IRC Logs for #crux-devel Friday, 2008-05-16

*** mwansa has quit IRC00:44
*** namenlos has joined #crux-devel01:00
*** predatorfreak has joined #crux-devel01:00
*** mike_k has joined #crux-devel01:15
*** mwansa has joined #crux-devel01:17
cptntilman: Qt4 already depended on mesa3d01:17
cptnI assume it could be removed when keeping the fallback to the nvidia headers01:18
cptnbut that's for the future01:18
Romsterhmm guess now is the time to nuke my hard work.02:33
Romsteri hadn't reviewed your changes yet either.02:33
Romsterand i'm also wondering i bet i'll be forced to give up some of my other ports later too. i heard mentaion of cmake and mpfr is required on later gcc versons too.02:34
predatorfreakRomster: Keep yours in your personal repo.02:34
predatorfreakLord knows my personal repo differs from the core/opt stuff at times <<02:34
cptnRomster: you know, others are usually happy that they don't have to maintain ports anymore02:35
Romsteryeah i got a copy of it.02:35
Romsteri arn't like them, i adore keeping them working. :P02:36
Romsterwould you like someone to take away your hard work?02:36
cptnsure02:37
predatorfreakRomster: Not really :D02:37
cptnI mean, both you and sip used my qt4 as base, right?02:37
predatorfreakWorkaholic gene runs in my family02:37
cptnI was happy that someone picked it up02:37
Romsterand the other case of soemone pinching a port and removing the packger line (known yhafri, and i've seen jue do that too)02:37
Romsteri used qt3 as base.02:37
cptnmmmh, are you _sure_?02:37
Romsterand what packager line did i have...02:37
Romsteryes i used qt3 from contrib.02:38
cptnthen why does your qt4 port have the exact same README as my t4 port?02:38
Romsterah i think i looked at some personal repo of qt4 aswell for some alterations but then i took it form there and looked though the source and altered it more on my own.02:38
cptnand why am I the packager in there?02:38
cptnI've never packaged qt302:38
Romsterhmm02:38
cptnyou may want to think it over...02:39
Romsteri must of copied.... ah i know i must of copied some personal qt4 port that wasn't in opt.02:39
cptnyeah02:39
Romsterit was ages ago and wasn't important i keept the packager line.02:39
Romsteri gave credit where it was due, i do not remove the packager line like i've seen yhafri and jue yes jue i have evidence02:40
Romsteri'm sure there are others that are guilty but i have't found out.02:40
cptnI've adapted the packager line before too02:40
*** sepen_ has joined #crux-devel02:40
Romsterthat's bad..02:40
cptnthat's common practice if you do enough change02:40
*** sepen_ has quit IRC02:40
cptnno it's not02:40
Romsterhmmz i've never changed the packager line ever unless it's my own work from scratch.02:41
cptnI have to admit, with the changes you did for qt4 I would have prefered if you had claimed packaging for that port02:41
cptnI think crediting is important02:41
Romsterok where do you draw the line and change the packager line...02:41
cptnI did that for opt/qt4 in the commit message02:42
predatorfreakcptn: Romster is a humble person.02:42
predatorfreakMost of the time02:42
Romster:) i like to be nice where i can but sometimes i'm a total arsehole.02:42
predatorfreakNow back to updating hal/hal-info02:42
cptnI change it whenever I do major changes I don't expect the original packager would want to support02:43
*** sepen_ has joined #crux-devel02:43
cptnor not necessarily be able to support02:43
predatorfreakOh god damn it.02:44
predatorfreakNot another fucking PAM-like thing.02:44
predatorfreakGO AWAY POLICYKIT02:44
predatorfreakHAL needs it for full functionality though <<02:45
Romsterugh...02:45
Romsterpam is invading.02:45
Romsterpam is full of holes.02:45
predatorfreakPolicyKit doesn't need PAM02:48
predatorfreakBut it appears "useful"02:48
predatorfreakHowever, hal supposedly can operate fine without it.02:48
predatorfreakSo I'm going to disable it.02:48
predatorfreakand if anyone notices regressions.02:48
predatorfreakI'll package it.02:48
Romsterk02:48
predatorfreakGentoo force disable it.02:49
predatorfreakSo we should be fine02:49
Romsteryeah i'd try without.02:49
Romstersounds good.02:49
Romsterthose other distros are great for seeing if they patched a port02:49
Romsterremove X11R6 from QMAKE_{INC,LIB}DIR_X11 (inspired by contrib/qt4)02:54
Romsternice one cptn yeah it's my goal to eradicate the old path in favour of the newer xorg paths.02:55
cptnI guessed that that was your goal02:55
Romsterany of my ports i see that i fix that up.02:55
cptnmaybe you want to double check those02:55
cptnsince there's no /usr/X11 directory...02:56
teKI had to do exactly the same for a binary (ati driver?) vim -b rocks *g*02:56
Romsterdid i mess up..02:56
cptnyeah02:56
cptn 28 -QMAKE_INCDIR_X11       = /usr/X11R6/include02:56
cptn29 -QMAKE_LIBDIR_X11       = /usr/X11R6/lib02:56
cptn30 +QMAKE_INCDIR_X11       = /usr/X11/include02:56
cptn31 +QMAKE_LIBDIR_X11       = /usr/X11/lib02:56
Romster/usr/lib/X1102:56
Romsterah god damn it.02:56
Romstergrr02:57
Romsternow i feel bad.02:57
cptnwell, it didn't break anything in the first place02:57
Romsterand i try to do my best job and i do something so stupid like that.02:57
sepen_$ pkg-config --libs-only-L QtGui02:58
sepen_-L/usr/X11R6/lib02:58
Romsteri'm a perfectionist and i let that slip.02:58
sepen_is that right?02:58
Romstersepen_, yeah that path patch is meant to correct that.02:58
sepen_hmmm02:58
sepen_but which is the consequence?02:58
Romster-L/usr/lib/X1102:59
cptnsepen_: that could go away02:59
Romsterseems to not matter but I prefer to fix it.02:59
cptnsepen_: should be /usr/lib02:59
sepen_yeah, Im according to this02:59
Romstersince the old path doesn't exist anymore and that's just asking for Murphy to bite ya in the arse one day unsuspecting02:59
cptnI can add that, although it's all just cosmetical02:59
sepen_cptn, when you wrote me a mail around my beta qt4 port I noted this problem but never fixed that port for me03:00
Romsterit's cosmetic but it be nice to eventually eradicate the old paths.03:00
cptnsepen_: I'll look into it tonight03:01
Romsteri apparantly did a half arsed job i looked it up but i must of got to the /usr/X11/... that was in the existing code.03:01
sepen_cptn, hmmm03:01
sepen_cptn, but never was ok my port03:01
cptnsepen_: i think the one in opt is pretty clean now03:01
cptnsepen_: removed lots of stuff which was inherited from qt303:02
Romsterit's now /usr/include/X11/ and /usr/lib/X11/03:02
cptnand not needed anymore03:02
sepen_hmm yeah, its based on your port, right?03:02
Romsterso any of them X11R6 paths should be fixed.03:02
cptnsepen_: yep03:02
sepen_I need a cofee03:02
* sepen_ coffiiing03:03
predatorfreakDamn, hashdeep from md5deep would be VERY useful for pkgmk.03:03
predatorfreakIt'd make it possible to do md5, sha256, whirlpool whatever all in one pass03:04
predatorfreakwith one program03:04
cptnRomster: the libraries are in /usr/lib only03:04
Romsternow i got another 4 hour compile todo :P03:04
cptnwell, you can keep your port03:04
cptnit's the same version after all03:04
Romster/usr/lib/X1103:04
Romsterthe rest of the libs are found in /usr/lib as that is a standard path03:05
Romstertrue.03:05
Romsterjust looking now what you changed.03:05
cptnls -l /usr/lib/X11/libX*|wc -l03:05
cptnls: cannot access /usr/lib/X11/libX*: No such file or directory03:05
cptnls -l /usr/lib//libX*|wc -l03:05
cptn16803:05
predatorfreakSigh, I swear, one of these days I'll probably wind up rewriting pkgmk to use better tools <<03:06
predatorfreakI can only imagine how spectacularly that will fail to catch on.03:06
Romstercptn, damn it can you change it to bz203:06
Romsteryou missed that :P03:07
cptnah, yeah, will do03:07
Romster03ee0deb73908c84a7b4d9ceb5f8b3f6  qt-x11-opensource-src-4.4.0.tar.bz203:07
Romsterthanks :)03:08
Romsterin fact there should be a audit on all ports for if there is a bz2 archive for those still on gz03:08
Romsteri'd prefer lzma but that's a can of worms still03:09
cptnit is?03:11
predatorfreakcptn: tilman won't merge lzma until libarchive supports lzma.03:12
predatorfreakWhich at this rate could be a century03:12
cptnah03:13
Romsterwell last time i started putting lzma in my ports i  got the biggest bitching ever and gentoo even uses lzma now and yeah wating for it to go in either libarchive or the liblzma is finished it's rewrite.03:13
predatorfreakRomster: Portage supports lzma through hacky methods too.03:13
Romsterbut my alternitive time being method got shot down in flames.03:13
predatorfreakThey decompress using lzma and output to a pipe.03:13
predatorfreakThen untar the pipe.03:14
Romstermy method wasn't that bad it's like how you would uncompress a rar.03:14
predatorfreakRomster: I've peaked inside emerge/portage once or twice <<03:14
Romster:P03:14
predatorfreakOn Gentoo I still massively prefer paludis over portage though03:14
Romsterdepending on a 500kb lzma is bad apparently.03:14
predatorfreakInfinitely nicer configs and faster operation.03:14
Romsternever looked at paludis03:15
predatorfreakIt's basically a group up portage replacement by ciaranm and company.03:15
predatorfreakciaranm is still King Of The Gentoo Trolls though03:15
Romsterhmm should look into paludis for blackhole to gentoo03:15
teKimho introducing lzma in favour of tar.gz oder .bz2 is a really _minor_ improvment03:15
Romsterpredatorfreak, and i are working on a firewall script.03:16
cptnRomster: I think only the method of requiring lzma was shot down03:16
cptnnot lzma03:16
Romstershaves off MB's03:16
predatorfreakRomster: Paludis is portage basically.03:16
predatorfreakOnly better.03:16
predatorfreakWrite an ebuild and it'll work with both.03:16
Romsterpredatorfreak, cool i've bookmarked it.03:16
Romstersweet.03:16
predatorfreakI use paludis on my server because portage is slow as fuck on there.03:16
predatorfreakand paludis is MUCH faster.03:17
predatorfreakNot to mention it has nicer features and smarter dependency handling03:17
Romstergee you'd think they would swap over tools.03:17
Romstercptn, ah my hackish method over the libarchive transparently way.03:18
Romsteri have to admit my hackish method was solid not some half baked thing.03:18
cptnsure03:18
predatorfreakRomster: Gentoo developers prefer to keep working on portage.03:18
cptnbut IIRC you provided the lzma yourself03:18
predatorfreakPaludis is a total reimplementation in C++.03:18
Romsterand it was to tie between the time of now and when libarchive.03:18
cptnso you could have provided a bz2 too, without requiring any extra tools03:18
predatorfreakBut IMO, Paludis utterly decimates Portage on so many levels.03:19
Romsterand i had a patch to add to pkgmk to allow extraction that tilman rejected.03:19
cptnanyway, I think lzma will become a part of pkgmk at some point in time03:19
cptngiven that gnu.org replaces bz2 with it03:19
Romsterpredatorfreak, gentoo should really think to swap to Paludis  than there Portage.03:19
Romsteryeah there are slowly.03:20
cptnRomster: well, would you switch to prt-get2 if it was written by yhafri?03:20
predatorfreakRomster: Ciaranm is a community troll, even though him and his cohorts code well.03:20
cptneven if it was nicer03:20
Romsterid' like to see xdelta in pkgmk too i'm still working on a patch for that.03:20
predatorfreakFat chance.03:20
Romstercptn, eww maybe not.03:20
predatorfreakThey've formed their little community bases03:20
Romsteri don't trust himwith his hacks.03:20
cptnRomster: that's the situation with paludis03:20
Romsteri can trust predatorfreak and my hacks03:21
cptnit's definitely not a technical argument03:21
predatorfreakEven though I agree that paludis is better than portage.03:21
*** mwansa has quit IRC03:21
predatorfreakcptn: Well, if yhafri coded it well.03:21
predatorfreakand it wasn't a reflection of his Pkgfile's.03:21
predatorfreakYes.03:21
cptnheh03:21
Romsteri'd have to look over the tool before i dared to use it.03:21
predatorfreakcptn: But his Pkgfile's are terrible.03:21
predatorfreakThe difference between Paludis and Portage is both are coded well.03:21
predatorfreakIt's just that the head of the Paludis camp is a giant fucking asshole :D03:22
*** mwansa has joined #crux-devel03:22
cptnokay, I didn't really know any equivalent to ciaranm in our community :-)03:22
Romsteri think some of my Pkgfiles need improving more too. but yhafri's beats the lot.03:22
predatorfreakcptn: Han? :D03:22
predatorfreakBefore he ran off.03:22
Romsterwhat does yhafri use them all on anyways.03:22
Romsteri miss Han he was always entertaining.03:22
cptnpredatorfreak: I think his pkgutils were used by some :-)03:22
predatorfreakcptn: I modified them at one point.03:23
RomsterI picked up some of hans packages when he left.03:23
predatorfreakHacked on some of my own modifications like SHA256 and stuff.03:23
*** mwansa has quit IRC03:23
tilmanstupid lzma wining03:23
predatorfreakThen I dumped CRUX for Arch for awhile.03:23
tilmanffs03:23
predatorfreaktilman: LZMA LZMA LZMA LZMA LZMA LZMA03:23
Romstercptn has seen my pkgmk and prt-get hacks i think.03:23
predatorfreaktilman: :D03:23
Romsterrofl.03:23
Romsterwhy is it stupid?03:24
cptnwhining usually is03:24
Romsterhe said wining not whining.03:24
cptnwhat would that mean?03:24
Romstera lot of difference.03:25
*** mwansa has joined #crux-devel03:25
cptn"to drink wine"03:26
cptnaccording to merriam-webster03:26
cptnI'm confident he meant whining03:26
Romstertrue but he typed it wrong and tilman has the decency to say my english is bad :P03:27
Romsterany way moot point.03:27
Romsteris there ever gonna be a email for contrib over viewer? I'm wating to apply for the position :P03:28
Romsterwaiting*03:28
cptnah, yeah, didn't get around to send it03:28
cptnbut if you like the idea, you can propose it yourself03:29
cptnboth tilman and me aren't part of contrib after all :-)03:29
cptnotherwise, I'll send one later today03:29
Romsterah ok i'll propose it.03:29
Romsteror would it be best i wait for you to send one?03:29
teKoh noz Romster becoming my boss03:29
Romsteri don't like stepping on toes.03:30
RomsterteK, rofl.. not much would change teK.03:30
Romsteri'd be making sure contrib ports are well maintained though.03:30
cptnmmmh, okay03:30
cptnthat wasn't really our intention03:31
Romsterand to report to the core team too.03:31
cptnI think you should do that already, i.e. everyone in contrib :-)03:31
Romsterthat's the main intention.03:31
Romstercptn, i do now.03:31
Romsteri email and use FS more now too.03:31
teKRomster: I'm perfectly fine with this (additionally to the current situation)03:32
cptnit's more about the administrative stuff03:32
cptnI'll fire a mail right awy03:32
Romsterk03:32
cptn*away03:32
Romsteri'd like to know what tasks.03:32
Romsterbut i got a general idea.03:32
Romstercptn, shouldn't you nuke qt3 from opt now too?04:03
Romsteror are we having a grace period.04:03
cptnI have to admit I've not tried to build any of my qt3 apps against qt404:03
Romsteralso i thought you'd change the path in qt4 to -bindir /usr/bin04:04
Romsterqt4 has qt3 compatibility so it should work.04:05
cptnwell, source level compatibility only, right?04:05
Romsterany qt3 program i believe should work with qt4 that's the idea of qt3 compatibility.04:07
cptnany qt3 app can be compiled against qt4 maybe04:07
Romsteri haven't tested but i think i'll change your path. to /usr/bin and try removing qt304:07
cptnbut it has to be recompiled04:07
Romsteri think it should.04:07
Romsteryeah.04:07
Romsterno biggie.04:07
cptnpersonally I'd keep the too as long as there are plenty of qt3 apps04:09
cptnno need to install the whole qt4 port if you have a working qt3 around04:09
*** predatorfreak has quit IRC04:11
Romsteri thought you wre gonna change the qt4 paths to match the ones in qt3 and nuke qt3 off opt/04:12
Romsterhttp://pastebin.ca/101988104:12
Romsterwere*04:12
cptnI know you thought that04:12
cptnyou mentioned it a couple of times already04:12
Romsterproviding qt4's qt3-compatibility is working ok04:12
cptnI've never thought about it myself yet04:13
Romsterguess i'll test that myself.04:13
Romsteri'll use your qt4 port and hack the paths to the ones in qt304:13
Romsterand remove qt3 install qt4 and see if all my qt3 stuff works ok.04:13
cptni'd guess that if it really works, you wouldn't have to adjust the pathes04:14
cptnyou can just remove qt3, right?04:14
*** mwansa has quit IRC04:15
*** teK has quit IRC04:15
Romsterhmm i'd prefer to adjust the paths, since the ones currently in qt4 are hacked so it will not conflict with the qt3 name psace.04:15
cptnthey're the same as in Qt304:16
cptnbefore qt4 existed04:16
cptnthat's how Qt3 lived in a file system04:16
cptnyou had a QTDIR with include and lib04:16
cptnso no, it's not hacked04:16
Romstereh.. not what i just looked at in my chroot after a ports -u as seen in that paste.04:16
Romsterqt4 used qt3 paths with workarounds so it wouldn't name clash04:17
cptnokay04:17
cptnyou obviously know it better04:17
Romsterthe major one being -prefix /usr/share/qt4 and -bindir /usr/bin04:17
Romsterqt to qt4 isn't much to worry about though.04:18
*** mwansa has joined #crux-devel04:18
*** teK has joined #crux-devel04:18
cptnyou're right, I guess we could fix the qt3 port to be more consistent04:19
Romsterthe hacked bit was move prefix and symlink binarys to /usr/bin with foo-qt404:19
Romsteri was thinking as long as qt4's qt3 compatability worked was to fix prefix remove that symlink for the /usr/bin and test without qt3 on the system.04:20
Romsterand have qt3 only in kde tree which it is there too.04:20
cptnso that kde users couldn't install opt/qt4?04:20
Romsteralthough that would cause a issue for kde users that wanted kde404:20
cptnyep04:20
Romsterwell the kde repo could adapt the opt port as it is now as a override untill kde4 takes over.04:21
Romsteri see that as a viable option.04:21
Romsterthen when kde moves to kde4 later kde can use opt qt404:22
cptnor we could keep qt3 and qt404:22
cptnand be done already now04:22
cptnonce there are no qt3 apps anymore, qt3 just dies04:22
Romstereh if you like the current way.04:22
Romsterit was my understanding you were gonna remove qt3 out of opt.04:22
Romstersince it's in the kde3 repo.04:23
cptnthat was sip's plan04:23
Romsterah.04:23
cptnyou're confusing us04:23
Romsteri must of got confused..04:23
Romsterdamn it.04:23
cptnthat said, I see your point04:23
cptnI'd rather run qmake instead of qmake-qt404:23
Romsteryeah04:24
Romsterthat's what i was expecting to be done.04:24
Romstersorry to be a pain.04:24
Romsteronly part i haven't tested is qt4's compatability.04:24
Romsterso i should come up with some facts if it works or not.04:24
Romsterso i'll test and report back on that.04:25
cptnokay, great04:25
cptnI'll try to run some tests as well04:25
Romsterjust to be sure,04:25
cptnit would be good to learn from Alan what his plans are for kde04:25
Romsternot sure about the opengl stuff and why it had nvidia and never ati04:25
Romsteri think he watns to stick to kde3 for longer but i'd like to help him out with kde404:26
Romsterall alan got done was http://www.mizrahi.com.ve/crux/kde4/ and this is my bordom efforts that doen't work right yet http://romster.dyndns.org:8080/linux/ports/crux/romster-kde/04:27
Romsteri'm hoping me and alan can collaborate on our efforts.04:28
Romstermaybe i've spent too much time studying ports :P04:28
Romsterbut not enough to realise i messed up them paths for xorg.04:29
Romsterbut as you said it was more of a cosmetic reason for qt4.pc04:29
cptntook me some time yesterday to clean it up too04:32
cptnthere were quite some workarounds from qt3 which were not needed anymore04:32
Romsteryeah it's quite a beast of a port now you know why i wasn't happy when it was taken off my hands all the effort i put into it.04:32
cptnwell, no :-)04:33
Romsterah i shouldn't of studied it more... feels like i didn't do a good enough job.04:33
cptnas I mentioned, I'm usually happy if others do the work :-)04:33
Romstershould of*04:33
Romster:)04:33
Romsterhmm should the include/qt4 and lib/qt4 be kept as qt4 or just qt again.04:34
Romsteractually i think qt again no need for qt3 if this all works04:34
cptndepends on whether alan has to keep a qt3 legacy port I guess04:35
cptnin that case, it would be nice if he moved it to /usr/share/qt304:36
cptnand opt/qt4 would be /usr/share/qt404:36
cptnyou know what the python guys say, explicit is better than implicit :-)04:37
Romsteri'd sugest if anyone in kde land wants to use qt4 to adopt qt4 as it is now before it gets changed.04:37
Romsterhehe04:37
Romstercptn, http://rafb.net/p/6WVzGx97.html04:39
Romsterwhat i changed and believe is correct.04:39
Romsteri'll start compiling i shoudln't be that long i got ccache from the last compile.04:40
cptnwould be nice to get rid of the symlink for loops04:40
Romsterlike i just did.04:41
cptnyeah, I was refering to the paste :-)04:42
Romsterthat was the hack i was talking about to get around qt3 on system that we shouldn't need anymore.04:42
cptnrunning qmake-qt4 on a qt3 project didn't do the right thing here04:42
Romsterok compile started at 19:4204:42
Romsterlets see if i can beat 4 hours since it's cached in ccache.04:43
*** namenlos has quit IRC04:43
Romstergood job on the clean up even with the work i did it would of been a huge task.04:43
cptnmmmh04:43
cptnhow well have you investigated the compatibility?04:44
Romstermaybe i should of been more aggressive with removing stuff like you did.04:44
Romsteronce it's built i'll remove qt3 and solely use qt4 and try out qt3 programs04:44
cptnwell you know04:45
*** sepen_ has quit IRC04:45
cptnthe include file names changed04:45
cptnso qscrollview.h -> q3scrollview.h04:45
*** sepen_ has joined #crux-devel04:45
cptnsure, you can patch all the Qt3 apps outthere04:45
cptnbut I'm not sure if this is worth the effort04:46
Romsterthe point was the qt3 compatibility was meant to make qt3 apps work with qt4 as is.04:46
Romsterwithout any changes.04:46
Romsterthat's what i'm going to confirm if it does indeed work as is.04:47
cptnwell, my 500 LOC app doesn't build04:47
Romsteri got opera and skype and eh i can find more qt apps to try.04:47
Romsterhmm04:47
cptnI'd be surprised to see any real app work04:47
Romstermoved headers maybe04:47
cptnopera is closed source anyway04:47
cptnso you need the proper libraries04:47
Romsteryeah but should work as is.04:47
cptnwith qt3 libraries04:48
Romsteri got it as dynamic along with skype. and i can test building kde4 agenst it.04:48
cptnnot qt404:48
Romsteryes but qt3 compatibility is there to allow all qt3 programs to work with qt4 until they update and the qt3 compatibility mode becomes depreciated.04:49
Romsteror am i missing a point here.04:49
cptnyes04:49
cptnto be rebuild with very little change04:49
cptn*rebuilt04:49
cptnnot "to work as is"04:49
Romsterhmm i need to dig the docs then...04:49
Romsterso all the stuff i discussed is worthless then04:50
Romsterand we still need qt3 when qt4 provides qt3 compatibility.04:50
Romsterlogic there does not add up.04:50
cptnsay you have a large code base04:51
cptnwith lots of qt3 code04:51
cptnand you want to move to qt4 for the new features04:51
cptnthe developer can use the qt3support module04:51
cptnto move to qt4 with few manual changes04:51
cptnthere's a tool to do some of the work for you04:52
cptni.e. it renames classes from QFileDialog to Q3FileDialog04:52
Romsterso this implies 'few manual changes' so it depends if the developers of the qt3 apps have used the qt3support module yet.04:52
cptnyes04:53
Romsterdamn it, i thought it was a translation layer that worked transparntly. :/04:53
cptnbut for closed source apps, like opera, that wouldn't work anyway04:53
cptnsince they're dynamically linked against libqt.so.304:54
cptnwhich isn't provided in qt404:54
Romsterhmm well we could still provide the libqt.so.3 for compatability without the headers.04:54
Romsterlike libstdc++-compat does04:55
cptnyeah, or use the statically linked opera04:55
Romsteror that option04:55
cptnhowever qt4 is not source compatible with Qt304:56
cptnthus qt3 apps have to be ported04:56
cptnwith a shortcut when using the compatibility layer04:56
Romsterwell that clears up my confusion i had the wrong idea.04:57
Romstermy theory was good until you busted it with the facts <<04:57
Romsterhope i haven't wasted your time cptn.04:58
Romsterdunno if you learned anything from me about it all. :)04:59
Romsteri think it's mostly you show me my flaws.04:59
Romsters/show/showing04:59
Romsterdamn distcc...05:00
Romsteryou need a work around for that.05:00
Romsterdistcc[11231] Warning: failed to distribute /var/cache/ccache/unixmake2.tmp.arkanoid.11211.ii to fire, running locally instead05:00
Romsteractually slows the compile down waiting..05:00
cptnheh, no, it's good to really understand the problem05:00
Romsterpre-install to disable distcc i think.05:00
Romstertrue.05:01
cptnhave you investigated why it fails?05:01
Romsteri jsut hope no one sees me as a don't know what his on about type of person.05:01
Romsterhmm i did ages ago gave up i think it's it's bootstraping it uses qmake and qmake is not on all my systems05:02
Romsterso it fails05:02
cptnah05:02
Romsterunless you seperate qmake from qt405:02
cptncare to file a bug?05:02
Romsterlike i did with boost-jam and boost.05:02
Romstersure.05:02
cptnthe trolls have a compile farm tool on their own05:02
Romsterthat's why i did that hack in my port.05:03
Romstero.05:03
cptnI'd expect that Qt4 would distribute well05:03
Romstersweet.05:03
Romsterit would speed up that 4 hours since i can distribute to 3 machines.05:03
Romsteryou wanna do the work or me, no idea if you got distcc setup there to use,05:04
Romsteri tend to hit bugs before others because of distcc.05:04
cptncheck again that packager line in distcc :-)05:04
Romsterhttp://crux.nu/bugs/index.php?do=details&task_id=27505:10
Romsterah yes you made it originally :P05:10
cptnthanks05:11
Romsterbut then you left someone else took it then the later dropped it i think that was tilman then i picked it up as i was using it.05:11
Romsteractually that bit of error might be due to time delay taking too long on the timeout to return the compiled object not too sure.05:12
Romsterand i closed the ticked assigned to me for qt4 in contrib.05:12
Romsterfact it didn't work and wished it worked with distcc. but i also know it wont ever work unless i put qmake on every host.05:13
Romsterand it needs the post-install masquerade directory like i did with the other compilers.05:13
Romsterhmm so how do ya only make qmake without the rest of qt4.05:14
cptnwell, qmake is not a compiler05:14
cptnso it shouldn't be distributed05:14
cptnbut I'll look into that05:14
cptnalso, it doesn't need a post-install05:14
cptnwe can leave some setup bits to our users05:15
Romsteractually i tried bjam ages ago with distcc only to find out it failed... h right it's a make tool not a compiler d'oh.05:15
Romsterto do the distribute boost back when i didn't know so much.05:16
cptnactually qmake is like automake05:16
Romsterfor now i'll just comment that line out of my pkgmk.conf to build qt405:16
cptni.e. an Makefile generator05:16
Romsteroh05:16
cptnwe run make afterwards05:16
Romsterah much like scons05:17
Romstermake the Makefile.05:17
cptnisn't scons a remplacement for make05:17
Romsteryes05:17
Romsterso is cmake05:17
Romsterso many popping up it arn't funny.05:17
cptnyep :-)05:17
cptnanyway, time for lunch05:17
Romsterk05:18
cptnI'll try to find some time this weekend05:18
cptnsounds like a fun issue05:18
Romsteri'll test some stuff anyways and confirm it.05:18
Romsterah friday night :P05:18
Romsteryeah weekend fun time.05:18
Romsteri hate to give you more todo :P05:19
Romsterthorough and perfection sometimes gets in my way. and the need to improve and learn.05:20
Romsteri had some hot dogs earlier but i think i'll get something else to eat now too.i didn't fancy anything big.05:22
*** sepen_ has quit IRC05:30
*** sepen_ has joined #crux-devel05:30
*** mike_ has joined #crux-devel05:57
*** mike_k has quit IRC05:58
*** sepen_ has quit IRC06:25
*** sepen_ has joined #crux-devel06:26
tilmancptn: the contrib people should probably elect their overview person themselves06:35
pitilloI answered to the mail proposing romster and sepen, but sepen was right about english level needed06:37
pitillo(I can't see here romster's mail, I can see only at webpage :?)06:37
tilmansorry, i didn't even see the mails yet06:38
tilmandon't have imap access here06:38
pitillotilman, take a look to the web if you can06:38
pitillotilman, http://lists.crux.nu/pipermail/crux-contrib/2008-May/thread.html06:38
tilmanyeah, i'm readin the archive now06:40
tilmanhaha06:41
tilmani'd say that core/opt maintainers are unsuitable for that job06:42
Romsterhmm predatorfreak is in contrib too.. didn't think of that he is in opt as well07:03
pitillotilman, do you mean predatorfreak?07:18
tilmanyes07:19
pitilloif he has one foot in opt and in contrib too, I can see him like a good candidate too07:19
pitillobetter than sepen (he told the reson in his mail about english skill)07:19
pitillo*reason07:20
tilmanfunny that you didn't think of mike_k yet :)07:23
Romsterhmm07:24
Romsterhe would also be a good candidate.07:24
aonagreed07:24
pitilloright07:24
pitillobit by bit are shown good candidates07:25
tilmanpitillo: having an opt member do it might feel a bit weird :)07:25
Romsteri know my english is not the best but i got experience. that's the only reason why i replied.07:25
aoni don't think english skills are the key point here anyway07:26
Romsterit's people skills i assume and who's helpful.07:26
Romsterand friendly.07:26
pitillotilman, well, he is maintaining contrib ports too (1 foot in opt and the other in contrib)07:26
Romsteri don't see that as a problem personally.07:27
pitilloaon, for someone who must be a by-pass between contrib-coremembers? From my point of view skills are important07:27
Romsterbut we have yet to see who else replies.07:27
pitilloand more knowing sepen... who doesn't understand at all some sentences (in my case is the same, or missunderstood)07:27
tilmanpitillo: i understand. IMO the job should be done by someone who's in contrib only. but well, it's not my choice ;)07:28
aonpitillo: yeah, but it doesn't have to be perfect07:29
pitillotilman, ok. Btw the choice is finally in the contrib member too.07:29
pitilloaon, not perfect, but you know us and I think our skill can be a problem to make things go fluent(?)07:30
aonyeah, i'm not commenting on any of the suggested candidates really, just in general sense07:31
pitilloaon, yeah, it's great to know your opinion too07:31
Romsterpitillo, as i see it you make effort to learn english and ports and stuff and that's what counts the most.07:33
Romsterno pain no gain.07:34
pitilloRomster, yes, that is why I am here... learning a lot (english and stuff) and for me isn't really a pain, I think the pain can be for you (to keep me here...) :P07:36
*** Viper_ has joined #crux-devel07:47
Romsterhehe pitillo07:49
Romsterhi Viper_07:49
Viper_hey romster :)07:50
*** jue has joined #crux-devel07:51
*** pedja has quit IRC08:10
*** pedja has joined #crux-devel08:12
tilmanpitillo: of course :)08:41
*** pitillo has quit IRC09:13
*** mike_ has quit IRC10:08
*** mwansa has quit IRC11:09
*** sepen_ has quit IRC11:21
RomsterViper_, can ya take a look to your email i fired one off earlier about getting my portdb entry changed.11:33
Romstersince prologic is unreliable now i've moved them.11:34
*** sepen has quit IRC12:46
*** sepen has joined #crux-devel12:47
*** cptn has quit IRC12:55
*** cptn has joined #crux-devel12:55
*** ChanServ sets mode: +o cptn12:56
*** mike_k has joined #crux-devel12:56
*** jue has quit IRC13:25
*** cptn has quit IRC13:37
*** cptn has joined #crux-devel13:40
*** ChanServ sets mode: +o cptn13:40
*** cptn has quit IRC13:43
*** cptn has joined #crux-devel13:46
*** ChanServ sets mode: +o cptn13:46
*** cptn has quit IRC14:15
*** predatorfreak has joined #crux-devel14:18
*** mike_k has quit IRC16:44
*** DarkNekros has joined #crux-devel16:59
*** Viper_ has quit IRC18:24
*** predatorfreak has quit IRC18:37
*** Romster has quit IRC20:25
*** predatorfreak has joined #crux-devel20:45
*** DarkNekros has quit IRC21:16
*** predatorfreak has quit IRC21:33
*** mwansa has joined #crux-devel21:36
*** Romster has joined #crux-devel21:54
*** mwansa has quit IRC22:54
*** mwansa has joined #crux-devel23:03
*** mwansa has quit IRC23:34

Generated by irclog2html.py 2.11.0 by Marius Gedminas - find it at mg.pov.lt!