All times shown according to UTC.

Time Nick Message
00:05 iamjayakumars joined #nightingale
01:30 stevew joined #nightingale
05:02 Mook joined #nightingale
05:03 Mook1 joined #nightingale
05:32 stevew joined #nightingale
05:36 AndrewLuecke joined #nightingale
10:02 mib_06k4sw joined #nightingale
10:04 ando joined #nightingale
10:11 freaktechnik joined #nightingale
10:12 Topic for #nightingale is now The topic for #nightingale is: Nightingale Media Player || http://getnightingale.com || Current release: 1.12.1 || Offtopic #chirping || Please idle and wait for a reply! || Channel logs: http://bit.ly/ngaleirc || Dev-Meeting today: http://hbsl.ch/ngdm3
10:13 ando left #nightingale
10:33 freaktechnik hm, I wanted to as Mook something but I forgot what...
10:33 *ask
10:34 oh, yeah, is there a reason to not return the rv variable and instead hardcode NS_OK?
10:51 :O http://developer.getnightingal[…]_lib_hal_ctx.html
11:21 clokep joined #nightingale
11:35 freaktechnik so mediaitems leak, as I leak more memory when my library is bigger...
12:05 BAM
12:05 ngalebot joined #nightingale
12:05 ngalebot [13nightingale-hacking] 15freaktechnik pushed 2 new commits to 06sb-trunk-oldxul: 02http://git.io/RTdVGg
12:05 13nightingale-hacking/06sb-trunk-oldxul 147981078 15Martin Giger: Equalizer Presets. Backend is working (at least I think it is), while the front-end (the binding) is broken for some reason.
12:05 13nightingale-hacking/06sb-trunk-oldxul 14814f619 15Martin Giger: Minor fixes & improvements to mpris & internalsearch
12:05 ngalebot left #nightingale
12:11 freaktechnik why can't scratch just save the newlines at the end of the files?
12:11 WHY
12:11 anyways, away
12:13 travis-ci joined #nightingale
12:13 travis-ci [travis-ci] nightingale-media-player/nightingale-hacking#153 (sb-trunk-oldxul - 814f619 : Martin Giger): The build is still failing.
12:13 [travis-ci] Change view : https://github.com/nightingale[…]26...814f619dca20
12:13 [travis-ci] Build details : http://travis-ci.org/nightinga[…]g/builds/27011417
12:13 travis-ci left #nightingale
16:31 freaktechnik joined #nightingale
17:18 Erikcht joined #nightingale
17:18 Erikcht hello
17:19 The meeting will happening, isn't ?
17:19 freaktechnik yes, as written in the topic
17:20 Erikcht htx
17:20 thx
17:23 frektechnik, may I ask something to you ?
17:24 freaktechnik sure
17:24 Erikcht thx
17:25 I note that the oldSongbird fanpage from facebook has a lot of follows.
17:25 16.000 likes
17:26 freaktechnik mhm
17:26 Erikcht Do you think prudent , tell then , now that the project of then is over, to follow the nightingale fanpage and pass to know our project that is the fork from this project ?
17:27 freaktechnik well, most of them already know, or only liked it because of songbird.me
17:28 as we were also mentioned in their blogpost
17:28 (the last one)
17:28 Erikcht but, i think 16.000 OR MORE, still don't knowing what happening...
17:28 freaktechnik I doubt that
17:28 also they wouldn't see it
17:29 Erikcht It would not take a try?
17:30 ngalebot joined #nightingale
17:30 ngalebot [13nightingale-hacking] 15freaktechnik pushed 1 new commit to 06sb-trunk-oldxul: 02http://git.io/-vSBBQ
17:30 13nightingale-hacking/06sb-trunk-oldxul 140b1fb0f 15Martin Giger: Fixed applying presets from the eq-preset-list
17:30 ngalebot left #nightingale
17:30 freaktechnik well, you can try it.
17:30 Erikcht thx
17:30 who were the "admins" from the old songbird, that you know that were abble to post on songbird fanpage ?
17:31 freaktechnik well, probably nobody out there who still can
17:31 as the company running it doesn't exist anymore.
17:31 Erikcht the old CEO ?!
17:32 freaktechnik I doubt that he still has access
17:32 or that he ever
17:32 had
17:32 that's social media people's thing to do
17:32 Erikcht Eric Wittman ...  Robert Lord ...
17:39 travis-ci joined #nightingale
17:39 travis-ci [travis-ci] nightingale-media-player/nightingale-hacking#154 (sb-trunk-oldxul - 0b1fb0f : Martin Giger): The build is still failing.
17:39 [travis-ci] Change view : https://github.com/nightingale[…]20...0b1fb0f491bf
17:39 [travis-ci] Build details : http://travis-ci.org/nightinga[…]g/builds/27024981
17:39 travis-ci left #nightingale
17:44 Erikcht well,  i tried...
17:46 i post something cool , giving thx to the the pioneers and telling to then to post something to nightingale
17:52 look what I found...
17:52 https://www.facebook.com/perma[…]657&id=2224650920
17:53 freaktechnik Andrew is not a songbird dev^^
17:53 he's around here sometimes.
17:53 he's just like us, a poweruser ;)
18:03 Erikcht great
18:03 i remember he from old times :D
18:07 Freaktechnik, i have done a install fest Linux + Steam in Rio de Janeiro.
18:08 freaktechnik cool
18:08 Erikcht on the week of FLISOL 2014,
18:08 do you know UFRJ ?
18:09 freaktechnik no
18:10 Erikcht http://www.ufrj.br/
18:11 Universidade Federal do Rio de Janeiro  ( "Federal University of Rio de Janeiro")
18:12 I did a workshop workshop focus on installing linux.
18:13 5 people , corious, coming to this install fest . After install Linux and Steam, I installed the Nightingale
18:14 all then become satisfied with this project.
18:14 by the "lyrics" and the app "soundcloud" .
18:15 freaktechnik those are indeed pretty nice add-ons
18:15 and pretty unique too
18:16 Erikcht The people of laboratory  UFRJ "campos praia vermelha" wat to repeat this install fest (not only on FLISOL) . they never saw something like a install fest
18:17 freaktechnik a university where people don't know linux and install fests...
18:17 Erikcht yes, those addons are awesome..
18:17 no
18:17 your pcs use linux
18:17 all then
18:17 10 pcs on laboratory
18:17 all with ubuntu "12.04"
18:17 freaktechnik but your it departement doesn't do install fests for you?
18:17 urgh
18:18 Erikcht but they are lower users, only use softwares of "radio opensource and other thinks"
18:18 no, i was invited to do this job
18:19 with that, They gave me the freedom to pursue future installations.
18:19 freaktechnik nice
18:21 Erikcht they focus ar lectures, free radio and ciências sociais
18:21 "social science"
18:21 freaktechnik don't you have like IT people who do nothing but set up all the computers and help people use them?
18:22 Erikcht UFRJ has it, but, they are focus on TEACH
18:23 I think they apear on 2 from 2 years.
18:26 I have a friend who specializes in Debian. Not to program, but to be user.I took it that install fest to help me with possible Debian installations that may appear (and yes, it appeared, only 1 person wanted Debian, this person was a user of slackware)
18:27 freaktechnik o.O slackware, haven't heard of that in a long time...
18:29 Erikcht rs
18:31 This person was using slackware with 9 partitions. I do not know why ... The boy was curious, very good people.
18:32 PS:(R.S is the same of L.O.L)
18:32 freaktechnik hm, a root, a boot, a home, a var, a swap, and what were the other four for?
18:33 Erikcht Save pictures of cats. Photos of dogs
18:33 Save photos Farmer
18:33 freaktechnik so home split up in multiple partitions like some people do
18:48 ngalebot joined #nightingale
18:48 ngalebot [13nightingale-hacking] 15johnmurrayvi pushed 6 new commits to 06master-xul-9.0.1: 02http://git.io/P8ZECQ
18:48 13nightingale-hacking/06master-xul-9.0.1 1488b4d59 15John Murray: only register libs which implement a mozilla module
18:48 13nightingale-hacking/06master-xul-9.0.1 14daedcdc 15John Murray: misc clean up, look at XXX note in sbGStreamerPipeline later
18:48 13nightingale-hacking/06master-xul-9.0.1 14ad05a5d 15John Murray: don't prefix the value entry with "service,"
18:48 ngalebot left #nightingale
18:59 travis-ci joined #nightingale
18:59 travis-ci [travis-ci] nightingale-media-player/nightingale-hacking#155 (master-xul-9.0.1 - 5eff892 : John Murray): The build passed.
18:59 [travis-ci] Change view : https://github.com/nightingale[…]73...5eff892c34b2
18:59 [travis-ci] Build details : http://travis-ci.org/nightinga[…]g/builds/27028797
18:59 travis-ci left #nightingale
19:43 Erikcht hi, johnmurrayvi is you travis-ci ?
19:44 freaktechnik travis-ci is a bot from the travis continuous integration
19:44 what it does is try to build nightingale after we comitted changes to it.
19:54 Erikcht ok...
20:12 stevew joined #nightingale
20:12 rsjtdrjgfuzkfg joined #nightingale
20:21 freaktechnik hi rsjtdrjgfuzkfg
20:21 rsjtdrjgfuzkfg hi freaktechnik
20:33 freaktechnik now the only thing not working is just some fiddly front-end menulist stuff...
20:33 ngalebot joined #nightingale
20:33 ngalebot [13nightingale-hacking] 15freaktechnik pushed 1 new commit to 06sb-trunk-oldxul: 02http://git.io/r_MD5Q
20:33 13nightingale-hacking/06sb-trunk-oldxul 14cbad1be 15Martin Giger: Fix saving and deleting and make the gain indicators change when a preset is applied
20:33 ngalebot left #nightingale
20:37 Erikcht Hello everybody. ...
20:37 rsjtdrjgfuzkfg hi Erikcht
20:37 Erikcht hello
20:37 i have a friend, that know build .deb coming to here today...
20:43 travis-ci joined #nightingale
20:43 travis-ci [travis-ci] nightingale-media-player/nightingale-hacking#156 (sb-trunk-oldxul - cbad1be : Martin Giger): The build is still failing.
20:43 [travis-ci] Change view : https://github.com/nightingale[…]bf...cbad1be1e1be
20:43 [travis-ci] Build details : http://travis-ci.org/nightinga[…]g/builds/27033599
20:43 travis-ci left #nightingale
20:47 travis-ci joined #nightingale
20:47 travis-ci [travis-ci] nightingale-media-player/nightingale-deps#40 (xul-9.0.1 - 06e6848 : John Murray): The build passed.
20:47 [travis-ci] Change view : https://github.com/nightingale[…]fc...06e6848aa436
20:47 [travis-ci] Build details : http://travis-ci.org/nightinga[…]s/builds/27032950
20:47 travis-ci left #nightingale
20:52 Ronald_Granthon joined #nightingale
20:53 Ronald_Granthon Hey, Erick... I'm here...
20:55 johnmurrayvi joined #nightingale
20:55 freaktechnik hi john
20:55 johnmurrayvi hello!
20:55 freaktechnik and hi ronald
20:55 Erikcht AEEEEEEEEEEEEEEEEEEE
20:55 * freaktechnik pings GeekShadow
20:55 rsjtdrjgfuzkfg hi *
20:55 Erikcht finaly RD2 ....
20:56 RD2 Hi freaktechnik...
20:56 johnmurrayvi oh yay, the logs server is back up. just enough time to catch up
20:57 Erikcht people
20:57 freaktechnik well, it's mainly my dairy in the past few days...
20:57 Erikcht RD2 is my friend from Rio de Janeiro (Brazil) that know build .deb
20:59 johnmurrayvi lol freaktechnik I've seen your commits without a new line at the end, so I feel your pain about not being able to set that in the editor. luckily sublime text has a setting for it, but I'm not sure how much longer I'll be using it
20:59 freaktechnik well, it magically removes them since some lib upgrade
20:59 so yeah
20:59 aaaaaaaaaanyways.
21:00 Meeting notes: http://wiki.getnightingale.com[…]eloper_meetings:3
21:00 so welcome everybody.
21:00 to our third meeting
21:01 johnmurrayvi (I wonder if ilikenwf is catching up on sleep lol)
21:01 freaktechnik I added the tracklist interface to Nightingale's mpris
21:01 it isn't completely implemented, but turns out doing stuff related to the current sequence is pretty complicated with ngale
21:02 RD2 To build a .deb only need know some libs is necessary to Nightingale works and source code
21:02 freaktechnik RD2: we already can build debs. See our debian directory in our github repo.
21:03 RD2 Are you know some libs is necessary to Nightingale works freaktechnik ?
21:03 freaktechnik as I said, it's all in our debian directory. And now please, stick to the meeting notes.
21:03 else I'll have to turn on +v
21:04 so, taglib.
21:04 we'll actually talk about taglib in the core & deps now, becausse that's kind of related.
21:05 I think johnmurrayvi has been trying to fix the taglib tests by poking around, and ended up reapplying POTIs changes to a current taglib version
21:05 johnmurrayvi yeah... basically for the trunk section, the important note is that taglib causes every (I think) failed test
21:05 freaktechnik Now I personally don't think we should use a patched taglib.
21:05 Yes, there are issues with vanilla taglib, but there are issues with a patched one too
21:05 which is why rsjtdrjgfuzkfg created the new implementation
21:05 and if I had to choose between the two options I'd prefer a vanilla taglib.
21:06 * rsjtdrjgfuzkfg too
21:06 freaktechnik yes, we currently use an inofficial API with vanilla taglib, but still...
21:06 rsjtdrjgfuzkfg however, I'm not up to date with taglib versions
21:06 afaik they wanted to fix that so we can use a proper api
21:06 freaktechnik well, I guess johnmurrayvi knows a little more -> see the deps section in the notes
21:06 johnmurrayvi well I patched up to version 1.7.2. the 1.8.0 version provides FileStream and IOStream classes, which could be used in place of the FileIO class POTI had implemented
21:07 rsjtdrjgfuzkfg the workaround we're having right now is about tags
21:08 johnmurrayvi I'm all for using vanilla deps, but at the same time, it isn't like we could have a "system taglib" build on mac or windows... and both of those platforms have higher usage than linux
21:08 rsjtdrjgfuzkfg as back then when I updated a tagunion did not allow access to all tags
21:08 freaktechnik it still doesn't
21:08 mac is actually about the same...
21:08 Erikcht joined #nightingale
21:08 freaktechnik but it's just... easier to maintain.
21:08 johnmurrayvi with the patched versions I've tested, it fixes a large chunk of issues (let me find the list)
21:09 freaktechnik it fixed the tests + the whole file access issues.
21:09 johnmurrayvi https://github.com/nightingale[…]pdated&state=open
21:09 freaktechnik however the vanilla implementation fixed some issues we probably don't have documented
21:09 like, hanging while importing on windows
21:09 johnmurrayvi (except #243)
21:09 freaktechnik or more mp4 issues
21:09 Erikcht people
21:10 johnmurrayvi that was mainly why I was updating the patches to newer taglib versions
21:10 for example, in 1.6.1, POTI had to implement FlacPicture
21:10 Erikcht do you knwo the problems that is troble the nightingale to show on ubuntu software center ?
21:11 rsjtdrjgfuzkfg wouldn't it be possible to fix that issue on our side?
21:11 johnmurrayvi but in the 1.7.2 version, taglib has a FLAC::Picture class
21:11 freaktechnik Erikcht: still the same as I wrote in the forum post.
21:11 Erikcht ...
21:11 freaktechnik Erikcht: please do not disturb the meeting procedure defined by the notes.
21:11 johnmurrayvi rsjtdrjgfuzkfg one of POTI's changes was exporting the TagUnion class
21:11 freaktechnik *our
21:12 rsjtdrjgfuzkfg I haven't looked at all their changes in detail, however I had the impression that it should be possible to fix it at our side
21:12 johnmurrayvi so it would be a lot more painful to fix things on our side, especially when taglib is already at 1.9.1
21:12 rsjtdrjgfuzkfg expecially once they roll out virtual stuffz
21:12 johnmurrayvi (and 2+ will have virtual functions in TagUnion)
21:13 yeah
21:13 rsjtdrjgfuzkfg Updating to a recent taglib fixed some issues back then
21:13 freaktechnik well, the main issue we had, which lead to the whole idea/rediscussion is mp4 albumart reading not working
21:13 rsjtdrjgfuzkfg that is correct, and flac writing is missing as well afaik
21:13 johnmurrayvi the other thing is that there are some changes that deal with itunes tags
21:14 and the flac album art I believe
21:14 freaktechnik and I'd guess if we would take the time to use taglib's filestreams most  of the redwrite issues would probably be fixable.
21:14 if we discover an actual bug I vote for fixing it in taglib and pull it to taglib.
21:15 * rsjtdrjgfuzkfg too
21:15 johnmurrayvi I was thinking, at least for 1.12.2, it seems rational to fix a lot of issues by reverting to the patched version (1.7.2), then for 1.13 we could work on unpatched
21:15 freaktechnik would the patched version change our implementation?
21:15 (of the metadata handler)
21:16 johnmurrayvi I agree too. 1.12 was the first release with the vanilla taglib, so it makes sense to fix the 1.12.x branch imo
21:16 yes, but it's not too many commits. I already did it actually
21:16 when I was testing the patched versions
21:16 freaktechnik are you sure it does not (re)introduce other regressions?
21:17 johnmurrayvi not that I've been able to tell, but I haven't gotten to try it on mac or windows yet
21:17 https://github.com/johnmurrayv[…]ched-taglib-1.7.2
21:17 rsjtdrjgfuzkfg the taglib changes build on windows?
21:18 as I had some issues with that if I remember correctly
21:18 johnmurrayvi I haven't gotten to try compiling them on win yet, but I can start up my vm really fast
21:18 freaktechnik I think that was one of the resons why you wanted to get rid of the patches...
21:19 rsjtdrjgfuzkfg however, we know that johnmurrayvi is a master when it comes to windows building ;)
21:19 freaktechnik mhm
21:19 rsjtdrjgfuzkfg so he might get it to work
21:19 johnmurrayvi oh, duh that's where I put the changes, I completely forgot I forked taglib lol
21:19 https://github.com/johnmurrayvi/taglib
21:20 lol I think I just get super determined because of how annoying and frustrating it is
21:20 * freaktechnik knows that feel
21:20 freaktechnik I didn't do what I wanted to today, so I could get the freaking EQ presets to work
21:20 johnmurrayvi I set up multiple cross compilation systems and even with an old page from Mook, a system that uses the MSVC tools on linux through wine lol
21:21 in terms of deps, we will have to release new mac ones because of the dock stuff
21:22 freaktechnik mhm
21:22 johnmurrayvi so if we do revert to a different taglib, it would be easier to do it now
21:22 freaktechnik I still don't believe you built xulr on mac...
21:22 first I want to see testing builds for the revert on each platform.
21:22 So I can crash ot.
21:22 *it
21:22 johnmurrayvi lol it surprisingly wasn't too bad... I was shocked. I think ilikenwf figured out most of the stuff
21:23 building on windows is SOOOOOO much worse than mac...
21:23 rsjtdrjgfuzkfg ^^
21:23 freaktechnik so, can we agree on "Need testing builds for reverted taglib"?
21:24 johnmurrayvi I can get testing builds together. I just didn't want to spend time working on that stuff before discussing it here
21:24 yep!
21:24 rsjtdrjgfuzkfg if you guys want, yes
21:24 however, also: vanilla taglib for futurez!
21:24 freaktechnik sure
21:24 johnmurrayvi yes, definitely. maybe target it at 1.13?
21:24 freaktechnik but if the patching johnmurrayvi fixes stuff, we should prefer that for now
21:25 as that's less resources spent...
21:25 (and is immediate)
21:25 rsjtdrjgfuzkfg if it works on all platforms, yes
21:25 freaktechnik yeah, that's why I want to see testing builds
21:26 johnmurrayvi should have some soon, although I'm not sure if I've built the 1.9.2 deps on win
21:26 so I may only build taglib and replace the version in the deps already
21:26 freaktechnik that's how that works, yes
21:27 if you get xulr 1.9 to build, you're a genious ;
21:27 okay, so moving on to "Add vendor deps to git history"
21:28 johnmurrayvi ahh yes, so the "vendor" dependencies are a mashup of different pieces that have a few hacks in them
21:28 and not nearly all of it is built
21:28 freaktechnik what are they needed for?
21:28 to build against certain browser components?
21:29 johnmurrayvi but I figured, since there won't be anymore "vendor" dep releases, it seems like we should just add them to git
21:29 yeah
21:29 AFAICT, I'm not especially knowledgeable on the browser stuff
21:30 but it gives use fuel (fuel.xpt) and the base browser components (browsercompsbase.xpt)
21:30 freaktechnik oh, wait some of that stuff isn't available outside of xulr/gecko, I think
21:30 but fuel should be public
21:31 though possibly not before xulr2
21:31 johnmurrayvi it all comes from xr/gecko, and is version specific (hence vendor-xul9.0.1)
21:32 freaktechnik I am sure that fuel is public these days, but if we reimplement an interface or something, we'd need those idls then
21:32 so yeah, I think ilikenwf just ripped out everything in the depndencies folder, and that's why they're separate?
21:32 johnmurrayvi so the contents wouldn't change that often... since we aren't upgrading xr versions at super speed lol
21:33 freaktechnik well, those ifs are quite stable anyways
21:33 johnmurrayvi I think poti provided the vendor.zip folder initially
21:34 freaktechnik hm, okay
21:34 johnmurrayvi but the reason I mention adding it, was because of the parallel building stuff. I tweaked some of the makefiles in the "vendor" components, but those changes aren't tracked
21:34 freaktechnik I know that the whole deps stuff was in tree for the ubuntu builds
21:34 oh, they're actually no where in our tree right now?
21:35 johnmurrayvi the vendor deps?
21:35 freaktechnik mhm
21:35 johnmurrayvi they are only in the zip file that gets downloaded
21:35 I had to hack together the xr 9 vendor deps
21:36 freaktechnik "oh, that file's missing, lemme add that"
21:36 so where would you put them`?
21:37 johnmurrayvi I'm not sure. it kinda makes sense to leave them in the dependencies dir, and instead of downloading them from SF, they just get checked out with the tree
21:39 the first checkout after adding would require a "-f" though if someone already has a copy there
21:39 which isn't terrible, and I figure most people building from source would be able to figure it out
21:40 or, in the "community update" we could mention it
21:41 freaktechnik well, I guess it makes the most sense, and slowly remove the use of build.sh, which I despise.
21:41 (I dispise build.sh...)
21:41 * rsjtdrjgfuzkfg kinda dislikes it as well, however we should have some scripüt downloading the deps
21:42 johnmurrayvi yeah... I always use make instead of it, but I can understand having it
21:43 that seems to be a decent segway into the next section freaktechnik lol
21:43 freaktechnik well, you need it when switching from --enable-debug to not --enable-debug
21:43 johnmurrayvi: shush, you're not leading this ;)
21:43 but yeah, johnmurrayvi's been poking around the buildsystem to get it to build parallel
21:44 johnmurrayvi ahhh I didn't think about that. I honestly just downloaded both the release and debug deps and merged them lol
21:44 hahah okay :x
21:45 so I have been trying to get stuff to build with parallel job
21:45 s
21:45 hence the hacking in the vendor deps which was the first breakage point
21:46 from here (https://github.com/nightingale[…]ild/rules.mk#L117), everything builds alright with only ".NOTPARALLEL export" afaik
21:46 but when you remove it completely, that's when the problems really start
21:47 freaktechnik becuas eof the secondpass stuff?
21:48 johnmurrayvi I haven't gotten too far into it, but the changes to vendor do allow for the rest of the tree to start building in parallel, but it'll break when it's trying to build a lib that depends on another in the tree that hasn't been built yet
21:48 freaktechnik well, yeah, that's the main issues with parallel...
21:49 johnmurrayvi I'll hopefully have more info on second expansion and parallel builds for next meeting, but adding the vendor dir seems to be the first step
21:49 freaktechnik cool, looking forward to that...
21:49 so, you poked around windows a little more?
21:50 johnmurrayvi until I ran into the lib dependency issue, the build was going a lot faster
21:50 yep, I've been working on windows a fair amount again.
21:51 the major issue currently is still gettext, glib, libiconv, etc dropping msvc support
21:51 and not everything building with mingw yet
21:51 so I went back to working with msvc
21:52 and used the older versions of those deps which did support msvc
21:52 I've gotten the first run window up
21:53 having some bizarre issues, particularly with *possible* heap corruption when "secmod.db" is trying to be created
21:54 and deadlock around the same time
21:55 they're pretty frustrating problems, but hopefully will get resolved
21:56 I actually get through the first run wizard, but the problems seem to occur when trying to fetch the recommended addons and when the main window is opening (at "final-ui-startup")
21:56 freaktechnik is that why you moving some components around?
21:57 johnmurrayvi partially. all of the component registration stuff really confused me when I first started working on it last summer. I threw all of the binary component contracts and categories into jscomponents.manifest
21:58 so now I'm trying to take those out
21:58 freaktechnik well, mdn did say that all the kcomponents now needed to be registered in a manifest...
21:58 and I still don't understand that...
21:58 but, okay
21:58 johnmurrayvi and get all of the binary module registration correct, as well as ALL of the category entries for the binary and js components
21:59 yeah, that's what I read too... but I think that's only for js components
21:59 freaktechnik hm, possible
21:59 but why would you then have to register interfaces?
22:00 johnmurrayvi let me find this link
22:00 I think it's mainly just registering the .xpt files
22:00 like
22:00 freaktechnik so interfaces
22:00 johnmurrayvi yeah, but I think it's so it doesn't load extraneous stuff
22:01 freaktechnik oklay
22:01 well, let's move on to...
22:01 FREETYPEEEEE
22:01 johnmurrayvi woohoo.. freetype 2.5 broke the xr builds
22:02 for 1.9.2, 9.0.1, and 15.0.1 (all the ones I've built recently at least)
22:02 freaktechnik what do you suggest? Not updating ft?
22:02 johnmurrayvi there's a patch in the xul-9.0.1 branch that needs to be applied to the vanilla xr9 code
22:03 well I patched the 1.9.2 xr already (since it's in the tree)
22:03 freaktechnik oh, okay, great, so it's fixed and all :)
22:03 johnmurrayvi the 9.0.1 code isn't in the tree, so that patch has to be applied separately
22:03 yep!
22:03 (hopefully lol)
22:04 freaktechnik oh, I just realized I didn't add eq presets to the notes.
22:04 johnmurrayvi (quick, edit while no one is looking)
22:04 freaktechnik but, as we now come to a related topic anyways I'll include that...
22:04 johnmurrayvi oh wait
22:04 one sec
22:04 freaktechnik it's locked :P
22:04 johnmurrayvi link about registration stuff https://developer.mozilla.org/[…]nges_in_Gecko_2.0
22:05 and http://mxr.mozilla.org/mozilla[…]sSampleModule.cpp
22:05 just wanted to put that in there, since I mentioned trying to find the link before
22:05 freaktechnik so, we do now have the whole eq presets AOI in trunk
22:05 *API
22:06 I have implemented everything, there is still a few bugs and a few wontfixes
22:07 trying to find my gist
22:07 https://gist.github.com/freaktechnik/8941892
22:07 this is more or less the API
22:07 now, it's a little different to what we worked out earlier, mainly for efficiency resons.
22:08 for one I added ngIEqualizerPresetCollection, which offers array search methods which are much efficienter when performed on the underlying arrays than on an nsIArray
22:08 secondly I added a presets attribute to the manager, so clients don't have to aggregate the presets on their own and the aggregated presets list is identical accross all the implementations
22:09 what's left to fix is setting the textfield's value in the equalizer presets list on save and delete correctly
22:09 with what's currently in our source it gets cleared on those two actions
22:09 the itnerfaces in the source code are a lot better documentated, than that gist, btw.
22:10 now there is two wontfix issues, one of them being disregardable.
22:10 the first one is, that the initial value of the presetname will only be correct when nightingale is english, else it will not match it, as that's a localized preset.
22:10 I have no idea yet how to best fix that, but it's doable.
22:11 so not really a wontfix actually
22:11 second is that user set presets of a localizable rpeset (wso it changes its name when the application lang changes) don't change theirs too
22:11 but I guess that's expected, as the same is the case for example for genre tags.
22:12 so yeah. eq  presets.
22:12 I also move some strings from the dtd to a properties file and added three string to the dtd, but I guess that's more things that concern GeekShadow....
22:13 (that's a wall of text for you :P)
22:13 rsjtdrjgfuzkfg but all in all that looks fine for me
22:13 the first issue is, as you said, fixable
22:13 and I don't care for the second one
22:14 freaktechnik well, just the default value of the pref is not localizable, I guess, but we could make it switch to the localized counterpart on first init or something, yes.
22:16 good. I also worked oin the website a little, mainly updated it to be cutting edge again and merged it into the official website repo
22:16 now, I need some images for the features page, but I forgot to fix it, as I was playing with security stuff, so I can't show you the features page...
22:16 I guess I'll have to live with placeholders for a little longer...
22:17 more importantly I stumbled over anXSS issue with launch.getnightingale.com
22:17 it's all fixed now though, and I fixed launch in more recent browsers...
22:18 johnmurrayvi quick question (maybe I should have asked before this section, since it's about the wiki), should there be a link to the developer_meetings page somewhere on the main wiki page or "contribute" page or something?
22:19 (i.e. http://wiki.getnightingale.com[…]ings&do=backlink)
22:19 freaktechnik maybe, haven't found a good place for it yet, presonally, but feel free to add it where you think it fits
22:19 optimally it would be like on the front page, but as we do have no date/calendar stuff in the wiki, it's hard..
22:20 johnmurrayvi hm. well there is that empty cell in the bottom right hand corner of top table
22:20 freaktechnik you mean in the menu, yes.
22:20 johnmurrayvi ah yeah, that haha
22:21 really quickly, also related to the wiki, I posted http://wiki.getnightingale.com[…]=bugs:taskmarkers
22:22 which has a list of the current XXX/TODO/FIXME comments in the sb-trunk-oldxul code
22:22 * freaktechnik added some more //TODO stuff
22:22 freaktechnik oh, speaking of todo, the reset/Delete buttons don't switch correctly yet.
22:23 new issues... there's only two: 292 and 294. Also https://github.com/nightingale[…]ing/pulse/monthly
22:25 dont' think we need to talk about those two, as they were filed by the team...
22:25 johnmurrayvi (neither are terribly important)
22:25 yeah
22:26 freaktechnik according to the vote results on http://forum.getnightingale.com/thread-1041.html
22:26 we should make a blogpost covering the interesting new things we learn from this meeting
22:26 so, yay.
22:26 johnmurrayvi maybe for this first one, we should talk about the first two as well?
22:27 freaktechnik I say ilikenwf has to write it, so he reads back the logs ;)
22:27 johnmurrayvi hahah
22:27 freaktechnik yeah, that'd make sense
22:27 rsjtdrjgfuzkfg lol
22:28 freaktechnik seriously, anyone up for writing it?
22:28 johnmurrayvi so.. ilikenwf over all three haha
22:28 (last blog post was 12 jan 2014)
22:28 freaktechnik I know
22:28 johnmurrayvi I'd be willing to help write stuff, I'm just not exactly sure what to say.
22:28 freaktechnik that's why I brought it up, and that's also the main idea behind that community updates thing
22:29 do you have a blog account?
22:29 johnmurrayvi I don't think so
22:29 freaktechnik great
22:29 ...do I have the rights to add users....
22:30 can you pm me an e-mail?
22:30 johnmurrayvi (hm. clicked on the stats link and getting an error/warning at the top: http://stats.getnightingale.co[…]day&idDashboard=1)
22:30 rsjtdrjgfuzkfg freaktechnik: you should havre
22:30 johnmurrayvi yeah
22:30 freaktechnik I'll help writing it, and I'm sure ilikenwf will also help...
22:30 that error is from piwik being retarded when they integrated zen mode...
22:31 johnmurrayvi ahh
22:31 freaktechnik okay, yoou should get an email with your login infos
22:31 to sign in just click the signing button in the sidebar...
22:31 good.
22:31 so, misc.
22:32 I'd like to replace RMLL with ubuntu software center for Erikcht
22:32 Erikcht :)
22:32 freaktechnik (as GeekShadow doesn't seem to be around, I guess I'll speak with him about RMLL later)
22:33 so in order to get into the default repos for ubuntu the guys at cannonical suggested that we get a debian package into the debian repos.
22:33 and in order to get into the debian repos, we would need a maintainer there.
22:34 Erikcht freak
22:34 freaktechnik but I don't know much about how the whole debian package ecosystem works. So. Anyone volunteering to maintain us in the debian repos is welcome
22:34 we do build more or less from our source on debian, so not much maintaining is needed in theory
22:35 Erikcht my friend, RD2 could help, he know build .deb . For any help you need...
22:36 freaktechnik as I said we can build .deb files. From our source. Without big modifications. See our wiki article about building
22:36 Erikcht wow, ok....
22:36 freaktechnik what we would need is a) somebody to maintain nightingale as package in the debian repos and b) we would have to fixup some stuff, maybe
22:36 Erikcht ok
22:37 the time each update ?!
22:37 freaktechnik (as we currently fail the .deb linting tests)
22:37 johnmurrayvi Erikcht we can build the .debs, but we need to get nightingale added to the debian repositories so that we can possibly be added to the ubuntu software center
22:37 Erikcht ok
22:37 johnmurrayvi ohhh, I remember Ukikie was asking me about building the deps with hardening
22:38 freaktechnik well, nightingale is on the wanted packages list for debian, so this basicly means all we'd need is a maintainer for the debian package in the debian package repositories...
22:38 johnmurrayvi I need to do that still
22:38 freaktechnik the whole norelo stuff too
22:38 and the 644 premissions
22:38 but else we're in good shape
22:38 johnmurrayvi yeah
22:38 Erikcht YOU NEED TO FIX 644 PERMISSIONS ?
22:38 johnmurrayvi are those the only real issues with the lint?
22:38 linting*
22:38 freaktechnik well, the permissions aren't an issue with the lint, as we patch those while building now, but that's a workaround
22:39 afaik, yes
22:39 Erikcht wow... ok
22:39 freaktechnik Erikcht: well, mozilla sources are strange sometimes.
22:39 johnmurrayvi hmm
22:39 freaktechnik optimally images wouldn't be 755
22:39 (not only images though)
22:39 johnmurrayvi the libs?
22:40 freaktechnik idk
22:40 mainly jars and such things
22:40 you can see a list of the file extensions we currently patch in the dependencies makefile, I think
22:40 johnmurrayvi what files should be 755?
22:41 freaktechnik only executables, afaik
22:41 and not the resources too
22:41 johnmurrayvi ah, alright, I need to look at the linting output again
22:42 freaktechnik hm, I think acutally the only one from the deps that should be 755 is the run-mozilla.sh, looking at the makefile...
22:42 https://github.com/nightingale[…]20885a19155c208b1
22:42 johnmurrayvi does it complain about anything in the compiled/dist/xulrunner dir? because I think I had mainly been looking at  https://github.com/nightingale[…]le-deps/issues/17 and https://github.com/nightingale[…]le-deps/issues/18
22:42 (the first one being caused by https://github.com/nightingale[…]er-tarball.sh#L42)
22:42 freaktechnik I don't exactly remember...
22:43 well, djsj linked the debian spec
22:43 http://lintian.debian.org/tags[…]ecutable-bit.html
22:44 johnmurrayvi ohhh yeah okay the dependencies/Makefile.in changes definitely look like they're caused by the make-xulrunner-tarball.sh file
22:45 freaktechnik joined #nightingale
22:45 freaktechnik urgh
22:46 johnmurrayvi I'm not sure why the permissions are changed in the first place
22:46 freaktechnik there shouldn't be any reason that I'd know of...
22:47 while working on the EQ presets I've noticed that sb and mozilla use different naming schemes for contractIDs
22:48 I've gone the middle way, by doing an uppercase Nightingale followed by the lowercase "normal" contract id, mozilla style.
22:48 now, I'm not sure how we should do it.
22:48 johnmurrayvi sorry to bug you Mook, but is there a reason why the files from xulrunner are set to 0755? (specifically, here https://github.com/nightingale[…]er-tarball.sh#L42)
22:48 freaktechnik mozilla does it all lowercase and without the appname in the contract id
22:49 johnmurrayvi (I wasn't sure if there were permissions issues on mac and windows)
22:49 freaktechnik while POTI always uppercased the single parts of the contract id and it always starts with @songbirdnest.com/Songbird
22:49 and honestly, I have no idea what we should do, but we should always use the same scheme
22:49 so let's define one!
22:49 (this is the interactive part)
22:50 rsjtdrjgfuzkfg I do not see the difference tbh
22:50 @mozilla.org/some/stuff;Version
22:50 vs @songbirdnest.com/some/stuff;Version
22:50 isn't it?
22:50 freaktechnik Mozilla: "@mozilla.org/mediacore/manager;1" Songbird: "@songbirdnest.com/Songbird/Mediacore/Manager;1"
22:50 (the mozilla one doesn't exist)
22:51 rsjtdrjgfuzkfg that's the same syntax
22:51 freaktechnik no
22:51 rsjtdrjgfuzkfg I mean upper- or lowercase is not that important for me
22:51 freaktechnik one is uppercased and has the appname in it
22:51 well, it's a mess if it changes everywhere...
22:51 rsjtdrjgfuzkfg the idea of contracts is that each domain is allowed to specify its own naming schema
22:51 johnmurrayvi hmm... not all of the sb stuff has "Songbird"
22:51 https://github.com/nightingale[…].manifest.in#L113
22:51 freaktechnik I personally also doubt that we'd need the app name in it
22:51 rsjtdrjgfuzkfg at least that's what I think of that
22:52 so I'm for @getnightingale.com/XXX;Version with XXX being lowercase
22:52 freaktechnik urgh, so yeah, POTI has it randmly...
22:52 rsjtdrjgfuzkfg but we can choose that freely
22:52 freaktechnik also, no /nightingale, plz
22:52 rsjtdrjgfuzkfg imgo
22:52 *imho
22:52 freaktechnik well, that contract ID list johnmurrayvi linked makes me cringe
22:53 because it makes it harder to remember CIDS
22:53 johnmurrayvi freaktechnik this made me cringe, I had to double check that it was correct: https://github.com/nightingale[…].manifest.in#L190
22:53 that's a lower case "songbird"
22:53 freaktechnik ...and that's why we should have some basic rules...
22:54 rsjtdrjgfuzkfg yes, I agree on that part
22:54 johnmurrayvi oh, that list is going to be split up eventually...
22:54 rsjtdrjgfuzkfg but imho we should not need to change old ones
22:54 johnmurrayvi nah I agree
22:54 freaktechnik well, I'll change my newer ones, like the mpris and the eq presets ones
22:54 but what's already done stays, yes.
22:54 johnmurrayvi too many to change
22:54 freaktechnik shouldn't change the API if it's not absolutely necessarty
22:54 johnmurrayvi may break addons too?
22:54 rsjtdrjgfuzkfg if you want to change them, you can change them. I mainly wanted to express that I don't see a reason to change everything :D
22:55 johnmurrayvi: yes
22:55 freaktechnik changing the SB ones is nonsensical
22:55 johnmurrayvi yeah
22:55 It looks like GeekShadow may be back?
22:55 freaktechnik why?
22:55 oh, he's massrtetweeting, I see
22:55 johnmurrayvi his name became colored (unmarked away?)
22:56 freaktechnik oh
22:56 my client doesn't indicate that...
22:56 good, so lowercased cids
22:56 johnmurrayvi works for me
22:57 freaktechnik then, mozilla switched from NS_ENSURE macros to if(NS_SUCCESS(rv)) and similar
22:57 rsjtdrjgfuzkfg yes
22:57 GeekShadow yo
22:57 freaktechnik because the NS_ENSURE macros hide returns
22:57 rsjtdrjgfuzkfg h GeekShadow
22:57 GeekShadow sorry ^^ I was out tonight
22:57 freaktechnik I personally prefer the NS_ENSURE macros...
22:57 hi GeekShadow
22:57 they make code easier to read, as there is no unnecessary blocking
22:57 johnmurrayvi (GeekShadow can help with the massive blog post too :) )
22:58 freaktechnik yes, it makes understanding code when you don't know the macro yet harder, but when working on code, it's way slicker than having if cluases and stuff for the error handling
22:58 GeekShadow not sure what is this about
22:58 johnmurrayvi I've only seen NS_SUCCESS really... when did they switch?
22:58 freaktechnik yes
22:58 lemme search the mdn article
22:59 but it's basicly writing out parts of the NS_ENSURE macro in code
23:00 johnmurrayvi can't the rv value indicate specific errors, like errno does?
23:00 freaktechnik yeah, but NS_SUCCESSFUL(rv) only is true if it's NS_OK
23:00 for example
23:01 so instead of using NS_ENSURE_SUCCESS(rv, rv) you'd do if(NS_FAILED(rv)) { return rv; } or if(NS_SUCCESSFUL(rv)) { /* code */ return NS_OK; } return rv;
23:02 * rsjtdrjgfuzkfg prefers the NS_ENSURE_SUCCESS stuffs
23:03 freaktechnik https://developer.mozilla.org/[…]e_the_nice_macros
23:03 so we'll stick to NS_ENSURE, good
23:03 johnmurrayvi ohhh, I was mixing up what I had been seeing... e.g. https://github.com/nightingale[…]rService.cpp#L173
23:03 freaktechnik I personally don't think hidden returns are bad...
23:04 well, there's three macros you'lll see a lot. NS_ENSURE_ARG_POINTER, NS_ENSURE_TURE and NS_ENSURE_SUCCESS, with the last one being the most popular...
23:04 rsjtdrjgfuzkfg I think hidden returns are bad if they occur on the expected execution path
23:04 but for ensuring that is fine with me
23:04 (I even think it is nicer...)
23:04 freaktechnik as I said, it doesn't break up the normal code flow with unnecessary blocks
23:05 and it's a bit like doing a throw...
23:05 johnmurrayvi yeah
23:05 I'm okay with whatever, I think there are cases for both types
23:06 freaktechnik well, NS_SUCCESS and NS_FAILED are useful too, for example when iterating
23:06 rsjtdrjgfuzkfg yes
23:06 johnmurrayvi yeah
23:06 I like the NS_WARN_IF macros though
23:07 "Previously the NS_ENSURE_* macros were used for this purpose, but those macros hide return statements and should not be used in new code."
23:07 freaktechnik well, that's basicly NS_ENSURE_TRUE(somethingthatshouldbeture, NS_ERROR_WHATEVER_ERROR_WARN_GIVES)
23:08 oh wait, the point is it doesn't return I guess
23:08 it just NS_WARNS
23:09 johnmurrayvi seems like it could be pretty useful when debugging
23:09 freaktechnik well, it's a bit like an assert, yes.
23:10 but in the end you get a trace to the error too when using NS_ENSURE
23:10 soo
23:10 Erikcht People, to remember...
23:11 The shoutcst addon isn't showing the radios
23:11 freaktechnik yep, bug is filed
23:11 Erikcht and... the recomendations , on ubuntu , inst't downloading the recomendations addons
23:11 freaktechnik that depends on the build you're using
23:12 if you're on nightly you won't get anything
23:12 Erikcht wow...
23:12 if ... installing by that ?!
23:12 freaktechnik oh, also XPCOM error of the month is NS_ERROR_XPC_GS_RETURNED_ERROR
23:12 because it's the dumbest thing ever
23:12 johnmurrayvi hahah I've gotten that a lot....
23:13 Erikcht if ... installing by that !?
23:13 freaktechnik it gets thrown when an error gets thrown while constructing an XPCOM component. But instead of giving you the error, it only gives you this boilerplate error
23:13 same for _CI_
23:13 johnmurrayvi there is a ppa for the nightly builds Erikcht
23:13 freaktechnik Erikcht: if you install the nightly build you will not get add-on recommendations
23:13 Erikcht sorry,,,
23:13 freaktechnik in the normal build they _should_ work
23:13 Erikcht sudo add-apt-repository ppa:nightingaleteam/nightingale-nightlysudo apt-get updatesudo apt-get install nightingale
23:14 freaktechnik but it's possible I broke them...
23:14 uhm, GeekShadow, anything to say re RMLL?
23:14 I'll be there saturday evening, iirc
23:14 johnmurrayvi I've gotten both of those errors quite a bit with the category registration stuff
23:15 rsjtdrjgfuzkfg have fun there ;)
23:15 GeekShadow freaktechnik, for RMLL i'll try to have your fee included in Mozilla
23:15 not sure yet
23:15 because i'll lead the organization of Mozilla booth this year
23:16 freaktechnik oh wow
23:16 GeekShadow I have stickers left
23:16 we may also prepare print stuff ?
23:16 johnmurrayvi freaktechnik why do you think you broke the addon stuff?
23:16 freaktechnik hm, we should probably start writing the RMLL blogpost too
23:17 johnmurrayvi: because I get a warning every time I launch ngale, that it couldn't update the recommendedAddOnsBundle, and I'm always hoping that the normal one still works
23:17 but it should, as I only nuked songbirdnest references
23:18 johnmurrayvi (in case I look back at this again if I forget, "-fstack-protector" and LDFLAGS += "relro")
23:18 freaktechnik god, so I think that's it for today.
23:19 *good
23:19 GeekShadow lol
23:19 it sounded like you were exhausted ?
23:19 johnmurrayvi haha
23:19 freaktechnik I am
23:19 it's fucking hot here
23:19 rsjtdrjgfuzkfg ^^
23:19 GeekShadow yeah same here
23:20 freaktechnik oh, wait, one thing.
23:20 next one's one the 5th of July, live from RMLL
23:21 rsjtdrjgfuzkfg yay!
23:21 GeekShadow oooh yep
23:21 Ukikie Oh my goodness, so much backlog.  So I read a little context on the highlight, lintian output is just some things that may need fixing, it's not really the big problem so much.  The big problem is the bundled libs, sqlite, xul, etc.
23:21 Topic for #nightingale is now Nightingale Media Player || http://getnightingale.com || Current release: 1.12.1 || Offtopic #chirping || Please idle and wait for a reply! || Channel logs: http://bit.ly/ngaleirc || NExt Dev-Meeting: 5.July live from RMLL
23:21 freaktechnik xul will always be bundled, ukikie
23:21 Ukikie (Might be wrong on that one, but sqlite and stuff.)
23:22 Hardening is just a feature, but AFAIK it's still not a blocker.
23:23 johnmurrayvi Ukikie do you know anyone who could be a Debian maintainer? I'd do it, but I have no idea what I'm doing there haha.
23:23 Ukikie Also a maintainer, better if it's from pkg-mozilla, but someone to maintain it in Debian.
23:24 johnmurrayvi ahh
23:26 Ukikie http://qa.debian.org/developer[…]alioth.debian.org - http://lists.alioth.debian.org[…]zilla-maintainers *may* be a good choice?
23:26 freaktechnik but our icon has a strange license. *hides*
23:26 Ukikie But, err, not me. :P
23:26 Hah, oh dear.
23:27 freaktechnik need to create icegale.
23:27 GeekShadow lol
23:27 they are sure going to remove the DRM stuff in iceweasel
23:27 Ukikie Haha.:D
23:27 freaktechnik about:drm
23:28 ?
23:28 johnmurrayvi Ukikie how do you call lintian on the ng debs? any specific flags to get the most info? I haven't used it much
23:28 freaktechnik also, I easteregged. But you've probably seen that all in the diff...
23:29 Ukikie johnmurrayvi: lintian -I -E --pedantic nightingale_1.12.1-1_amd64.changes (do it on the changes file and it'll check both source package and binary.)
23:29 GeekShadow: Can't say I blame them fully there, though.
23:30 johnmurrayvi sweet, thank you!
23:31 freaktechnik well, I have to say, mozilla's approach is the best one while tstill trying to be marketrelevant...
23:31 johnmurrayvi OH before I forget, and this is still being logged
23:31 freaktechnik which is something iceweasel doesn't have to do
23:31 johnmurrayvi I got the deps and ng-h to build with clang and gold
23:31 just wanted to throw that in there
23:32 Ukikie freaktechnik: I haven't compared, not really seen them all.
23:32 And, nice, clang. :D
23:33 johnmurrayvi it'll be way better with parallel builds
23:33 * freaktechnik wants mach
23:33 johnmurrayvi gold also. (gold = multithreaded ld)
23:34 freaktechnik those moz.build files just look sexy
23:34 GeekShadow gold is just a linker no ?
23:34 johnmurrayvi yeah it is, but it's quite a bit faster than ld
23:35 freaktechnik oh, johnmurrayvi (and anybody else who understands our c++ stuffs), why do we build the base mediacore trice, instead of linking to the built objects?
23:35 johnmurrayvi hmm?
23:35 freaktechnik most of the base .cpp get rebuilt in gstreamer and manager/handler
23:36 johnmurrayvi does that subdir get built three times?
23:36 oh
23:36 oh wait
23:36 freaktechnik those inherit from the base stuff
23:36 johnmurrayvi yeah
23:36 so it shouldn't be rebuilding the base files?
23:36 is it?
23:37 freaktechnik http://forum.getnightingale.co[…]t-183.html#pid183 this is what manko said about the lincensing of our logo, and as he made ours, I think we should respect that. So icegale
23:38 johnmurrayvi: https://github.com/nightingale[…]kefile.in#L56-L67
23:38 I was pretty confused, as I had to add a header include so sbBaseMediacoreMultibandEqualizer.cpp would build...
23:38 but yeah, we compile some stuff twice or thrice.
23:38 *trice
23:39 now I wasn't sure if we could omit that by importing the compiled libraries when linking, as those are all classes inherited by GStreamer implementations
23:40 GeekShadow freaktechnik, what's wrong with the licence ?
23:40 freaktechnik well, it's not trademarked, so we should get through, right
23:42 johnmurrayvi freaktechnik that's a good question about the base mediacore stuff, I'm not sure to be honest. I'll see if I can figure anything out
23:44 freaktechnik same thing here btw: https://github.com/nightingale[…]kefile.in#L45-L52
23:45 johnmurrayvi I would think it should just link to the already compiled base objects too
23:46 freaktechnik so adding the base to DYNAMIC_LIBRARY_EXTRA_IMPORTS or something
23:46 johnmurrayvi or DYNAMIC_LIB_EXTRA_OBJS
23:46 freaktechnik idk all these makefile constants by heart...
23:46 I always have to looke them up in the rules.mk
23:47 johnmurrayvi haha I looked up that one...
23:47 freaktechnik GeekShadow: re print stuff at RMLL: yeah, maybe, no idea thouugh
23:47 johnmurrayvi I thought I had seen it before
23:47 freaktechnik not sure if it's worth it to import the objects one by one or just import the whole folder...
23:47 as they both import at least half of the stuff from base
23:50 johnmurrayvi I was thinking just import the path and change the CPP_EXTRA_SRCS to DYNAMIC_LIB_EXTRA_OBJS then changing the ".cpp" to "$(OBJ_SUFFIX)"
23:50 Erikcht people, i am gone out.
23:50 freaktechnik mh, that's coo
23:51 Erikcht thx
23:51 freaktechnik by Erikcht
23:51 johnmurrayvi bye
23:51 looks like it might be working
23:52 freaktechnik cool, that's the kind of optimization I like
23:52 ...
23:53 johnmurrayvi oh wait... hmm, why am I still getting sbBase*.o's...
23:53 freaktechnik make clobber
23:53 ngalebot joined #nightingale
23:53 ngalebot [13nightingale-hacking] 15freaktechnik pushed 1 new commit to 06sb-trunk-oldxul: 02http://git.io/uYo0Uw
23:53 13nightingale-hacking/06sb-trunk-oldxul 1476b8a5d 15Martin Giger: Change contract IDs to agreed format
23:53 ngalebot left #nightingale
23:53 freaktechnik aka from the last build
23:54 johnmurrayvi well I had removed them... so I'm not sure what rule re-built them
23:55 GeekShadow freaktechnik, looks like I have to move all eq strings
23:55 johnmurrayvi oh, that may have been why, I think I cleaned out the base dir too? maybe. Idk
23:55 GeekShadow I'll script that
23:56 freaktechnik GeekShadow: mhm
23:56 sorry about that, but dtds don't like js and the other way around...
23:56 GeekShadow but there is a way to use dtd from js
23:57 freaktechnik makefiles were regenerated, johnmurrayvi (I know they should, but ergh, make"
23:57 GeekShadow: yes, but it's not cool
23:57 GeekShadow lukas wrote the script
23:57 freaktechnik plus it doesn't make sense
23:57 johnmurrayvi I had changed the Makefile.in file in gstreamer/src... hmm

← Previous day | Index | Server Index | Channel Index | Today | Next day → | Atom Feed | Search | Google Search | Plain-Text | plain, newest first