--- Log opened Wed Jun 15 00:00:31 2005
00:01 -!- asparagui [~billy@mo-71-1-109-215.dhcp.sprint-hsd.net] has quit ["compiling"]
00:07 -!- regeya [~shane@adsl-sp3-cdale176.micgi.com] has quit [Nick collision from services.]
00:09 -!- regeya_ [~shane@adsl-sp3-cdale176.micgi.com] has joined #fink
00:12 -!- You're now known as RangerAway
00:21 < knghtbrd> about the size of the average net gamer, apparently  =)
00:28 -!- KraMer_ is now known as KraMer
00:44 -!- eno-away is now known as eno
00:48 < dmacks> Well that's nice...http://prdownloads.sourceforge.net/libytnef/libytnef-1.5.tar.bz?download
00:53 < dmacks> throws an endless loop of "file could not be found" with a "go back" link that takes one to the same page.
00:58 -!- cirdan_ [~chris@pcp04356153pcs.glstrt01.nj.comcast.net] has joined #fink
00:58 -!- cirdan [~chris@68.45.134.216] has quit [Read error: 54 (Connection reset by peer)]
01:14 < dmacks> cirdan_?
01:14 < knghtbrd> dmacks: SF sucks  ;)
01:14 < dmacks> knghtbrd: True true.
01:15  * knghtbrd wishes fink didn't depend on SF for its files  =/
01:16 < dmacks> That particular one isn't even in part of Fink's site. *If* we could get a verified copy, we could host it ourselves. But SF is so broken even FinkMirrors can't find the file to mirror it!
01:18 -!- sugoi [~Sugoi@71-35-125-2.tukw.qwest.net] has joined #fink
01:18 < cirdan_> dmacks: 
01:18 -!- cirdan_ is now known as cirdan
01:18 < sugoi> if one forgets his passwd ... can an install disc help reset it w/o reinstalling everything?
01:19 < cirdan> yes
01:19 < sugoi> for macosx  ... =) .. i didnt know where else to ask
01:19 < cirdan> if you convince the cd you are really u, and u know the scecret handshake ;-)
01:19 < sugoi> haha
01:20 < sugoi> is it an option? in all honesty.. this isnt me, its my buddy calling for help (he doesnt even know what irc is, let alone the power of google)
01:20 < cirdan> u can
01:20 < cirdan> ues
01:20 < cirdan> it's not hard
01:20 < sugoi> an option when the cd starts up that is
01:20 < sugoi> ok
01:20 < sugoi> ill tell him
01:22 < dmacks> cirdan: debconf-i18n's deps look pretty suspicious.
01:22 < dmacks> Files field too.
01:26 < dmacks> Is that pkg really portable enough to auto-sense perl5.8.1 vs perl5.8.4, and whether to use Locale::Gettext vs Text::Iconv? And do I really need perl5.8.1/5.8.4 on Tiger?
01:27 < dmacks> It's just manpages and .mo files, no? Why does it need any perl at all?
01:28 < dmacks> Also triggers two validator checks (and by eye I see one is a general problem in the SplitOffs as well)
01:30 < dmacks> It appears many/most debian mirrors don't have the file yet, but I assume it'll propagate eventually?
01:35 < cirdan> dmacks: i fixed it
01:35 < cirdan> it should be anyway
01:35 < cirdan> it's an older release...
01:35 < cirdan> bb in the morning
01:36 < dmacks> Okay:)
01:36 < cirdan> tkae a lok at the debian package...
01:36 < cirdan> similar deps and stuff :-p
01:37 < cirdan> it most likely doesnt meet strict policy
01:37 < cirdan> since that stuff is new
01:37 < cirdan> have some fun w/it if u feel bored
01:37 < cirdan> :-p
01:37 < dmacks> Fink!=Debian...they can make whatever mistakes and cruftification they like:)
01:38 < dmacks> I'll look maybe...not now though...gotta go watch Family Guy
01:38 -!- dmacks is now known as dmacks_away
02:00 -!- broeken [~broeken@hst32165.phys.uu.nl] has joined #fink
02:06 -!- Omni|Work is now known as Omni|Sleep
02:07 -!- Omni|Sleep [~hopper@Omnifarious.sustaining.supporter.pdpc] has left #fink []
02:21 -!- eno is now known as eno-zzz
02:28 -!- theid is now known as theid_away
02:42 -!- kane_ [~kane@perl.xs4all.nl] has joined #fink
03:34 -!- Fang [~Fang@2002:56c1:e583:1:0:0:0:1] has joined #fink
04:02 -!- sugoi [~Sugoi@71-35-125-2.tukw.qwest.net] has left #fink ["Leaving"]
04:25 -!- Feanor [~astrange@feanor.developer.opendarwin] has quit [Read error: 145 (Connection timed out)]
05:11 -!- rudy [~rudy@ip68-224-184-250.hr.hr.cox.net] has joined #fink
05:11 -!- chris01 [~chris01@212.126.165.246] has joined #fink
06:46 -!- Feanor [~astrange@mp1-248-96.dialup.emory.edu] has joined #fink
06:51 -!- rudy [~rudy@rudy.growl] has quit []
07:57 -!- canllaith [~dessa@jhall.developer.kde] has joined #fink
07:59 -!- akh [~akhansen@ldx3.psfc.mit.edu] has joined #fink
08:00 -!- canllaith [~dessa@jhall.developer.kde] has left #fink []
08:07 < knghtbrd> hmm, fink's postfix seems not to be working right  =/
08:09 < broeken> does fink use the proxy information in the preference pane to set/detect proxies?
08:11 < akh> broeken: I think so.
08:11 < akh> knghtbrd: Send but not receive?
08:12 < broeken> akh, so what's the use of settings proxies in fink(commander)?
08:12 < akh> Let me check...
08:13 < knghtbrd> akh: can connect to port 25, but it doesn't do anything.
08:13 < knghtbrd> no response
08:13 < akh> I've had problems with it myself.
08:14 < knghtbrd> yeah, problems like this are why went and compiled fink's postfix
08:14 < knghtbrd> so if you find the problem, lemme know  =)
08:16 < akh> broeken: On second thought, maybe fink doesn't use the prefpane stuff.  I've never used proxies here myself, so I don't know.
08:16  * akh was conflating with "cvs-proxy"
08:16 < broeken> there's a fc bug with the proxies and at first i thought it was a fink bug
08:17 < akh> knghtbrd: the maintainer seems to be accessible if you haven't contacted him.
08:17 < broeken> (un)fortunately it isn't
08:17 < akh> That's bad.
08:27 < knghtbrd> conflicting?
08:27 < knghtbrd> er, I assume not me  =)
08:28 < knghtbrd> akh: unfortunately, I now cannot send email
08:28 < knghtbrd> email sent to myself seems not to get delivered
08:28 -!- baba [~baba@YahooBB220041001026.bbtec.net] has joined #fink
08:29 < akh> hmm...
08:30 < knghtbrd> this might be due to NAT issues and the fact that postfix doesn't know that it needn't actually send anything anywhere
08:30 < knghtbrd> but even if that's the problem, no email sent could get a reply
08:31 < knghtbrd> my only option, I suppose, is to try and make Apple's postfix work again (given the effort required to get it to STOP RUNNING even after you tell launchd to stop, unload, and disable it..)
08:35 < akh> ugh
08:36 < knghtbrd> yeah, stop means stop, unless you're launchd
08:37 < knghtbrd> if you're launchd, it means "ignore me"
08:37 < knghtbrd> and of course kill means restart
08:37 < knghtbrd> even if you have already stopped, unloaded, and disabled
08:37 < akh> It's a "feature" ;-)
08:37 < knghtbrd> at some point you rename the daemon and then kill it
08:38 < knghtbrd> restart that, bitch!  ;)
08:38 < knghtbrd> (it promptly gave up at that point.)
08:39  * akh will try fink's again, probably.
08:39 < akh> And nag the maintainer mercilessly until I get it working.
08:41 < broeken> akh, wouldn't it be nice if fink did use the prefpane for proxies?
08:41 < akh> Probably.  
08:41 < akh> I'm not sure exactly why it doesn't.
08:42 < akh> Maybe it's a carryover from 10.0
08:42 -!- ylon [~none@24.53.138.133] has quit [Client Quit]
08:45 < knghtbrd> actually, I think I know what I'm gonna do
08:45 < knghtbrd> will daemonic protest if I make one of its startupitems just go away?
08:52 < akh> Dunno--you could always use "daemonic disable" first.
09:00 < knghtbrd> did that after I cp -rp'd
09:00 < knghtbrd> kinda pointless, the fink StartupItem made by daemonic is pretty lame
09:00 < knghtbrd> much better would be to just use the one Apple made for Panther
09:01 < knghtbrd> yes, you can poke at master directly.  You're not supposed to, though.
09:02 < knghtbrd> Wietse has fits when people fiddle about with that stuff (especially when it doesn't work and they come crying to him about it)
09:03 -!- dreamind [~dreamind@C2107.campino.wh.tu-darmstadt.de] has joined #fink
09:04 < knghtbrd> his viewpoint (and I think I agree with him) is that he has created a safe/sane wrapper program that verifies that everything works and starts master appropriately after setting up its environment.  Doing it any other way breaks certain intended features.
09:04 -!- Feanor [~astrange@feanor.developer.opendarwin] has quit [Read error: 110 (Connection timed out)]
09:04 < knghtbrd> for example, postfix can be run in chroot if you're extra paranoid
09:05 < knghtbrd> The combination of (usually) /usr/sbin/postfix and /etc/postfix/postfix-script facilitate doing this or not doing it, as you prefer.
09:16  * akh just follows orders.  ;-)
09:21 < knghtbrd> hmm, is there a canned cmdline PackageMaker?
09:22 < akh> That, I don't know.  
09:22 < knghtbrd> if not, some silly person (I hear knghtbrd is a silly person) should go out and make one.
09:26 -!- You're now known as RangerRick
09:26 < knghtbrd> answer: yes, /Developer/Tools/packagemaker (which is a symlink to /Developer/Applications/PackageMaker.app/Contents/MacOS/PackageMaker)
09:28 < akh> Ah--I wondered.
09:29 < akh> Now, if only there were a command-line way for me to switch my Airport card on and off...
09:33 < akh> !lart Tiger's slow dlsym
09:33  * Melian takes a big bite out of Tiger's slow dlsym's jugular vein
09:33 < akh> (unrelated to Airport)
09:33 < pogma> akh: 10.4.2 will fix it
09:35  * akh hopes so.
09:35 < baba> hmm NeoOffice/J doesn't run on 10.4 (not fink related)
09:36 < akh> baba:  it does for me.
09:38 < baba> give me ur Mac, then
09:39 < akh> Did you install it after installing Tiger?
09:40 < akh> (I installed it before)
09:42 -!- silas- [~mccune@ford.math.ohio-state.edu] has quit ["leaving"]
09:46 < baba> yes
09:46 < baba> i'll stick to KOffice
09:50 -!- newmanbe [~xchat-ssl@f2ef3500496f6760.session.tor] has joined #fink
09:51 -!- broeken [~broeken@hst32165.phys.uu.nl] has quit []
09:53 -!- Macrobyte [~macrobyte@pcp06495808pcs.nstnig01.ct.comcast.net] has joined #fink
09:54 < knghtbrd> !alicekill launchd
09:55 < akh> baba:  Really odd, then.  Is this 1.1RCpatch8?
09:55 < knghtbrd> aww  =(
09:55 < newmanbe> Melian?
09:55 < akh> !lart launchd
09:55  * Melian brandishes Excalibur! "With this sword, I vanquish thee, launchd!" and lops off launchd's head
09:56 < akh> !comfort knghtbrd
09:56 < knghtbrd> that's sufficient  =)
09:56 < baba> akh: 1.1 Release Candidate thing
09:57 < RangerRick> baba: right, but did you install the additional patches?
09:58 < baba> RangerRick: nope
09:58 < RangerRick> bet that would help...
09:58 < baba> alright
09:58 < baba> that's more than me in beer
09:59 < akh> heh
09:59 -!- TheSin [~TheSin@iphost-64-56-130-194.edm.wiband.net] has joined #fink
10:00 < newmanbe> Melian appears to be tired, not even able to comfort people.
10:00 < Melian> newmanbe: I give up, what is it?
10:00 < newmanbe> !comfort knghtbrd
10:00 < TheSin> !botsmack
10:00 < akh> Melian needs a nap
10:00 < Melian> akh: I wish you would RTFM.
10:01 < baba> Melian needs a better brain
10:01 < Melian> baba: I give up, what is it?
10:01 < akh> Melian, /me _W_TFM!
10:01 < Melian> akh: bugger all, i dunno
10:01 < baba> bugger..!!!
10:03 -!- Macrobyte [~macrobyte@pcp06495808pcs.nstnig01.ct.comcast.net] has quit []
10:05  * pogma watches baba get arrested for buggering Melian 
10:07 < baba> no i didn't, i mean, i did, but not i didn't mean it, oh please
10:07 < akh> ewwww
10:08 < akh> with a bot no less
10:08 < newmanbe> Fink Project v. baba, newmanbe presiding
10:09 < newmanbe> baba: You are charged with buggering Melian.
10:09 < baba> Melian is guilty!
10:09 < newmanbe> baba: What is your plea?
10:09 < Melian> parse error: dunno what the heck you're talking about, baba
10:09 < pogma> baba: bugger has a couple of meanings, one of which, in case you did not know, is sodomy
10:09 < baba> all fallacy
10:09 < newmanbe> baba: This court find you guilty of buggering Melian.
10:10 < baba> oh no!
10:10 < newmanbe> The community is welcome to submit suggestions for baba's punishment.
10:10 < baba> i'm only an innocent fink packager!
10:10 < RangerRick> yay!  open-source torture!
10:11 < baba> i cannot go to Todai and steal iMac
10:11 < baba> no, never!
10:12 -!- mode/#fink [+o pogma] by ChanServ
10:12 < newmanbe> baba: This court sentences you to find a replacement for ever SourceForge.net resource used by the Fink Project.
10:12 -!- mode/#fink [+o baba] by pogma
10:12  * newmanbe is done.
10:12 -!- mode/#fink [-o pogma] by pogma
10:13 < pogma> I suggest the judged judge the judge :)
10:14  * baba begs
10:14  * baba prays to Goddes of Fink
10:14 < newmanbe> Goddess?
10:14 <@baba> yep
10:14 < newmanbe> Is that Melian?
10:15 <@baba> i dunno
10:15 <@baba> Fink is demale, isn't it?
10:15 < pogma> hmm, not woking out as I planned
10:15 < akh> !fink
10:15 < akh> oh, man--she even forgot that one.
10:16 < newmanbe> Melian: OpenBSD is if you wanted a secure system, you wouldn't add any software.
10:16 < Melian> okay, newmanbe
10:16  * newmanbe requires that baba complete his sentence in six months.
10:17  * pogma wonders if baba realizes that he has ops
10:17 < newmanbe> He could have oped himself.
10:17  * baba is tipsy, very much
10:18 -!- Melian was kicked from #fink by baba [baba]
10:18 -!- Melian [~blootbot@pcp04356153pcs.glstrt01.nj.comcast.net] has joined #fink
10:18 < akh> !fink
10:18  * baba kick Melian kekekke
10:18 -!- Melian was kicked from #fink by baba [baba]
10:18 -!- Melian [~blootbot@pcp04356153pcs.glstrt01.nj.comcast.net] has joined #fink
10:19 -!- Feanor [~astrange@mp1-248-35.dialup.emory.edu] has joined #fink
10:19 < akh> Kicking a mentally deficient bot--that's harsh.
10:19 <@baba> hi Feanor 
10:19 < Feanor> hi
10:19 < newmanbe> Hi.
10:20 -!- mode/#fink [+b *!*blootbot@*.glstrt01.nj.comcast.net] by baba
10:20 <@baba> shut up Melian 
10:20 <@baba> Melian: are you happy now?
10:20 < newmanbe> megahal: Is baba crazy?
10:20 < megahal> newmanbe: Tony: i'm telling you, that was crazy.
10:21 <@baba> ...
10:21 < newmanbe> Hmm, was megahal reset again?
10:21 < Feanor> yes
10:24 < Feanor> its initial script is http://db.gamefaqs.com/console/ps2/file/xenosaga_script.txt and it takes a while for it to stop quoting it
10:25 -!- Macrobyte [~macrobyte@pcp06495808pcs.nstnig01.ct.comcast.net] has joined #fink
10:26 -!- Macrobyte [~macrobyte@pcp06495808pcs.nstnig01.ct.comcast.net] has left #fink []
10:31 < newmanbe> That person who wrote that script had too much time on their hands.
10:31 < newmanbe> Or transcribed it.
10:31  * newmanbe goes to investigate a computer that just turned on its monitor.
10:32 < Feanor> gamefaqs is made entirely out of people with too much time on their hands
10:32 < bbraun> the intarweb is made out of people with too much time on their hands
10:32 -!- jtyler [~jtyler@iphost-64-56-130-194.edm.wiband.net] has joined #fink
10:33 < TheSin> hey jtyler 
10:34 < jtyler> hey TheSin ! :D
10:35 < pogma> bbraun: and irc is made up of...?
10:35 < bbraun> irc is part of the intarweb, of course.  =)
10:37 < pogma> sigh, do we have a docbook2man binary anywhere? What pacakge?
10:38  * pogma tries docbook-utils
10:38 < newmanbe> IRC is not part of the intarweb.
10:38  * newmanbe does not know how many times he has had to tell bbraun that.
10:38 < newmanbe> It is part of the Internet.
10:39 < bbraun> I can irc from the intarweb, therefore it is part of teh intarweb.
10:39 < newmanbe> How so?
10:39 < bbraun> ?
10:39 < newmanbe> How can you IRC (not a verb) from the intarweb?
10:40 < bbraun> they have those funky irc client webapp thingies.
10:41 < newmanbe> Those are downloaded via the intarweb, but use the Internet.
10:41 < bbraun> my web browser is my view to the intarweb, and it is in my browser, therefore must be part of the intarweb.
10:42 < newmanbe> Some web browsers can go the the intargopher, does that mean it is part of the intarweb?
10:42 < bbraun> yes.
10:42 < bbraun> everything is part of the intarweb
10:42 < bbraun> *everything*
10:42 < pogma> is bbraun part of the intarweb?
10:43 < bbraun> yup
10:43 < bbraun> i'm on irc, which is part of the intarweb
10:43 < newmanbe> Fink Project v. bbraun, newmanbe presiding
10:43 < pogma> the proof seems somewhat circular, but hey, okay :)
10:43 < newmanbe> bbraun: You are accused of being stubborn.
10:43 -!- eno-zzz is now known as eno-away
10:44 < newmanbe> bbraun: What is your plea?
10:44 < bbraun> um.
10:44 < bbraun> guilty!
10:44 < newmanbe> bbraun: This court accepts your plea.
10:44 < newmanbe> bbraun: You are sentenced to allowing the Fink PDB to be moved to sancho.opendarwin.org.
10:45 < bbraun> ok.  what's the holdup?
10:45  * newmanbe shrugs.
10:45 < pogma> bbraun: same as usual, it isn't quite broken enough for anyone to expend the effort
10:46 < bbraun> hehe
10:46 -!- mode/#fink [+b #fink!*@*] by baba
10:47  * newmanbe contemplates changing his nick to #fink.
10:47 < pogma> baba: stop dinking or deop yourself :)
10:47 < pogma> drinking even
10:47 -!- mode/#fink [-b #fink!*@*] by baba
10:47 <@baba> heh
10:47 < pogma> mayhaps /me should stop drinking too :)
10:47 -!- mode/#fink [-b *!*blootbot@*.glstrt01.nj.comcast.net] by baba
10:48 < bbraun> stop drinking?  clearly you haven't had enough
10:48 -!- mode/#fink [-b *!*Nexus330@*.168.29.124.mad.wi.charter.com] by baba
10:48 < pogma> heh, baba unbans people he didn't ban
10:48  * baba opens another botthle
10:49 -!- mode/#fink [+b *!*Nexus330@*.168.29.124.mad.wi.charter.com] by baba
10:49 < akh> Wonder who/what that is--it seems always to be banned.
10:49 < Feanor> be careful or you might unban that guy that used to argue politics all the time
10:50 -!- regeya_ [~shane@adsl-sp3-cdale176.micgi.com] has left #fink ["Leaving"]
10:50 <@baba> he didn't know how to change identd or something?
10:50 < pogma> yeah, can't unban that bbraun person :)
10:51 < bbraun> yeah, don't unban bbraun, he's a dork
10:51 -!- baba [~baba@YahooBB220041001026.bbtec.net] has quit [Remote closed the connection]
10:51 -!- baba [~baba@YahooBB220041001026.bbtec.net] has joined #fink
10:52 < baba> safer without op
10:53  * newmanbe adjourns the Court of Fink Project.
10:53 < akh> We're all safer when baba doesn't have @.  ;-)
10:53  * baba kicks akh
10:54 -!- akh [~akhansen@ldx3.psfc.mit.edu] has quit [Read error: 131 (Connection reset by peer)]
10:57 -!- akh [~akhansen@ldx3.psfc.mit.edu] has joined #fink
10:57 < baba> wb akh in danger
10:58 < akh> grrr...stupid GWOD (gray window of death)
11:03 < akh> and I was in the middle of a fscking KDE update, too.   :x 
11:07  * akh sent a report on, of course.
11:09 < RangerRick> d'oh
11:12 < baba> kde! kde!
11:12 < newmanbe> !kde
11:12 < akh> I think it was just kdeweb3
11:13 < akh> Rather than fighting the dependency engine for my Tiger upgrade, I decided to update manually starting from the bottom of my "fink list"
11:13 < akh> (umbrello)
11:13 < akh> That way I don't have to scroll back.
11:15 < RangerRick> hehe
11:15 < RangerRick> the alphabet strikes again!
11:15 -!- baba [~baba@YahooBB220041001026.bbtec.net] has quit ["This computer has gone to sleep"]
11:22 < runelind> greetings
11:23 < akh> hi
11:23 < runelind> how's it going akh 
11:23 < akh> Other than a kernel panic, OK.
11:23 < runelind> akh: did you give the kernel a brown paper bag to breath into?
11:24 < akh> Nah, slapped it around.  ;-)
11:25 -!- dsias_ [~dsias@adsl-068-153-207-210.sip.bct.bellsouth.net] has joined #fink
11:29 -!- theid_away is now known as theid
11:35 < akh> grrr...stupid stale buildlock.
11:44 < akh> hmm...kdebase3.  That'll take a day or so.
11:56 -!- Macrobyte [~macrobyte@pcp06495808pcs.nstnig01.ct.comcast.net] has joined #fink
11:59 -!- Macrobyte [~macrobyte@pcp06495808pcs.nstnig01.ct.comcast.net] has left #fink []
12:23 -!- mdmonk [~mdmonk@dftech.org] has joined #fink
12:24 -!- chris01 [~chris01@212.126.165.246] has quit ["bye"]
12:26 -!- mdmonk [~mdmonk@dftech.org] has quit [Client Quit]
12:27 -!- mdmonk [~mdmonk@dftech.org] has joined #fink
12:28 -!- mdmonk is now known as mdmonk_away
12:52 -!- Murr [~neeri@A17-202-20-71.apple.com] has joined #fink
12:54 -!- miga [~miga@easyconnect2121138-139.clients.easynet.fr] has joined #fink
12:54 < miga> Hi everybody.
12:57 < newmanbe> Hello.
12:57 < miga> I would like to know if it has sense to use the RuntimeVars field to add DYLD_FALLBACK_LIBRARY_PATH:%p/lib, so that a package could work with ssl variant of another packages?
12:57 < miga> hello newmanbe.
12:58  * newmanbe always thought the best way to do that was "Depends: package-ssl | package".
12:58 < miga> It seems to work around the missing symbols and pass the --build-as-nobody build. The only thing that is maybe annoying is that the user should restart X11 so that the generated scripts could be taken into account.
13:00  * newmanbe doesn't get what you are saying and commands someone else to help.
13:00 < miga> newmanbe that's not the problem here, the problem is that whenever you install any ssl variant of any package in 10.4 Xcode 2.1, any other package refuses to work because of missing symbols.
13:01 < newmanbe> Any other package being a package that does not depend on it too?
13:01 < miga> actually not any other package, but any other package that depends on those ssl variants.
13:02 < newmanbe> Sounds like something (else) in XCode 2.1 is broken.
13:03 < miga> Yes, but meanwhile (that is it can be that Apple takes some time to repair it) it can be a work around.
13:03 < newmanbe> Or to tell people to not use XCode 2.1
13:04 < miga> That has no sense, you could not demand people not to install it. I mean that has no sense for programmers.
13:05  * newmanbe reconvenes the Court of Fink Project.
13:05 < newmanbe> Fink Project v. Apple Computer, newmanbe presiding
13:05 < newmanbe> Apple, for not showing up at the hearing, has been found guilty of breaking things.
13:06 < newmanbe> Apple has been sentenced to repair everything it has ever made that has been broken and provide the replacement for free.
13:06 < newmanbe> It has six months to complete the sentence.
13:07 < miga> Maybe, I will not argue against it, but that does not make things progress, if you understand what I mean.
13:07 < newmanbe> Yes, I thought about it.
13:07 < newmanbe> But I want them to fix balloon help to make it less annoying.
13:07 < miga> That was on 8, no?
13:08 < newmanbe> I know it was in System 7.
13:08 < newmanbe> I think Mac OS 9 too even.
13:09 < newmanbe> That was /me
13:09 < newmanbe> 's third case today.
13:10  * miga could not check, has lost a cupper switch on my g3 while tentatively overclocking it, so that it does not boot anymore. 
13:11 < newmanbe> Apple has to fix that now.
13:11 -!- Murr [~neeri@A17-202-20-71.apple.com] has quit [Read error: 145 (Connection timed out)]
13:11 < miga> not any more under warranties for long long time now.
13:12 < newmanbe> Neither is System 7.
13:12 < newmanbe> But it is now.
13:17 -!- Macrobyte [~macrobyte@pcp06495808pcs.nstnig01.ct.comcast.net] has joined #fink
13:18 -!- Murr [~neeri@A17-202-20-71.apple.com] has joined #fink
13:19 -!- jerwin [~jeremy@66-44-54-154.s154.tnt1.lnhva.md.dialup.rcn.com] has joined #fink
13:19 < miga> Speaking about that, do you know if it will be possible to access the data on this g3 by connecting it to the imac and booting in target mode or is it impossible due to this missing switch?
13:20 < newmanbe> What is that switch supposed to do?
13:21 < newmanbe> Special order from the Court of Fink Project: It has a new name: High Court of Fink Project.
13:21 < jerwin> agh. apt-get is hitting the slow repositorory in preference to the fast, local repository.
13:22 < miga> Maybe that's not the right word in English. Wait a moment I search for the page.
13:22 < RangerRick> switch?
13:22 < newmanbe> Second special order from the High Court of Fink Project: Apple will have miga's computer fixed.
13:22 < RangerRick> miga: you hold down "t" while booting to boot into target mode
13:22 < RangerRick> it should work as long as the hardware itself isn't messed beyond it's ability to read it  :)
13:22 < jerwin> does apt-get have a verbose mode?
13:22 < newmanbe> It's not +T?
13:22 < newmanbe> jerwin: try man apt-get
13:23 < RangerRick> newmanbe: I think it's just "t"
13:23 -!- Feanor [~astrange@feanor.developer.opendarwin] has quit [Read error: 60 (Operation timed out)]
13:24 < miga> here: <http://www.macbidouille.com/article.php?id=13&page=5>
13:24 < miga> See the image with the red circle, that's it.
13:24 < RangerRick> oh, you're overclocking?  yeah, then all bets are off... you could try target mode, but...
13:27 < miga> That's not sure, eh? I've successfully overclocked, but then I'd tried something higher, it did not work, I carefully keep the switch, then I forgot where I put it, it probably felt on the carpet, from here in the vacuum cleaner.
13:29 < miga> RangerRick what do you think if I use a RuntimeVars field to add DYLD_FALLBACK_LIBRARY_PATH:%p/lib for problem with ssl? Bad or good idea?
13:31 < akh> Maybe a wrapper script?
13:32 < RangerRick> miga: I don't really understand the consequences of that, I think pogma is the only one who does, I'd suggest talking to him when he wakes up  
13:32 < akh> BTW, for those keeping score, I had the -ssl problem using XCode 2.0
13:33 < akh> So it's not a 2.1ism
13:33 < akh> (_Evp_idea_cbc)
13:33 < miga> Yes, it has been here since the beginning.
13:34 < miga> That's a Tiger thing.
13:34 < RangerRick> no, it's a tiger dyld issue
13:34 < akh> pogma filed a radar on it.
13:34 -!- jerwin [~jeremy@66-44-54-154.s154.tnt1.lnhva.md.dialup.rcn.com] has quit ["Leaving"]
13:34 < RangerRick> yeah, and he's having a hard time convincing the dyld guy to fix it
13:34 < RangerRick> :(
13:35 < RangerRick> last I heard
13:35 < miga> Yes, but meanwhile it is or not a good idea to use a RuntimeVars field to put it automatically in the info file?
13:35 < akh> Ah--that bites.
13:35 < akh> miga:  best bet is to ask pogma.
13:35 < RangerRick> miga: like I said, I don't know the consequences of putting it in runtimevars (which essentially overrides *all* packages, not just yours)
13:35 < RangerRick> hence ask pogma
13:35 < miga> It works apparently, I've tested with bluefish and the new epiphany.
13:36 < RangerRick> I'm not comfortable with that without knowing what pogma thinks of it
13:36 < miga> OK, I'll ask him. Thanks.
13:37  * akh debates a temporary FAQ item telling people to set that variable at runtime...
13:38 < miga> akh, do not debate any more put it in the faq :-)
13:39 < akh> It's easy enough to unset later, I guess.
13:41 < akh> Or, better yet, set it just for that process
13:41 < akh> Yeah, that's the ticket.
13:42 < miga> Yes, exactly as pogma said on users yesterday.
13:42 < akh> So I guess this isn't going to be fixed in 10.4.2?
13:44 < miga> probably.
13:44 -!- Feanor [~astrange@mp1-248-35.dialup.emory.edu] has joined #fink
13:47 < RangerRick> doesn't seem like it
13:48 < miga> akh can you print from browser while under X11?
13:51 < akh> I got an error concerning Xprint servers.
13:53 < miga> OK, me too, and kevin horton also.
13:53 < miga> The problem is that at least with mozilla and epiphany, the want xprint, which is not ported to fink, and could not right now because it uses gnome 2.8.
13:55 < akh> ah
13:55 < akh> !lart GNOME2.8
13:55  * Melian brandishes Excalibur! "With this sword, I vanquish thee, GNOME2.8!" and lops off GNOME2.8's head
13:56  * akh summons a GNOME guru to take over maintainership.
13:56 -!- Murr [~neeri@A17-202-20-71.apple.com] has quit [Read error: 131 (Connection reset by peer)]
13:56 < akh> * sound of crickets *
13:57 -!- Macrobyte [~macrobyte@pcp06495808pcs.nstnig01.ct.comcast.net] has quit []
13:57 -!- Murr [~neeri@A17-202-20-71.apple.com] has joined #fink
14:07 -!- Feanor [~astrange@feanor.developer.opendarwin] has quit [Read error: 104 (Connection reset by peer)]
14:08 -!- Feanor [~astrange@mp1-248-35.dialup.emory.edu] has joined #fink
14:12 < newmanbe> Fink Project v. akh; newmanbe presiding
14:12 < newmanbe> akh: You are charged with dereliction of duty. How do you plead?
14:13 -!- ringerc [~craig@dsl-202-72-144-62.wa.westnet.com.au] has joined #fink
14:14 < theid> I've got a significant portion of Gnome 2.11.x built, but I can't chat right now; bbl
14:14 -!- theid [~theid@207.177.103.77] has quit []
14:17 < newmanbe> Yes.
14:18 < newmanbe> Fink Project v. newmanbe; newmanbe presiding
14:18 < newmanbe> newmanbe: You have been charged with typing something in the wrong window. How do you plead.
14:18 < newmanbe> Guilty.
14:18 < newmanbe> newmanbe: I sentance you to be a fair judge.
14:23 -!- Murr [~neeri@A17-202-20-71.apple.com] has quit [Read error: 131 (Connection reset by peer)]
14:24 -!- Murr [~neeri@A17-202-20-71.apple.com] has joined #fink
14:24 < akh> newmanbe: <whew> no verdict.
14:25  * akh pleads $JOB
14:30 < akh> (for all crimes)
14:30 < akh> $25
14:30 < akh> oops
14:30 < akh> !$25
14:31 < akh> d'oh! 
14:31 < akh> !403
14:31 < akh> !x11sdk
14:31 < akh> man
14:31 < akh> ...
14:31 < akh> What does Melian know?
14:36  * akh calls for a mental competency trial for Melian.
14:45 < miga> akh would you mind if I correct some misspelling in fix-upgrade?
14:47 -!- ringerc [~craig@dsl-202-72-144-62.wa.westnet.com.au] has quit ["Whoops, someone let the magic smoke out!"]
14:47 < akh> Go ahead.
14:48 < akh> I found a missing .
14:48 < akh> ooh, and 'selupdate'
14:49 < miga> Yes, also selupdate and change pre in code.
14:49 < akh> Ugh--I was even using a tag-completing editor.
14:49 < miga> Don't worry, I'm translating it into French, so that all the same for me to make the changes in English at the same time.
14:50 < akh> And nobody reads the docs anyway.  ;-)
14:50  * newmanbe declares akh innocent by reason of insanity.
14:51 < miga> That's not true, more and more people read it. But sometimes it's difficult to either understand it or find the right place.
14:51 < akh> miga:  I was of course exaggerating.
14:51 < akh> newmanbe:  Thanks!  Purple monkey dishwasher.
14:52 < newmanbe> The High Court of Fink Project opens the competency trial for Melian.
14:52 < akh> I'd like to call Melian to the stand.
14:52 < akh> !faq
14:53 < newmanbe> Melian: How do you plead?
14:53 < Melian> wish i knew, newmanbe
14:53 < newmanbe> Melian needs a counsel.
14:53 < Melian> newmanbe: parse error: dunno what the heck you're talking about
14:53 < miga> akh: what a surprise, you enter my universe? :-)
14:53 < akh> heh
14:54 < akh> newmanbe: appoint one.
14:54 < akh> e.g megahal
14:54 < akh> (just to make the logs extra surreal)
14:55  * newmanbe appoints megahal as counsel for Melian.
14:55 < newmanbe> megahal: How does your client plead?
14:55 < megahal> newmanbe: Shion: that's very kind of computer do you plead.
14:55 -!- Fang [~Fang@2002:56c1:e583:1:0:0:0:1] has quit [Remote closed the connection]
14:56 < newmanbe> megaha: I said, how does your client plead?
14:56 < newmanbe> megahal: I said, how does your client plead?
14:56 < megahal> newmanbe: How do you plead.
14:56  * newmanbe declares megahal in contempt of court.
14:56  * newmanbe kickbans megahal.
14:57 < akh> the State submits that neither the client nor her counsel are mentally competent.
14:57 < newmanbe> The court agrees.
14:57 < newmanbe> The court appoints bbraun as counsel for Melian.
14:58 < newmanbe> bbraun: How does your client plead?
14:58 < bbraun> all guilty all the time
14:59 < newmanbe> The court accepts the plea of guilty.
14:59 < newmanbe> This court declares Melian not competent.
14:59 < newmanbe> Case adjorned.
15:00 < akh> (though this really should be a "hearing" rather than a trial)
15:00 < newmanbe> Yeah, I was thinking about that.
15:00 < newmanbe> Too late.
15:00 < bbraun> my client would like to appeal on the grounds of incompetent councel
15:00 < bbraun> council
15:00 < newmanbe> I suppose Melian (or her counsel) could appeal.
15:00 < akh> !appeal
15:01 < bbraun> my client would like to submit the counsel's lack of spelling ability as evidence.
15:01 < newmanbe> Hmm, first the counsel has to be declared incommpetent.
15:01 < newmanbe> The High Court of Fink Project opens the competency hearing of bbraun.
15:01 < newmanbe> bbraun: Would you like a counsel appointed for you?
15:02 < bbraun> no
15:02 < newmanbe> Do you wish to argue that you are competent.
15:02 < newmanbe> ?
15:03 < bbraun> no
15:03 < newmanbe> The High Court of Fink Project find bbraun incommpentent.
15:03 < newmanbe> s
15:07 -!- LawjoskarAway [~larry@cpe-024-168-176-124.nc.res.rr.com] has left #fink []
15:09 < miga> This is new: cvs [update aborted]: connect to cvs1(10.5.1.7):2401 failed: Connection refused
15:09 < newmanbe> It is probably not new.
15:09 < miga> It is very new for me.
15:10 < newmanbe> Wait, that doesn't look like it is sf.net.
15:10 < miga> What could it be?
15:11 < newmanbe> The High Court of Fink Project should declare SourceForge.net broken; But I have had enough entertainment for right now.
15:11 < newmanbe> Has your firewall rules changed recently on that computer.
15:12 < akh> It's sf.net
15:12 < miga> I only rely on Apple ones, and I did not change anything recently.
15:12  * akh got the same thing during an 'update.sh'
15:12 < akh> (from shell.sf.net)
15:12 < miga> Are they mad, now???
15:12 < akh> Yes.
15:12 < newmanbe> 10.5.1.7 does not look like a sf.net IP address.
15:12 < akh> Keep trying, it'll work eventually.
15:13 < akh> newmanbe: it may be some internal address.
15:13 < akh> You can see it on some of the failures on auto-logs, too.
15:13 < miga> Trying 4 times, works at the 4th one.
15:14 < akh> That's been my experience, too.
15:14 < miga> How do you use auto-logs?
15:15 < akh> I use NNTP and subscribe to gmane.os.apple.fink.auto-logs
15:15  * newmanbe declares IRC stupid and orders everyone to join fink@conference.jabber.org .
15:15 < newmanbe> gmane got it now? /me goes to subscribe.
15:16 < miga> I've searched two days ago, but I think, not sure, that it is archived only after a month, or am I wrong?
15:16 < akh> It just started on GMANE yesterday.
15:16 < miga> Oh, OK, I'll check it, thanks.
15:16 < akh> (so there's only 5 messages)
15:17 < miga> I don't see it on gmane.
15:17 < cirdan> afternoon all
15:18  * akh isn't sure how far back GMANE goes--you may be right.
15:18 < miga> Afternoon circan, well night here.
15:18 < newmanbe> cirdan; The High Court of Fink Project declared Melian incompetent.
15:18 < newmanbe> She keeps forgetting stuff.
15:18 < miga> Maybe I should update the newsgroup.
15:18 < cirdan> !test
15:18 < cirdan> Melian: reload
15:18 < Melian> reloading...
15:19 < Melian> reloaded:
15:19 < cirdan> !test
15:19 < cirdan> Melian: test
15:19 < miga> Yes, that's it, I see it now.
15:19 -!- Melian [~blootbot@pcp04356153pcs.glstrt01.nj.comcast.net] has quit ["adios amigos"]
15:19 < cirdan> she's loosing the connection to mysql somehow
15:20 -!- Melian [~blootbot@pcp04356153pcs.glstrt01.nj.comcast.net] has joined #fink
15:20 < newmanbe> !test
15:20 < Melian> Yes, newmanbe, you're still online.
15:20 < cirdan> newmanbe: i am a dissenting opinion in the High Court :-p
15:21 < newmanbe> !faq
15:21 < Melian> faq is, like, at http://www.finkproject.org/faq/
15:21 -!- kane_ [~kane@perl.xs4all.nl] has quit []
15:21 < newmanbe> No you're not; your're not a part of the high court.
15:21 < newmanbe> You could appeal, the last counsel was declared incompetent too.
15:22 < cirdan> newmanbe: i have articles of incorperation that say i am ;-)
15:22 < newmanbe> I could make you a junior-justice.
15:23 < cirdan> heh
15:23 < cirdan> and dust puppie?
15:24 < newmanbe> The only requirement right now is that you have to also be in fink@conference.jabber.org.
15:24 < cirdan> ooo
15:25 < newmanbe> And be approved by me.
15:26 < cirdan> !trout slap newmanbe 
15:26  * Melian takes the trout and shoves it up slap newmanbe's left nostril.
15:26 < cirdan> !fishslap newmanbe 
15:26 < Melian> slaps newmanbe upside the head with a wet fish.
15:26 < newmanbe> !trout slap Jabber haters
15:26  * Melian takes the trout and shoves it up slap Jabber haters's both nostrils.
15:26 < cirdan> always confuse those
15:26 < cirdan> !poetrysmite newmanbe 
15:26 < Melian> newmanbe: Oh freddled gruntbuggly thy micturations are to me As plurdled gabbleblotchits on a lurgid bee. Groop I implore thee my foonting turlingdromes.
15:26 < cirdan> :-)
15:27 < akh> heh
15:27 < newmanbe> cirdan: You need to get blootbots working with Jabber.
15:27 < newmanbe> I know eggdrop works.
15:28 < cirdan> heh
15:28 < newmanbe> Feanor: Get megahal working with Jabber.
15:28 < newmanbe> Anyone: Get a pastebot and a loggerbot that works with Jabber.
15:30 < newmanbe> Hello?
15:30 < newmanbe> !test
15:30 < Melian> No, newmanbe, you're not online any more.
15:30 < akh> heh
15:31  * akh has to get a client that works with Jabber, first.
15:31 < cirdan> iChat! gAim!
15:32 < newmanbe> JBother!
15:33 < newmanbe> If you would be an admin (op), JBother is probably the best.
15:33 < cirdan> newmanbe: YBother!
15:34 < miga> I never succeeded in getting iChat work with Jabber.
15:35 < newmanbe> cirdan: Because it is Jabber.
15:35 < newmanbe> I am supposed to tell you have good JBother is.
15:36 < newmanbe> "(it's nice, supports logging, time, animated(!) smilies, excelent MUC support, GPG....) :)"
15:37 -!- Albie [~ambs@bl5-161-131.dsl.telepac.pt] has joined #fink
15:38 < akh> newmanbe: so fink@conference.jabber.org ?
15:44 < newmanbe> Yes.
15:49 -!- [Raccoon [~raccoon@adsl-63-193-149-212.dsl.lsan03.pacbell.net] has joined #fink
15:49 < cirdan> TheSin: yoyo
15:50 -!- [Raccoon [~raccoon@adsl-63-193-149-212.dsl.lsan03.pacbell.net] has quit [Client Quit]
15:52 < newmanbe> Scratch the loggerbot, the chat room can be logged automatically.
15:52 < akh> That's an advantage right there.
15:53 < cirdan> heh
15:53 < cirdan> but then you can't
15:53 < cirdan> !lart newmanbe 
15:53  * Melian resizes newmanbe's terminal to 40x24
15:53 -!- mdmonk_away [~mdmonk@dftech.org] has quit ["gotta go."]
15:53 < akh> Yah, that
15:53  * akh 's true.
15:53 < akh> hmm
15:53 < newmanbe> That is why you would need to make blootbots work with Jabber.
15:53 < akh> Nothing on google?
15:53 < cirdan> newmanbe: timRicker said patches would be welcome :-p
15:53 -!- waltman [~waltman@adsl-207-245-72-170.cust.oldcity.dca.net] has left #fink []
15:55 < newmanbe> Or you could switch to eggdrop, which works with Jabber (it may be a fork though).
15:55 < newmanbe> And when you first log in, you get part of the log printed so you can jump right in to the conversation, without having to find the logs yourself.
15:58 < newmanbe> akh: There are bots made for Jabber.
15:58 < akh> I thought there might be.
15:59 < newmanbe> C'mon; Are none of you as brave as akh?
16:03 < akh> heading out...
16:03 -!- akh [~akhansen@ldx3.psfc.mit.edu] has quit []
16:06 < newmanbe> Y'all are scary cats.
16:06 < newmanbe> scard'y 
16:11 -!- Murr [~neeri@A17-202-20-71.apple.com] has quit [Read error: 145 (Connection timed out)]
16:22 < newmanbe> Sigh, sometimes the easiest way to get apt to do what you want is to make it do what you don't want it to do.
16:22 -!- das_ is now known as Applehater
16:22 -!- Applehater is now known as das_
16:25 -!- theid [~theid@207.177.103.77] has joined #fink
16:42 < TheSin> hey ci
16:42 < TheSin> err  cirdan 
16:45 -!- Albie [~ambs@bl5-161-131.dsl.telepac.pt] has quit ["Leaving"]
17:08 < newmanbe> !test
17:08 < Melian> No, newmanbe, you're not online any more.
17:11 < newmanbe> Melian: join fink@conference.jabber.org
17:11 < newmanbe> Melian: It is too a valid channel name.
17:11 < Melian> I think you lost me on that one, newmanbe
17:11 < newmanbe> !lart personal-messanging bots
17:11  * Melian pries personal-messanging bots's back open with a screwdriver and flashes a new bootldr to personal-messanging bots
17:12 < newmanbe> Melian: You do need a new "boot-loader", to be able to use Jabber.
17:12 < Melian> newmanbe: KCI error, or a problem with the Keyboard-Chair Interface.
17:13 -!- Feanor [~astrange@feanor.developer.opendarwin] has quit [Read error: 110 (Connection timed out)]
17:16 < newmanbe> megahal: Do you like Jabber?
17:16 < megahal> newmanbe: My secret jouhou password file doesn't have the file yet, but i like in the end of the intarweb, therefore it is jabber.
17:18 -!- newmanbe [~xchat-ssl@f2ef3500496f6760.session.tor] has quit ["Leaving"]
17:18 -!- newmanbe [~xchat-ssl@c99ee869f8d08b7e.session.tor] has joined #fink
17:20 -!- Feanor [~astrange@mp1-249-168.dialup.emory.edu] has joined #fink
17:26 -!- asparagui [~billy@mo-71-1-109-215.dhcp.sprint-hsd.net] has joined #fink
17:37 -!- htodd [htodd@i8u.org] has quit [Remote closed the connection]
17:42 -!- htodd [htodd@i8u.org] has joined #fink
17:42 -!- [Raccoon [~raccoon@adsl-63-193-149-212.dsl.lsan03.pacbell.net] has joined #fink
17:42 -!- [Raccoon [~raccoon@adsl-63-193-149-212.dsl.lsan03.pacbell.net] has quit [Client Quit]
17:45 -!- Murr [~neeri@A17-202-20-71.apple.com] has joined #fink
17:47 -!- You're now known as RangerAway
17:49 -!- vasi [~vasi@modemcable214.145-70-69.mc.videotron.ca] has joined #fink
17:49 -!- KraMer [~mark@adsl-70-240-234-88.dsl.hstntx.swbell.net] has quit [Connection timed out]
18:01 -!- dmacks_away is now known as dmacks
18:02 < dmacks> !seen akh
18:02 < Melian> akh <~akhansen@ldx3.psfc.mit.edu> was last seen on IRC in channel #fink, 1h 59m 14s ago, saying: 'heading out...'.
18:04 < dmacks> miga: Which part(s) of gnome2.8 are needed for xprint?
18:05 < miga> dmacks: hi, wait a minute, I'll check it.
18:15 < vasi> hey, apparently we could have used '.noindex' instead of '.build'
18:16 < miga> Apparently, it requires an Xprt server and configuration files that are broken in xfree86
18:16 < miga> It is available from source at <http://xprint.mozdev.org/>
18:17 < miga> Also on debian.
18:17 < miga> Client side Xprint is available on all platforms.
18:17 < dmacks> miga: Hm, this sounds more complicated than "just upgrading a few gnome library packages"
18:18 < miga> Yes, that is.
18:18 < dmacks> Bummer.
18:18 < dmacks> vasi: Interesting!
18:19 < vasi> er, is anybody actually working on gnome anymore?
18:19 < miga> vasi: what do you mean by working?
18:20 < vasi> in the sense of getting us to ${most-recent-gnome}
18:20 < dmacks> You'll note that the basic parts (ones used by packages outside of gnome itself) are now up to the versions in 2.10 in 10.3
18:20 < vasi> that's good :-)
18:20 < miga> theid has said this afternoon he had worked on gnome 2.11, already almost packages ready.
18:21 < dmacks> jfm is working on some of the more gnome-centric parts (when he gets back from vacation)
18:21 < dmacks> So much has changed 2.6->2.10 I'm going slowly...we're finding all kinds of changed-deps and things-moved-among-packages
18:22 < vasi> yeah
18:22 -!- TheSin [~TheSin@iphost-64-56-130-194.edm.wiband.net] has quit ["Client exiting"]
18:22 -!- Feanor [~astrange@feanor.developer.opendarwin] has quit [Connection timed out]
18:22 < dmacks> (unfortunately, a lot of the commits happened when syncmail wasn't working:(
18:23 < vasi> did commits like that go through?
18:23 < miga> Is there a way to know what was committed?
18:23 < dmacks> syncmail is called *after* the actual commit happens.
18:24 < vasi> miga, 'cvs log'?
18:24 < vasi> or maybe cia, though i have no idea whether there's a way to browse past stuff
18:24 < dmacks> I tried to use date comparisons for cvs diff, but it didn't seem to work.
18:24 < vasi> dmacks, CVS is picky about date formats i believe
18:26 < dmacks> (I also don't know if a syncmail crash blocks ciabot.pl (I think "no") or if ciabot.pl worked at that time (since it also uses mail IIRC))
18:27 < vasi> good point
18:27 -!- jtyler [~jtyler@iphost-64-56-130-194.edm.wiband.net] has quit ["Leaving"]
18:33 < dmacks> We should have a PDB feature that dumps a table of all versions in all trees, hilighting version difference weirdness (stable>unstable; 10.3>10.4T)
18:33 < dmacks> (I guess 10.4T>10.3 would be good too, to see what needs to be back-ported or have a .patch adjusted to be more general)
18:35 < vasi> maybe let's make that NOT dependent on the PDB actually working :-)
18:36 < vasi> unless there's an intermediate stage to the PDB stuff that dumps the contents of a tree into a text file or something, that might be useful
18:36 < dmacks> ha!
18:36 < dmacks> There is indeed such a thing.
18:37 < dmacks> foreach dist {cvs update; dump index to file; update sql db from file}; adjust "most recent", "package count", and other flags in sql db
18:39 < dmacks> So if one cvs-update fails, only that tree will be out-of-sync in the web pdb; if connection times out during flag-fix, all verison-data is correct but sometimes search or other things that operate on data outside of individual PkgVersion objects fails.
18:40 < dmacks> OTOH, nobody's stopping anybody from running the pdb on one's personal machine and analyzing the data himself:)
18:40 < dmacks> What the heck is our compare-trees.pl script for?
18:41  * dmacks wonders if we can add a debian-version-strings sort method to mysql without having to recompile the server.
18:42 < vasi> ooh i dunno if that's a smart way to do it
18:43 < vasi> i think sqlite can actually use perl subs as comparison methods
18:43 < dmacks> Me neither...just pondering.
18:43 < dmacks> I think so too (didn't we already consider that back when contemplating using sqlite as pdb back-end?)
18:49 -!- baba [~baba@YahooBB220041001026.bbtec.net] has joined #fink
18:49 < vasi> no idea dmacks, the pdb is a mystery to me
18:50 < dmacks> (by "pdb back-end" I meant in fink core indexing, not "the web pdb")
18:52 -!- KraMer_ [~mark@adsl-70-240-193-143.dsl.hstntx.swbell.net] has joined #fink
18:53 < vasi> oh yeah
18:54 < vasi> i wonder how the dports folks manage to get ok performance for indexing with a sqlite backend?
18:54 -!- KraMer_ is now known as KraMer
18:54 < vasi> i'll have to look at their system more closely some time
18:59 -!- baba [~baba@YahooBB220041001026.bbtec.net] has quit ["This computer has gone to sleep"]
18:59 < vasi> hmm, they use the same kind of field => value table i was considering...which i guess means they can't handle complicated values? or does tcl take care of that?
19:04 < dmacks> I doubt the latter is true.
19:04 < vasi> would help if i actually could READ tcl code
19:05 < dmacks> What??? I thought the whole reason everyone loves dports>fink is that tcl is the perfect programming language, even for non-experts!
19:05 < vasi> oh, it looks like they do a one port => one file method like we do (in HEAD)
19:05 < dmacks> nice:)
19:05 < vasi> er, but then how does their searching work?
19:05 < vasi> hmm
19:05 < dmacks> (/me has actually heard that dbports/tcl argument made)
19:08 < vasi> heh
19:12 -!- dreamind [~dreamind@C2107.campino.wh.tu-darmstadt.de] has quit []
19:21 < vasi> argh i can't read tcl
19:22 < vasi> where did they find all these tcl hackers? so strange...
19:24  * dmacks looked at doing some tcl work in grad school. /me decided against it:)
19:29 < vasi> ah, they do use an sqlite db...and creating the db *IS* extremely slow
19:30 < vasi> it looks like they have some way of syncing to a central DB over http
19:30 < dmacks> aaahhhh
19:30 < pogma> yes,  kvv did that iirc
19:30 < vasi> so that's why users don't see much of a hit, though i guess it makes using different trees kinda impossible.
19:31 < pogma> he also put an sqlite db in his darwinbuild scripts
19:31 < vasi> well not impossible, just more difficult
19:31 < miga> hi pogma, did you receive the mail I sent to you apropos missing symbols?
19:31 < vasi> so how does it deal with complex types?
19:31 < pogma> mig, yes, I did.
19:32 < miga> And what is the answer?
19:32 < pogma> miga: I think asking users to do "DYLD_FALLBACK_LIBRARY_PATH=: foo" is sucky, but the best response
19:32 < miga> In info file?
19:33 < pogma> miga: it does not have to be /sw/lib, as long as it does not contain /usr/lib/
19:33 < pogma> miga: if it breaks during build, then you need it in the info file too
19:33 < pogma> miga: I am hoping it will be fixed in a software update
19:33 < pogma> sometime...
19:33 < miga> No, that's the problem it compiles perfectly, it just breaks at run time.
19:34 < pogma> miga: you will have to suggest to users that they need to do 'env DYLD_FALLBACK_LIBRARY_PATH=: app' then
19:34 < pogma> miga: please do not set this in runtimevars
19:35 < miga> Yes, but may I put it in DescUsage or should I wait someone complaints in any mailing list to tell it?
19:35 < pogma> miga: never know what a user has on their system, they may have libraries in /usr/local/lib which do not have the full path in the install_name (or even have the wrong full path in the install_name)
19:35 < pogma> miga: put it in DescUsage if you like
19:35 < miga> Yes, I understand that.
19:36 < miga> OK, thanks. I'll do it.
19:44 -!- dsias_ [~dsias@adsl-068-153-207-210.sip.bct.bellsouth.net] has quit [Read error: 113 (No route to host)]
19:49 -!- Murr [~neeri@A17-202-20-71.apple.com] has quit [Read error: 145 (Connection timed out)]
19:49 < cirdan> hi
19:50 < dmacks> 'hlo
19:50 < vasi> hmm ok i get it now
19:50 < vasi> they use a very similar system to HEAD
19:50 < vasi> they have a non-sql db-like index file
19:51 < vasi> (text format)
19:51 < vasi> which points to the Portfiles
19:51 < vasi> the index file only contains basic info....like our index.db
19:51 < cirdan> heh
19:51 < cirdan> miga: make a shellscript to call the app and set the variable
19:51 < vasi> but they don't use an intermediate step with the (\d+).db files like i do
19:52 < cirdan> for now
19:52 < cirdan> crap
19:53 < cirdan> i gotta update debconf or somehting
19:53 < cirdan> !lart missing sources
19:53 < dmacks> Yeah, there's all kindsa complaints about it.
19:53  * Melian gets a hotmal account and SPAMs missing sources
19:53 < cirdan> well, bbl then  :-)
19:54 < vasi> k, curiosity satisfied, back to real work now
19:56 < dmacks> If you want to procrastinate further, consider switching the boatload of numbered files to a single Tie::File file.
19:59 < dmacks> (or even Tie::File::AsHash, so the db itself handles %f lookups directly instead of routing through a numeric intermediate key.
20:00 < dmacks>  /nick dmacks...Destroyer-Of-Productivity
20:01 -!- dmacks is now known as dmacks_away
20:06 -!- akh [~akhansen@68-118-244-23.dhcp.oxfr.ma.charter.com] has joined #fink
20:08 < akh> ooo--Tide Coldwater can save $63/yr in energy bills.
20:32 -!- brion [~brion@adsl-69-230-182-162.dsl.irvnca.pacbell.net] has joined #fink
20:33 < brion> stupid question time... is there really no current gimp package? :(
20:37 < akh> yup.  
20:37 < akh> (meaning there isn't one)
20:37  * brion cries
20:39 < akh> There was a problem with 2.2.x that led it to be pulled, and I guess the maintainer hasn't had a chance to get back to it yet.
20:39 < brion> hrm
20:39 < brion> the Gimp.app package doesn't seem to be working on 10.4 so I'm gimpless until I get it compiled. :)
20:40 < akh> I believe 2.0.x works OK.
20:41 < akh> 2.0.6
20:41 < brion> it's not in stable; i don't feel like switching over to unstable just to get outdated software
20:42 < akh> Fair enough.
20:42 < brion> :)
20:52 < theid> is there any kind of online resource for searchable porting tips, aside from the fink.sf.net porting section and numerous Google searches?
20:52 < theid> if not, I'd be willing to set up a database for that kind of thing
20:53 < miga> that will be very useful.
20:53 < vasi> sure go ahead :-)
20:53 < theid> right now I'm relying on grep through the fink dist files, Google, and maybe some mailing lists
20:53 < vasi> you might want to look at msachs' buildfink scripts....they try to categorize error messages for failed builds
20:54 < theid> but it occurred to me a few hours ago that my best resource was the IRC logs, and those were only accessible through Google's cache now
20:54 < vasi> so you can go through the list of categories, they should be good guides for "what can go wrong"
20:54 < vasi> er, is b9 down?
20:54 < vasi> !logs
20:54 < Melian> logs are at http://meme.b9.com/cview.html?channel=fink&date=today or http://fink.aquaflux.org (currently down)
20:54 < theid> aquaflux
20:54 < vasi> well don't use aquaflux, it's borked...use meme.b9
20:54 < theid> I'm sure b9 doesn't have nearly as much (hasn't been going as long, right)
20:55 < vasi> i guess, but it's more recent and up-to-date
20:55 < theid> yeah, I had to go back to 2002 for the _environ fix, and that wouldn't be on b9
21:08 < theid> okay, I've got a mysql account ready to go... I'll probably get to that this weekend
21:12 < pogma> _environ fix? You could have just asked :)
21:13 < theid> I hate to ask lots of old questions... I tend to get told to ____(fill in the blank) instead of ask here
21:13 -!- dmacks_away is now known as dmacks
21:15 < dmacks> Not sure which _environ fix you have in mind but a quick google for things like ["os x" environ fix] gets several distinct sets of familiar-sounding patches:)
21:16 < theid> mostly for python, and the fix for that wasn't what I needed
21:16 < dmacks> ah
21:17 < dmacks> Someone (no, and donoughta 
21:18 < dmacks> Someone (no, and don't even think of suggesting it!) oughta update fink's porting page for 10.4 differences.
21:18  * akh nominates someone not currently here.
21:19 < dmacks> akh: Fink does not use the System's proxy settings, because it's technically difficult and functionally confusing for users (perhaps even more-so than our current fink.conf approach:)
21:20 < dmacks> (/me catching up on logs:)
21:20 < akh> dmacks: Ah
21:23 < dmacks> If users report problems getting sources from mastermirrors, check fink-auto-logs. Looks like there were problems mirroring lots of the new debconf stuff.
21:24  * dmacks really oughta put datestamps in the Subject, huh?
21:24 < akh> Yah
21:25 < cirdan> hmm
21:25  * cirdan looks to see where sources are stored on sancho...
21:27 < miga> cirdan: thanks for the advice to write a shell script, but as it may not be necessary in all cases, I'll just write what to do in DescUsage field.
21:27 < cirdan> miga: well, if it doesn't hurt the cases where it doesn't need to be set...
21:27 < cirdan> noone ever thinks to look in fink info foo
21:27 < cirdan> i even forget :-)
21:27 < cirdan> sometime
21:27 < cirdan> s'
21:28 < cirdan> bbraun: ping
21:29 < miga> I have no idea if it hurts or not, so the less the better for me.
21:29 < cirdan> dmacks: u know the path? /Volumes/src/distfiles?
21:30 < dmacks> cirdan: No idea. I've never needed to know:)
21:31 < dmacks> Alrighty...let's see how poorly I understand formail...
21:34 < vasi> miga, what happens if the user DOESN'T set the env var?
21:34 < akh> vasi:  there's a dylib error.
21:35 < miga> All gtk applications crashes at run time.
21:35 < vasi> uh, in that case please write a wrapper script
21:35 < miga> But there is three work arounds:
21:36 < vasi> it's not touch, you just do export FOO=blah exec /sw/bin/program-real
21:36 < vasi> er, tough even
21:36 < miga> First case: the user has installed bundle-gnome, then it is sufficient to define this variable in the .xinitrc and all go well.
21:37 < dmacks> First case pollutes user's shell env. Icky.
21:38 < miga> Certainly not, if the user launches X11, it does not pollute any shell outside of X11
21:38 < vasi> but it pollutes every shell within X11
21:38 < miga> Yes, but you cannot launch gnome-session without this env variable set, otherwise it crashes at launch time.
21:38 < dmacks> Goal should be minimum-necessary change, not just any sufficient change.
21:39 < cirdan> cat /sw/bin/gnome-session
21:39 < miga> In this case, the minimum change is to set it in .xinitrc.
21:39 < cirdan> #!/bin/sh
21:39 < cirdan> FOO=blah exec /sw/bin/gnome-session.real
21:39 < cirdan> :-)
21:40 < vasi> miga, what cirdan said
21:40 < miga> That's exactly the same
21:40 < cirdan> miga: until there is a fix, anyway
21:40 < cirdan> no, it's not
21:40 < cirdan> the variable is not exported
21:40 -!- Feanor [~astrange@mp1-249-55.dialup.emory.edu] has joined #fink
21:40 < cirdan> actuaclly, if you do FOO=blah in .xinitrc, i dont think it'll pollute, either
21:40 < miga> That does not work, precisely. I have tried it.
21:41 < cirdan> miga: which, a script?
21:41 < cirdan> you did have the #!/bin/sh at the top, right?
21:41 < miga> No, just it in the .xinitrc
21:41 < cirdan> ah
21:41 < miga> Yes.
21:41 < cirdan> it'd have to all be on a single line, the same as in the script
21:42 < miga> Actually I don't use a .xinitrc, except in the case I want to test bundle-gnome.
21:42 < vasi> pogma, does msachs know about this bug? maybe he can poke some people
21:42 < cirdan> i have icewm :-)
21:42 < miga> Yes, he knows. 
21:42 < cirdan> odd thing is, when i logout of it, it closes my term sessions but doesnt logout
21:43 < dmacks> If we're going to try to push a fix onto users (semi-)automatically, better to do it in a wrapper we can supply in a new revision of the affected package instead of telling users to keep editing their own dotfiles.
21:43 < cirdan> right
21:43 < dmacks> That way it's easy to undo if Apple fixes their bug, or change if we find better solutions.
21:43 < cirdan> like 10.4.2 ;-)
21:44 < miga> But it seems there is a problem with Apple's supplied libjpeg
21:44 < cirdan> anyone here test it?
21:44 < dmacks> miga: Isn't that a(nother) effect of the same underlying Apple bug?
21:44 < miga> It is the result of defining this variable, so that yes indirectly.
21:44 < dmacks> (and/or people being sloppy with DYLD_LIBRARY_PATH ?)
21:45 < miga> what is sloppy?
21:45  * dmacks means setting a global variable instead of restricting its effect to the program(s) that need it.
21:45 < miga> cirdan: just tell me what to do, I'll test it.
21:46 < cirdan> miga: sudo mv /sw/bin/gnome-session /sw/bin/gnome-session.real
21:46 -!- newmanbe [~xchat-ssl@c99ee869f8d08b7e.session.tor] has quit ["Leaving"]
21:46 < cirdan> sudo echo "#!/bin/sh" > /sw/bin/gnome-session
21:46 -!- Odemia [~tim@142-165-171-178.sktn.hsdb.sasknet.sk.ca] has joined #fink
21:47 < cirdan> sudo echo "DYLD_LIBRARY_PATH=someting exec /sw/bin/gnome-session.real" > /sw/bin/gnome-session
21:47 < cirdan> sudo chmod +x /sw/bin/gnome-session
21:47 < aaron42_away> itym ">>"
21:47 < cirdan> oops
21:47 < cirdan> yeah
21:47 < cirdan> sudo echo "DYLD_LIBRARY_PATH=someting exec /sw/bin/gnome-session.real" >> /sw/bin/gnome-session
21:47 < cirdan> not the other one ;-)
21:48 < miga> Are you sure the DYLD_LIBRARY_PATH=, should not it be DYLD_LIBRARY_PATH=:?
21:48 < cirdan> no idea
21:49 < cirdan> whatever you need it to be :-)
21:49 < miga> pogma told me DYLD_LIBRARY_PATH=:
21:49 < cirdan> thatn that
21:50 -!- __jt__ [~james@69-162-30-40.stcgpa.adelphia.net] has joined #fink
21:54 < theid> Does anyone know if "u_int32_t" is equivalent to "uint32_t"?
21:54 < cirdan> prolly...
21:54 < Feanor> i can't see what else it would be
21:54 < cirdan> not sure
21:55 < theid> works one way and not the other, so for now I'll use it
21:55 < aaron42_away> depends on your shell, iirc
21:55 < aaron42_away> for sh code, no colon
21:56 < vasi> aaron42_away, i don't think that's right
21:56 < vasi> theid, weird
21:56 -!- aaron42_away is now known as aaron42
21:56 < aaron42> is the colon supposed to be the path separator?
21:56 < aaron42> if so, it still makes no sense at the beginning
21:56 < vasi> yeah
21:57 < vasi> aaron42, it's parsed by ld not sh
21:57 < aaron42> ah, nm then
21:57 < theid> vasi: I think it may have something to do with ipv6, but I'll leave it for later
22:05 -!- rpop [~slice@face-00000002.user] has quit [Read error: 131 (Connection reset by peer)]
22:07 < miga> ok, it works, but then all gtk applications should have also that script, since they can work outside of the gnome environment.
22:07 < miga> Not that script but a similar one.
22:10 < cirdan> hmm
22:10 < cirdan> not fun
22:10 < cirdan> just hack the code to set the variable :-)
22:10 < cirdan> in the apps themselves
22:10 < cirdan> ;-)
22:10 < miga> Yes, exactly, that's why I think pogma solution is better.
22:10 < cirdan> miga: not if it messes other stuff up
22:10 < miga> No, that's dangerous.
22:10 < cirdan> someone try to fetch debconf
22:10 < dmacks> theid: different headers define that type with and without the underscore.
22:11 < aaron42> is autoconf2.5 known to be broken?
22:11 -!- aaron42 [~agdavi01@tiger.slug.louisville.edu] has quit [Remote closed the connection]
22:11 < lisppaste> aaron42 pasted "autoconf2.5 failure" at http://paste.lisp.org/display/9155
22:12 < miga> And another inconvenience is that you can no more launch the apps by double-clicking it.
22:12 < dmacks> stdint.h:typedef u_int32_t uint32_t; /* u_int32_t is defined in <machine/types.h> */
22:12 -!- aaron42 [~agdavi01@tiger.slug.louisville.edu] has joined #fink
22:12 < aaron42> is autoconf2.5 known to be broken?
22:12 < miga> Yes, and it depends on architecture, hence signedness.
22:13 < dmacks> aaron42: Works for me.
22:14 < aaron42> any ideas from my paste?
22:14 < miga> uint is the gnu unsigned int when used in package, and it generates a lot of signedness warnings, not good at all when dealing with corba.
22:16 < aaron42> i'm thinking maybe the texi2html erros mean no docs at all get made
22:16  * dmacks builds....
22:19 < aaron42> ok, this is weird
22:19 < dmacks> I don't get any diagnostics from 'texi2html -split_chapter ./autoconf.texi'
22:19 < aaron42> why would texi2html in /sw/bin/ not be found by my shell?
22:19 < aaron42> it looks like it was using /usr/bin
22:20 < dmacks> Failure to correctly include fink init script in shell dotfile?
22:20 < aaron42> when i moved that aside, it couldn't find it at all
22:20 < aaron42> it sees everything else in there!
22:20 < dmacks> Weird.
22:20 < aaron42> path: /usr/local/scripts:/sw/bin:/sw/sbin:/usr/bin:/bin:/usr/sbin:/sbin:/usr/X11R6/bin:/usr/local/bin:/usr/local/cross-tools/bin:/Developer/Tools:/Users/agdavi01/bin:/Users/agdavi01/scripts:.
22:20 < dmacks> Broken alternatives?
22:21 < cirdan> could be
22:21 < cirdan> dmacks: what u building?
22:21 < aaron42> ah, there we go
22:21 < dmacks> cirdan: read back ~1 screen.
22:21 < cirdan> debconf, or autofoo?
22:21 < aaron42>  /sw/bin/texi2html -> /sw/etc/alternatives/texi2html -> /sw/bin/texi2html.tetex-base
22:21 < dmacks> autofoo
22:22 < aaron42> and /sw/bin/texi2html.tetex-base doesn't exist
22:22 < dmacks> That'd be a problem, yes.
22:22 < cirdan> sudo update-alternatives
22:22 < aaron42> dumps a help screen
22:22 < cirdan> well, read it :-p
22:23 < cirdan> dmacks: can u try to fetch debconf?
22:24 < cirdan> aaron42: try sudo update-alternatives --auto texi2html
22:24 < dmacks> Not found at ftp.{us,at,se}.deb; master.finkmirrors has it.
22:25 < dmacks> (debconf_1.4.42.tar.gz)
22:25 < cirdan> ok, as long as master has it now
22:26 < dmacks> cirdan: regarding autoconf2.5, why build+install+delete .elc instead of just not building them in the first place?
22:26 < aaron42> cirdan: th, that seems to have done it
22:26 < cirdan> dmacks: didnt know u could do that
22:27 < cirdan> someone fixed it in unstable, iirc
22:27 < cirdan> dmacks: also drm wanted it built
22:28 < cirdan> but it shouoldnt be built by default
22:28 < dmacks> cirdan: Sure you aren't thinking of gettext, which everyone was yelling about for weeks?
22:29 < cirdan> yes, i am :-)
22:29 < cirdan> dmacks: dunno
22:29 < cirdan> lots of people maintain autofoo*, i'm just the contact point ;-)
22:30 < dmacks> 'k. I used to think everyone had at least "some" emacs, but now I'm not so sure, and no reason to build stuff we ain't gonna use.
22:30 < cirdan> right
22:31 < cirdan> if you know how to turn it off, then lets :-)
22:31 < dmacks> aaron42: Debian and GNU do several texinfo things quite differently, and are unable/unwilling to find a common ground.
22:34 -!- rpop [~slice@face-00000002.user] has joined #fink
22:35 < aaron42> i'm seeing 
22:36 < aaron42> lots of version churning in autofoo, db, and gettext
22:36 < cirdan> yeah...
22:36 < aaron42> devery time i do another update-all, it goes db41->db-3->db42->db43
22:36 < cirdan> eberythign needs to move to new gettext
22:36 < aaron42> before doing antyrhing else
22:37 < cirdan> and many things dont need to spec a specific version of autofoo
22:37 < vasi> !seen msachs
22:37 < Melian> msachs <~msachs@17.255.99.114> was last seen on IRC in channel #fink, 1d 1h 30m 24s ago, saying: 'Doesn't make sense to me that ssl == non-crypto'.
22:37 < aaron42> it's telling me that autoconfs 6-9 will be installed as prereqs
22:37 < dmacks> aaron42: Yeah...only one of those can be installed at at time, fink has no idea which one is needed first, many packages each require a (different) specific db##
22:38 < dmacks> (perhaps automake1.[6-9] ?)
22:38 < cirdan> yeah, some tings need older autoconfsto build...but not many should need it at all  
22:38 < aaron42> yeah, somethign like that
22:39 < cirdan> heh, i cant type
22:39 < aaron42> anyway, it makes doing a ~200 package upgrade a pain in the ass
22:39 < cirdan> yup
22:40 -!- akh [~akhansen@68-118-244-23.dhcp.oxfr.ma.charter.com] has quit []
22:40 < aaron42> i'm sure finding the optimal order is NP-Complete, but couldn't some improvements be made?
22:41 < aaron42> like, check the prereqs and implement them, before crashing out?
22:41 < cirdan> aaron42: have fun
22:41 < cirdan> cvs diff to give us the patch :-)
22:41 < aaron42> hah
22:44 < dmacks> Fink 0.25 will be better about auto-swapping build-time dependencies as-needed.
22:45 < aaron42> good to hear
22:46 < dmacks> (given a list of packages and their deps, it's trivial to determine the order to do them; we just don't do it very well:(
22:46 < dmacks> (otherwise 'make' wouldn't exist:)
22:46 < vasi> dmacks, it's not exactly trivial to do it quickly :-)
22:46 < vasi> (make doesn't have conflicts...)
22:46 < aaron42> make doesn't have negative deps
22:47 < dmacks> Yeah, that's the only problem.
22:47 < vasi> or choices!
22:47 < aaron42> i'm not at all sure it's actually solveable
22:47 < vasi> oh, actually with TheSin's SHLIB_DEPS stuff it's going to be worse
22:47 < dmacks> We could have a positive anti-dep that gets satisfied by removal of the package, but make and the like freak out when previously-satisfied deps become un-satisfied later.
22:47 < aaron42> it sounds like a satisfiaility problem
22:47 < vasi> because then we technically don't know what the deps will be before it's built
22:47 < aaron42> and most of those are np-complete
22:48 < vasi> aaron42, there's a finite number of packages, i think it's pretty solvable
22:48 < aaron42> ok, feasibly then
22:48 < vasi> i think it's pretty feasible, apt does a pretty good job
22:48 < aaron42> if it takes the age of the universe to calculate, it's not a useful solution
22:49 < vasi> apt doesn't take nearly that long ;-)
22:49 < dmacks> Why? Not like compiling KDE is gonna get done before then anyway.
22:51 < vasi> heh
22:52 < vasi> if our algo was more recursive it would be easier to hack on
22:52 -!- Macrobyte [~macrobyte@pcp06495808pcs.nstnig01.ct.comcast.net] has joined #fink
22:52 < vasi> ie: allow rewinding changes and such
22:52 -!- Macrobyte [~macrobyte@pcp06495808pcs.nstnig01.ct.comcast.net] has left #fink []
22:52  * vasi hates dependency resolution
22:52 < cirdan> !summon thesin
22:52  * 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 thesin's coordinates.
22:52 < cirdan> :-)
22:53 < vasi> oy
22:53 < aaron42> seriously tho, i think it's a sat prob in many vars: assign all vars such that an expression is true
22:53 -!- vasi is now known as vasiRuns
22:53 < aaron42> e.g., ((foo && automake6_foo && automake9_foo) && (bar && aotmake6_bar && automake9_bar))
22:53 < aaron42> or something ablong those lines
22:54 < vasiRuns> aaron42, given the knowledge of the problem domain, i don't think that it works the other way though, and that's necessary for np-completeness
22:54 < dmacks> If we didn't insist on a 3-pass approach (plan and ask about alts; download all; build/insall/whatever) it would be a *lot* easier.
22:54 < vasiRuns> ie, there would need to be a way to turn any SAT problem into a dependency system
22:54 < vasiRuns> dmacks, true dat
22:55 < cirdan> u see this? http://thomas.loc.gov/cgi-bin/query/z?c109:H.J.RES.24.IH:
22:55 -!- vasiRuns is now known as vasiGone
22:55 < vasiGone> oooh, a blank page
22:56 < cirdan> night
22:56 < aaron42> hmm, perhaps
22:56 < dmacks> If we really wanted to do it consolidatedly, we could create a shadow installation (no actual files, just status db and file lists)...do a dry run there and prompt to resolve any issues that arise, instead of trying to guess what might happen. But that's hella painful.
22:56 < aaron42> mostly i'm going on the intuition that everything interesting is NP-Complete :)
22:57 < aaron42> at best
22:57 < dmacks> No, there's a very simple solution to "get me a beer":)
22:58 < aaron42> the *really* interesting stuff is undecidable
22:58 < dmacks> Right: what kind of beer?, for example
22:58  * aaron42 doesn't like beer
22:58  * aaron42 prefers cider
23:00 < dmacks> In college a coupla friends of mine made their own. It pretty much tasted like apple juice. It kinda had a different effect.
23:01 < aaron42> i saw online where you can make high-quality vodka by filtering cheap vodka through water filters
23:01 < dmacks> Yup.
23:01 < aaron42> turn $5 vlad into $30 grey goose
23:01 < Feanor> it only works sometimes
23:02 < Feanor> the original of that was on the sa forums, i think, and one of the results ended up with cheap vodka that tasted like activated charcoal
23:02 < cirdan> dmacks: recipe please1
23:02 < cirdan> !
23:02 < dmacks> Kinda funny that the guy who owns Grey Goose dropped out of college.
23:02 < aaron42> one bottle cheap vodka, one brita filter, one hour
23:03 < aaron42> pour vodka through filter six times
23:03 < cirdan> suck filter dry, drink vodka
23:03  * dmacks was only involved in the consumption not creation of said cider.
23:03 < cirdan> find said friends and steal secret recipe
23:04 < cirdan> :-p
23:04 < cirdan> ok, sleeping for real now
23:04 < cirdan> 11:11pm
23:04 < aaron42> assuming the filter can be reused, this should be cheaper than the alternative:
23:04 < dmacks> For vlad, passing through an old sock is probably a cost-effective treatment too:)
23:04 < aaron42> one bottle everclear, one bottle distilled water: mix and drink
23:04 < cirdan> hehe
23:04 < cirdan> right
23:04 < cirdan> water optional
23:05 < cirdan> anyone wanna do some shots ;-)
23:05 < cirdan> iShots!
23:05 < aaron42> 190/2=95
23:05 < dmacks> One of the math programs uses calculation of cost of [alcohol content] in various forms as its exmaples.
23:06 < aaron42> heh
23:06 < dmacks> I oughta port it to fink just for that reason.
23:06 < cirdan> sounds good
23:07 < aaron42> bottled water comes in metric bottles, right?
23:07 < aaron42> a 750 of evian + a 750 of everclear = eVodka
23:19 < dmacks> http://futureboy.homeip.net/frinkdocs/ "More Liquor"
23:26 < aaron42> heh
23:31 -!- Netsplit orwell.freenode.net <-> irc.freenode.net quits: chdevers, gzl
23:33 -!- Netsplit over, joins: gzl
23:34 -!- vasiGone [~vasi@modemcable214.145-70-69.mc.videotron.ca] has quit []
23:38 -!- chdevers [~cdevers@pool-68-163-153-167.bos.east.verizon.net] has joined #fink
23:41 -!- Netsplit orwell.freenode.net <-> irc.freenode.net quits: bbraun, gzl, Airo, runelind, statico, miga, neo, knghtbrd, usataway, Feanor,  (+1 more, use /NETSPLIT to show all of them)
23:43 -!- Netsplit over, joins: gzl, Feanor, miga, knghtbrd, runelind, statico, neo, Airo, bbraun, usataway
23:43 -!- miga [~miga@easyconnect2121138-139.clients.easynet.fr] has quit ["Bye, I'm leaving"]
23:44 -!- Netsplit over, joins: cmeme
23:49 < aaron42> how about this as a workaroudn for the dep mess for the moment:
23:49 < aaron42> for pkg in `cat pkgs`; do fink install pkg || exit; done
23:50 < aaron42> make that fink -y
23:50 < pogma> cat pkgs ?
23:51 < dmacks> Yeah...those are envelopes used to deliver kittens.
23:51 < aaron42> assuming i've saved a list of packages to install in a file called pkgs
23:52 < dmacks> The problem is in the ordering of builds.
23:52 < aaron42> yes, but this will build every packages separately
23:52 < dmacks> ?
23:52 < pogma> not necessarily
23:52 < aaron42> i've seen at least some cases where hat helps
23:52 < aaron42> that
23:53 < aaron42> right now, frex, i'm building somethign that depends on autmoake18
23:53 < aaron42> it didn't work in an update-all
23:53 < aaron42> but it works fine on its own
23:54 < dmacks> Starting from a clean fink, 'fink build bundle-kde-ssl' will crash numerous times with dependency problems:)
23:54 < aaron42> ok, true enough, but bundle-anything is a nightmare
23:54 < pogma> just repeatedly typing 'fink install foo' will get there eventually
23:54 < dmacks> But "on their own" each lower dep works fine, for sufficiently low enough levels of "lower".
23:54 < aaron42> most of the ones i'm doing don't have hundred-pkg dep list
23:54 < aaron42> s
23:55 < dmacks> If you choose a restricted enough set, there are many non-generalizable solutions:)
23:55 < aaron42> true
23:55 < aaron42> i'm not expecting mine to work perfectly, just to work better
23:56 < dmacks> The problem is that it sounds like it doesn't.
23:56 < aaron42> i am currently compiling a case where it would
23:57 < dmacks> Our build scripts do a fairly decent job of eventually getting everything done.
23:57 < aaron42> btw, are there any plans to make fink jobs able to run in parallel?
23:58 < aaron42> it'd be nice to be able to do multiple simultaneous installs sometimes
23:59 < dmacks> That is currently possible.
--- Log closed Thu Jun 16 00:00:13 2005