm the only one who thinks having two lists would be better :) |
| [11:36:18] | carmony (carmony!~justin@c-24-10-194-172.hsd1.ut.comcast.net) has joined #mythtv | |
| [11:37:00] | stuarta: | i thought they were |
| [11:37:53] | stuartm: | heh, microsd cards are just too small, opening the packaging on a new one, it's pinged across the room and disappeared |
| [11:38:02] | stuarta: | doh |
| [11:38:18] | pheld (pheld!~heldal@cl-5.osl-01.no.sixxs.net) has joined #mythtv | |
| [11:38:21] | stuarta: | i'm sure your vacuum cleaner will find it |
| [11:43:50] | stuartm: | found it, it had bounced off the wall and fallen in a 1/2" gap between a bookcase and desk, fished it out with a ruler |
| [11:44:59] | stuarta: | :) |
| [11:49:05] | jannau: | stuartm: I would prefer two lists too even if it means I have to change my filter rules again |
| [11:49:35] | stuarta: | mythtv-bugs? |
| [11:55:44] | stuartm: | mythtv-tickets? mythtv-tracker? |
| [11:59:10] | stuarta: | well i've just proved you can't combine mythweb's rewrite rules with lighttpd and a HTTP["url"] clause |
| [11:59:36] | stuarta: | basically lighttpd's rewriting doesn't work within a HTTP["url"] clause |
| [11:59:47] | stuarta: | dropping the url clause and it works well |
| [12:00:30] | stuarta: | nice and fast compared to apache |
| [12:10:57] | stuarta: | why is the branch in the commit messages always coming out as temp/newaudiosetup??? |
| [12:25:27] | stuartm: | stuarta: because all those commits were merged into the newaudiosetup branch ... and yes, this is a problem, but it was apparently the cause of a long argument last night and I've already raised it this morning :( |
| [12:26:03] | stuarta: | bah |
| [12:26:14] | stuarta: | how chumptastic |
| [12:26:20] | stuartm: | only a couple of those commits were original, the rest are just repeating the log messages of commits made to trunk/master when a merge occurred |
| [12:26:26] | stuartm: | stuarta: welcome to git |
| [12:26:57] | stuarta: | hey, i like the idea of git (and it's still an idea for me since i haven't actually tried doing anything in it) |
| [12:27:11] | stuarta: | it's just a tad messy |
| [12:31:39] | stuartm: | I also like the idea of git, or many of the features at least, but I don't think that overall it's very well designed they need someone with experience of user interfaces and usability to step up and manage the development |
| [12:34:56] | stuartm: | as I noted in an email to the -developer list, there seems to be no filter, no editing, every patch seems to have been accepted and every new idea seems to have been implemented |
| [12:37:00] | stuartm: | apparently they actually _removed_ the ability for remote changes to merge with local uncommitted changes which makes no sense, it's something that SVN for example does without issue – so for example if you spot and fix a typo or compile error but someone has beaten you to it, when you 'pull' git will barf |
| [12:37:52] | stuarta: | yeah svn goes, oh they are the same, i'll flag it but continue |
| [12:39:35] | stuartm: | svn just merges them in, marks it with an M in the summary of changes but that's all, git refuses to do anything until you've reverted the changes in your local copy |
| [12:42:59] | stuartm: | I'm pretty certain that xris and Beirdo are sick of my moaning, so I'll shut up now ;) |
| [12:49:02] | stuartm: | oh, is Udo still insisting there are backend leaks? Sure, the last time he was correct, danielk22 and I found and fixed them, what's changed since then? |
| [12:53:05] | Dave123 (Dave123!~dave@cpe-74-74-222-96.rochester.res.rr.com) has joined #mythtv | |
| [12:53:32] | Dave123 (Dave123!~dave@cpe-74-74-222-96.rochester.res.rr.com) has quit (Read error: Connection reset by peer) | |
| [12:53:38] | Dave123-road (Dave123-road!~dave@cpe-74-74-222-96.rochester.res.rr.com) has joined #mythtv | |
| [13:11:59] | stuartm: | ok, so there are some genuine leaks in that log, nothing major, but still worth fixing |
| [13:22:46] | stuarta: | 0 direct, 0 indirect |
| [13:24:44] | stuartm: | stuarta: there are several logs in that file, and some do show genuine direct leaks |
| [13:25:13] | stuarta: | ah |
| [14:23:01] | RDV_Linux (RDV_Linux!~doug@CPE001195554bb4-CM00252eac6f40.cpe.net.cable.rogers.com) has quit (Remote host closed the connection) | |
| [14:24:06] | Goga777 (Goga777!~Goga777@shpd-95-53-165-153.vologda.ru) has joined #mythtv | |
| [14:24:23] | Prost (Prost!~lets.pros@who.knows.what.possessed.us) has quit (Ping timeout: 272 seconds) | |
| [14:27:17] | RDV_Linux (RDV_Linux!~doug@CPE001195554bb4-CM00252eac6f40.cpe.net.cable.rogers.com) has joined #mythtv | |
| [14:27:18] | stuartm: | some in the hateful recordingprofile code |
| [14:27:33] | stuartm: | I have a patch somewhere which starts to refactor that |
| [14:28:01] | Goga777 (Goga777!~Goga777@shpd-95-53-165-153.vologda.ru) has quit (Remote host closed the connection) | |
| [14:29:35] | danielk22: | stuartm: I think we talked about the recordingprofile code last time.. it needed significant work to fix a pretty small leak. |
| [14:33:36] | stuartm: | yeah, my work wasn't aimed at fixing the leak, that would just be a bonus |
| [14:35:54] | danielk22: | Hmm, m |
| [14:37:06] | stuartm: | I can't really remember how far I got, it's not been worked on in a while and I'm not even sure if it would fix the leak |
| [14:37:20] | danielk22: | I think we may need to require rebase before commit if we can't get mythtv-commits sorted out. |
| [14:38:49] | stuartm: | I think we can filter out the rubbish, but I'm having a hard time selling that idea since there are some that want all those emails |
| [14:39:40] | danielk22: | two lists? "must read" and "following branches" |
| [14:40:02] | stuartm: | that's probably the answer |
| [14:40:29] | danielk22: | There is no way we can require someone to be up to date with mythtv-commits before asking a question here or in mythtv-dev with this volume.. |
| [14:40:51] | danielk22: | + it's hard for us to follow. |
| [14:45:39] | danielk22: | iamlindoro: wrt #9216, the fglrx drivers aren't stable or fast. I would expect the open source drivers to work better if they work at all. Really we need someone interested in ATI hardware to step up and optimize the code for that HW. |
| [14:45:50] | Captain_Murdoch: | could we put the filtering/munging in Beirdo's code? have his code put in something to indicate that the commit is against mythtv/master or mythtv/fixes/* and add a tag or header in those cases. then those of us who don't want to see anything else can just look for that tag/header. |
| [14:46:39] | danielk22: | iamlindoro: I've given up on ATI hardware after multiple expensive attempts to use it for MythTV. |
| [14:48:37] | jannau: | Captain_Murdoch: we still need something to ignore the duplicate commits caused by merging |
| [14:49:16] | stuarta: | when i bought my last ATI card, nVidia were refusing to do either drivers or programming info for their cards |
| [14:49:16] | jannau: | rebasing wouldn't help much since |
| [14:51:29] | gigem_ is now known as gigem | |
| [14:58:28] | Captain_Murdoch: | danielk22, I agree, I don't expect everyone to spend time reading all the -commits emails, I'd rather they be able to spend that time coding. the same applies for IRC, we can't expect everyone to read every line of scrollback to know what's been going on or discussed. on a related note, I don't want to start making 'project plans', but sometimes it would be nice if we had brief ideas of what people were working on or intend to |
| [14:58:28] | Captain_Murdoch: | work on. that could be an email or just a simple wiki page or something. |
| [14:59:02] | ** Captain_Murdoch wonders if he's the only one who wants to tell Udo that the last svn revision is the one with the highest number since he insists on keeping mentioning not knowing what it is. ** | |
| [15:06:32] | danielk22: | Captain_Murdoch: you don't expect someone who has run into a difficulty to look at the commit's list before asking a question? (Other than "Does anyone know if has X changed in last Y weeks?") |
| [15:07:07] | stuarta: | only sphery knows exactly what's changed recently |
| [15:07:19] | stuarta: | and he can quote the last 3+ months as well |
| [15:10:04] | danielk22: | Heh, well sphery has a fantastic memory. I obviously don't expect that of anyone. But I do expect some level of self-help before someone asks a question in a development context. |
| [15:15:25] | jannau: | I think git log --stat — path/to/filez.c or git log -S'foo' — path/ are much more useful for that than searching in -commits |
| [15:17:35] | danielk22: | jannau: I think that is a valid alternative for you. But I don't expect that of Joe User, nor would it work very well on my cell phone. |
| [15:25:45] | Captain_Murdoch: | danielk22, not sure of whether that was really a "don't you" or "you don't?" :) they should do their homework first yes. if it's a question like your example, that's more of a -users question which we shouldn't/wouldn't field in here. |
| [15:28:35] | Captain_Murdoch: | just because someone is a dev doesn't mean that they get get a pass all the time. we're not here to do that level of hand-holding. :) |
| [15:28:59] | Captain_Murdoch: | not saying anyone has needed that, just saying it's not what I want to do with my time. |
| [15:35:07] | ** iamlindoro overreacted on this topic last night due to being very ill, and low on sleep and patience last week ** | |
| [15:35:45] | iamlindoro: | I recognize that it's hard to treat all commit messages with the same level of rapt attention. Hope we can make the commits message a useful tool again, that's all-- I rely on it to get my info and I'd like to go on doing so if it's possible |
| [15:36:24] | iamlindoro: | I'd also like to have it as leverage to hold trunk users who *do* expect to be spoon fed to the fire a bit ;) |
| [15:36:46] | jannau: | danielk22: me neither, my answer was for the "in development context" |
| [15:38:57] | jannau: | iamlindoro: a simple solution would be just send mails for commits in master and fixes/* |
| [15:39:21] | jannau: | we should do that asap |
| [15:40:58] | danielk22: | jannau: Right. I'm am a little concerned about the bar being set too high for casual contributors following trunk, but I haven't really spent enough time getting comfortable with the git workflow to comment very intelligently yet. |
| [15:42:55] | danielk22: | Commits e-mails are something I can read on my cell phone in the subway where there is no internet. Web pages and gtk/kde apps that people have been proposing as alternatives aren't really viable for me. |
| [15:42:55] | jannau: | there is no point arguing that -commits has too much message with very little useful information |
| [15:45:32] | jannau: | danielk22: yes, I agree. terse commit emails is one of the best way to stay up with developmnet contineously |
| [15:49:27] | gigem: | markk_: ok on waiting on the tv thread changes for 0.24. i'll keep using it for a few more days and report if i see any problems. |
| [16:04:42] | gregL (gregL!~greg@cpe-74-76-125-87.nycap.res.rr.com) has quit (Quit: Leaving) | |
| [16:07:20] | clever (clever!~clever@142.167.220.215) has quit (Ping timeout: 245 seconds) | |
| [16:08:56] | gigem (gigem!~david@host137.12.intrusion.com) has quit (Remote host closed the connection) | |
| [16:09:03] | gregL (gregL!~greg@cpe-74-76-125-87.nycap.res.rr.com) has joined #mythtv | |
| [16:09:05] | clever (clever!~clever@142.167.220.215) has joined #mythtv | |
| [16:09:21] | gigem (gigem!~david@host137.12.intrusion.com) has joined #mythtv | |
| [16:13:13] | iamlindoro: | paul-h: I did some research/work on converting iPlayer to remote control last night-- right now I've just taken the half-step of converting it to use the fullscreen mythflash player, but without remote so far |
| [16:13:37] | iamlindoro: | paul-h: But I've learned that there is a javascript API, albeit undocumented, so I'm going to try to reverse engineer how that works and get the whole shebang working |
| [16:14:31] | iamlindoro: | paul-h: I won't be able to test much, but I should be able to tell if it works, as I should be able to issue a play command and then have it tell me that I can't play it-- hopefully you can tell me the rest about how the experience is |
| [16:18:38] | j-rod|afk is now known as j-rod | |
| [16:21:31] | laga_ is now known as laga | |
| [16:21:39] | laga (laga!~laga@h1626373.stratoserver.net) has quit (Changing host) | |
| [16:21:39] | laga (laga!~laga@pdpc/supporter/student/laga) has joined #mythtv | |
| [17:18:36] | stuartm: | jamesba: you wouldn't happen to know if there's a public API we could or should be using for iPlayer integration? |
| [17:20:58] | iamlindoro: | None that they've chosen to document TTBOMK ;) |
| [17:23:06] | jamesba: | I know that there currently isn't, but I can pass your email address on to person who'd be responsible for one if it did become public, if you see what I mean |
| [17:23:26] | jamesba: | what email address shall I send him? |
| [17:23:39] | iamlindoro: | jamesba: Send him mine, robert.mcnamara at gmail |
| [17:24:43] | iamlindoro: | And if it's any reassurance, all we currently do is aggregate the public RSS feeds, we'd just like users to also be able to use our new capability of running it with the remote (and ideally, fullscreen) |
| [17:24:52] | stuartm: | iamlindoro: not using your mythtv.org address? |
| [17:24:55] | iamlindoro: | We definitely don't want to violate any terms of service |
| [17:25:13] | iamlindoro: | sorry, stuartm is right, rmcnamara at mythtv.org |
| [17:25:26] | iamlindoro: | goes to the same spot-- need to get used to using it |
| [17:28:45] | jamesba: | sorry, have already emailed him with the first address you gave |
| [17:28:50] | iamlindoro: | No problem |
| [17:28:56] | iamlindoro: | Thanks for doing that nonetheless |
| [17:29:03] | jamesba: | least I can do |
| [17:29:10] | jamesba: | unfortunately, also the most I can do ;) |
| [17:29:21] | iamlindoro: | hah, it's appreciated regardless of result |
| [17:42:01] | gigem_ (gigem_!~david@host137.12.intrusion.com) has joined #mythtv | |
| [17:46:14] | gigem (gigem!~david@host137.12.intrusion.com) has quit (Ping timeout: 276 seconds) | |
| [17:46:43] | natanojl (natanojl!~jonatan@c83-252-237-63.bredband.comhem.se) has joined #mythtv | |
| [17:47:49] | Steve_Goodey (Steve_Goodey!~steve@host86-145-237-137.range86-145.btcentralplus.com) has joined #mythtv | |
| [18:15:47] | kormoc_afk is now known as kormoc | |
| [18:25:51] | NightMonkey (NightMonkey!debian-tor@pdpc/supporter/professional/nightmonkey) has joined #mythtv | |
| [18:33:54] | Beirdo: | stuartm: how is 45k of leaked memory possibly "major" priority, other than to shut Udo up? |
| [18:35:50] | stuartm: | there multiple valgrind logs in the file, not just one, with varying levels of leakage shown some of which have the potential to be more serious over an extended period |
| [18:36:32] | Beirdo: | I see two reports, did I miss one? |
| [18:37:21] | stuartm: | there seems to be some ill will towards Udo, but lets not forget that the last time he complained and was ridiculed he turned out to be right, I found and fixed multiple leaks |
| [18:37:43] | Computer_Czar (Computer_Czar!~dcooper@69.4.155.83) has quit (Ping timeout: 272 seconds) | |
| [18:39:06] | Beirdo: | the first report had NO leak, the second report had minimal leaks. Yeah, we should look into them, but I would disagree with major priority, but go for it |
| [18:39:30] | stuartm: | I'm not going to treat this any less seriously just because the guy pisses people off, in fact quite the opposite – if you want him to shut up then what better way is there than to fix the issue? |
| [18:40:16] | Beirdo: | heh, there is that, but we don't want people thinking that they can get us to do what they want just by being a royal pain in the ass, either, do we? |
| [18:41:16] | Beirdo: | if it wasn't Udo that posted it, I still personally wouldn't give it that high of a priority for such a small leak. Certainly would keep it open until fixed though. |
| [18:41:34] | Beirdo: | but, your ticket, your priorities, I gues s:) |
| [18:41:55] | gigem_ is now known as gigem | |
| [18:41:57] | Beirdo: | if you need a hand squashing memory leaks, let me know, I'd be happy to help when I have time. |
| [18:47:36] | stuartm: | I only bumped the priority because you commented on it, not to piss you off, but because I'm trying to reach out to the guy and improve our user-relations a little – the first thing I did was to lock the ticket and assign it to myself because I knew it would otherwise attract negative feedback from others here |
| [18:49:41] | Goga777 (Goga777!~Goga777@shpd-95-53-165-153.vologda.ru) has joined #mythtv | |
| [18:49:53] | Beirdo: | well, it seems you have some smallish leaks in RecordingProfile, and one in DVBCam::SetPMT, i think it was |
| [18:50:13] | Beirdo: | the DVB one may be problematic |
| [18:50:56] | stuartm: | it does no harm to thank a user for reporting a bug and to give the impression that we want to fix it, it brings a little goodwill – the opposite just causes lasting ill-feeling to MythTV devs in my experience |
| [18:51:40] | stuartm: | Beirdo: the DVBCam one is more interesting, I've not seen that before and it could well be a more serious issue that none of the devs see because we don't use CAMs |
| [18:51:58] | Beirdo: | right. But for all his whining, he proved what we've been telling him. There is no appreciable leak. Now that we have his report, we can fix the ones he found. |
| [18:52:13] | SteveGoodey (SteveGoodey!~steve@host86-145-237-137.range86-145.btcentralplus.com) has joined #mythtv | |
| [18:52:15] | Beirdo: | yeah, that one could be a real issue over time, agreed |
| [18:52:44] | Beirdo: | the others seem to be the "allocate once and never deallocate due to deleteLater" variety |
| [18:52:55] | Beirdo: | but that's just a quick skim over it |
| [18:54:50] | stuartm: | the recording profile one might in isolation be small, but given the fact that it leaks every single time a recording starts it can easily become noticable for someone like Udo running 24/7 recordings on several channels at once – that might not be the intended use of MythTV but it's still leaking |
| [18:55:26] | stuartm: | i.e. I can see why it matters to Udo and no-one else |
| [18:55:55] | Beirdo: | is it leaking for every recording, or just leaking the last one that was in memory when you hit Ctrl-C because we used deleteLater, and the actual deletion never happened during the shutdown? |
| [18:56:14] | sphery: | FWIW, if you guys go to fix the RecordingProfile leaks, please don't restore https://github.com/MythTV/mythtv/commit/2e1e2ae2 (which came out of the last round of "fix it for Udo" changes in #6734, as it took me about 4 hours to find it when debugging http://code.mythtv.org/trac/ticket/9104 . TTBOMK, it only looks like a leak to valgrind because of the deleteLater() |
| [18:56:36] | stuartm: | Beirdo: leaks every time, we don't cache the RecordingProfile value |
| [18:56:46] | Beirdo: | ah, well, that's a paddlin |
| [18:58:08] | stuartm: | [27315] |
| [18:58:08] | MythLogBot: | SVN 27315: (branch fixes/0.24) https://github.com/MythTV/mythtv/commit/2ca5fd80 |
| [18:58:57] | SteveGoodey (SteveGoodey!~steve@host86-145-237-137.range86-145.btcentralplus.com) has quit (Remote host closed the connection) | |
| [18:59:33] | stuartm: | Beirdo: that code is fubar because to access the RecordingProfile data we have to construct an entire UI object, it's pretty much identical to the code I recently fixed in PlayGroup |
| [18:59:45] | Beirdo: | ahhh, yeah, that mess |
| [19:00:10] | sphery: | stuartm: yeah, a refactor is fine if you want it fixed properly |
| [19:00:22] | sphery: | just needs something other than a couple deletes |
| [19:03:29] | stuartm: | sphery: what I did in PlayGroup was to split it into two parts, when just accessing the playgroup info we use PlayGroup and in the PlayGroupEditor we use PlayGroupConfig which uses PlayGroup to load/save |
| [19:03:57] | stuartm: | iirc that's the fix I started on for RecordingProfile, I need to check what I did there assuming I can find the patch |
| [19:04:39] | sphery: | stuartm: cool--I agree the design needs fixing, but I wrote it off as "will get fixed when we do settings properly" |
| [19:05:25] | stuartm: | Playgroup was just this simple – https://github.com/MythTV/mythtv/commit/c0c432a |
| [19:09:05] | stuartm: | a simple refactor such as that won't stop the leak, but it will reduce it's appearance to the frontend and normally only once a run |
| [19:09:49] | sphery: | cool... I'm pretty sure, though, that with the current design, we're still not leaking id/name--it only looks that way because of the deleteLater() stuff |
| [19:11:31] | sphery: | i.e. ConfigurationGroup's destructor explicitly calls deleteLater() on every child, so since id and name are both added as children, they're deleted--valgrind just never sees the actual delete because the "later" doesn't happen before backend/valgrind is stopped |
| [19:13:09] | stuartm: | sphery: I'm looking at that now, it will take me a while to trace it through but I'm not as sure that we aren't leaking ID, unlike Name it's doesn't seem to be parented to RecordingProfile either directly or indirectly |
| [19:13:38] | stuartm: | s/it's/it/ |
| [19:13:52] | stuartm: | ah, sorry, there it is |
| [19:14:26] | sphery: | the Name is parented to an "anonymous" VerticalConfigurationGroup created at https://github.com/MythTV/mythtv/blob/master/ . . . le.cpp#L1169 line 1179 |
| [19:14:33] | sphery: | oops, you just found it :) |
| [19:15:07] | sphery: | ok, maybe not anonymous, buy generic |
| [19:15:51] | stuartm: | messy |
| [19:16:44] | sphery: | yes, maybe we should just fix this by working the settings rewrite... The code we use for backend settings should also be usable in frontend settings (even in stages), |
| [19:18:56] | stuartm: | sphery: long term maybe, I don't see us using that stuff for the frontend in the next couple of years, that would just be too optimistic :) |
| [19:20:25] | sphery: | heh, I'm still an optimist (even though I haven't been too productive, myself, lately) |
| [19:23:06] | Goga777 (Goga777!~Goga777@shpd-95-53-165-153.vologda.ru) has quit (Ping timeout: 265 seconds) | |
| [19:30:04] | sphery: | FWIW, anyone with filters on their commits may want to check said filters. Seems the List-Id header changed from "CVS commits list for mythtv <mythtv-commits.mythtv.org>" to "Git commits list for mythtv <mythtv-commits.mythtv.org>" |
| [19:31:58] | iamlindoro: | Any of our UK devs sitting at a box running trunk from today or so? |
| [19:32:00] | stuartm: | sphery: I've always filtered on ".*<mythtv-commits.mythtv.org>" to avoid precisely that problem |
| [19:32:36] | sphery: | yeah, seems I should, too |
| [19:32:59] | stuartm: | iamlindoro: dunno when I last rebuilt, I can't really tell from --version anymore |
| [19:33:40] | iamlindoro: | stuartm: I'm mostly looking for something post-my-commits to MNV/paul's to mythbrowser/mythuiwebbrowser... so under 24 hours old for sure |
| [19:33:51] | stuartm: | maybe we need to start including a build date line |
| [19:36:04] | stuartm: | iamlindoro: give me a few minutes and I'll be running head |
| [19:37:08] | iamlindoro: | stuartm: OK... no pressure. I just have a replacement iplayer HTML file that I'd like someone to try to see if the content still plays-- I don't expect remote control to work, I just want you to pull it up from a browser, confirm you get an embedded player, and click play |
| [19:38:11] | stuartm: | markk_: "tv_play.cpp:4670: warning: ‘tmpSpeed’ may be used uninitialized in this function" |
| [19:39:00] | stuartm: | danielk22: your warnings script isn't pulling from git? |
| [19:39:57] | danielk22: | stuartm: no, not yet. I haven't even moved my regular dev stuff over yet. |
| [19:40:59] | stuartm: | ok, np |
| [19:41:44] | stuartm: | Beirdo should have the built-bot setup by this time tomorrow* and we can refer to that instead |
| [19:42:09] | danielk22: | cool |
| [19:42:14] | stuartm: | * that might just be wishful thinking on my part |
| [19:42:25] | danielk22: | hehe |
| [19:42:46] | danielk22: | well I'll probably get around to it this weekend regardless. |
| [19:47:14] | Beirdo: | tomorrow?! More likely the weekend :) |
| [19:56:04] | danielk22: | Is it safe to just cp a clone or should I run several clone's if I want a dozen or so clones of the repo? |
| [19:56:14] | danielk22: | (^ git) |
| [19:59:13] | stuartm: | danielk22: gigem came up with something that's worth a look, it allows multiple working directories to be run from a single clone |
| [20:00:29] | danielk22: | git-new-workdir ? |
| [20:00:49] | stuartm: | that's the one, I was digging through the log to find the reference |
| [20:02:21] | gigem: | danielk22: yes, it was git-new-workdir. |
| [20:03:11] | stuartm: | I was going to try it out today but I didn't have the time |
| [20:04:27] | gigem: | danielk22: cp'ing instead of cloning might work depending on how you intend to work. if everything is independent and you intend to push to master on github, it would probably work. if instead you need to move changesets around in your own branches/trees, you should proably use git-new-workdir or local clones. |
| [20:05:28] | gigem: | stuartm: i've only played with it a little so far, but i did quite a bit |
| [20:05:52] | gigem: | of googling and i'm pretty sure it's what i'm going to do. |
| [20:06:31] | stuartm: | gigem: yeah, everything I read made it sound like exactly what I want |
| [20:07:41] | danielk22: | I do both. I have 4–5 independent trees + a couple sets trees that should all have the same code. It looks like this would help with second set of trees (for testing things in my "production" environment). |
| [20:13:19] | gigem: | there's no reason you can't do both (clone and git-new-workdir) at the same time. you just need to make very sure you don't violate the restriction that g-n-w requires or you could lose data. |
| [20:18:30] | NightMonkey (NightMonkey!debian-tor@pdpc/supporter/professional/nightmonkey) has quit (Remote host closed the connection) | |
| [20:24:54] | stuartm: | one thing I can't find is whether g-n-w allows for a sub-directory to be copied and not the full tree |
| [20:26:04] | stuartm: | g-n-w would in theory save at least 184MB per working directory vs clone, but that might not be enough for kormoc and his 9+ mythweb checkouts |
| [20:35:00] | castlec1 (castlec1!~christoph@cpe-76-181-105-97.insight.res.rr.com) has quit (Quit: Leaving.) | |
| [20:40:53] | xris: | kormoc needs to change his workflow a bit. ;) |
| [20:41:16] | NightMonkey (NightMonkey!debian-tor@pdpc/supporter/professional/nightmonkey) has joined #mythtv | |
| [20:42:09] | xris: | but I think splitting mythweb out was an ok idea/compromise |
| [20:42:14] | xris: | it makes my life easier, too... |
| [20:43:26] | brfransen (brfransen!~brfransen@adrianDHCP-47.216-254-250.iw.net) has quit (Ping timeout: 255 seconds) | |
| [20:49:32] | sphery: | Did this really change (purposefully or by accident) as described? http://code.mythtv.org/trac/ticket/6948#comment:15 If so, it would be breaking the Avoid conflicts between Live TV and scheduled shows setting--not to mention changing the basic premise of "always use the best tuner unless avoid conflicts is set". |
| [20:49:38] | brfransen (brfransen!~brfransen@adrianDHCP-47.216-254-250.iw.net) has joined #mythtv | |
| [20:51:05] | iamlindoro: | Heh, the patch also reverts a proto bump |
| [20:51:30] | danielk22: | sphery: The patch hasn't been applied, but it really shouldn't have that effect. |
| [20:52:42] | sphery: | yeah, seems a strange effect for that patch... I thought I had remembered someone else mentioning that behavior before, elsewhere... Maybe it's a bug that went in elsewhere. |
| [20:53:49] | danielk22: | iamlindoro: hmm, looks like it also applies an unnecessary DB change... dirty patch |
| [20:54:48] | ** iamlindoro gives up trying to find out if "3875641D" is danielk22's home phone number ** | |
| [20:56:05] | sphery: | lots of people mention applying that patch... yay for corrupt schemas! |
| [20:57:12] | sphery: | though maybe most of those people who applied it will have applied it to 0.24 or higher (meaning they'll already have 1264) |
| [20:58:30] | iamlindoro: | stuartm: Don't suppose you've gotten up to date on trunk? |
| [20:59:23] | xris: | has trunk been updated in the last week? |
| [20:59:26] | stuartm: | iamlindoro: trying, mythbrowser is refusing to build atm |
| [20:59:48] | stuartm: | xris: v.funny |
| [20:59:52] | iamlindoro: | trunk, master, head, pointy thing at the top, whatever |
| [21:00:01] | xris: | heh |
| [21:00:09] | ** xris likes "pointy thing at the top" ** | |
| [21:00:52] | iamlindoro: | stuartm: failed to make distclean in mythtv? |
| [21:01:08] | iamlindoro: | (ie, the new changes to libmythtv.pro might not have taken and thus playgroup.h might not have installed) |
| [21:01:54] | stuartm: | iamlindoro: nah, full distclean in both mythtv and plugins |
| [21:03:13] | iamlindoro: | hmm... then outside of that, whatever it is is probably my fault |
| [21:03:15] | stuartm: | I'll make it behave soon enough |
| [21:04:55] | SteveGoodey (SteveGoodey!~steve@host86-145-237-137.range86-145.btcentralplus.com) has joined #mythtv | |
| [21:04:58] | xris: | hmm, github commit rss feed is annoying. puts the file list at the top... |
| [21:05:01] | stuartm: | meanwhile I'm trying to cram my music collection onto my mp3 player, it's not easy, in size terms it's not going to fit and it's not even hard to figure what to keep and what to bin but it's time consuming |
| [21:06:21] | stuartm: | this is why I wanted a playlist export feature for mythmusic, build up a playlist in mythmusic then select 'copy this playlist to device' and it will copy over the files |
| [21:07:19] | xris: | hmm. need to come up with a commit message convention for git, too... first lines are way too long on a bunch of recent things |
| [21:07:19] | stuartm: | transferring 20GB onto a class 6 SD card is painfully slow :( |
| [21:07:46] | xris: | or... it's a fisheye bug. odd |
| [21:07:50] | stuartm: | Beirdo: thanks |
| [21:09:41] | MaverickTech (MaverickTech!~MaverickT@111.86.233.220.static.exetel.com.au) has quit (Ping timeout: 264 seconds) | |
| [21:09:44] | lapion (lapion!~axion@axion.xs4all.nl) has joined #mythtv | |
| [21:10:21] | lapion: | hello, does anyone know how to use both displays in a dual header system for mythtv ? |
| [21:11:30] | ** kormoc points lapion at the topic ** | |
| [21:11:45] | lapion: | sorry.. |
| [21:11:58] | lapion: | forgot about mythtv-users |
| [21:12:05] | lapion: | +# |
| [21:17:02] | lapion (lapion!~axion@axion.xs4all.nl) has left #mythtv ("Ex-Chat") | |
| [21:27:45] | brfransen (brfransen!~brfransen@adrianDHCP-47.216-254-250.iw.net) has quit (Ping timeout: 245 seconds) | |
| [21:30:45] | brfransen (brfransen!~brfransen@adrianDHCP-47.216-254-250.iw.net) has joined #mythtv | |
| [21:30:50] | Beirdo: | stuartm: for the header thingy? no problemo. It makes perfect sense, it just didn't cross my mind at the time :) |
| [21:34:56] | Kunalagon (Kunalagon!~Kunalagon@212.200.240.135) has joined #mythtv | |
| [21:34:57] | brfransen (brfransen!~brfransen@adrianDHCP-47.216-254-250.iw.net) has quit (Read error: Connection reset by peer) | |
| [21:35:05] | danielk22: | Bierdo: Looking at the git docs it sounds like git rebase is the closest analog to svn merge, before I go using rebase for everything...am I missing something? The use-case here is updateing the mythtv-rec branch with commits in "remotes/origin/master" |
| [21:35:22] | Beirdo: | no, you want merge |
| [21:36:00] | Beirdo: | rebase will base your branch on the other one (i.e. move the far end of your branch onto a different spot on master) |
| [21:36:20] | Beirdo: | what I'm pretty sure you'd want is "git merge master" |
| [21:36:26] | Beirdo: | while on your branch |
| [21:36:33] | brfransen (brfransen!~brfransen@adrianDHCP-47.216-254-250.iw.net) has joined #mythtv | |
| [21:36:55] | Beirdo: | and yes, it will spew lots of superfluous commit messages (I'm trying to determine a way around that) |
| [21:37:26] | Beirdo: | if you look on the network graph on github, that's what both jya and I have done with our branches |
| [21:37:28] | danielk22: | heh, well it will once I resolve all the conflicts and commit :) |
| [21:37:37] | Beirdo: | hehe |
| [21:37:43] | Beirdo: | yeah, git mergetool :) |
| [21:37:54] | Beirdo: | uses vimdiff (if vi is your friend) |
| [21:38:52] | danielk22: | infadel! emacs is my friend ;] |
| [21:40:06] | Beirdo: | heh, it may have a setting to use emacs too, i never checked |
| [21:42:11] | SteveGoodey (SteveGoodey!~steve@host86-145-237-137.range86-145.btcentralplus.com) has quit (Remote host closed the connection) | |
| [21:43:41] | RDV_Linux (RDV_Linux!~doug@CPE001195554bb4-CM00252eac6f40.cpe.net.cable.rogers.com) has quit (Remote host closed the connection) | |
| [21:47:27] | jannau: | Beirdo: kdiff3, tkdiff, meld, xxdiff, emerge, vimdiff, gvimdiff, ecmerge, diffuse, tortoisemerge, opendiff, p4merge and araxis |
| [21:47:47] | RDV_Linux (RDV_Linux!~doug@CPE001195554bb4-CM00252eac6f40.cpe.net.cable.rogers.com) has joined #mythtv | |
| [21:47:50] | Beirdo: | qhite a choice :) |
| [21:47:55] | Beirdo: | quite rather |
| [21:48:16] | Beirdo: | I use vimdiff, personally. Others may use whatever works for them :) |
| [21:50:10] | danielk22: | Beirdo: So when I do push this it will resend the last month of changes to the mailing list? |
| [21:50:27] | Beirdo: | :( yeah, it seems to want to |
| [21:50:46] | Beirdo: | that's what I'm trying to decipher the best way to squelch |
| [21:51:08] | Beirdo: | I'm thinking of SHA1 of the comments + author + files |
| [21:51:29] | Beirdo: | but I'm worried about getting a false positive in places I haven't thought of yet |
| [21:52:21] | Beirdo: | I think I'll try to implement that tonight, but I'm not sure it will quite do what we expect. |
| [21:53:01] | danielk22: | Well I need to test this merge anyway.. but I will want to commit/push it at some point.. |
| [21:53:08] | Beirdo: | right |
| [21:53:24] | Beirdo: | if I can get that in before, then it might keep the noise down |
| [21:53:52] | Beirdo: | the other issue with that method is getting all the old commits in there |
| [21:53:58] | jannau: | Beirdo: I think as workaround we should disable commit mails for all branches except master and fixes/* |
| [21:54:45] | xris: | jannau: that was my thought, too |
| [21:54:45] | Beirdo: | I don't like that, but as a workaround... it's also the second person with that idea |
| [21:54:48] | Beirdo: | so... |
| [21:54:55] | Beirdo: | Let me go do that right now |
| [21:55:22] | xris: | Beirdo: like I said at lunch. people using shared branches other than those will probably be less often, and can either communicate amongst themselves or set up other means of tracking (e.g. RSS) |
| [21:55:54] | xris: | heck, I think I'm going to move to RSS monitoring anyway.. something about email prevents me from reading it (probably too much noise from trac stuff) |
| [21:56:00] | Beirdo: | yeah. It should do for the time being while we consider other possibilities that may not even work |
| [21:56:01] | jannau: | a single mail for each push for the other branches might be a ggod idea |
| [21:57:45] | jannau: | xris: we were speaking about moving the trac ticket mails to a seperate list |
| [22:00:10] | kc (kc!~Casper@unaffiliated/kc) has quit (Ping timeout: 240 seconds) | |
| [22:00:11] | sphery: | though it's easy enough to sort them yourself on X-Trac-Project:.*MythTV |
| [22:05:28] | xris: | jannau: that would be very handy |
| [22:06:06] | xris: | more stuff for me not to read... and wish that trac was better about just sending emails to ticket owners |
| [22:10:26] | Beirdo: | now.. back to tweaking... stupid real work keeps getting in hte way |
| [22:12:07] | jannau: | sphery: if people weren't changing the list id just to update the several years outdated VCS name |
| [22:12:07] | Beirdo: | jannau: I'll look at the single (much shortened) email for a non-master, non-fixes branch this evening |
| [22:12:40] | jannau: | Beirdo: thanks |
| [22:12:41] | Beirdo: | for now I'll just disable the email fro non-master, non-fixes |
| [22:12:45] | Beirdo: | sound good? |
| [22:12:50] | stuartm: | sphery: yes it can be sorted, but filters can be fragile and not easily applied when you're trying to monitor a list from a mobile phone or similar, and if people really want both tickets and commits going to the same inbox then it's easy enough to subscribe to both |
| [22:13:25] | j-rod is now known as j-rod|afk | |
| [22:14:21] | stuartm: | fwiw I've always thought it was an odd fit for ticket traffic to go to a commits list, a mythtv-ticket list makes a lot of sense |
| [22:14:44] | danielk22: | Beirdo: That means the 215 commits for the merge will not print anything both when I push them and when I later merge the mythtv-rec branch to master? Or only that the current push won't cause a mail storm? |
| [22:14:52] | stuartm: | in fact there's not reason that a third mythtv-everything list can't be setup which gets commits AND tickets |
| [22:15:17] | Beirdo: | danielk22: it would keep the current merge from emailing anything at all |
| [22:15:23] | Beirdo: | once I have it in place |
| [22:15:29] | Steve_Goodey (Steve_Goodey!~steve@host86-145-237-137.range86-145.btcentralplus.com) has quit (Remote host closed the connection) | |
| [22:15:33] | danielk22: | that sounds good to me :) |
| [22:16:18] | xris: | submitting a bug to atlassian about making their fisheye RSS feeds usable |
| [22:16:32] | Beirdo: | and it's in place |
| [22:16:55] | xris: | their post format blows github away, but they use the entire commit message for the subject, which renders them half-unusable |
| [22:17:54] | Beirdo: | tonight I'll work on the "summarize the commits" on non-master, non-fixes |
| [22:18:13] | Beirdo: | which will tidy things up, and still give us an idea of what's going on |
| [22:28:10] | stuartm: | danielk22's idea about two lists for commits, one with the essential (master/fixes) commits and one with everything was a good idea |
| [22:29:46] | Beirdo: | I like the idea of mythtv-commits for the core commits, mythtv-tickets for ticketing and mythtv-all-commits or something for all of the above? |
| [22:29:58] | Beirdo: | and including non-core commits |
| [22:32:28] | Beirdo: | actually, I've always been a fan of nntp, but most aren't :) |
| [22:33:23] | sphery: | if we break it out, it makes more sense to me to have 3 non-overlapping ones |
| [22:33:37] | Beirdo: | yeah |
| [22:33:45] | Beirdo: | that could work well |
| [22:33:56] | sphery: | so mythtv-commits for master/fixes commits, mythtv-tickets for ticketing and mythtv-branch-commits, maybe? |
| [22:34:05] | Beirdo: | and as has been said, people can subscribe the same email to all |
| [22:34:11] | stuartm: | mythtv-noisy-commits |
| [22:34:18] | Beirdo: | heheh |
| [22:35:14] | danielk22: | I like sphery's suggestion.. |
| [22:35:41] | Beirdo: | I think we have general agreement :) |
| [22:35:43] | Beirdo: | yay! |
| [22:36:35] | Kunalagon (Kunalagon!~Kunalagon@212.200.240.135) has quit (Quit: Leaving.) | |
| [22:37:08] | danielk22: | AFAIK Chutt chose to re-use the commits list for tickets because tickets were expected to be a low volume trickle of updates... obviously that isn't really the case anymore. |
| [22:37:28] | Beirdo: | hehe |
| [22:37:35] | Beirdo: | yeah, not so much anymore |
| [22:41:36] | xris: | makes sense |
| [22:42:20] | jpharvey (jpharvey!~jpharvey@host86-135-81-74.range86-135.btcentralplus.com) has quit (Ping timeout: 245 seconds) | |
| [22:44:50] | Beirdo: | I guess we know what tonight's fun will be :) |
| [22:45:07] | Beirdo: | yay for more mailing lists to move |
| [22:54:57] | jpharvey (jpharvey!~jpharvey@host86-185-253-248.range86-185.btcentralplus.com) has joined #mythtv | |
| [22:58:20] | gkeen316 (gkeen316!~gkeen316@d75-156-147-11.abhsia.telus.net) has joined #mythtv | |
| [23:09:25] | natanojl (natanojl!~jonatan@c83-252-237-63.bredband.comhem.se) has quit (Ping timeout: 272 seconds) | |
| [23:24:01] | stuartm: | heh, spammers have already found my mythtv.org alias |
| [23:29:41] | pheld (pheld!~heldal@cl-5.osl-01.no.sixxs.net) has quit (Ping timeout: 272 seconds) | |
| [23:29:56] | sphery: | stuartm: Remember how you moved default menu fallback and files to themes/defaultmenu? One of the users in the other channel mentioned a bunch of menu xml files in MythTV's sharedir. It seems that the mythtv make install puts setup.xml in there and the plugins make install puts archivemenu.xml, archiveutils.xml, game_settings.xml, musicmenu.xml, music_settings.xml, netvisionmenu.xml, videomenu.xml, video_settings.xml, and ... |
| [23:30:03] | sphery: | ... weather_settings.xml in there. Should those be going to defaultmenu? |
| [23:31:16] | stuartm: | yeah they should, I'm sure that was part of my original patch ... maybe I missed that from the commit? |
| [23:31:53] | sphery: | yeah, or some parts got reverted or something... Not worth digging into old commits to find out :) |
| [23:34:26] | sphery: | kudos to castlec1 for noticing |
| [23:35:57] | pheld (pheld!~heldal@cl-5.osl-01.no.sixxs.net) has joined #mythtv | |
| [23:36:26] | sphery: | stuartm: the mythtv-related part of the change was at https://github.com/MythTV/mythtv/commit/882fa1a4 (way back in the [19925] days) , but I don't know if there was another plugins-related commit |
| [23:36:26] | MythLogBot: | SVN 19925: (branch master) https://github.com/MythTV/mythtv/commit/882fa1a4 |
| [23:39:25] | CppIsWeird (CppIsWeird!~user@unaffiliated/cppisweird) has quit (Ping timeout: 245 seconds) | |
| [23:50:50] | skd5aner: | !seen simonkenyon |
| [23:50:50] | MythLogBot: | simonkenyon has not been seen here |
| [23:50:51] | iamlindoro: | !seen simonkenyon |
| [23:50:51] | MythLogBot: | simonkenyon has not been seen here |
| [23:50:56] | iamlindoro: | bah |
| [23:50:56] | skd5aner: | doh |
| [23:51:13] | iamlindoro: | He was in -theming at 0229 |
| [23:51:19] | iamlindoro: | ah, wait |
| [23:51:24] | iamlindoro: | !seen simonckenyon |
| [23:51:24] | MythLogBot: | simonckenyon was last seen 13 hours 21 minutes 47 seconds ago |
| [23:51:30] | skd5aner: | ah |
IRC Logs collected by
BeirdoBot.
Please use the above link to report any bugs.