All times shown according to UTC.

Time Nick Message
01:18 stevew joined #nightingale
07:57 stevew joined #nightingale
17:31 freaktechnik :S https://nightingale.pastebin.mozilla.org/6352847
17:48 \o/ http://forum.piwik.org/read.php?2,119615
19:52 johnmurrayvi joined #nightingale
20:12 freaktechnik hi johnmurrayvi
20:12 johnmurrayvi hello!
20:12 it looks like it's been super quiet in here recently..
20:12 freaktechnik well, it's turned into my second twitter account again, yeah
20:12 johnmurrayvi haha
20:12 freaktechnik but I thought ngissuesbot a new trick :3
20:12 ngissuesbot freaktechnik:Hm?
20:13 freaktechnik however he forgets too much
20:13 johnmurrayvi hahah I saw a couple of them in the logs
20:13 freaktechnik (or I launched him with a different config)
20:13 johnmurrayvi I still need to set up an irc bouncer/whatever
20:13 freaktechnik (I guess I launched him from my root or home this time, I'd hope home)
20:14 it seems GeekShadow's bouncer died about ten hours ago...
20:14 johnmurrayvi that's pretty unfortunate. especially given the timing
20:15 freaktechnik mhm
20:20 johnmurrayvi soooo what happens if the meeting leader isn't here? lol
20:25 freaktechnik bug him on fb, else carry on, I guess^^
20:31 johnmurrayvi ahh :( well at least there hasn't been a ton of stuff since the last meeting
20:50 Mook joined #nightingale
21:00 freaktechnik :/
21:00 johnmurrayvi lameeeeeee
21:02 rsjtdrjgfuzkfg joined #nightingale
21:04 freaktechnik hi rsjtdrjgfuzkfg
21:04 rsjtdrjgfuzkfg hi freaktechnik
21:07 johnmurrayvi lol I just noticed the title of the meeting
21:08 freaktechnik <_<
21:08 johnmurrayvi fitting title..
21:10 freaktechnik I did that two weeks ago...
21:10 in my inability to calculate a week before this week.
21:10 johnmurrayvi hah wow.. I'm apparently oblivious
21:11 freaktechnik so, tried to reach GeekShadow via Twitter, Fb and GTalk. Not sure if it's worth to spend the money for an SMS...
21:11 johnmurrayvi lol nah
21:11 freaktechnik oh, wait, geekshadow changed it^^
21:11 johnmurrayvi check the meeting notes now lol
21:12 freaktechnik he changed it from "Where's GeekShadow" to "Back to School"
21:12 kay
21:12 johnmurrayvi ohhhhh, yeah that's what I was mentioning lol
21:12 freaktechnik (just going to throw this in there anyways https://github.com/nightingale[…]ng/pulse/monthly)
21:12 johnmurrayvi slow month...
21:13 rsjtdrjgfuzkfg ^^
21:13 freaktechnik djcj month
21:13 johnmurrayvi I'll be the meeting leader again (that was the edit I made to the notes)
21:14 so I guess if that's okay, we can start?
21:15 rsjtdrjgfuzkfg mhm
21:15 freaktechnik go ahead, I guess GeekShadow is loast at some ffOS event, judging his tweets
21:16 johnmurrayvi welcome to the sixth meeting everyone (who made it)
21:16 freaktechnik even Mook made it!
21:16 johnmurrayvi yeah! haha
21:16 not ilikenwf or GeekShadow though..
21:17 judging from: https://github.com/nightingale[…]ing/pulse/monthly
21:17 it'll be a pretty short meeting
21:17 the notes are here: http://wiki.getnightingale.com[…]eloper_meetings:6
21:17 freaktechnik well, you might want to highlight this one, though: https://github.com/nightingale[…]eps/pulse/monthly
21:17 ;)
21:18 johnmurrayvi first up, some debian/lintian stuff
21:19 oh hah wow I didn't check the deps pulse
21:19 freaktechnik debian: We added the ignoring rules for things we won't fix, and that's about it, iirc.
21:19 (from the full diff of this month)
21:20 wait, no that was last month even, I guess
21:20 johnmurrayvi https://github.com/nightingale[…]..sb-trunk-oldxul
21:21 mostly cleanups it seems
21:21 freaktechnik well, yeah, the makefile stuff is fixing the makefile to comply better with the makefile format conventions.
21:22 johnmurrayvi makes sense to do that
21:23 does the autogeneration of the makefiles cause problems for that?
21:24 freaktechnik uhm, it'd be easier for the main makefile, if it was autogenned too, but that's not how the mozilla buildsystem works
21:24 johnmurrayvi ahh true
21:25 well, moving onto dependencies..
21:25 I guess there was more work done than I thought lol
21:26 rsjtdrjgfuzkfg from the notes I assume windows is the "only" thing missing?
21:26 johnmurrayvi it all relates to the first topic in the notes
21:26 rsjtdrjgfuzkfg yes, that's pretty much it lol
21:26 rsjtdrjgfuzkfg ^^
21:26 * rsjtdrjgfuzkfg remembers windows pains
21:27 johnmurrayvi gettext / other gnu libs dropped MSVC support
21:27 so, that's one reason it's been so terrible
21:27 rsjtdrjgfuzkfg didn't you try to cross-compile or something?
21:27 johnmurrayvi (actually I have 139 commits this month in my fork: https://github.com/johnmurrayv[…]ps/pulse/monthly)
21:28 I did before, mainly for xr-9+, but the problem is that xr-1.9.2 builds mozcrt with the msvcrt code
21:29 (in mozilla/memory/jemalloc)
21:30 the CRT source is patched when it's built too, so that the CRT uses the jemalloc memory functions instead of the windows ones
21:31 using mingw, the libs try to link against a generic "msvcrt.dll", but to get the relocatability/congruency with other deps, mozcrt is used
21:31 Mook would not using jemalloc be a workaround? (kinda sucks, yeah...)
21:31 (that would involve taking out the hacks to use mozcrt, of course)
21:32 johnmurrayvi possibly? I haven't tried that yet. I spent some time updating your/poti's work with the msvc 2005 patch for msvc 2008 though
21:33 Mook (mostly preed's, IIRC)
21:33 johnmurrayvi ahh
21:34 well I was thinking (hoping) msvc 2008 would be easier to build the updated gettext, libiconv, glib, libffi etc with
21:34 as I had gotten xr 9 to work somehow, which used 2010
21:34 * rsjtdrjgfuzkfg remembers to try that a long time ago... without success, though...
21:35 johnmurrayvi hmm.. not encouraging lol
21:35 rsjtdrjgfuzkfg :D
21:35 but I did not even get xr to builg
21:35 so that is at least a bit encouraging, right? :P
21:35 johnmurrayvi haha I suppose so
21:37 I can't remember for sure (it's been a few days) but I think I may have gotten xr to build with msvc 2008 (with the patching)
21:37 more info here though: https://github.com/nightingale[…]ecomment-53739857
21:38 it's definitely worth trying a build without jemalloc though
21:38 I have mac and linux-x86_64 dep builds with gst1 up on SF though (in a staged dir)
21:39 rsjtdrjgfuzkfg so you expect to release in... 3 days? :D
21:39 or whatever the staged folders stage for
21:39 freaktechnik hehe
21:39 I'd suggest a beta first^^
21:40 rsjtdrjgfuzkfg alpha?
21:40 johnmurrayvi haha nah, I just wanted to put something up for an alpha version
21:40 freaktechnik rsjtdrjgfuzkfg: nightly builds are like alphas, I'd say.
21:40 (I mean, we verison them with a)
21:40 johnmurrayvi ahh
21:40 well, the gst-1 branch will need to be merged into trunk
21:40 * rsjtdrjgfuzkfg thinks of a good ol' testing build :P
21:41 johnmurrayvi assuming I can get the windows stuff working
21:41 freaktechnik yeah, we'll also need to regenerate langpacks first...
21:42 johnmurrayvi hm, alright
21:42 so, moving on.... taglib... what should we do for 1.13
21:42 I have the patched 1.7.2 version
21:43 and it actually passes all metadata tests on linux-x86_64
21:43 rsjtdrjgfuzkfg 1.7.2 does not yet have virtuals in it, right?
21:43 johnmurrayvi but that would be changing more in the mediacore
21:43 unfortunately it does not
21:43 I don't think they will until 2.0
21:43 rsjtdrjgfuzkfg some way we can use 1.8?
21:43 oh
21:44 johnmurrayvi 1.8 is what is currently being used
21:44 * rsjtdrjgfuzkfg thought they wanted that in the 1.7 area, so I'd expect 1.8 to have it
21:44 rsjtdrjgfuzkfg so we still need the nasty workarounds there
21:44 johnmurrayvi actually, with the patched taglib, they wouldn't be needed..
21:45 rsjtdrjgfuzkfg if you wish to patch taglib...
21:45 johnmurrayvi but it also means using a modified version of taglib
21:45 rsjtdrjgfuzkfg that's what I'm not liking
21:45 johnmurrayvi (I have https://github.com/johnmurrayvi/taglib already actually)
21:47 rsjtdrjgfuzkfg how much of a mediacore change are you thinking?
21:47 johnmurrayvi I can understand that. however, afaik, system taglib isn't usable right now anyways... plus mac and windows don't really have a "system taglib" option anyways, so they'd still have issues
21:48 I think this was all that was needed: https://github.com/johnmurrayv[…]ched-taglib-1.7.2
21:48 rsjtdrjgfuzkfg does it still work? I remember that some stuff did not work with the old code
21:49 not sure what exactly, though
21:49 (total tracks or something)
21:49 freaktechnik rsjtdrjgfuzkfg: me and johnmurrayvi checked various things
21:49 rsjtdrjgfuzkfg do we use the same rating field?
21:49 freaktechnik but as your patching is still mostly in there we couldn't reproduce any of ti
21:49 that one I do not know
21:49 rsjtdrjgfuzkfg as I changed it back then iirc
21:49 johnmurrayvi I think we checked the unicode(?) characters, 900+ import limits, and having a corrupted song in batch import causing a hang
21:49 freaktechnik but checked for the batch import limit and the utf8 stuff
21:50 rsjtdrjgfuzkfg and I think our users should have persistent fields, to not loose ratings
21:50 freaktechnik yeah, I agree on that
21:50 rsjtdrjgfuzkfg except if you implemented standard-compliant popm usage
21:50 but I'd assume you didn't.
21:51 freaktechnik main reason a patched taglib would be better right now, is because the unpatched one doesn't build against half of the system taglibs for linux distros out there, for osme reason (hardening?)
21:51 rsjtdrjgfuzkfg agree on that
21:51 johnmurrayvi it's the included header file that declares the virtual functions I think
21:51 rsjtdrjgfuzkfg johnmurrayvi: afaik I did not use a custom header, did I?
21:51 johnmurrayvi but the unpatched taglib has had issues with metadata in read-only files too I think?
21:52 freaktechnik rsjtdrjgfuzkfg: ilikenwf included that to build agains system taglibs
21:52 rsjtdrjgfuzkfg ah.
21:52 johnmurrayvi https://github.com/nightingale[…]ib/src/tagunion.h
21:53 rsjtdrjgfuzkfg Hm, while I don't have anything against bundling a patched taglib, I'd prefer to either fork taglib to make it clean (virtual methods and stuff) or bundle a vanilla taglib.
21:53 johnmurrayvi it's not normally exported in a taglib build
21:53 rsjtdrjgfuzkfg However, if you have already something that works, why not.
21:54 Some day somebody (TM) will need to change the whole thing anyways once we properly handle multiple values per tag
21:54 johnmurrayvi that does seem like a good idea, I'm just not sure how much extra work that would be right now
21:54 true
21:54 freaktechnik also, it seems taglib2 will bring some nice improvements
21:54 like better m4a support out of the box
21:54 rsjtdrjgfuzkfg is taglib2 "out" yet?
21:54 freaktechnik and FLAC, too I think
21:54 nah, still in dev
21:54 rsjtdrjgfuzkfg if so, we could consider bundling it imho
21:55 freaktechnik (at least last I checkeD)
21:55 rsjtdrjgfuzkfg I could live with some kind of beta, if it is more reliable than the current solution :P
21:55 johnmurrayvi yeah, the latest release is 1.9.1
21:55 freaktechnik I am not sure the taglib2 branch is even stable yet...
21:56 (I mean, kind of stable)
21:56 rsjtdrjgfuzkfg then that is not a good idea :P
21:56 johnmurrayvi rsjtdrjgfuzkfg there are some testing builds I put up here: https://sourceforge.net/projec[…]ib-1.7.2-testing/
21:57 freaktechnik so we need to make sure we're still using the same rating field then
21:57 johnmurrayvi hmm
22:00 I'm not sure I know enough about that to say yes or no
22:00 rsjtdrjgfuzkfg what question are you referring to?
22:01 johnmurrayvi whether the rating fields would be compatible
22:01 rsjtdrjgfuzkfg you can simply try
22:01 I think I used NIGHTINGALE-RATING
22:02 but not sure
22:02 or RATING-NIGHTINGALE
22:02 or something like that
22:02 johnmurrayvi for which file type? also, I was about to say I don't think I've tried the patched version with a profile that was previously using the unpatched version
22:03 freaktechnik ID3, I'd think
22:03 rsjtdrjgfuzkfg I think I switched to unified fields
22:03 so, the internal field names were equal for everything
22:03 for some special fields, taglib does the conversion
22:04 other fields are handled as custom fileds with the given name (in all tags supporting that)
22:04 afaik the rating tag is one of them
22:04 johnmurrayvi one option could be a migration handler, but I'm not sure how to use them tbh
22:04 * rsjtdrjgfuzkfg would prefer not to use migration unless necessary
22:04 rsjtdrjgfuzkfg e.g. on a database change
22:04 freaktechnik I wouldn't think migration should be necessary for that
22:05 johnmurrayvi oh wait
22:06 so could I just change this: https://github.com/johnmurrayv[…]id3v2tag.cpp#L219
22:06 rsjtdrjgfuzkfg likely
22:07 johnmurrayvi here's the current: https://github.com/nightingale[…]rTaglib.cpp#L2440
22:08 ape tags too: https://github.com/nightingale[…]rTaglib.cpp#L3336
22:08 I think in the patched version, they're all just "rating"?
22:08 rsjtdrjgfuzkfg possible
22:08 but that is more likely to cause conflicts
22:09 freaktechnik also against most metadata specs, right?
22:09 rsjtdrjgfuzkfg yes
22:09 but our current field valid, but only a comment
22:09 the correct way would be popm
22:09 johnmurrayvi ahh, okay, I didn't know that
22:10 hmm
22:10 freaktechnik tagging is a science on its own, apart from media management and playback^^
22:10 johnmurrayvi yeah haha
22:11 I'll look into the popm field, and change all of the other rating fields
22:12 and I guess I'll merge in the mediacore changes after that and after I put up new dep builds with the patched taglib
22:13 and possibly merging in the gst1 stuff
22:13 rsjtdrjgfuzkfg if you merge after it works on windows, that's fine with me :P
22:13 (if I should forget it: good luck with that)
22:14 johnmurrayvi haha sounds good! if you'd like to try the older test build, they should work except for the ratings
22:15 anyways... onto the website? what happened to piwik freaktechnik?
22:15 freaktechnik I updated it
22:15 and it broke
22:15 to put it short
22:15 we now have a huge stack trace showing everytime you got to stats.getnightingale.com
22:15 rsjtdrjgfuzkfg yay?
22:16 freaktechnik already asked in the piwik channel and posted in the appropriate forum thread.
22:16 (http://forum.piwik.org/read.php?2,119615)
22:16 seems we're not alone
22:16 rsjtdrjgfuzkfg downgrade?
22:16 freaktechnik I swear, I always update my personal piwik first, which always works fine
22:16 then I update ngale's and it breeaks
22:16 EVERY. DAMN. TIME.
22:16 johnmurrayvi that's good at least. hopefully they'll push a new update soon
22:16 hahaha
22:17 freaktechnik rsjtdrjgfuzkfg: might look into that
22:17 johnmurrayvi so, in regards to the new website, more info on the donation page?
22:18 freaktechnik well, I added content to it, I think I already showed you, johnmurrayvi
22:18 http://lab.humanoids.be/ngale/[…]le.com/donate.php
22:18 I also discovered a bug in the bountysource API...
22:18 (which was immediately fixed)
22:18 johnmurrayvi really? what was it?
22:19 freaktechnik they forgot a header, which basically broke it in our case.
22:19 johnmurrayvi huh...
22:19 freaktechnik (as it's client-side js)
22:19 johnmurrayvi two quick things: https://dl.dropboxusercontent.[…]018%3A19%3A04.png
22:19 freaktechnik yeah, chrome being weird
22:20 johnmurrayvi yeah...
22:20 freaktechnik also, the footer isn't great currently
22:20 (well it is too great, in other words)
22:20 johnmurrayvi second thing: the xr9 bounty has 170.45? I didn't know they did $0.xx
22:21 freaktechnik hm, that's interesting actually, as it says 170 on their page
22:21 lemme check the API response
22:21 maybe due to conversion...=
22:21 rsjtdrjgfuzkfg is bountysource considering feedback? Their page is displaying wrong stuff with js turned off / not completely on...
22:21 freaktechnik (of currencies)
22:21 rsjtdrjgfuzkfg: yes, their source is on gh, just file a bug
22:22 (or look for the matching one)
22:22 johnmurrayvi https://github.com/bountysource/frontend/issues
22:22 freaktechnik yeah, the API gives back 170.45
22:22 johnmurrayvi huh, that's interesting
22:22 rsjtdrjgfuzkfg https://github.com/bountysourc[…]ontend/issues/447
22:23 known :P
22:23 freaktechnik https://www.bountysource.com/i[…]utm_source=github <- press on the big thumbs up button ;)
22:23 oh wait, that won't works :P
22:23 *work
22:24 johnmurrayvi lol
22:24 so, last topic: new issues
22:24 freaktechnik so yeah, if you have any ideas for additional text or content just tell me.
22:24 johnmurrayvi sounds good, will do!
22:24 from https://github.com/nightingale[…]ing/pulse/monthly
22:24 two new issues
22:24 freaktechnik (both opened by me, yay)
22:25 johnmurrayvi haha
22:25 rsjtdrjgfuzkfg hm. If that is not ot now, I'd prefer the bounty thing not to be two lines in three columns
22:25 maybe one column put a bounty on an issue, and two columns donate to nightingale or something?
22:26 freaktechnik main reason I'm having put a bounty on an issue on full width atm is because of the bounty list
22:27 rsjtdrjgfuzkfg list?
22:27 freaktechnik well, if you had js enabled...
22:27 (don't do the bs  API request client side)
22:27 I meant it's not done server side
22:27 johnmurrayvi you can see it in the screenshot I linked
22:28 rsjtdrjgfuzkfg aah
22:28 got it
22:28 freaktechnik also, we have the same situation of little text spread out over three columns on dowload.php and nightlies.php, though I think I can compress that content to fit into a 2+1 layout
22:29 rsjtdrjgfuzkfg you are designer, not me ;)
22:29 *the
22:30 freaktechnik but yeah, I agree that it's ugly.
22:30 so, issues?
22:31 johnmurrayvi 321 is reading aiff metadata
22:32 freaktechnik https://github.com/nightingale[…]acking/issues/321 is a reaction to https://twitter.com/getnightin[…]06069412018139138, I do not know if AIFF actually has any metadata fiels other than stuff like bitrate
22:32 johnmurrayvi seems reasonable. I *may* be able to take a look at it while looking at the other taglib stuff
22:32 freaktechnik but I know that taglib has a handler for ti
22:32 but our metadata handelr doesn't vote to know how to read aiff files
22:32 johnmurrayvi hm
22:33 freaktechnik but first we'd have to investigate AIFF's tags
22:33 as "apparently there isn't a standard for AIFF tagging"
22:33 rsjtdrjgfuzkfg -.-
22:33 johnmurrayvi sweet....
22:33 lol
22:33 freaktechnik but there is a taglib class for it
22:33 so I figured why not
22:33 didn't look at what the class reads though
22:33 johnmurrayvi yeah I had seen that
22:34 freaktechnik because as I said, it could be like bitrate and duration or similar
22:34 johnmurrayvi I'll try to peek at it
22:34 freaktechnik cool
22:34 #322 is a curious one
22:34 ngissuesbot Issue #322: EQ Sliders don't load their band's value [03open] 14-- (Milestone: 1.13) https://github.com/nightingale[…]acking/issues/322
22:34 johnmurrayvi argh, I didn't put #321 hah
22:34 ngissuesbot Issue #321: Read AIFF Metadata [03open] https://github.com/nightingale[…]acking/issues/321
22:34 freaktechnik something about sbdataremotes not syncing with prefs in all cases, from what I know so far.
22:35 though it works for me, so yeah, it's a weird one.
22:35 johnmurrayvi odd
22:35 freaktechnik could be an XULR visual glitch, which would suck
22:36 johnmurrayvi I need to do a rebuild + update on the version I'm currently using for my main profile, so I'll check if it happens for me
22:38 hopefully we'll have GeekShadow (and maybe ilikenwf) for next meeting
22:38 it'll be 4 Oct 2014
22:38 freaktechnik not really anything to write about in the blogpost, so I guess we'll skip it?
22:38 johnmurrayvi yeah hah
22:39 maybe we should say we're going to do them bi-monthly?
22:39 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 || Dev-Meeting: 4.October
22:39 freaktechnik I think we said we'd do them whenever there was something to write about
22:39 johnmurrayvi makes sense
22:41 GeekShadow can be the leader next time :) lol
22:41 freaktechnik he'll have to add an appointment to his calendar though
22:41 johnmurrayvi yeah haha
22:42 I'm gonna go hop in the shower and get some food. I feel gross because I sat in the sun all day at a football (american football) game
22:43 so I'll see you all later!
22:43 freaktechnik see you
22:43 rsjtdrjgfuzkfg (too late) cya
22:43 :P
22:43 freaktechnik there's still SNAILBot
22:43 rsjtdrjgfuzkfg mhm
23:14 ilikenwf joined #nightingale
23:51 ezekiel joined #nightingale

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