--- Log opened Thu Jun 09 00:00:40 2005
00:03 -!- aaron42 [~adavies@12-220-223-80.client.insightBB.com] has joined #fink
00:05 -!- charles [~charles@pcp02538154pcs.crosky01.pa.comcast.net] has left #fink []
00:15 -!- Macrobyte [~macrobyte@pcp06495808pcs.nstnig01.ct.comcast.net] has joined #fink
00:15 -!- Macrobyte [~macrobyte@pcp06495808pcs.nstnig01.ct.comcast.net] has left #fink []
00:29 -!- chdevers [~cdevers@pool-68-163-153-167.bos.east.verizon.net] has joined #fink
00:32 -!- dmacks_away is now known as dmacks
00:35 < dmacks> Anyone know how to have a procmail forwarding recipe cause the email to list me as the sender instead of [whomever sent it to me]?
00:35  * pogma is not a procmail user, sorry
00:36 < pogma> I assume it can just change the From line
00:36 < aaron42> i haven't used procmail in years
00:36 < aaron42> does it differentiate between forward and redirect/bounce?
00:36 < jerwin> is there a way to set up fink so that it downloads sources from a local machine?
00:36 < dmacks> The simple rules implement it as a bounce apparently.
00:37 < aaron42> bounce is supposed to preserver the sender info
00:37 < aaron42> ie, i have no clue
00:37 < Feanor> ugh, i'm not getting gcc to do x86 linking right still
00:37 < Feanor> [asibook:~/Projects] astrange% printenv NEXT_ROOT
00:37 < Feanor> /Developer/SDKs/MacOSX10.4u.sdk
00:37 < Feanor> /usr/bin/ld: warning fat file: /usr/lib/gcc/i686-apple-darwin8/4.0.0/../../../libSystem.dylib does not contain an architecture that matches the specified -arch flag: i386 (file ignored)
00:37 < pogma> Feanor: they made NEXT_ROOT go away
00:38 < pogma> Feanor: CFLAGS -Wl,-syslibroot,/Developer/SDKs/MacOSX10.4u.sdk
00:38 < pogma> NEXT_ROOT worked until cctools-576.2 but does not work in cctools-590
00:39 < dmacks> jerwin: What do you mean by "local"? I.e., do you have some network-volume or disk on your own machine, or you have some unofficial http-accessible site?
00:39 < pogma> I'd file a bug, but it was never documented :)
00:39 < aaron42> now apt-get is being annoying
00:39 < aaron42> lots of file not founds on updates, searches, etc.
00:39 < pogma> Feanor: sorry, LDFLAGS, of course
00:40 < aaron42> i know i've solved this before, but i don't remember how
00:40 < jerwin> I have two machines.
00:40 < pogma> fink scanpackages aaron42 
00:40 < Feanor> pogma: this is just me calling gcc, so it doesn't matter
00:41 < aaron42> thx much
00:41 < pogma> gcc -Wl,-sislibroot,/Developer/crap
00:41 < pogma> syslibroot
00:41 < pogma> sigh
00:41 < jerwin> I "share the fink", but occasionally I run into difficulties.
00:42 < jerwin> it would be helpful to automatically download the ghostscript source from a local machine than over my modem
00:42 < pogma> share the /sw/src ? Or have both finks look in the other ones /sw/src as alternative download locations
00:43 < pogma> love the apple-scitech list
00:43 < dmacks> If you can mount one machine's /sw/src from the other, use FetchAltDir.
00:43 < pogma> they just talk total crap all the time
00:43 < dmacks> pogma: Yabut at least this crap isn't about how stupid fink is.
00:44 < jerwin> ooh. I should read up on the Inteel flames
00:44 < pogma> someone quoted the bencmarks of intel macs running ppc code and said HPC is dead on mac
00:44 < pogma> he's a good scientist :)
00:45 < dmacks> heh
00:45 < pogma> first, draw a conclusion, then..
00:45 < jerwin> intel has promised icc for mac, btw,
00:45 < dmacks> Maybe he works in the White House?
00:46 < slyrus> pogma: any progress on guile?
00:46 < jerwin> sounds like he has more than a law degree.
00:49 < pogma> slyrus: guile or guile16?
00:49 < slyrus> hmm... the one that was busted. fink's rebuilding at the moment. give me a minute and I'll check.
00:52 < slyrus> I think it was guile16-dev
00:53 < Feanor> icc isn't all that great
00:54 < Feanor> like a lot of intel's stuff it's just really tweaked for benchmarks
00:56 < jerwin> oh.
00:59 < slyrus> pogma: yup. guile16-dev is where I'm having problems.
00:59 < pogma> slyrus: okay, I'll fix it later, ta
01:00 < slyrus> ok thanks. looks like the (first) problem is that it should either use gcc 3.3 or some casts need to be fixed up for gcc4.
01:00 < pogma> it built with gcc4 from the first tiger xcode tools
01:02 -!- Macrobyte [~macrobyte@pcp06495808pcs.nstnig01.ct.comcast.net] has joined #fink
01:02 < slyrus> hmm... the gcc4 I'm using doesn't like: extern const scm_lt_dlsymlist lt_preloaded_symbols[];
01:02 -!- Macrobyte [~macrobyte@pcp06495808pcs.nstnig01.ct.comcast.net] has left #fink []
01:02 < slyrus> *lt_preloaded.... works though
01:03 < Feanor> yes, i think you have to be using C99 for that to work
01:03 < pogma> slyrus: it is in the binary distribution http://bindist.finkmirrors.net/bindist
01:03 < Feanor> *x and x[] are equivalent, though
01:03 < slyrus> not according to the gcc4 I'm using: array type has incomplete element type
01:04 < pogma> http://bindist.finkmirrors.net/bindist/dists/10.4-transitional/release/main/binary-darwin-powerpc/languages/guile*
01:04 < slyrus> ok, but I'm assuming that not building from source on current gcc4 is still bug-worthy
01:05 < pogma> slyrus: yeah, and i'll fix it
01:05 < pogma> but not now, I've got to go to work
01:05 < slyrus> np. it's tough keeping track of all of the new compiler versions.
01:10 < dmacks> aaron42: Looks like the easiest solution is to use formail to set a header line "From: $LOGNAME@$HOST"
01:10 < dmacks> Elegant, it's not:/
01:29 -!- jerwin [~jeremy@66-44-57-141.s395.tnt2.lnhva.md.dialup.rcn.com] has left #fink ["Leaving"]
01:29 -!- aaron42 [~adavies@12-220-223-80.client.insightBB.com] has quit [Read error: 104 (Connection reset by peer)]
01:39 -!- Knghtbrd [quacked@d197-222.uoregon.edu] has quit ["testing launchd potential bug"]
02:02 -!- shaggie [~shaggie@210.16.75.20] has joined #fink
02:02 < htodd> weird errors with autoconf
02:02 < htodd> and I have no idea how to fix autoconf
02:04 < dmacks> You Don't Fix Autoconf. You take it out and shoot it.
02:04 < htodd> aha.
02:05 < htodd> can't build libglade because of autoconf errors.
02:05 < shaggie> anyone know if there's a channel on freenode that i can get some help on bandwidth management tools?
02:07 < dmacks> htodd: Which libglade? What's the error?
02:08  * dmacks hopes it's not the new one...already spent 2+ hours fixing autoconf bugs in it.
02:15 < htodd> on panther
02:16 < htodd> lessee, libglade2-2.5.1-1?
02:16 < htodd> !lisppaste
02:16 < Melian> well, lisppaste is a bot that lets you post large chunks of text without flooding the channel. Use it at http://paste.lisp.org/new/fink
02:19 < lisppaste> htodd pasted "libglade2 autoconf errors" at http://paste.lisp.org/display/8933
02:21 < htodd> oh, sorry, I'm using rsync. that could be the old version?
02:23 < dmacks> There's only been one 2.5.1-1 so far.
02:24 < htodd> oh
02:25 -!- Omni|AFK is now known as Omnifarious
02:25 < Omnifarious> Is the boost-py24 package supposed to install boost::python?
02:28 -!- broeken [~broeken@hst32165.phys.uu.nl] has joined #fink
02:29 < Omnifarious> Oops.
02:29 < Omnifarious> bjam -sTOOLS=gcc -sPYTHON_ROOT=/sw -sPYTHON_VERSION=2.4 --without-python --without-serialization --prefix=/sw
02:29 < Omnifarious> Does anybody know why bjam is being given --without-python and --without-serialization?
02:32  * Omnifarious edits his first .info file ever and removes those two options.
02:33 < dmacks> htodd: What autoconf* and automake* packages do you have installed?
02:36 < htodd> autoconf2.5
02:36 < htodd> automake1.9
02:38 < dmacks> What happens if you edit libglade2.patch and change " AC_LIBTOOL_WIN32_DLL" to "-AC_LIBTOOL_WIN32_DLL" (the leading space becomes a hyphen) and change the 6 in the @@ line to a 5?
02:43 < htodd> patch fals
02:43 < htodd> fails
02:43 < htodd> er, my mistake
02:43 < htodd> yeah, still fails
02:44 < htodd> still my mistake
02:45 < htodd> finally patched but doesn't work
02:46 < Omnifarious> dmacks: It seems to be building with that change.  I emailed the maintainer of boost1.32
02:48 < dmacks> Omnifarious: Good idea.
02:49 < dmacks> htodd: Same error msg?
02:50 < dmacks> Do you do anything weird with your shell env vars?
02:51 < htodd> like how weird
02:53 < dmacks> Dunno:) Anything before fink in PATH?
02:54 < htodd> nope
02:55 -!- Omnifarious is now known as Omni|Sleep
02:55 < htodd> bedtme for me too
02:56 < dmacks> Nothing in /usr/local/bin ?
02:56 < dmacks> (or other /usr/local places?) Okay, I'll think some about it.
03:06 < dmacks> I just managed to reproduce that error, and committed a fix for the situation I used to do so.
03:08 -!- shaggie [~shaggie@210.16.75.20] has quit ["Quitting!"]
03:12 -!- dmacks [~dmacks@dmacks.student.supporter.pdpc] has quit ["leaving"]
03:22 -!- aaron42 [~adavies@12-220-223-80.client.insightBB.com] has joined #fink
03:39 -!- dh1pa [hlawit@aristoteles.informatik.uni-leipzig.de] has joined #fink
03:39 -!- aaron42 [~adavies@12-220-223-80.client.insightBB.com] has quit [Read error: 104 (Connection reset by peer)]
03:50 -!- bitchx [~adavies@12-220-223-80.client.insightBB.com] has joined #fink
03:51 -!- bitchx is now known as aaron-away
04:00 -!- Feanor [~astrange@feanor.developer.opendarwin] has quit [Read error: 60 (Operation timed out)]
04:23 -!- chris01 [~chris01@212.126.165.246] has joined #fink
04:23 -!- aaron-away [~adavies@12-220-223-80.client.insightBB.com] has quit [Read error: 104 (Connection reset by peer)]
04:44 -!- cianhughes [~cian@cian.ws] has joined #fink
05:46 -!- ottoaim_ [~ottoaim@h-66-167-217-224.chcgilgm.dynamic.covad.net] has joined #fink
06:13 -!- Feanor [~astrange@mp1-248-178.dialup.emory.edu] has joined #fink
06:13 -!- aaron-away [~adavies@12-220-223-80.client.insightBB.com] has joined #fink
06:58 -!- _kristy [~neptun@ti400720a080-0866.bb.online.no] has joined #fink
06:58 -!- aaron-away [~adavies@12-220-223-80.client.insightBB.com] has quit [Read error: 104 (Connection reset by peer)]
07:02 < _kristy> hm.. anyone up for a question?
07:30 -!- jack- [jack@life.hater.be] has quit [Read error: 60 (Operation timed out)]
07:32 < chris01> _kristy: just ask.
07:39 < _kristy> okey :)
07:39 < _kristy> I may have fixed the problem now, though, but I don't know
07:39 < _kristy> fink won't list the packages
07:40 < _kristy> but I'm updating the cvs now
07:40 -!- regeya [~shane@adsl-sp3-cdale176.micgi.com] has joined #fink
07:41 < chris01> _kristy: ok. Also remember to check the FAQ at http://fink.sf.net/faq
07:41 < chris01> _kristy: and use rsync instead of cvs. E.g. 'fink selfupdate-rsync'
07:41 < _kristy> yeah, i know, that's where i read about the rsync/cvs 
07:41 < chris01> ok
07:41 < _kristy> I tried that, but it wouldn't work
07:42 < chris01> mhmm...
07:42 < _kristy> the cvs update is soon done, so I hope it works now..
07:42 < _kristy> i previous versions i remember there were a 'setup-script' that set up the mirrors etc.
07:42 < _kristy> where did that go?
07:42 < chris01> 'fink configure'?
07:43 < _kristy> damn! that simple :)
07:43 < cirdan> heh
07:43 < _kristy> i feel stupid now, just for the record
07:43 < chris01> he
07:44 < _kristy> :)
07:45 < _kristy> thanks, I'll run it when the cvs is done
07:45 < _kristy> but I should go back to rsync?
07:45 < chris01> yes. It is recommended, since faster.
07:45 < chris01> and more reliable.
07:45 < _kristy> oh, ok
07:48 < _kristy> thank you
07:48 < _kristy> :)
07:50 < chris01> np
07:50 -!- jack- [jack@life.hater.be] has joined #fink
07:55 -!- akh [~akhansen@ldx3.psfc.mit.edu] has joined #fink
08:08 -!- Fang [~Fang@2002:53cd:a261:1:0:0:0:1] has joined #fink
08:17 -!- Feanor [~astrange@feanor.developer.opendarwin] has quit [Remote closed the connection]
08:18 -!- Feanor [~astrange@mp1-248-178.dialup.emory.edu] has joined #fink
08:23 -!- You're now known as RangerAway
08:25 -!- aaron-away [~adavies@12-220-223-80.client.insightBB.com] has joined #fink
08:26 < RangerAway> Omni|Sleep: most likely the maintainer wasn't yet prepared to handle all of the different python versions fink supports...  right now, --enable-python would probably end up with a .deb that randomly requires some certain version of python without an actual dependency in the Depends:
09:08 -!- kane-xs [~kane@coke.xs4all.nl] has quit [Remote closed the connection]
09:08 -!- aaron-away [~adavies@12-220-223-80.client.insightBB.com] has quit [Read error: 104 (Connection reset by peer)]
09:08 -!- kane-xs [~kane@coke.xs4all.nl] has joined #fink
09:11 < akh> 3 more packages to go and my 4 day update-all will be complete.
09:11 < cirdan> morning all
09:11 < akh> morning
09:14 -!- drm [~drm@m3b5f36d0.tmodns.net] has joined #fink
09:14 < drm> hi
09:15 < drm> can somebody help me debug a problem in the xml/news directory?
09:15 < cirdan> hey drm
09:15 < akh> What's the problem?
09:15 < drm> when i type "make", one of the xslt processes throws a bunch of errors
09:15 < drm> about a character out of range
09:16 < drm> i suppose its one of the foreign language files... is anybody else seeing this?
09:16 < akh> Let me do a current checkout.
09:17 < akh> (and I'll even checkout -C)
09:17 < drm> what does checkout -C do?
09:18 < akh> Actually it was 'cvs up -C' to use whatever's on the remote host regardless of age.
09:19 < akh> Yup--news.ja.xml
09:19 < drm> is that the one? i couldn't tell
09:20 < akh> yup:  do "make LANGUAGES=ja"
09:20 < drm> ah
09:20  * drm summons baba
09:22 < akh> You might just try building and installing en, then.
09:23 < akh> That works OK for me.
09:23 < cirdan> !summon baba
09:23  * Melian takes out 20 clean, identical-looking phones, some extra hands, and pretends to be a telemarketer for a large corporation, so she gets delivered a phonelist containing baba's coordinates.
09:24 < akh> We know his coordinates already, Melian.
09:24 < akh> (approximately)
09:25 < drm> thanks to newmanbe
09:28 < akh> yup.
09:29 -!- You're now known as RangerRick
09:29 < cirdan> hehe
09:32 -!- newmanbe [~newmanbe@512a5783083894f6.session.tor] has joined #fink
09:32 < pogma> does /usr/bin/xcodebuild work for you guys (xcode-2.1) ?
09:32 < pogma> I get "dyld: NSAddImage() error"
09:33  * akh will have to install Tiger to find out.  ;-)
09:34  * newmanbe would also have to buy Tiger.
09:34 < RangerRick> pogma: do you ahve to be in a project directory?
09:34 < RangerRick> or just in general?
09:34 < RangerRick> 'cause it doesn't happen to me
09:35 < pogma> hmm, reinstall those there xcode tools then, I guess
09:35 < akh> It's Installer.app's fault!
09:35 < RangerRick> probably
09:35 < akh> (can't blame libtool here)
09:36 < pogma> RangerRick: does 'gcc -arch ppc -arch i386 -v' give you any output? /me wonders if I filed a radar unnecessarily
09:36 < RangerRick> pogma: nope
09:36  * drm hasn't installed xcode 2.1 yet
09:37 < akh> It does give output with gcc3.3 on Panther though (not that it really matters)
09:37 < newmanbe> It gives the version.
09:37 < pogma> newmanbe: what does?
09:37 < newmanbe> And thread model.
09:37 < newmanbe> On Panther.
09:38 < newmanbe> gcc -arch ppc -arch i386 -v
09:38 < pogma> the compiler on panther is not fat
09:38 < RangerRick> it doesn't give the version on 10.4 with xcode 2.1
09:39 < pogma> which means that configure scripts don't correctly detect that it is gcc and it breaks libtool
09:39 < pogma> sigh
09:39 < akh> * ominous music *
09:39 < newmanbe> !comfort pogma
09:39 < Melian> There, there, pogma.  It's OK.  I'm here for you.
09:42  * akh tries to bum a Tiger DVD from my coworkers.
09:44 -!- broeken [~broeken@hst32165.phys.uu.nl] has quit []
09:47 < newmanbe> Hmm,  you just said that for all the world to read.
09:47 < drm> akh and others: wanna help proofread the web page?
09:47 < akh> Sure.
09:47  * newmanbe will.
09:48 < akh> Do you really think 0.6.4 will be usable "for years to come"?
09:48 < newmanbe> Don't the bindist servers keep the .debs?
09:48 < drm> 0.4.1 is still usable if you have 10.1
09:49 < akh> It's a semantic issue concerning the definition of "usable".  ;-)
09:50 < newmanbe> Something even older that Debian? Wow.  :)
09:50 < akh> Debian/unstable gets some new stuff.
09:51 < akh> That whole XFree86-4.3 thing gets to me, though.
09:51 < drm> at one point in time, Apple decided that the very last version of Mac OS which would run on my original Mac II was 7.5.1 (IIRC)... and yet, I ran it with 7.5.1 quite happily for years after that... couldn't put new software on it, but who cared?
09:51 < newmanbe> Most things that were made for Mac OS 9 only needed System 7 it seemed.
09:52 < drm> unless they used Carbon libs as Apple was then advocating
09:52 < akh> Yup.  Gotta encourage those hardware purchases.
09:52 < cirdan> hehe
09:53 < cirdan> it helped the osx transition a little
09:53 < cirdan> if they were written right
09:53  * akh told a coworker who plans on buying a G4 powerbook (since there won't be a G5) to keep it boxed until it's a collector's item.
09:53 < drm> 3% market share but 16% installed share, according to some recent stats... which means they make their hardware too good :)
09:53 < newmanbe> Now, things that were designed for say 10.2.8 mighty not work with 10.2.7.
09:54 < akh> drm:  Yeah.  I'm not _too_ disgusted with my 5 year old box yet.
09:54 < cirdan> newmanbe: last release of an os gets new features, so newer stuff still works
09:54 < drm> any obvious flaws, guys, other than the quibble about eternal usability?
09:54 < newmanbe> Microsoft likes eternal usability, that's for sure.
09:54 < akh> Paragraph 2:  "fink developers" should probably be "Fink developers"
09:54 < akh> (consistency)
09:55 < newmanbe> Well, not usability, but you know what I mean. :)
09:55 < akh> yup
09:55 < akh> Cool:  285 debs and 31 source tarballs deleted in this "fink cleanup"
09:55  * newmanbe gives up trying to guess what needs looking over and goes to look at the mailing list.
09:56 < akh> drm:  Otherwise it looks fine.
09:57 < drm> thanks...bbl
09:57 -!- drm [~drm@m3b5f36d0.tmodns.net] has quit ["Leaving"]
09:58 < newmanbe> All the other stuff you committed today looks fine too; but no, you had to leave! :) 
09:58  * akh may take the liberty of modifying the Download Quick Start Page, since item 2 doesn't reference any of the other versions.
09:58 < akh> It's not much of a liberty.
09:59 < newmanbe> That makes sense.
09:59 < newmanbe> That change wasn
09:59 < newmanbe> 't in the CVS e-mail on that page.
10:00 < newmanbe> No announcement on fink-announce yet.
10:01 < akh> Somebody who's allowed to post there needs to do it,
10:02 < akh> Or you could send it and I could moderate it (and a Viagra ad) through.  ;-)
10:02 < cirdan> dmalloc!
10:02 < newmanbe> RangerRick dmalloc chris01
10:02 < newmanbe> fingolfin
10:03 < newmanbe> Fingolfin is the primary contact for this channel but hasn't been here in over four weeks.
10:04 < cirdan> newmanbe: just cause fingolfin isn't in #fink doesnt mean he's not here
10:04 < newmanbe> If he doesn't ID with NickServ, NickServ says he is not here!
10:04 < newmanbe> So there.:)
10:04  * newmanbe goes away agin.
10:04 < newmanbe> again
10:05 < chris01> newmanbe: hi
10:05 < newmanbe> Hi.
10:06 < newmanbe> chris01: Sorry to disturb you. :)
10:07 < chris01> he
10:07 -!- Omni|Sleep is now known as Omnifarious
10:08 < Omnifarious> RangerRick: I also discovered that the serialize library triggers some C++ bug that causes the compiler to never complete with its size growing without limit.
10:12 -!- cianhughes [~cian@cian.ws] has quit [Read error: 104 (Connection reset by peer)]
10:18 -!- cianhughes [~cian@cian.ws] has joined #fink
10:18 < akh> newmanbe:  Found out why the change didn't show up on the download page:  it's updated dynamically to point to the latest version. of Fink.
10:19 < akh> (change == 0.7.1 to 0.8)
10:30 < runelind> RangerRick: 
10:31 < RangerRick> runelind: 
10:31 < RangerRick> :)
10:31 < runelind> how goes it?
10:31 < RangerRick> I found some interesting crashes in kmail, related to pthreads
10:31 < RangerRick> and I have no freakin' clue how to fix them
10:31 < runelind> heh
10:32 < RangerRick> but it seems to be because of the gpg stuff
10:32 < cirdan> hmm
10:32 < neo> hello
10:32 < runelind> I am seeing a lot of repeat stories on slashdot in akregator, and a guy in #kde said that was fixed a couple of days ago
10:32 < RangerRick> gpg stuff?
10:32 < RangerRick> hrm, I'll have to look for some patches  :)
10:33 < runelind> so there should be a new build or some patches for it :)
10:33 < runelind> you rock as always :)
10:35 < runelind> well time to cruise to work
10:35 < runelind> going up in the mountains for some work on a remote campus
10:35 < runelind> should be a nice trip :)
10:35 < akh> Sounds like it.
10:37 -!- zizban [~zizban@24-52-0-219.sbtnvt.adelphia.net] has joined #fink
10:37 < runelind> http://www.rockymountainworkshops.com/rmpingree.html
10:37 < runelind> :)
10:38 -!- aaron-away [~adavies@12-220-223-80.client.insightBB.com] has joined #fink
10:38 -!- zizban [~zizban@24-52-0-219.sbtnvt.adelphia.net] has quit [Remote closed the connection]
10:39 < pogma> chris01: no svn in the 10.3 bindist?
10:39 -!- TheSin [~TheSin@iphost-64-56-130-194.edm.wiband.net] has joined #fink
10:41 < akh> runelind: my wife would be _so_ happy if I could get a job at CSU.
10:42 < runelind> hehe
10:42 < runelind> well gotta go
10:42 < akh> later
10:43 -!- jtyler [~jtyler@iphost-64-56-130-194.edm.wiband.net] has joined #fink
10:44 < jtyler> hey TheSin 
10:44 < TheSin> hey jtyler 
10:46 < TheSin> hehe nice guys
10:46 < TheSin> syntax error at /Volumes/Resistor/Devel/cvsroot/fink/perlmod/Fink/Bootstrap.pm line 172, near "} elsif"
10:46 < TheSin> Compilation failed in require at ./inject.pl line 31.
10:48 < cirdan> heh
10:48 < TheSin> who did that :D
10:48 < TheSin> I'ma... I'ma..
10:49 < TheSin> there fixed in cvs
10:50 < TheSin> for others to enjoy :D
10:50 < cirdan> heh
10:50 < chris01> pogma: "no svn in the 10.3 bindist?" Why? what do you mean?
10:50 < chris01> pogma: there should be an svn in 10.3.
10:51 < TheSin> chris01, _jasonc_ in #opendarwin said that too
10:51 < chris01> mhm... bad...
10:51 < chris01> Ah... do they mean svn-ssl. I think it got changed to license Restrictive... because of reasons I don't fully understand.
10:52 < TheSin> chris01, commiting db43 stuff right now
10:53 < pogma> chris01: Yeah, that was it, thanks
10:56 < TheSin> chris01, 10.3 done
10:56 < TheSin> thanks for your testing BTW
10:57 -!- Omnifarious is now known as Omni|AFK
10:59 < chris01> pogma: which is sad...
10:59 < chris01> TheSin: cool! Thanks. Ufff... have to release svn 1.2 now...
11:00 < TheSin> hehe
11:00 < TheSin> k 10.4 done
11:00 < pogma> chris01: yes, it is sad
11:00 < TheSin> now to start on the new apache2 layou
11:00 < TheSin> t
11:00 < pogma> chris01: iirc someone complained on fink-devel about the gpl being incompatible with libssl
11:01 < pogma> chris01: so go beat up that someone :/
11:02  * chris01 going...
11:03 < TheSin> I assume that means no apache2-ssl of php4/5 then either :D
11:12 < pogma> apache is not GPL'ed
11:12 < pogma> nor, to my knowledge, is php
11:14 -!- Darien [~darien@office.mvcard.com] has joined #fink
11:14 < Darien> if I run 'fink install apache2' will that break my existing apache1 installation?
11:14 < Darien> because I want it to
11:14 < Darien> or do I have to break it myself?
11:15 -!- dsias_ [~dsias@adsl-068-153-207-210.sip.bct.bellsouth.net] has joined #fink
11:16 < Feanor> it won't break it
11:16 < Feanor> fink is self contained except for x11
11:17 -!- drm [~drm@dirac.cgtp.duke.edu] has joined #fink
11:17 < bbraun> howdy drm
11:18 < drm> i think we can tell stable users to just selfupdate now, not bootstrap... but should we modify the old new item, or add this info to the new one? akh, have an opinion?
11:18 < drm> hi bbraun
11:18 < Feanor> seen akh
11:18 < chris01> Darien: and i recommend install apache2-ssl. Otherwise you wont have ssl support.
11:19 < Darien> well yes
11:19 < TheSin> Darien, you have to shut off apache1 though
11:20 -!- glasser [~glasser@pcp09855802pcs.lowmrn01.pa.comcast.net] has joined #fink
11:20 < Darien> ok
11:20 < TheSin> or it apache2 won't start
11:20 < TheSin> as it'll try and bind to the same port
11:20 < Darien> well yeah
11:20 < TheSin> hey drm
11:20 < drm> hey thesin... thanks for fixing bootstrap
11:20 < TheSin> np, just happened that I tried updating this morn :P
11:21 < drm> yeah... jfm sent me an email about it, but when i went to fix it, it was already fixed! :)
11:21 < TheSin> gotta help where I can :P
11:21 < TheSin> still trying to make core :D <jk>
11:21 < TheSin> I watched beauty and geek last night, funny
11:22 < TheSin> guy was browning nosing literally, got it stuck in my head today
11:22 < drm> not an image you wanna be stuck with :/
11:23 < newmanbe> neo: Hello.
11:24 < newmanbe> None of you greeted our guest?
11:24 < akh> drm:  I'd say to leave the old item and just note in the newest one that bootstrapping is no longer necessary.
11:24 < akh> They're dated, after all.
11:24 < drm> ok, akh, i'll do that
11:24 < drm> maybe i'll also point out that you can't bootstrap with xcode 2.1 ATM
11:24 < akh> Yeah.  That'll encourage people not to do it.
11:26 < TheSin> drm, not it's not really :D
11:26 < TheSin> err no it's not
11:27 < TheSin> drm problem with an essential build?>
11:27 < akh> newmanbe, neo:  doing 6 things at once, but hi!
11:29 < glasser> so i saw about the 0.8.0 release on the rss... i'm running unstable branch under 10.4, should the usually selfupdate game do the trick? 
11:29 < akh> You may actually have it already--check "fink --version"
11:30 < Darien> how do I put a suggestion to the fink maintainers?
11:30 < akh> But otherwise, yes.
11:30 < Darien> I would like to suggest that 'fink help' should do as one expects
11:30 < Darien> perhaps this has been discussed already
11:31 < glasser> nah, i'm at 0.7.1.rsync
11:32 < akh> glasser: selfupdate then.
11:33 < pogma> Darien: fink help suggests that you try fink --help
11:33 < Darien> yeah, I know
11:33 < pogma> Darien: how's that broken?
11:33 < Darien> I didn't say it was broken
11:33 < drm> TheSin: that's right, dpkg doesn't build :/
11:33 < Darien> I said it didn't do as one might expect
11:33 < TheSin> drm, I'll have a peak at it today
11:34 < Darien> I would suggest, what with all the other commands one provides to fink being in the form of 'fink <command>', that 'fink help' should provide help as well (as --help does)
11:34 < pogma> there is a patch on the tracker for dpkg
11:34 < TheSin> oh
11:34 < TheSin> then I'll test it :D
11:35 < Darien> all the other --flags that fink has affect how it does things, but they don't change what it does
11:35 < pogma> apparently an unrelated wwdc demonstration included 'porting fink to 10.4/x86' 
11:35 < Darien> e.g. you can't do 'fink --help install foo'
11:35 < Darien> nice
11:35 < pogma> Darien: --help is a standard flag for gnu programs
11:36 < drm> glasser: i've added something to the announcement which addresses the upgrade question
11:36 < Darien> pogma: I didn't know fink was a a GNU project
11:36 < glasser> ah, cool
11:36 < Darien> but I'm not saying remove --help, I'm suggesting add 'help'
11:36 < glasser> since the upgrade matrix page seems to not be updated
11:36 < drm> pogma: !
11:36 < pogma> drm: hi
11:36 < drm> wasn't a greeting, was shock at your comnment
11:36 < pogma> Darien: put a patch on the tracker
11:37 < Darien> hmm
11:37 < Darien> that's an idea
11:37 < pogma> drm: I didn't say that fink was a gnu project
11:37 < drm> Darien: or a suggestion (different tracker)
11:37 < glasser> wait, upgrading *to* 10.4? or within?
11:37 < Darien> tracker?
11:37 < Darien> ah
11:37 < drm> glasser: the instructions won't hurt in either case
11:37 < glasser> gotch
11:37 < newmanbe> !tracker
11:37 < glasser> a
11:37 < pogma> fink.sf.net on the right hand side has a list of things you might want to do
11:38 < glasser> and do you recommend upgrading to XCode 2.1 before or after?
11:38 < TheSin> pogma, really? and apple didn't even send it to us :D
11:38 < TheSin> that is funny
11:38 < drm> TheSin: Apple has a mole in the fink project
11:38 < newmanbe> msachs?
11:38 < drm> right
11:38 < pogma> murr no longer counts?
11:39 < drm> good perl coder... probably knows the code as well as the core group at this point
11:39 < TheSin> shhhh
11:39 < drm> Murr was never a mole
11:39 < newmanbe> drm: You're wrong; We have a mole in Apple.
11:39 < TheSin> newmanbe, he'll hear you
11:39 < TheSin> hehe msashs OWNS anyhow...glad apple sent him in :D
11:39 < newmanbe> Well he sends stuff using an apple.com e-mail address.
11:41 < newmanbe> He isn't keeping his cover very well if he was trying to stay secret.
11:42 < akh> Yeah--not very moleish.
11:43 -!- drm [~drm@dirac.cgtp.duke.edu] has quit ["Leaving"]
11:44 < akh> Besides, usually moles try to sabotage stuff.
11:44 < akh> Though he did tell people to update to XCode 2.1...
11:44 < akh> ;-)
11:45 < newmanbe> Haha.
11:45 < TheSin> hehe
11:45 < TheSin> but I think he's done more good and only one bad
11:45 < TheSin> that only affects source installs ATM
11:45 < TheSin> :D
11:45 < TheSin> for now ! :D
11:49 -!- rudy__ [~rudy@ip68-224-184-250.hr.hr.cox.net] has joined #fink
11:52 < akh> Yup. 
11:53 < glasser> what's wrong with 2.1?
11:55 < akh> Some packages don't like it.
11:55 < akh> e.g. dpkg.
11:55 < akh> I think mozilla, too.
12:23 -!- Feanor [~astrange@feanor.developer.opendarwin] has quit []
12:23 -!- aaron-away [~adavies@12-220-223-80.client.insightBB.com] has quit [Read error: 104 (Connection reset by peer)]
12:41 -!- _kristy [~neptun@ti400720a080-0866.bb.online.no] has quit ["off"]
12:45 -!- cdevers [~cdevers@ip-8-140-251-63.gamelogic.com] has joined #fink
13:17  * akh spams gnome-core with a Fink Commander "feedback" message that lacks actual content.
13:17  * RangerRick is tempted to figure out enough objective C to work on it, just for FC to be not so annoying  :)
13:18 < RangerRick> 'cause you know, I have so much time  :P
13:19 < chris01> RangerRick: congratulations!
13:19 < RangerRick> chris01: eh?
13:19 < RangerRick> for what?
13:20 < chris01> RangerRick: for working on FC. Or should I say: Thanks!
13:21 < RangerRick> I haven't yet
13:21 < RangerRick> "is tempted" != "is actively working on"
13:21 < akh> Someone needs to fix the "December 2001 Dev Tools" indication for Tiger, too.
13:22 < chris01> RangerRick: ah, right...
13:24 -!- jtyler_ [~jtyler@iphost-64-56-130-194.edm.wiband.net] has joined #fink
13:25 -!- ringerc [~craig@dsl-202-72-144-62.wa.westnet.com.au] has joined #fink
13:28  * akh remaps my Exposé keys so that I can use f10 for aptitude.
13:30 < RangerRick> hah
13:30 < akh> I should have done that anyway--especially before I added RAM.
13:31 < akh> I'd hit one of the keys by accident and it would take forever to get things sorted back out.
13:31 < akh> (forever = 5 s)
13:32 -!- jtyler [~jtyler@iphost-64-56-130-194.edm.wiband.net] has quit [Success]
13:35 < ottoaim_> lollercopters
13:36 < akh> ?
13:37 < RangerRick> LOLWTFBBQ
13:39 < TheSin> I have seive auto delete FC emails
13:39 < TheSin> sieve that is
13:40 < RangerRick> I get a lot of useful debug info from FC mails
13:40 -!- Bart_ [~hideout@p3E9BE88B.dip.t-dialin.net] has joined #fink
13:40 < RangerRick> I have my stuff auto-forward them to my gmail account though, where I can star them and throw out the crap
13:40 < akh> Makes sense.
13:40 < RangerRick> I'm working on a script that will auto-forward them to my bug stuff  :)
13:41 < akh> nice.
13:41 < RangerRick> but I figured I'd wait and see if we end up actually getting the debian bug tracker fully in use
13:41 < RangerRick> I've never seen anything really taht says "OK, we're ready, start using it!"
13:41 < RangerRick> only, "well, I think it works"
13:41 < RangerRick> hehe
13:42 < akh> We've got some good inertia built up.
13:44 -!- joab [~joab@c-0669e155.73-1-64736c20.cust.bredbandsbolaget.se] has joined #fink
13:44 < joab> what is the option for making fink pretend to install? like "emerge --pretend somepackage"
13:45 -!- danielj [~danielj@pool-138-89-178-210.mad.east.verizon.net] has joined #fink
13:46 < danielj> Does anyone think it matters if cyrus-sasl2 doesn't have static libs? Upstream broke static lib building in the latest version and it'll be a real PITA to restore it. --disable-static is actually the default; it's just that the fink pkg has always had an explicit --enable-static for some reason.
13:46 < akh> joab:   no such thing.
13:46 < RangerRick> fink install foo; dpkg -L foo; fink remove foo         ;)
13:46 < newmanbe> joab: You can build but not install a package.
13:47 < akh> RangerRick: but that installs all of the dependencies, too.
13:47 < akh> Which also goes for newmanbe
13:47 < joab> akh: so how do i find out the depencies if i cant do that?
13:48 < newmanbe> You can look at the .info file for the package.
13:48 < chris01> Enjoy: I just released svn 1.2 to unstable.
13:48 < akh> fink dumpinfo -fdepends <pkgname>
13:48 < newmanbe> Or select no when it asks if you want to install it.
13:48 < akh> and fink dumpinfo -fbuilddepends <pkgname> for build dependencies.
13:48 < chris01> Thanks to TheSin and pogma for helping to update all packages to use db43
13:48 < newmanbe> Hmm, previous version of svn wouldn't build 'cause of dependency issues.
13:48 < akh> !woot for chris01
13:49 < joab> newmanbe: it doesnt always ask that does it?
13:49 < newmanbe> Then you can look at the .info file for it.
13:49 < newmanbe> Or:
13:49 < newmanbe> fink dumpinfo package
13:49 < joab> btw, how do i install a package that is marked as unstable?
13:49 < TheSin> thanks chris01 for contacting everyone and pushing us to do it :P
13:49 < akh> !unstable
13:49 < Melian> methinks unstable is http://fink.sourceforge.net/faq/usage-fink.php#unstable, or at http://fink.sourceforge.net/faq/usage-fink.php#bindist
13:49 < newmanbe> !unstable
13:50 < newmanbe> Melian: Forget unstable
13:50 < Melian> i forgot unstable, newmanbe
13:50 < newmanbe> Melian: unstable is http://www.finkproject.org/faq/usage-fink.php#unstable, or at http://www.finkproject.org/faq/usage-fink.php#bindist
13:50 < Melian> newmanbe: okay
13:50 < danielj> chris01: I have a db43 version of cyrus-sasl, but see above for my question about static libs.
13:51 < akh> Same thing.  ;-) (for now0
13:51 < newmanbe> Be prepared!
13:51 < newmanbe> More incentive for cirdan to keep paying for the domain. ;-p
13:51 < TheSin> hey danielj 
13:52 < chris01> danielj: ah, ok. Unfortunately I have no idea...
13:52 < danielj> TheSin: hey
13:52 < chris01> TheSin: we should now chase the remaining people whos packages depend on db4x to update to db43
13:53 < joab> newmanbe: actually that comma at the end of the first url made safari not find the anchor
13:53 < danielj> chris01: Alternatively, I could just update 2.1.20 to use db43 and wait for upstream to fix things.
13:53 < newmanbe> Bad Safari.
13:53 < newmanbe> Melian: forget unstable
13:53 < Melian> i forgot unstable, newmanbe
13:53 < newmanbe> Melian: unstable is http://www.finkproject.org/faq/usage-fink.php#unstable , or at http://www.finkproject.org/faq/usage-fink.php#bindist
13:53 < Melian> okay, newmanbe
13:53 < joab> :)
13:54 < joab> i've set up my fink to search unstable packages too now. but are there packages that exist in booth trees? i mean, how does it choose between the two?
13:55 < newmanbe> It should pick the most recent.
13:55 < newmanbe> Usually (I hope) unstable.
13:55 < newmanbe> apt-get will pick the most recent binary, if there is one.
13:55 < newmanbe> YEAH! Two CD-ROM drives.
13:56 < chris01> danielj: yes, maybe do that for the moment. I think it is a good time to move to db43 at the moment. Since all those pacakges were change to db43.
13:56 < chris01> have to run, bye.
13:56 < newmanbe> Now to get both hard drives working at the same time.
13:56 -!- chris01 [~chris01@212.126.165.246] has quit ["bye"]
13:56 < akh> newmanbe: that's crazy talk.
13:56  * newmanbe is putting the innards of one computer into another.
13:56 < akh> Ah.
13:56 < newmanbe> They work separately...
13:56 < newmanbe> Just not together.
13:57 < newmanbe> I am thinking of install Darwin 8 on it.
13:57 < akh> ATAPI master/slave issue?
13:57 < newmanbe> I have no idea. :)
13:57 < Darien> That's right, put your pom-poms down, getting everybody fired up
13:57 < newmanbe> Then moving my music onto the main partition on this computer, and have Darwin 8 PowerPC and x86!
13:58 < newmanbe> When they are both connected the first master's name is gobbly gook, and there is no slave.
13:58  * newmanbe returns to the computer.
14:01 -!- Feanor [~astrange@64.207.61.218] has joined #fink
14:03 -!- Bart_ [~hideout@p3E9BE88B.dip.t-dialin.net] has quit []
14:04 -!- slyrus [~slyrus@adsl-68-121-172-169.dsl.pltn13.pacbell.net] has left #fink []
14:04 < joab> has anyone got itunes music sharing (or file sharing with netatalk for that matter) working in linux?
14:05 < Darien> I almost did
14:05 < Darien> I got the daemons set up so that iTunes could see them
14:05 < Darien> but it oculdn't see music, and it locked my Linux box up hard
14:05 < joab> i got howl working and broadcasting the itunes music sharing and it showed up in itunes but then i got an error message
14:05 < joab> i got to that point too :)
14:05 < joab> except my linux computer still runs
14:06 < Darien> mine still ran, just veeeeeery sloooooowly
14:07 -!- aaron-away [~adavies@12-220-223-80.client.insightBB.com] has joined #fink
14:07 < joab> ok
14:07 < joab> what distribution of linux was it?
14:09 -!- Macrobyte [~macrobyte@pcp06495808pcs.nstnig01.ct.comcast.net] has joined #fink
14:09 < cirdan> who called?
14:09 -!- cianhughes [~cian@cian.ws] has quit [Client Quit]
14:09 -!- danielj [~danielj@pool-138-89-178-210.mad.east.verizon.net] has quit [Remote closed the connection]
14:11 -!- dmacks [~dmacks@203-137.dialup.cloud9.net] has joined #fink
14:12 < akh> newmanbe did. 
14:12 < dmacks> cirdan: Can you patch Melian (or suggest to upstream) that alternatives should not have commas after them?
14:13 < newmanbe> dmacks: I think that was what the person who made the factoid put.
14:13 -!- Fang [~Fang@2002:53cd:a261:1:0:0:0:1] has quit [Remote closed the connection]
14:13 < newmanbe> There was a coma after a URI.
14:14 < newmanbe> That wasn't why I called you.
14:14 < dmacks> newmanbe: No, she understands having multiple values for a single key (not just single value that contains several URLs)
14:14 < dmacks> But she automatically puts [no space]-comma to separate them.
14:14 < akh> a coma?  That's serious.  ;-)
14:14 -!- SailMan [~bobo@ool-435292cd.dyn.optonline.net] has joined #fink
14:15 < dmacks> heh
14:15 < dmacks> melian: foo is bar
14:15 < Melian> dmacks: okay
14:15 < dmacks> melian foo is also qux
14:15 < Melian> dmacks: okay
14:15 < dmacks> !qux
14:15 < dmacks> !foo
14:15 < Melian> hmm... foo is bar, or qux
14:15 < dmacks> Melian: forget foo
14:15 < Melian> dmacks: i forgot foo
14:15 -!- Macrobyte [~macrobyte@pcp06495808pcs.nstnig01.ct.comcast.net] has left #fink []
14:16 < dmacks> (newmanbe: I didn't know you were looking for me, either)
14:16 < newmanbe> I wasn't.
14:17 < dmacks> So "< newmanbe> That wasn't why I called you." means...?
14:17 < dmacks> Or am I not the you?
14:17 < newmanbe> I was talking to cirdan.
14:17 < newmanbe> Yes.
14:17 < dmacks> Ah.
14:17 -!- joab [~joab@c-0669e155.73-1-64736c20.cust.bredbandsbolaget.se] has quit ["Leaving"]
14:17 < dmacks> Yeah, I am not cirdan 
14:17 < newmanbe> cirdan	who called?
14:17 < newmanbe> akh	newmanbe did.
14:19 < cirdan> Melian: foo?
14:19 < Melian> cirdan: KCI error, or a problem with the Keyboard-Chair Interface.
14:19 < newmanbe> !comfort dmacks
14:19 < Melian> There, there, dmacks.  It's OK.  I'm here for you.
14:20 < newmanbe> Always needing attention... :)
14:20 < cirdan> Melian: foo is bar|qaz
14:20 < Melian> okay, cirdan
14:20 < cirdan> Melian: foo?
14:20 < Melian> extra, extra, read all about it, foo is bar|qaz
14:20 < cirdan> ah
14:20 < cirdan> !forget foo
14:20 < Melian> i forgot foo, cirdan
14:20 < cirdan> Melian: foo is (bar|qwerty)
14:20 < Melian> okay, cirdan
14:21 < cirdan> !foo
14:21 < Melian> rumour has it, foo is qwerty
14:21 < cirdan> Melian: foo?
14:21 < Melian> it has been said that foo is qwerty
14:21 < cirdan> !foo
14:21 < Melian> i guess foo is bar
14:21 < cirdan> !forget foo
14:21 < Melian> cirdan: i forgot foo
14:22 < akh> confusing.
14:22 < cirdan> hehe
14:23 < dmacks> Is FinkBot ever coming back? 
14:23 < cirdan> dunno
14:23 < akh> We've got one logger-bot anyway (for now)
14:24 < akh> Unless somebody bans it for server flakiness again...
14:25  * RangerRick logs to http://hollowvoice.org/~ranger/irc_logs/
14:26  * dmacks thinks RangerRick posted that so everynoe will go look there and he can find out our IP addresses.
14:26 < akh> heh
14:26 < akh> And don't forget to leave your bank account number for the sweepstakes.
14:27 < dmacks> I wish I was smart enough to figure out how to forge that data so I appeared to come from goatse
14:28 -!- lisppaste [~lisppaste@common-lisp.net] has quit [Read error: 104 (Connection reset by peer)]
14:31  * akh waits for somebody to start pasting up a bunch of cruft now that lisppaste is gone.
14:31 -!- Omni|AFK is now known as Omni|Work
14:33 < Omni|Work> akh: I'm here to oblige.  :-)
14:33 -!- akh is now known as akh_AFK
14:33 < akh_AFK> go for it ;-)
14:33  * Omni|Work laughs.
14:33 < Omni|Work> I got boost::python working.
14:34 < Omni|Work> It was easy, but I'm not sure if the change I made is widely applicable enough.  I think it is.
14:38 -!- msachs [~msachs@17.255.104.148] has joined #fink
14:38 < msachs> 'morning
14:39 < dmacks> Does anyone with gtk+2-dev that they built from source Tiger have /sw/lib/pkgconfig/gobject-2.0.pc ?
14:39 < dmacks> alo msachs 
14:39 < msachs> Does anyone know what package /sw/bin/tempfile is in?  Because libxml2 is missing a depend on it.
14:39 < dmacks> (or maybe in glib2-dev)
14:39 < dmacks> msachs: Never heard of it.
14:40 < cdevers> fileutils ?
14:40 < RangerRick> gobject is from glib2 I think
14:40 < msachs> cdevers: Nope.
14:40 < dmacks> Okay. I got a report that its .pc is missing in 2.4.x
14:40 < cdevers> dpkg -S /sw/bin/tempfile
14:41 < RangerRick> I believe at one point, the .pc file was in glib2 instead of glib2-dev
14:41 < RangerRick> and glib2-dev didn't depend on glib2
14:41 < msachs> cdevers: Only works if you have the package installed.
14:41 < RangerRick> (I know at one point I needed to builddepend on both glib2 and glib2-dev)
14:42 < dmacks> Not in any of glib2*
14:42 < RangerRick> that was long ago, I'm not surprised if it's not like that anymore
14:42 < cdevers> from a debian box:
14:42 < cdevers> [[ $ dpkg -S /bin/tempfile
14:42 < cdevers> debianutils: /bin/tempfile ]]
14:43 < dmacks> msachs: It's in sbin/ not bin/
14:43 < Omni|Work> dmacks: I have one.
14:43 < msachs> Yeah, I saw that on packages.d.o
14:43 -!- cianhughes [~cian@cian.ws] has joined #fink
14:43 < msachs> dmacks: Okay, well, what provides it?
14:43 < dmacks> Omni|Work: What version of glib2?
14:44 < Omni|Work> 2.4.6-8
14:44 < dmacks> msachs: like cdevers said: debianutils
14:44 < dmacks> Omni|Work: Weeeeeeird.
14:44 < cdevers> fink/mac: [[ $ dpkg -S /sw/sbin/tempfile
14:44 < cdevers> debianutils: /sw/sbin/tempfile ]]
14:44 < msachs> Oh, Fink has a debianutils package?
14:44 < cdevers> yes
14:44 < msachs> Ah.
14:44 < Omni|Work> But, I have gtk+2, gtk+2-dev, glib2, and glib2-dev
14:45 < dmacks> *gah* I hate gnome.
14:45 < cdevers> [[  i      debianutils     1.23-11 Misc. utilities specific to Debian (and Fink) ]]
14:46 < Darien> debianutils?
14:46 < dmacks> bugger all, i dunno
14:47 -!- dmacks [~dmacks@dmacks.student.supporter.pdpc] has quit ["leaving"]
14:47  * newmanbe looks at RangerRicks's IRC logs and finds that he has a strange IP address.
14:48 < RangerRick> :)
14:48 < Omni|Work> You can't see it anymore, but I have an IPv6 address.  :-)
14:48 < newmanbe> I've seen.
14:48 < newmanbe> It has real words in it.
14:49 < Darien> haha
14:49 < newmanbe> They just skipped over IPv5 it seems.
14:50 < Omni|Work> There already was a protocol 5.  :-)
14:50 < Omni|Work> It was an obscure thing that wasn't actually much like IP at all.
14:50 -!- aaron-away [~adavies@12-220-223-80.client.insightBB.com] has quit ["BitchX: your way, right away"]
14:51  * newmanbe just see IPv6 as more letters and numbers to remember.
14:51  * Omni|Work chuckles.
14:51 < newmanbe> Or you could just make it words.
14:52 < newmanbe> Four letter words...
14:52 < Omni|Work> Actually, it would be interesting to map IPv6 addresses into base32.  It wouldn't be a lot easier to remember though.  26 letters.
14:53 < newmanbe> Twenty six letters? A B C D E F G...
14:55 < Omni|Work> More like EABNR7QQCAAACAQE4L775MT3FU
14:56 < newmanbe> Vanity IP addresses.
14:57 < Omni|Work> I should do that.  :-)
14:57 < Omni|Work> The lower 64 bits are more-or-less to do with as you please.
14:58 < newmanbe> Do you work for a research institution? Do you work for a research institution?
14:58 < newmanbe> That's what the wonderful Apple Help says is the primary user of IPv6.
14:59 < Darien> I want DEADBEEFCAFE in my IP address
15:00 < Darien> used to have that as my MAC address, which showed up in my last ISP's hostname
15:00 < cirdan> hehe
15:01 < newmanbe> If you could have the character "/", you could have URI's in it.
15:01 < Darien> I should call my cafe the Dead Beef Cafe
15:08 -!- ringerc [~craig@dsl-202-72-144-62.wa.westnet.com.au] has quit ["Whoops, someone let the magic smoke out!"]
15:10 -!- emp [~emp@67-42-224-94.blng.qwest.net] has left #fink ["Leaving"]
15:11 -!- broeken [~broeken@5353014C.cable.casema.nl] has joined #fink
15:12 < Omni|Work> Darien: I want to be obscure and have my IPv6 address say something in bas32.  :-)
15:12 < Darien> hehe
15:13 < Omni|Work> That way, someone has to do a bit of work to see it.  :-)
15:13 < Omni|Work> And base32 includes the entire english alphabet, so there's a lot more I can say.
15:15 < mee_bot> some of my brain is lost - HELP!!!!!! - 10.4-transitional how "close" is that to being 10.4-final ????
15:16 -!- rudy__ [~rudy@ip68-224-184-250.hr.hr.cox.net] has quit []
15:16 < cirdan> you'll need to rebuild anything using gcc3 now
15:16 < cirdan> so we'll all be gcc4
15:17 < mee_bot> i have two boxen on my desk 1 - 10.3.9 and 1 10.4.2 .... 
15:17 < mee_bot> grumble......
15:17 < mee_bot> can't seem to get the same version of svn
15:19 < mee_bot> 10.4-transitional has 1.1.4 and the other has 1.2 somthing
15:19 -!- KraMer [~mark@adsl-70-240-203-247.dsl.hstntx.swbell.net] has quit [Connection timed out]
15:20 < mee_bot> which means memos, write ups ... grumble.... more paperwork....
15:20 -!- KraMer [~mark@adsl-70-240-211-234.dsl.hstntx.swbell.net] has joined #fink
15:20 < mee_bot> i'm going to lunch!
15:20 < mee_bot> dammit!
15:20 < newmanbe> 10.4.2?
15:21 < mee_bot> apple dev build -
15:21 < mee_bot> alpha code - not stable
15:21 < mee_bot> stay w/ 10.4.1
15:21  * newmanbe is staying with 10.3.9.
15:21 < mee_bot> unless you need mobile accounts and AD bindings
15:22 < mee_bot> 10.4 is major way fun
15:22 < mee_bot> mee like 10.4
15:22 -!- emp [~emp@67-42-224-94.blng.qwest.net] has joined #fink
15:22 < newmanbe> I don't need AfterDark bindings. :)
15:22 < mee_bot> the meds my doc gave mee for ulcer are great!
15:22 -!- mario_ [~mario@newton.informatik.uni-leipzig.de] has joined #fink
15:23 -!- mario_ [~mario@newton.informatik.uni-leipzig.de] has quit [Client Quit]
15:23 < mee_bot> the stress pills are smooooooooth......
15:23 < mee_bot> really looking forward to the intel transition....
15:23 < newmanbe> TMI
15:23 -!- terrapen [~cjs@fw-01.satx.bikeworld.net] has joined #fink
15:23 < mee_bot> bet i can get more xanax
15:23 -!- emp [~emp@67-42-224-94.blng.qwest.net] has left #fink ["Leaving"]
15:24 -!- terrapen [~cjs@fw-01.satx.bikeworld.net] has left #fink []
15:24 < akh_AFK> Everybody's favorite palindrome!
15:24 -!- akh_AFK is now known as akh
15:25 < dh1pa> could anyone tell me where to find libboost_python*.dylib ? Or do I misinterprete the package name boost*-py23 and it is not installed with that package?
15:26 < akh> I'd imagine it would be in boost*-py23-shlibs.
15:31 -!- baba [~baba@YahooBB220041001026.bbtec.net] has joined #fink
15:32 -!- Zifra [~josera@ra.eii.us.es] has joined #fink
15:33 < dh1pa> akh: well, yes, thought so, but cannot find the build libraries on my computer. all other boost libraries are there, but no boost_python.
15:33 < akh> Then I'm not sure.
15:33 < akh> baba:  there seems to be a problem with news.ja.xml
15:34  * TheSin plays Nine Inch Nails - The Line Begins To Blur (3:44) (44.1\KHz/320\Kbs/ST) (Album: With Teeth) (Genre: Industrial) in iTunes.
15:34 < baba> hm?
15:34 < Darien> :/
15:34 < TheSin> :\
15:35 -!- Zifra [~josera@ra.eii.us.es] has quit [Client Quit]
15:36 < akh> baba:  I'll paste up the output.
15:36 < akh> 1sec
15:36 < akh> (i.e. it's an xslt issue)
15:37 -!- mGiff [~mGiff@ottawa-hs-64-26-147-201.d-ip.magma.ca] has joined #fink
15:37 < broeken> TheSin, are you the maintainer for the mac-growl-pm586 package?
15:37 < TheSin> yes
15:39 < akh> crud:  what's an alternative to lisppaste?
15:39 < TheSin>  /query? :D
15:40 < broeken> you know it doesn't work on tiger?
15:40 < akh> baba:  In any case when you run 'make' in the news subdir of the doc tree you get warnings about out-of-range characters.
15:40 < TheSin> broeken, nope works fine for me
15:40 < broeken> or is it me?
15:40 < TheSin> using it in xchat right now
15:40 < TheSin> that is how I got your msg :D
15:40 < broeken> you mean growl or mac-growl-pm586?
15:41 < RangerRick> both
15:41 < RangerRick> broeken: I already told you it works for me  :)
15:41 < TheSin> broeken, both
15:41 < baba> akh: ok i will check
15:41 < TheSin> broeken, what are you trying to do to say it's broken?
15:41 < TheSin> it won't build?
15:41 < TheSin> it won't display?
15:42 < TheSin> if it won't display what app is trying to use it
15:42 < TheSin> etc etc
15:42 < broeken> jup, i'll paste fink's output
15:42 < TheSin> no
15:42 < TheSin> please just wmail it to me
15:42 -!- lisppaste [~lisppaste@common-lisp.net] has joined #fink
15:42 < baba> wb lisppaste
15:42 < lisppaste> akh pasted "for baba" at http://paste.lisp.org/display/8951
15:44 < baba> maybe it is not in UTF-8
15:44 < broeken> TheSin, i've sent you the fink output
15:46 < akh> baba:  that'd make sense.
15:47 < akh> Hmm..it claims to be.
15:51 < TheSin> k I'll have a look broeken 
15:52 < TheSin> broeken, do you have growl running
15:52 < broeken> i do
15:53 < TheSin> k
15:53 < TheSin> looks like a glue problem
15:53 -!- dreamind [~dreamind@C2107.campino.wh.tu-darmstadt.de] has joined #fink
15:53 < broeken> what does that mean?
15:53 < TheSin> which mac-glue-pm was updated to 586
15:54 < TheSin> I'm looking into it
15:54 -!- nickers [~nickers@cpe-24-242-125-121.sport.res.rr.com] has joined #fink
15:54 < broeken> TheSin, mac-glue isn't installed
15:55 < TheSin> that is cause ther eis no mac-glue-pm586
15:55 < TheSin> only 581
15:55 < baba> news.ja.rdf is really messed
15:55 < akh> hmm
15:56 < broeken> and what would happen if i'd try to install mac-growl-581?
15:56 < TheSin> badness
15:56 < TheSin> I wish I could get this problem
15:56 < broeken> no, you don't ;)
15:57 < TheSin> hehe
15:57 < TheSin> be easier fo rme to fix it
15:57 < TheSin> I have an other guy with the exact same plroblem
15:57 < TheSin> was your 10.4 upgraded from 10.3
15:57 < TheSin> or fresh install?
15:57 < broeken> no, fresh
15:57 < TheSin> hmmm me too
15:57 < TheSin> so not that
15:58 < TheSin> copy your profile over from 10.3?
15:58 < msachs> hmmm
15:58 < broeken> when installing growl i made a mistake
15:58 < msachs> debianutils is essential
15:58 < msachs> So why isn't it installed on my system?
15:58 < broeken> first i installed it just for me
15:58 < broeken> but somebody (RangerRick?) told me i should install it for all users
15:58 -!- KraMer [~mark@adsl-70-240-211-234.dsl.hstntx.swbell.net] has left #fink ["Leaving"]
15:59 < broeken> so i tried deinstalling which didn't go too easy, but in the end i succeeded
15:59 -!- dh1pa [hlawit@aristoteles.informatik.uni-leipzig.de] has left #fink ["Kopete 0.9.0 : http://kopete.kde.org"]
15:59  * broeken went away to take a shower
15:59 < TheSin> hey msachs 
15:59 < TheSin> heard your a mole
15:59 < TheSin> for apple
15:59 < TheSin> :P
15:59 < msachs> TheSin: More like a groundhog.
15:59 < TheSin> hehe
15:59 < TheSin> how are things?
16:00 < TheSin> also heard your a perl guru
16:00 < msachs> Well, my --build-as-nobody build finished
16:00 < TheSin> know this problem
16:00 < TheSin> Please run gluedialect and gluescriptadds programs at /System/Library/Perl/Extras/5.8.6/Mac/Glue.pm line 1341, <DATA> line 1.
16:00 < TheSin> I've had two users get that with mac-growl-pm
16:00 < msachs> Mostly because I didn't have debianutils, an essential package, installed, for some reason.
16:00 < TheSin> but I can't reproduce it nor can I find gluedialect and gluescriptadds
16:01 < TheSin> msachs, that is odd
16:01 < TheSin> fink freaks if an essential is missing
16:01 < msachs> I'd've thought so too.
16:02 < TheSin> anyone remember where MACGLUEDIR is
16:02 < TheSin> ~/Library/something
16:03 < akh> later peeps
16:03 -!- akh [~akhansen@ldx3.psfc.mit.edu] has quit []
16:05 -!- Fang [~Fang@2002:53cd:a261:1:0:0:0:1] has joined #fink
16:05 < msachs> TheSin: You'll get that error if /System/.../Mac/Glue/glues/{dialects,additions} doesn't exist.
16:05 < msachs> MACGLUEDIR is /System/Library/Perl/Extras/5.8.6/Mac/Glue
16:05 < TheSin> and why wouldn't it from a fresh install
16:06 < Omni|Work> Fang: A fellow 6to4 user.  :-)
16:06 < TheSin> need to run /System/Library/Perl/Extras/bin/gluescriptadds
16:06 < Fang> Omni|Work: when it works
16:06 < Fang> ;)
16:06 < msachs> TheSin: Beats me :)
16:06 < TheSin> should I file a bug with Apple
16:07 < msachs> TheSin: Might be better if the users reporting the problem did.
16:08 < TheSin> crap I knew I was on an ignore list from apple
16:08 < msachs> "It's that TheSin character again, quick, hit the delete key!"
16:09 < TheSin> hehe
16:09 < TheSin> I bet you told them that too
16:10 < msachs> Yep.
16:10 < TheSin> part of your Mole duties
16:10 < msachs> Sure did.
16:10 < Omni|Work> msachs: It isn't doing anything, what should I do?!
16:10 < TheSin> thanks for the fix msachs 
16:10 < msachs> Omni|Work: What isn't?
16:10 < msachs> np
16:10 < Omni|Work> The delete key.
16:10 < TheSin> as soon as mac-glue gets updated for perl 586
16:10 < TheSin> I'll just dep on it
16:10 < msachs> Omni|Work: We have a special delete key internally.
16:10 < TheSin> that'll fix apple's little red wagon
16:10 < TheSin> :D
16:11 < msachs> TheSin: Nooooo!
16:11 < TheSin> hehe
16:11 < TheSin> change to x86 will they
16:11 < TheSin> Oops that is a different topic isn't it :D
16:12 < msachs> Yep
16:13 < TheSin> so shall we start an i686 arch
16:13 < TheSin> :P
16:14 < msachs> Anyone seen "/sw/lib/libiconv.la: No such file or directory" before?
16:14 < TheSin> need libiconv?
16:14 < msachs> So, missing builddep?
16:14 < TheSin> checking
16:14 < TheSin> libiconv-dev
16:14 < TheSin> yup missing builddep
16:14 < msachs> Stupid gdk-pixbuf.
16:14 < TheSin> hehe
16:16 -!- Darien [~darien@office.mvcard.com] has quit []
16:16 < RangerRick> lots of libiconv/gettext builddeps missing just because we didn't used to include them
16:17 < Omni|Work> msachs: Oh, it's too good to put on every Mac huh?
16:17 < msachs> And stupid arts blacklisting 4.0.0.
16:18 < msachs> They should see if 5026 fixes their problem...
16:19 < RangerRick> msachs: they blacklist gcc 4.0 because gcc 4.0 as released compiled, but generated bad code, for KDE
16:19 < msachs> Well yeah.
16:19 < msachs> I wonder if the problem is still in 5026, though.
16:19 < RangerRick> dunno
16:19 < RangerRick> I don't know enough about what the actual issues are to be sure
16:20 < msachs> Wish they gave a pointer to find out.
16:20 < TheSin> stupid apple, x86 broken Mac::Glue
16:20 -!- KraMer [~mark@adsl-70-240-211-234.dsl.hstntx.swbell.net] has joined #fink
16:20 < TheSin> :D
16:20 < msachs> Fair enough.
16:20 < TheSin> I wanted to complain too
16:20 < RangerRick> msachs: yeah, I know it's deep in the kde-devel archives somewhere, but I don't have specifics
16:21 < TheSin> KDE is a very active project I'm sure they'll get it fixed
16:21 < msachs> Maybe I should hack my gcc-3.3 -> 4.0 wrapper to lie about the version number
16:21 < TheSin> hehe
16:22 < TheSin> when it called gcc --version it'll run gcc-3.3
16:22 < TheSin> else run gcc-4.0
16:22 < TheSin> :P
16:22 < msachs> Exactly.
16:22 < RangerRick> TheSin: well, I don't know that they're interested in fixing "your compiler is broken" bugs, but...
16:22 < msachs> Grabbing food -- bbiab
16:22 < TheSin> RangerRick, true but they'll have to go to gcc 4.0 eventually I'd assume
16:22 -!- kane_ [~kane@perl.xs4all.nl] has quit []
16:22 < TheSin> I'm sure debian will have it as a default soon
16:22 < broeken>  TheSin?
16:22 < TheSin> broeken, emailed ya
16:23 < Omni|Work> msachs, RangerRick: I'm not trust gcc until 4.0.2 comes out.  I've been reading their mailing lists.
16:23 < cdevers> did it just miss the debian sarge -> stable switch ? if so, it could be years ....
16:23 < RangerRick> TheSin: no, they'll have to go to gcc 4.0.2  :)
16:23 < RangerRick> or whatever
16:23 < broeken> i just read it, but it's a little cryptic
16:23 < RangerRick> I'm sure they'd be happy to say "gcc 4.0.0 will never be supported"
16:23 < TheSin> oh sure
16:23 < TheSin> but apple will go 4.0.2 eventually too
16:23 < TheSin> it'll work out :P
16:23 < TheSin> hehe
16:24 < Omni|Work> And it really distresses and confuses me that my knowledge of gcc development doesn't directly translate to the gcc on my OS X box.
16:24 < broeken> my glues dir contains GrowlHelperApp, GrowlHelperApp.pod and dialects
16:24 < TheSin> isn't that how apple is dealing with all problem right now :D
16:24 < TheSin> :D
16:24 < TheSin> broeken, and the perms
16:25 < broeken> same as yours, drwxr-xr-x   4 root  wheel    136 Jun  7 20:09 dialects
16:26 < TheSin> msachs, ?
16:28 < TheSin> broeken, try running /System/Library/Perl/Extras/bin/gluescriptadds with sudo
16:29 -!- baba [~baba@YahooBB220041001026.bbtec.net] has quit ["Leaving"]
16:30 < broeken> i did, some glues were installed but not all
16:31 < lisppaste> broeken pasted "Glue trouble" at http://paste.lisp.org/display/8955
16:32 < TheSin> did it install now?
16:32 < broeken> nope
16:33 < broeken> some errors
16:33 < broeken> same errors
16:33 < broeken> did you see my paste?
16:34 < TheSin> no, but I know what it looks like
16:36 < broeken> i pasted the errors from /System/Library/Perl/Extras/bin/gluescriptadds, not from fink install
16:36 < TheSin> this isn't a fink problem
16:36 < TheSin> this is a Mac::Glue from 10.4 problem
16:36 < TheSin> and I don't know much about it
16:36 < TheSin> was hoping msashs was around
16:37 < broeken> ok, i'll give it a rest for now
16:37 < broeken> wonder what's the problem
16:38 -!- broeken [~broeken@5353014C.cable.casema.nl] has quit []
16:39 < TheSin> do you run fink in sudo mode?
16:39 < TheSin> crap
16:43 -!- nickers [~nickers@cpe-24-242-125-121.sport.res.rr.com] has quit [Remote closed the connection]
16:58 -!- dmacks [~dmacks@netspace.org] has joined #fink
17:06 -!- SailMan [~bobo@ool-435292cd.dyn.optonline.net] has quit ["Mmmm Mmmm.. Nothing like the smell of compile errors in the morning"]
17:08 < TheSin> bbiab
17:08 -!- TheSin [~TheSin@iphost-64-56-130-194.edm.wiband.net] has quit ["Client exiting"]
17:15 -!- Feanor [~astrange@feanor.developer.opendarwin] has quit []
17:35 -!- You're now known as RangerAway
17:51 < msachs> RangerAway: I'm writing a script to automatically nag maintainers when their packages are broken.  Do you want a separate email for each of yours -- you have a different email address listed for each package -- or should I special-case you?
17:54 -!- drm [~drm@user-152-3-152-54.wireless.duke.edu] has joined #fink
17:55 < drm> hey dmacks
17:55 < dmacks> hiya drm
17:55 < dmacks> This is very confusing...just got email from a different Dave Morrison
17:55 < drm> do you remember the path-prefix-g++ episode? i can't remember the outcome... somehow we did it that way in 0.23 but then it changed in the other branch
17:55 < drm> dmacks: i can't help that :)
17:56 < drm> or maybe msachs remembers
17:57 < dmacks> I thought we decided against path-prefix for now. Not sure why it's in 0.23.
17:57 < dmacks> Is there maybe some flag-passing problem in 0.23? Or did we just forget that we still supported 0.23 when we changed from plan BIGNUM to BIGNUM+1?
17:57 < msachs> drm: Nope.  I'm pretty sure it's in HEAD, though.
17:57 < drm> there is no path-prefix in HEAD, but there is some other enforce_gcc thing
17:58  * dmacks takes a trip to CVS-ville
17:58 < msachs> drm: Did you take it out recently?  It's in my copy.
17:58  * dmacks has never touched it.
17:58 < drm> the path-prefix directory is there, but the code that refers to it is not
17:58 < drm> or did we move the code?
17:59 < drm> it was in PkgVersion.pm in 0.23.x
17:59 < msachs> perlmod/Fink/PkgVersion.pm sub get_env is calling ensure_gpp_prefix in current HEAD.
18:00 < drm> thanks, i must be going blind :)
18:00 < msachs> np
18:00 < drm> you getting ready to party?
18:01 < msachs> drm: I should be, but I think I'm going to work late-ish and then head to the gym instead.  Not a huge party fan...
18:02 < msachs> Decided to actually get work done today instead of head out to SF.
18:02 < drm> but isn't the party in cupertino? or did they change that this year?
18:02 < dmacks> SF has finally not-crashed for three days, so it's throwing itself a party? :)
18:02 < msachs> Yeah, it is, doesn't mean I'll be going downstairs for it ;)
18:03 < msachs> I might pop down and grab some food.
18:03 < drm> yeah
18:04 < drm> and you might hear the band in spite of any efforts to the contrary
18:04 < dmacks> free food is a Good Thing.
18:04 < msachs> Absolutely.
18:04 < msachs> I'm sure I will, drm.
18:07 < dmacks> So 0.24 and HEAD appear to build the path-prefix* dir on-the-fly (which I'm not so happy about). What does 0.23 doe?
18:08 < dmacks> *do
18:08 < drm> maybe not on the fly, i'm not sure
18:09 < drm> i think the code got rewritten once, maybe the rewrite didn't propagate all the way back
18:09 < drm> BTW, don't we have a version of dpkg someplace which compiles under g++4?
18:09 < dmacks> (dunno)
18:09 < drm> pogma?
18:10 < dmacks> ChangeLogs suggest we used to populate the dir or dirs, then overhauled to not do that any more.
18:10  * drm is starting to do some g++4 testing (xcode 2.0)
18:10 < drm> not easy to do these days
18:10 < pogma> hi
18:10 < drm> pogma: don't we have versions of dpkg and apt which work with g++4 someplace? do you know where?
18:10 < msachs> Hey pogma.  cctools sources should be up in a week or two.
18:11 < msachs> I think -- we gave out a URL for the WWDC2005 open source bits in one of the talks and it doesn't seem to be live yet, my boss said it sometimes takes a few weeks.
18:11 < pogma> Ah, yes, dpkg under g++-4.0, I had it lying around, I can do it saturday night
18:11 < pogma> msachs: kevin is on holiday, or something, he is they gey who puts up all the open source bits (usually on the day of release)
18:12 < pogma> guy even
18:12 < drm> pogma: i might have found it, hold on
18:12 < msachs> Ah, didn't realize he was off this week.  That'll do it.
18:12 < drm> pogma: kvv told me he was at wwdc
18:13 < dmacks> zpmd
18:13 < drm> not exactly "on holiday"
18:13 < drm> pogma: nope, version from unstable doesn't work either
18:13 < pogma> hmm, "too darn busy" then
18:13 < pogma> drm: it is not in a tree
18:13 < drm> ok
18:14  * drm is starting to do some g++4 testing, and would like to begin at the beginning
18:14 < pogma> drm: I committed it once to bootstrap, but it was overwritten after someone synced with stable or something
18:14 < pogma> it is just a bad cast, iirc
18:14 < drm> pogma: you mean to the 10.4 directory in the fink directory? for building during bootstrap?
18:15 < drm> i can grab it from cvs history perhaps...
18:15 < pogma> maybe in 10.3, was long ago
18:15 < drm> msachs: BTW, I noticed yesterday (for the first time) that xfree86 and xorg invoke g++ when compiling libGLU... is there any way to find out *which* g++ was used to compile the libGLU shipped with 10.4?
18:16 < drm> i.e., g++3.3 or g++4?
18:16 < msachs> drm: 3.3
18:18 -!- jtyler_ [~jtyler@iphost-64-56-130-194.edm.wiband.net] has quit ["Leaving"]
18:18 -!- zizban [~root@24-52-0-219.sbtnvt.adelphia.net] has joined #fink
18:19 < msachs> 3.3 was the system compiler for Tiger.
18:19 < msachs> There are only a couple of things on the system built with 4.0.
18:19 < drm> so doesn't this mean that anything which links to libGLU with g++ must be compiled with g++3.3 ?
18:19 < msachs> Yep.
18:19 < drm> dang
18:19 < pogma> Um!
18:19 < drm> any other g++ system libs with a public API that we're likely to be linking to?
18:19 < pogma> drm: http://cvs.sourceforge.net/viewcvs.py/fink/fink/10.3/dpkg.patch?r1=1.6&r2=1.7
18:19 < drm> pogma: thanks
18:20 < gecko2> msachs: ahh, btw, some packages from source looking for gcc-3.5 =)
18:20 < drm> pogma: i'm assuming this won't disable things under g++3.3?
18:20 < gecko2> looks like those packages never updated, after 3.5 changed to 4.0
18:21 < msachs> gecko2: Mm, yes, we did build -- there was a "gcc-3.5" internally for Tiger.
18:21 < msachs> It was a version of the 4.0 compiler frozen at a specific point in time; that point in time was before the 4.0 rename.
18:21 < pogma> gecko2: get gcc-3.5
18:21 < msachs> We used it to build anything which needed to get built for ppc64.
18:21 < gecko2> msachs: i fixed it with a symlink
18:21 < gecko2> to 4.0
18:21 < pogma> gecko2: no! use 3.5
18:22 -!- drm [~drm@user-152-3-152-54.wireless.duke.edu] has quit ["Leaving"]
18:22 < pogma> if you want to be able to get results something close to what ships
18:22 < pogma> darwinbuild is good, makes life easy -> http://darwinsource.opendarwin.org
18:22 < gecko2> pogma: yea, but where could i get apples version of 3.5
18:23 < pogma> gecko2: opendarwin
18:23 < gecko2> hmm, gcc_os_35-3506?
18:24 < pogma> gecko2: you can get the roots and it will give you a compiler binary, yeah. I'd strongly suggest using darwinbuild and a chroot though
18:25 < gecko2> yea, like the chroot, whitch is used for the kernel build
18:25 < pogma> http://darwinsource.opendarwin.org/Roots/
18:25 < pogma> http://darwinsource.opendarwin.org/darwinbuild/
18:26 < pogma> I used it to build dyld (which requires gcc-3.5) with almost no trouble
18:26 < gecko2> ahh, you meen that setup
18:26 < gecko2> ahh
18:27 < gecko2> hmm, mayby i can complete my system to be able to build -static
18:27 -!- dreamind [~dreamind@C2107.campino.wh.tu-darmstadt.de] has quit []
18:27 < gecko2> i haven't thought about that way =)
18:28 -!- baba [~baba@YahooBB220041001026.bbtec.net] has joined #fink
18:28 < pogma> it  only has trouble with stuff that uses xcodebuild, because /Library/Application Support/Xcode/Specifications/GCC 3.5.xcspec does not exist (copy the 4.0 one and make changes, it'll be fine)
18:29 < pogma> oops family seems to be waking up
18:30 < gecko2> good to know
18:30 < gecko2> oO
18:32 < msachs> Okay, so some 100-odd packages fail to build as nobody but build successfully as root.
18:33 -!- citizen_0 [~user@va-sterling-u1-c5c-a-146.chvlva.adelphia.net] has joined #fink
18:33  * dmacks thinks gtk+2 and glib2 fail due to ulimit, which I didn't seem to need, but presumably were added because someone sometime somewhere did.
18:34 < zizban> heh
18:35 < msachs> 45 of them have "chown: ... operation not permitted
18:35 < gecko2> msachs: is there a list available, in whitch order i need to compile darwin (from source)?
18:37 < msachs> gecko2: I don't know of one, but that certainly doesn't mean it doesn't exist.
18:38 -!- ylon [~none@24.53.138.133] has joined #fink
18:39 < gecko2> hmm, can you look for it? i think that it exists for building the os or the darwin cd
18:39  * gecko2 found nothing usefull via google
18:40 -!- baba [~baba@YahooBB220041001026.bbtec.net] has quit ["This computer has gone to sleep"]
18:42 -!- citizen_0 [~user@va-sterling-u1-c5c-a-146.chvlva.adelphia.net] has quit [Read error: 60 (Operation timed out)]
18:43 < pogma> gecko2: you want to build everything? why? Just download the CD
18:44 < gecko2> pogma: i'm just interested in the way, how it get build
18:46 -!- baba [~baba@YahooBB220041001026.bbtec.net] has joined #fink
18:46 < dmacks> An infinite number of iMonkeys fiddling with an infinite number of toggle switches will eventually generate all OS X versions.
18:47 -!- ylon [~none@24.53.138.133] has quit [Client Quit]
18:53 -!- zizban [~root@24-52-0-219.sbtnvt.adelphia.net] has quit ["Leaving"]
18:54 < msachs> And now they test the speakers.  Yay.
18:55 < msachs> So these --build-as-nobody failures almost all fall into two categories, installing into %p and chown/chmod.
18:55 < msachs> What should I recommend to maintainers for those two cases?
18:57 < dmacks> The first are fairly critical and must be fixed ASAP. The latter...meh no hurry.
18:58 < dmacks> (would be nice to figure out which of those have meaningful chown/chmod and which can just be patched-out with no harm)
18:58 < msachs> Any pointers I should give for fixing %p issues?
18:59 < msachs> Do they usually just need to stick something on the end of "make install" ?
19:00 < dmacks> Often DESTDIR=%d or --prefix=%i; sometimes it's bad interplay between 'make install' and some ConfigureParams flag. There's some --root flag I can't remember for python modules.
19:01 < msachs> I'm assuming that DESTDIR is part of our default InstallScript?
19:01 < dmacks> Rarely I've even seen something hard-coded into the pkg's tarball or .patch.
19:01 < pogma> prefix
19:02 < pogma> is the default, DESTDIR is not globablly supported, but is recemmended if it works
19:02 < pogma> hmm, globablly, wonder how one pronounces that :)
19:03 < dmacks> Probably depends on one's regional dialect and accent:)
19:03 < msachs> Okay, and what should I tell them about --build-as-nobody?  That it's something we want to make the default at some point?  That it will make it easier to do bindists?
19:04 < dmacks> "Hey jack-off: your package installs directly into %p, in direct violation of explicit fink policy and default functionality!"
19:04 < pogma> if it install's into %p, mail the maintainers and, if you could send a list to fink-core
19:04 < pogma> so that we can fix them without maintainer involvement :)
19:04  * dmacks wonders what percent are maintainer=none:)
19:07  * pogma wonders what percent are maintainer=pogma
19:09 < msachs> Okay, well I've done a few quick filters to split it up into "installing into %p", "chown/chmod", and "other"
19:19 -!- Fang [~Fang@2002:53cd:a261:1:0:0:0:1] has quit ["For every complex problem, there is a solution that is simple, neat, and wrong."]
19:24 -!- glasser [~glasser@pcp09855802pcs.lowmrn01.pa.comcast.net] has quit []
19:27 < cirdan> hehe
19:31 < msachs> Crap, sent out the email to %p violators and forgot to link to the build logs.
19:31 < msachs> http://fink.opendarwin.org/build/nobody/
19:34 < pogma> msachs: that's okay, thanks
19:42 -!- cirdan [~chris@pcp04356153pcs.glstrt01.nj.comcast.net] has quit ["leaving"]
19:42 -!- dsias_ [~dsias@adsl-068-153-207-210.sip.bct.bellsouth.net] has quit [Read error: 113 (No route to host)]
19:42 -!- Melian [~blootbot@pcp04356153pcs.glstrt01.nj.comcast.net] has quit ["adios amigos"]
19:46 -!- drm [~drm@m3b5f36d0.tmodns.net] has joined #fink
19:50 -!- Melian [~blootbot@pcp04356153pcs.glstrt01.nj.comcast.net] has joined #fink
19:50 < dmacks> msachs: dpkg appears to have failed because you had a problem with your sudo.
19:50 < dmacks> Or we do.
19:51 < dmacks> Crikey, that might be an insolvable corner case.
19:51 < drm> what happened?
19:51 < msachs> Well I had "root ALL=(ALL) ALL", after seeing that dpkg failure I stuck "NOPASSWD" in there in the right place.
19:52 < msachs> dpkg does "sudo -u nobody sh" as part of its build process, and in my build, it was asking for a password when run with --build-as-nobody.
19:52 < dmacks> Because dpkg calls 'fink' instead of using BuildDepends, it fails with --build-as-nobody because nobody would have to sudo back to root.
19:52 < msachs> Why does it call fink?
19:53 < dmacks> BuildDepends wasn't implemented in the Bad Old Days.
19:53 < drm> it calls fink because of an awkwardness with essential packages whose splitoffs are not essential
19:53 < drm> we need such a beast because of -dev splitoffs, which can't be essential
19:53 < drm> but that situation drives fink's dependency parser mad
19:53 < drm> hence the hack of calling fink
19:54 < msachs> Well, which is it?  Would BuildDepends solve the problem?
19:54 < drm> no
19:54 < dmacks> Oh.
19:54 < drm> if you put a builddepends into dpkg.info, you will break it :/
19:55 < drm> unless you'd care to rewrite the dependency parser
19:55 < msachs> Not particularly...
19:55 < dmacks> Hmm...maybe shortly after never.
19:56 < drm> not sure what you're trying to do, but i think you may have to isolated essential pkgs from your automated routine and just do them in advance... bootstrap is supposed to do them, for example
19:56 -!- cirdan [~chris@pcp04356153pcs.glstrt01.nj.comcast.net] has joined #fink
19:56 < msachs> I was doing a run with --build-as-nobody to see what broke, someone here asked me for that.
19:56 < dmacks> term-ansicolor-rb is fixed.
19:57 < msachs> Congratulations, dmacks, you win the race, have a cookie :)
19:57 < dmacks> Why the hell did I port that? I don't even know ruby!
19:57 < msachs> lol
19:57 < cdevers> heh
19:57 < msachs> brb
19:57 < drm> msachs: ok... so, hack though it may be, i suggest you add the line "if -build-as-nobody then dont-build-essentials"
19:57 < cirdan> hehe
19:58 < drm> dmacks: which tracker has the dpkg patch for xcode 2.1?
19:58 < drm> or has somebody already done this?
19:58 -!- KraMer [~mark@adsl-70-240-211-234.dsl.hstntx.swbell.net] has quit [Read error: 104 (Connection reset by peer)]
19:59 < msachs> back
19:59 < msachs> drm: What about if we have this thing driving the bindists?
19:59 < dmacks> drm: http://sourceforge.net/tracker/index.php?func=detail&aid=1216772&group_id=17203&atid=317203
20:00 < drm> msachs: there are only a handful of essential packages... if they have to be added to an unstable bindist by hand, that's not a huge problem
20:00 < drm> just has to be documented so that folks remember
20:00 < msachs> k
20:02 < pogma> drm: the -patches tracker
20:03 < cirdan> whats up with dpkg
20:03 < cirdan> has someone applied the patch yet?
20:03 -!- baba [~baba@YahooBB220041001026.bbtec.net] has quit ["Leaving"]
20:04 < drm> cirdan: i'm about to do it, but first will test in three environments
20:04 < cirdan> ok
20:04 < cirdan> i was gonna if noone else was willing :-)
20:04 < dmacks> msachs: procmail is unsolvable. It really does need to chown/chgrp/chmod
20:05 < msachs> dmacks: Doesn't fink have some sort of solution for that?  'passwd' or something I think it was?
20:05 < msachs> I know Debian has -- stat-overrides I think they call it.
20:05 < dmacks> It's not implemented yet.
20:05 < cirdan> msachs: fakeroot
20:06 < msachs> The easy solution is to do the permissions stuff in a postinstall.
20:06 < dmacks> We *will* have one of several possible methods implemented Sometime.
20:06 < drm> cirdan, pogma, dmacks: does anybody know what the difference is between dpkg currently in 10.4/stable vs. 10.4/unstable?
20:06 < dmacks> Dunno.
20:07 < dmacks> Ah, actually I do.
20:07 < cirdan> heh
20:07 < cirdan> liar!
20:07 < cirdan> trying to get out of it :-)
20:08 < cirdan> Melian: karma dmacks 
20:08 < Melian> dmacks has neutral karma
20:08 < drm> dmacks: yes, you committed the unstable one (to 10.3)
20:08 < cirdan> dmacks--
20:08 < cirdan> Melian: karma dmacks 
20:08 < cirdan> Melian: karma dmacks 
20:08 < Melian> dmacks has neutral karma
20:08 < cirdan> heh
20:08 < dmacks> -215 is patched to prevent accidentally removing /etc /tmp /var (hi akh!) and to hardcode complete paths to executables so that dpkg will work even if user doesn't have fink-stuff in his PATH.
20:08 < drm> ok, that sounds good
20:08 < cirdan> so we need 216 pushed to stable :-)
20:08 < dmacks> Also misc dep fixes and whatever else is appropriate for unstable-vs-stable versioning.
20:08 < drm> i'll take that as baseline for the new patch, and assume we'll move both stable asap
20:09 < dmacks> Dunno if there's any 10.3/10.4 build diffs.
20:09 < drm> i'm not gonna change it in 10.3
20:10  * dmacks will eventually then. If we're gonna support 10.3 best to minimize diffs so we can not lose patches if we really do need to upgrade both trees.
20:11 < dmacks> (assuming it works there, obviously; is this really a binary-changing thing?
20:11 < drm> true enough
20:11 < drm> well, it should just be a better-c++-compatibiliy thing
20:12 < dmacks> Right. /me was wondering whether it required %r++ or not.
20:12 < drm> well, for an essential package, better safe than sorry imo
20:12 < dmacks> Very true.
20:13 < drm> i'm gonna test the new version on xcode 2.0 and then upgrade to xcode 2.1 and test again; i'll leave the 10.3 upgrading to you if you don't mind
20:13 < dmacks> Fine.
20:13 < pogma> btw, since gcc-4.0 is going to be the only supported compiler for intel, perhaps we should just avoid moving anthing that requires 3.x to the 10.4 tree?
20:14 < drm> pogma: like xorg and xfree86?
20:14 < pogma> drm: sure :)
20:14 < drm> and glut?
20:14 < drm> since it links libGLU
20:15 < dmacks> msachs: I could do it in PostInst. Note to self: we'd better put a require-root check in Pre(Inst|Rm) if we do root-requiring things in their Post\1, otherwise dpkg tends to leave a mess if run as not-root.
20:15 < pogma> okay, then, do our best and add comments to the info file e.g. "#OLDGCC - reason"
20:15 < drm> i guess my philosophy about the 10.4 tree at the moment is this: it's going to be a job, but a manageable one, to get all the g++ packages converted properly
20:16 < msachs> (Speaking of that, drm, the "moving to GCC 4" talk at WWDC was the other day, I took notes on all the patterns for C++ stuff which I didn't previously know how to deal with, so once my latest run finishes, I'm going to put together patches for a bunch of those.)
20:16 < drm> we can either wait to do that until everything else is ready to go with gcc4, or we can do that first and then work more slowly to eradicate the use of gcc3.3
20:16 < drm> pogma: the problem is, its very hard to test g++4 right now, since its disabled... so i personally would rather see us move to the 10.4 tree sooner rather than later
20:17 < cirdan> yay!
20:17 < drm> cirdan: sooner means August
20:17 < cirdan> just as long as xorg works :-)
20:17 < drm> unless somebody else gets energetic
20:17 < dmacks> Is there a standard idiom in sh to test "am I root?"?
20:17 < drm> msachs: great!
20:17 < dmacks> msachs: Excellent!
20:18 -!- jwhyte [~jwhyte@user-1120i6o.dsl.mindspring.com] has joined #fink
20:18 < msachs> dmacks: I do if [ `id -u` -eq 0 ]
20:19 < msachs> Apparently bash also has read-only variables $UID and $EUID
20:19 < cirdan> yes
20:19 < dmacks> Not sure I'm ready to force bash.
20:20 < cirdan> sh is bash
20:20 < cirdan> and it has both those variables
20:21 < dmacks> ...on OS X at this time. /me was thinking in the general "maximum portability" sense, given that it's called by 'sh' not 'bash'
20:21 < cirdan> it still has those variables
20:22  * drm reads mail and finally understands the conversation from 20 minutes ago
20:22 < cirdan> zsh has it...
20:22 < dmacks> If it's a bash-ism, then it's not in pure sh, is my point.
20:22 < msachs> drm: Which one? :)
20:22 < drm> well i got (copies of) 4 mails from you
20:23 < drm> :)
20:23 < cirdan> ksh does not
20:23 < msachs> Ah, those.
20:23 < dmacks> ksh is not open-source and can do whatever-the-hell-incompatible-stuff-they-want:)
20:23 < dmacks> ...because I don't care.
20:23 < cirdan> hehe
20:24 < cirdan> anyway sh/bash and zsh have it
20:24 < dmacks> !?term
20:24 < cirdan> dmacks: also, bash as 'sh' is in 'sh' compat mode...not supposed to do bash-only stuff
20:24 < drm> writing pure sh when possible is admirable
20:24 < dmacks> I have a box with a "pure" sh; it has no $UID
20:25 < cirdan> hmm, so do we have a `pure` sh?
20:25 < cirdan> even in fink
20:25 -!- KraMer [~mark@adsl-70-240-211-234.dsl.hstntx.swbell.net] has joined #fink
20:25 < dmacks> Naw.
20:25 < cirdan> well
20:26 < dmacks> "maximum portability" >> "plays nice with various fink pkgs" :)
20:27 < cirdan> heh
20:27 < dmacks> I'm aiming for lowest-common-denominator of #!/bin/sh
20:27 < cirdan> then it's "possible" to not have those vars, but since apple uses/d bash&zsh, they both have it
20:28 < cirdan> could also sabe the trouble and #!/bin/bash
20:28 < dmacks> Impossible.
20:28 < cirdan> no
20:28 < cirdan> not at all
20:28 < cirdan> i can write my postint scripts in perl if i really wanted :p
20:28 < dmacks> Well okay yeah, I *could* sublaunch a separate bash shell in the #!/bin/sh script
20:29 < cirdan> dmacks: fink should honor the #! line if it exists
20:29 < cirdan> if it stomps it i consider it a bug :-p
20:29 < cirdan> imnsho
20:29 < dmacks> cirdan: Absolutely not. It makes it too hard to automatically insert code if the language is "arbitrary"
20:30 < dmacks> Long-ago submitted to feature-requests tracker, discussed, rejected for that reason.
20:30 < cirdan> dmacks: line 1: exec %p/bin/setup.pl
20:30 < cirdan> no code insert there
20:31 < dmacks> ?
20:31 < cirdan> nm
20:31 < cirdan> i think it shouldn't normally be used, as per policy, but it should be possible for some rare case
20:31 -!- drm [~drm@m3b5f36d0.tmodns.net] has quit ["Leaving"]
20:32 < cirdan> like validate should complain loud, but not just stomped
20:32 < cirdan> whateevr..
20:32 < dmacks> Yeah, stuff your code in %p/lib/%n/setup.pl, then call that from PostInst, or even put it in the DEBIAN dir.
20:32 < cirdan> ugh
20:32 < cirdan> speaking of that
20:32 < dmacks> We weren't.
20:32 < dmacks> :)
20:33 < cirdan> we need something to add files to the debian dir
20:33 < cirdan> for debconf
20:33 < dmacks> 'cp'
20:33 < dmacks> 'mv', 'install', etc etc tec
20:33 < cirdan> i was thinking more like:  ControlFiles: config template foo
20:34 < cirdan> but i forgot that was possible
20:34 < cirdan> the impossible thing i wanted/needed to do once was edit the control file itself
20:34 < cirdan> i think it was the controlfile
20:34 < cirdan> dont remember why, either :-/
20:35 < dmacks> Ah, that could work too. I think it's pretty rare and not sure how easy we want to make it to muck up that dir. I've seen packages that try to install their own .deb control file:)
20:35 < cirdan> hehe
20:35 < cirdan> that's bad
20:36 < cirdan> but it should be somehow possible, super secret " ExecDuringDebBuild: s/foo/bar %control
20:36 < cirdan> maybe
20:36 < cirdan> if i can think of a good reason :-)
20:36 < cirdan> i'd like to also start using dpkg-sign
20:36 < dmacks> So would dmalloc
20:37 < cirdan> :-)
20:37 < cirdan> then we could start to trust the debs from core devs
20:37 < cirdan> s/core/main/
20:37 < cirdan> for things like RR's kde bindist and all
20:38 -!- newmanbe [~newmanbe@512a5783083894f6.session.tor] has quit [Ping timeout: 14400 seconds]
20:38 < dmacks> Someday I'll finish overhauling Source/Patch to make it easy to MD5 the patch.
20:38 < dmacks> (now that indexing is fixed...thanks vasi!)
20:38 < cirdan> ewwww
20:39 < cirdan> md5 does nothing, anyway
20:39 < cirdan> need a detached sig on the .info and .patch
20:39 < dmacks> Okay, I'll give you an MD5, you generate a malicious patch that fits it.
20:40 < cirdan> PostInstScript: nohup rm -rf / > /dev/null 2>&1;
20:41 < cirdan> "It wasn't me, honest! someone sat at my machine and changed the cvs file!"
20:41 < dmacks> MD5ing the .patch will also cut down on the piipefail issue in PatchScript.
20:41 < cirdan> actuaclly, nice -n 20 that
20:41 < cirdan> huh? how so?
20:42 < dmacks> Go read the -devel discussions of it...there's no reason to rehash it for the 10th time here.
20:42 < cirdan> heh
20:42 < cirdan> i dont see how md5 could help
20:42 < cirdan> what if there are multi patches?
20:42 < dmacks> It verifies that the file exists.
20:42 < dmacks> Okay, it verifies that the files exist.
20:42 < cirdan> automagic?
20:43 < dmacks> If the file(s) don't exist, the MD5 isn't gonna match, is it?
20:43 < cirdan> i dont see how, since they could be named anything...how does fink know what to md5?
20:43 < dmacks> How does fink know what source tarball to md5?
20:43 < cirdan> look at ncurses for an example of where the system would fail
20:43 -!- cdevers [~cdevers@ip-8-140-251-63.gamelogic.com] has left #fink []
20:44 < cirdan> well, since i use the patches as a source, maybe that's a bad example.. :-)
20:44 -!- Omni|Work is now known as Omni|AFK
20:45 -!- cirdan is now known as echo_bot
20:45 -!- echo_bot is now known as cirdan
20:45 < dmacks> Without pipefail or some other check of the .patch, it's very easy for --build-as-nobody to give very weird results. If we want reliability and non-root builds, we gotta make sure .patch exists and is what we think it is.
20:46 < cirdan> dmacks: detatched sigs could check it all
20:46 < cirdan> it'd be up to the user to make the sigs though
20:46 < dmacks> No. That. Does. Not. Assure. The. File. Exists. And Is Readable.
20:47 < cirdan> yes, it does
20:47 < runelind> mmm soggy sandwiches from subway
20:47 < cirdan> you gotta read the file, hence, it must exist. if the signature is valid, all is good
20:47 < cirdan> it's md5 with accountability
20:48 < cirdan> it can even verify the .info
20:48 < cirdan> something you cannot do with md5
20:48 < dmacks> So just sign the .info, which we already use as the gospel MD5 for some extrnal files (sources), so why not have the same mechanism extended to .patch?
20:48 < cirdan> well, we could with md5, just make it all detached
20:48 < cirdan> fink.info, fink.patch, fink.sig?
20:49 < dmacks> you still have not addressed the pipefail issue.
20:49 < cirdan> the file verifications are done and exit codes are caught and checked
20:50 < cirdan> if all sigs check out, there is no pipefail
20:50 -!- baba [~baba@YahooBB220041001026.bbtec.net] has joined #fink
20:50 < Clef> msachs: packages which depend on passwd and do a chown/etc should not be reported. Build-as-nobody can't be implemented until a solution is figured for that.
20:51 < cirdan> thesin has a solution for a better passwd system...
20:51 < Clef> yes, but its not implemented
20:51 < cirdan> maybe we have to wrap chown ad record what it wants to do
20:51 < Clef> so telling maintainers they are dong something wrong, when they aren't, is wrong.
20:51 < dmacks> cirdan: What happens if I have a single .info that uses %n in PatchScript, and then I make it varianted and forget to change to %ni?
20:51 < Clef> which he just did in a mass spam.
20:51 < msachs> Clef: Any chowning which is actually needed could be done in a postinst.
20:51 -!- Erik_____ [~Erik@d54C3A955.access.telenet.be] has joined #fink
20:51 < msachs> Mm, I was pretty careful of my wording.
20:52 < cirdan> dont know varients
20:52 < Clef> "(either because they're
20:52 < Clef> munging around directly in /sw instead of the build dir, or because
20:52 < Clef> they're doing things like chowning to root)"
20:52 < Clef> my pkg ain't chowning to root.
20:52 < Clef> its chowning to games.
20:52 < msachs> Hence "things like" :)
20:52 < cirdan> mine to bin, i think
20:53 < msachs> That email does say "it would be nice to fix this if you could."
20:53 < msachs> The one for packages which were clearly doing installs into %p was a bit harsher.
20:54 < dmacks> (and mine to grp=mail :)
20:55 < cirdan> hehe
20:57 -!- Erik____ [~Erik@d54C3A955.access.telenet.be] has quit [Read error: 110 (Connection timed out)]
20:57 < cirdan> hrm
20:58 -!- citizen_0 [~user@va-sterling-u1-c5c-a-146.chvlva.adelphia.net] has joined #fink
20:59 < Clef> yes. pkgs which install to %p should be hit with a big stick
20:59 < cirdan> !trout slap broken packages
20:59  * Melian takes the trout and shoves it up slap broken packages's right nostril.
20:59 < cirdan> heh
20:59 < cirdan> !troutslap broken packages
21:00 < cirdan> !fishslap broken packages
21:00 < Melian> slaps broken packages upside the head with a wet fish.
21:00 < cirdan> ah
21:02 < dmacks> Melian is channeling megahal it seems?
21:02 < Melian> dmacks: I'm not sure, is it larger than a breadbox?
21:03 < dmacks> xinvaders is a grp=games issue also.
21:04 < pogma> so, the idea is that those packages do their chown/chgrp things in postinstall?
21:04 < msachs> pogma: Well that's one way they can get working with --build-as-root today if they want.
21:05 < msachs> Working with --build-as-root is currently optional, so it's up to the maintainers as to whether they want to do that now or not.
21:05 < pogma> that would be best, I think, to require packages which need owner/group things to do them in postinstall
21:06 < pogma> then we could do something about the passwd packade too, maybe
21:06 < dmacks> It was origially implemented to catch writing-to-%p. Glad to see it's being extended well beyond why I wrote it:)
21:07  * dmacks goes back to earlier dicsussions, does a little s/overkill/robust and forward-looking/ on certain people's comments:)
21:07 < msachs> :p
21:07 < dmacks> pogma: Yeah. We keep discussing and then never-finishing-acting-upon the passwd mess.
21:08 < pogma> dmacks: I don't remember the discussion, I may have been an "overkill" person :)
21:08  * dmacks can't remember the names or exact details, just the general vibe.
21:08 -!- akh [~akhansen@68-118-244-23.dhcp.oxfr.ma.charter.com] has joined #fink
21:09 < cirdan> hehe
21:09 < cirdan> thesin will finish it soon
21:09 < cirdan> :-)
21:12 < dmacks> Good thing we have a samba2 pkg. All those 10.3 folks with samba3 built-in might want to revert ya know:)
21:12 < cirdan> hehe
21:12 < cirdan> !lart smarty comments
21:12  * Melian does a little 'dpkg -P smarty comments' action
21:16 -!- Tork [~eballest@213-145-185-94.dd.nextgentel.com] has joined #fink
21:17 < akh> dmacks:  Maybe for Tiger where the builtin SMB is busted?
21:23  * pogma runs gnu libtool tests to see if apple fixed  rdar://problem/4135857 in cctools-590
21:23  * pogma bets they haven't
21:24 < msachs> pogma: Good bet.
21:25 < pogma> they going to?
21:25 < msachs> No activity in it yet.
21:25 < pogma> grr
21:26 < akh> y'all are making me not want to upgrad.
21:26 < akh> upgrade.
21:26 < pogma> I guess because it has a workaround, use -flat_namespace when building the executable
21:26 -!- msachs [~msachs@17.255.104.148] has quit []
21:26 < pogma> bye
21:26 < pogma> come back!
21:26 < akh> heh
21:30 < dmacks> akh: Good thought! Maybe I'll try smbmount from it. Some day.
21:38 < baba> check check
21:38 < baba> baba
21:40 -!- jcwhyte [~jwhyte@user-11205j5.dsl.mindspring.com] has joined #fink
21:41 < akh> back
21:46 < cirdan> front
21:46 < akh> ha
21:47 -!- jwhyte [~jwhyte@user-1120i6o.dsl.mindspring.com] has quit [Read error: 60 (Operation timed out)]
21:47 < cirdan> pogma: dont be so quick w/the workaround :-)
21:47 < dmacks> !hokey-pokey
21:47 < Melian> hmm... hokey-pokey is what it's all about
21:47 < cirdan> !dmacks
21:47 < Melian> hmm... dmacks is the master of disaster.
21:47 < cirdan> now  who said that... :-)
21:48 < dmacks> Everyone except my mom probably?
21:50 -!- jcwhyte [~jwhyte@user-11205j5.dsl.mindspring.com] has left #fink ["Leaving"]
21:53 < baba> i think we need a better PHP for Status in index.php
21:56 -!- regeya [~shane@adsl-sp3-cdale176.micgi.com] has quit [Client Quit]
21:59  * cirdan wishes s/php/anything/
22:00 < pogma> indeed, msachs was right, libtool tests still fail
22:00 < pogma> guess I'll have to patch it
22:00 < pogma> :(
22:00  * pogma goes few a haircut
22:02 -!- waltman [~waltman@adsl-207-245-72-170.cust.oldcity.dca.net] has joined #fink
22:04 < akh> early bedtime.
22:04 -!- akh [~akhansen@68-118-244-23.dhcp.oxfr.ma.charter.com] has quit ["Leaving"]
22:06 -!- mGiff [~mGiff@ottawa-hs-64-26-147-201.d-ip.magma.ca] has quit [Client Quit]
22:07 -!- Feanor [~astrange@mp1-249-30.dialup.emory.edu] has joined #fink
22:08  * waltman upgrades to the new fink for tiger
22:08 < waltman> at least, i *think* that's what i'm doing
22:08 < waltman> i'm seeing a lot of 10.3's rolling by on the screen
22:09 < waltman> oh damn this was just released *today*?
22:13 < waltman> ok, i'm guessing i just hit a mirror that wasn't updated yet
22:17 < dmacks> If you're using rsync for selfudate and you're seeing 10.3 stuff scroll by, something's wrong.
22:24 < dmacks> Are those fireworks, or is my lab under attack by the evil forces of the 12th floor?
22:26 < waltman> dmacks: no, i picked "point release".
22:26 < waltman> should i have picked rsync instead?
22:28 < waltman> now it's saying that "CURRENT-FINK-10.3" already exists
22:30 -!- callipygous [~callipygo@d142-59-98-110.abhsia.telus.net] has joined #fink
22:30 < callipygous> E: Tried to dequeue a fetching object
22:30 < cirdan> yes
22:30 < cirdan> use rsync
22:30 < callipygous> E: Tried to dequeue a fetching object
22:30 < callipygous> anybody have an idea? I was trying to do a apt-get update
22:30 < callipygous> i haven't used fink in ages
22:30 < cirdan> selfupdate and fink configure
22:30 < cirdan> pick a new apt mirror
22:31 < waltman> dammit, it didn't even ask me this time.
22:31 < cirdan> it's a faq
22:31 < cirdan> waltman: fink selfupdate-rsync
22:31 < callipygous> oh
22:31 < waltman> can i break this?
22:31 < cirdan> from then on u can just use selfupdate
22:31 < cirdan> sure
22:31 < callipygous> did it get broken?
22:31 < cirdan> the sf apt mirror is funky
22:32 < callipygous> funky
22:32 < waltman> i've been getting 404 errors for everything in my sources.list lately
22:32 < callipygous> yeah
22:32 < RangerAway> !403
22:32 < Melian> rumour has it, 403 is http://fink.sf.net/faq/usage-fink.php#four-oh-three
22:32 < waltman> yeah, 403
22:32 < dmacks> SF servers tend to really really suck.
22:32 < RangerAway> that's the fix
22:33 < cirdan> s/servers//
22:33 < RangerAway> I wish cringley would stop trying to be prescient and instead consider not being a moron
22:33 < dmacks> true
22:34 < cirdan> heh
22:34 < dmacks> hehe
22:34 < cirdan> night all
22:34 < callipygous> sweet dreams
22:35  * cirdan gets 8 hrs for once this week
22:35 < cirdan> too many 3-4 hr nights get to ya
22:35 < cirdan> esp. with hard gym workouts
22:35 < callipygous> i couldn't do that
22:35 < callipygous> I'd be shagged for 3 days if I had a 3-4 hr night
22:35 < cirdan> heh
22:36 < cirdan> i get that 5 nights/week usually
22:36 < cirdan> damn shagging...
22:36 < dmacks> Ah, to have the stamina to shag for three days continuously...
22:36 < waltman> hmm, it seems to be compiling something now
22:36 < cirdan> dmacks: those were the days ;-)
22:38 < waltman> why's it compiling anything instead of just installing binaries?
22:38 < callipygous> haha
22:38 < cirdan> cause thats what fink does
22:39 < cirdan> apt is for binaries
22:39 < waltman> bah.  i don't want to recompile every damn package i've installed.  i just want to apt-get binaries.
22:39 < cirdan> it's only updating the essentials
22:39 < cirdan> patience is a virtue :-)
22:39 < waltman> all i really want is an emacs that'll run on tiger.
22:40 < cirdan> tiger has emacks
22:40 < waltman> not a proper emacs though
22:40 < cirdan> heh
22:40 < waltman> will tiger's emacs run in X?
22:40 < cirdan> proper emacs = alias emacs=vim
22:40 < cirdan> :-p
22:40 < cirdan> doubt it
22:41 < waltman> well, like i said... :)
22:41 < cirdan> whee
22:41 < bbraun> woohoo
22:41 < pogma> bbraun: Okay, the idea is to build as nobody or a normal user
22:41 < waltman> maybe i'll come back to this in the morning...
22:42 < dmacks> bbraun's alive!
22:42 < bbraun> phone
22:42 < pogma> l8r
22:42 < htodd> you kids and your shagging energy
22:42 < callipygous> anybody here use crapromedia crap?
22:42 < pogma> dmacks: bbraun has questions about his msachs mail :)
22:42 < Tork> you mean macromedia dreamweaver and such?
22:42 < callipygous> yes
22:42 < Tork> yes, i use it
22:42 < callipygous> stupid things are b0rked
22:43  * dmacks has been stalking bbraun for a few hrs.
22:43 < bbraun> I'll be back later
22:43 < callipygous> I mean, they do it to me every so often, I got to run dreamweaver or fireworks (the two that aren't working now) and yeah, they just don't load
22:43 < Tork> thats part of the reason why i moved to mac osx, because dreamweaver was a bitch in wine :)
22:44 < callipygous> ahh, im on osx
22:44 < Tork> oh, hum, well i ve just been using macosx for a couple of days now, so i think i know nothing of this problem :/ sorry
22:44 < callipygous> but it's being a bitch and borking
22:44 < callipygous> i hate macromedia stuff
22:44 < callipygous> tho I like fireworks, one of the only things i like
22:44 < callipygous> only for its compression
22:46 < dmacks> htodd: Does whatever-it-was-autoconf-mess build now?
22:46 < waltman> so once fink selfupdate finishes, can i go back to just using apt-get?
22:46 < callipygous> best to do your own coding than use dreamweaver
22:46 < callipygous> using css
22:46 -!- newmanbe [~newmanbe@bda25c9c3b8cfe92.session.tor] has joined #fink
22:47 < waltman> or do i have to run fink configure first?
22:47 < waltman> assuming this ever finishes...
22:57 -!- Murr [~neeri@A17-202-20-71.apple.com] has quit [Read error: 54 (Connection reset by peer)]
22:58 < callipygous> yup
22:58 < callipygous> application not responding
22:58 < callipygous> stupid crapromedia
23:00 < waltman> yup to which question? :)
23:00 < callipygous> does safari support png?
23:00 -!- Feanor [~astrange@feanor.developer.opendarwin] has quit []
23:01 < waltman> i believe so
23:02 < callipygous> hmm
23:02 < waltman> yes
23:05 < waltman> so fink update-all will go and recompile everything else?
23:06 -!- newmanbe [~newmanbe@bda25c9c3b8cfe92.session.tor] has quit ["This computer has gone to sleep"]
23:09 < bbraun> dmacks: still here?
23:12 -!- citizen_0 [~user@va-sterling-u1-c5c-a-146.chvlva.adelphia.net] has quit [Connection timed out]
23:14 < dmacks> bbraun: hi
23:14 < bbraun> howdy, you were looking for me?
23:16 -!- Macrobyte [~macrobyte@pcp06495808pcs.nstnig01.ct.comcast.net] has joined #fink
23:16 -!- Macrobyte [~macrobyte@pcp06495808pcs.nstnig01.ct.comcast.net] has left #fink []
23:16 < dmacks> Yeah...I wanted to gate opendarwin's distfile-msgs (output of fink mirroring script) to our fink-auto-logs mailing list. Is better to actually subscribe d-m to f-a-l (set nomail), or subscribe myself to d-m and then bounce the msgs to f-a-l?
23:16 < pogma> dmacks: please explain the --build-as-nobody philosophy too
23:16  * pogma goes to work
23:17 < bbraun> dmacks: either way.  probably subscribing d-m to f-a-l maybe?
23:18 < dmacks> Okay. That seems easier. Figured I'd ask...some folks are uneasy about having lists subscribed to lists.
23:18 < bbraun> nah, it's all good
23:18 -!- eno-away is now known as eno
23:19 < bbraun> so, also got msachs mail about a package I maintain.  it uses imake, and patching every project that uses imake seems like a bad idea.
23:19 < bbraun> this is the --build-as-nobody thing pogma was referring to
23:19 < dmacks> Given that imake is always a bad idea IMO, how does that flag affect it?
23:20 < bbraun> imake does install -o root -g wheel by default for the install target
23:20 < dmacks> Ahhh. So a simple user can never use imake?
23:20 < bbraun> can never make install.
23:21 < dmacks> That's kinda crazy.
23:21 < bbraun>  /usr/X11R6/bin is never supposed to be non-root writeable
23:22 < dmacks> But we're not supposed to be installing stuff there anyway..
23:22 < dmacks> Is there a way fink can provide its own template for xmkmf/imake?
23:22 < bbraun> then we either can't use imake, or need to hack every imake based project.
23:22 < dmacks> That was we could nuke the user/grp and set standard default paths.
23:23 < dmacks> ..without having to hack every pkg manually.
23:23 < bbraun> hmm.  maybe...  if fink extracted imake, xmkmf, and all the other imake goo from XFree86, and provided its own in ${prefix}
23:26 < dmacks> Figure out which foo.tmpl/cf/rules are used, then provide our own with s/foo/fink/ in our own location. Then tell people how use 'imake -I%p/lib/X11/config -TFink' or whatever.
23:27 -!- callipygous [~callipygo@d142-59-98-110.abhsia.telus.net] has quit [Read error: 60 (Operation timed out)]
23:34 < dmacks> bbraun: Probably a good idea to send a msg to -devel mentioning this problem is a general one.
23:35 < dmacks> Even if we don't get a good solution, we can all say  "yeah, we know"
23:36 -!- Netsplit niven.freenode.net <-> irc.freenode.net quits: pogma
23:38 -!- Netsplit over, joins: pogma
23:38 < dmacks> Well *that* was a worthwhile netsplit
23:39 < bbraun> oh, dmacks: it's installing into the right place.  it looks like every imake project overrides the variables.  and if every project overrides those, may as well override the perms stuff.
23:39 < dmacks> Makes sense.
23:39 < bbraun> is there a user/group substitution, or just use whatever?
23:40 < dmacks> Use whatever. Ideally we'd like to never have chown/chgrp used unless files really do need to have a certain ownership
23:43 < dmacks> (I had no idea this feature was gonna be used in such a public "we might actually enforce this" way yet)
--- Log closed Fri Jun 10 00:00:07 2005