--- Log opened Thu Sep 22 00:00:05 2005 00:01:04< vasi> i thought files already had conditionals? 00:01:18< vasi> could be i just wanted it to have them, and convinced myself that it did 00:01:29 * dmacks reads.............. 00:01:36< dmacks> It does! 00:02:48< dmacks> Maybe [everyone except you] should go drink a beer and you can just wish really hard the IBD and the engine and Shlibs gets implemented, and It Will Be So? 00:10:27< vasi> hehe 00:10:36< vasi> TheSin is back, and promising to finish up Shlibs 00:10:46< dmacks> great! 00:11:30< vasi> btw, i agree with you that the dep engine's "do a bad preflight, then kinda-sorta do what was planned" strategy sucks 00:11:55< dmacks> heh...nice synopsis of it there:/ 00:12:33< vasi> if i could go back in time, and learn perl a few years earlier, and then bump into chrisp before he wrote fink.... 00:12:57< dmacks> yeah 00:12:59< vasi> well, then chrisp would have written more code while drunk ;-) 00:13:41< vasi> given that writing a new dep engine would be a ton of work, best bet would be to have an already-working dep engine we could steal 00:13:47< vasi> but there aren't many of those around :-( 00:13:50< dmacks> ya 00:14:27< vasi> i intend to eventually play with attaching fink to Apt and/or Smart 00:14:35< vasi> see if we can hijack an existing engine 00:14:58< vasi> (the programmer behind Smart sounds very enthusiastic about adapting it to a source-based system) 00:16:00< dmacks> source-based isn't much different than binary...just sometimes we look at Depends and other times we look at [family Depends and BuildDepends] 00:16:43-!- vasi [n=vasi@modemcable133.147-70-69.mc.videotron.ca] has left #fink ["Client exiting"] 00:17:39-!- vasi [n=vasi@modemcable133.147-70-69.mc.videotron.ca] has joined #fink 00:19:03< vasi> it's actually very different 00:19:39< vasi> because a Real Dep Engine (tm) basically resolves "what should the system look like , to satisfy what the user asked us" 00:19:45< vasi> so it finds a single solution state 00:20:06< vasi> but a source-based system needs to resolve multiple states....one for each build, then a final state 00:21:28< vasi> yech :-/ 00:21:35< dmacks> It's all just a question of ordering the installations. And sometimes one has to stop installing and start building for a little while. 00:22:10< dmacks> The less one pre-plans, the easier it all is, becuase you can just go along satisfying little pieces recursively. 00:22:34< vasi> yeah exactly 00:23:00< vasi> (well, except for the prefetching...not sure how many people care about that) 00:23:48< dmacks> The crappier one's network, the more one cares:) 00:23:53< vasi> anyway, if we do end up ever wanting to write our own dep engine, the SysState module is designed to make that pretty possible 00:24:51< vasi> yeah, probably the best solution is to do a complete preflight, then do a prefetch, then do a recursive install....if things get weird (cuz you're running multiple finks, for example), then some things may be unfetched, but most cases should work 00:26:49< dmacks> I keep toying with the idea of having a shadow status file that one could actually play with state-to-state and trap "what do we need to build?" along the way. But it's not much easier or more efficient than rewriting the engine. 00:30:37< vasi> dmacks, SysState is basically that 00:30:45< vasi> you can install stuff, remove stuff, whatever 00:30:54< vasi> and it will tell you what's left unsatisfied 00:31:01< vasi> and it has undo! 00:31:54 * dmacks should get around to reading that .pm, huh? 00:32:51< vasi> yep 00:34:12< vasi> what's the status of your buildlocks destruction stuff? 00:34:19< dmacks> Ain't looked at it. 00:36:02< vasi> alright 00:36:10 * vasi is kinda impatient to show the world 0.25 :-) 00:36:22< vasi> which reminds me, i have to test the FinkCommander patch 00:36:52 * dmacks doesn't really expect near-term buildlock changes to have much (any?) user-visible change. 00:37:13< asari> vasi: oh, what is the patch like? 00:37:51< vasi> asari, it enables FinkCommander to work efficiently with incremental indexing 00:37:53< asari> smartpm.org is too slow... 00:38:10< asari> vasi: sounds interesting 00:38:47< vasi> asari, not particularly interesting, it has no user-visible change 00:39:03< vasi> all it does is make FinkCommander not-extremely-slow with the new unreleased fink 00:39:14< vasi> asari, is smart really that slow? 00:39:30-!- vasi [n=vasi@modemcable133.147-70-69.mc.videotron.ca] has left #fink ["Client exiting"] 00:39:33< asari> I mean the website 00:39:37-!- vasi [n=vasi@modemcable133.147-70-69.mc.videotron.ca] has joined #fink 00:40:03< asari> I can't read even "README" 00:40:54< asari> ah, actually is located linux-br.conectiva.com.br 00:41:04< vasi> http://smartpm.org/ works fine for me 00:41:23< vasi> er yeah, except the readme :-/ 00:41:23< dmacks> From here as well. 00:42:15< dmacks> http://zorked.net/smart/doc/README.html 00:42:16< vasi> i have the source on-disk, so i can paste the readme for you 00:42:39< dmacks> Looks like the site moved but they forgot to change the absolute href links 00:42:44< asari> thanks dmack, and vasi :) 00:42:54< vasi> i wish apt had better bindings.... 00:43:02< dmacks> It has bindings? 00:43:11< vasi> it has perl and python bindings 00:43:14< vasi> but very minimal 00:43:29< dmacks> qx/apt-get install thing/ :) 00:43:42< asari> vasi: no problem, let's use C 00:44:03< vasi> asari, if i'm hacking on a dep engine, i want to be able to play 00:44:17< vasi> ie: please god, not manual memory management 00:44:27< asari> just a joke ;) 00:44:35< dmacks> vasi: hehe:) 00:45:19< asari> bbl 00:45:22< vasi> there's perl for apt-pkg: http://packages.debian.org/unstable/perl/libapt-pkg-perl 00:45:31< vasi> but i think the dep resolver is in apt-inst 00:46:06< dmacks> Why don't we use RPC/SOAP/etc to execute the actual processing on another machine (hi RangerAway!), then we can leak like a f'in' sieve and...eh so what? 00:46:17-!- asari [n=asari@p2145-ipbf706marunouchi.tokyo.ocn.ne.jp] has quit ["Quitting!"] 00:46:37< vasi> :-) 00:48:25< vasi> ah, here's the binding stuff http://cvs.debian.org/?cvsroot=deity 00:49:14< dmacks> Cool...something else for /dev/null^Wgnome-core to do! 00:53:43< vasi> heh, i thought you nominated yourself to be gnome-core? :-P 00:53:53< dmacks> Shut up, just shut up. 00:53:59< dmacks> :) 00:54:53< dmacks> I'm just the baby-sitter for the procmail->/dev/null recipe. 00:56:23< dmacks> Pretty soon I fear we may have to fork gtk+2 and pango1 into new Package names, and make up new .dylib install_name locations. Woo hoo! 00:57:56< vasi> er, they're doing gtk+3? 00:58:46< dmacks> No, but we'll need cairo, which will mean having to do an instantaneous rev-up and Depends tweak to every package that uses gtk+2. 00:59:25< dmacks> Or else we make it a new package, allow concurrent installation of its new-fangled -shlibs, and can migrate more gradually. 01:00:13< dmacks> Unless we're certain no cairo symbols are exported? 01:01:49< vasi> hmm, just to add Depends: gtk+2 (>= first-vers-with-cairo) ? 01:01:58< dmacks> Ayup 01:02:16< vasi> how many ppl actually downgrade? 01:02:43< dmacks> No clue. 01:02:57< vasi> i'd rather just script a change to every .info file... 01:03:21< vasi> so that every gtk package gets yummy cairo goodness 01:04:23< dmacks> I'm more concerned about forcing everyone to rebuild half of their packages for this silly change, not the mechanics of making the change from our end. 01:04:46-!- asian [n=Sugoi@71-178.69-92-cpe.cableone.net] has joined #fink 01:05:19< asian> has anyone thought of a handy script to open a file using an application in X11??? 01:05:21< vasi> oh i don't really care, their fault for using unstable :-P 01:05:36< vasi> open-x11 01:05:37< dmacks> asian: open-x11 ? 01:05:40< vasi> it's already on your system 01:05:45< vasi> (if you have apple's x11) 01:05:55< asian> but say i want a file to open with kwrite? 01:06:04< asian> i am using apple's x11 01:06:07< dmacks> That's not even close to what you asked:) 01:06:15< asian> i was getting there 01:06:18< dmacks> Sort-of 01:06:23< asian> i said "an application in x11" 01:06:31< dmacks> Can you 'kwrite some-filename' ? 01:06:34< asian> then i was going to say: such as kwrite =) 01:06:42< asian> i can in xterm 01:06:54< asian> but, if i could do it with less steps 01:06:57< vasi> well there are various systems 01:07:08< asian> even if i could open xterm with an argument to run when it opened 01:07:23< vasi> there are some scripts floating around on the WWW to auto-detect what DISPLAY should be 01:07:37< dmacks> 'xterm -e kwrite filename' 01:07:55< asian> dmacks: seriously? or just a guess? 01:07:58-!- shres [n=sshreyas@202.144.95.244] has joined #fink 01:08:43< dmacks> xterm -e has existed for many years. 01:09:37< vasi> dmacks, open-x11 doesn't understand arguments :-( 01:10:08< dmacks> Yeah...it sucks like that (uses /usr/bin/open IIRC) 01:10:47< brendan> export DISPLAY=${DISPLAY:=:0} ? 01:11:30< dmacks> Ayup, brendan...that's the basic one; assumes X11 is running and that one is not using fast-user-switching. 01:14:01< asian> brendan: yeah, thats the problem, i want to open a source.cpp file with kwrite w/o x11 already running from terminal 01:14:47< shres> to get around that i create an embedded launcher which does this if([[NSWorkspace sharedWorkspace] launchApplication:@"X11.app"]) 01:14:48< shres> system("/opt/gnome-2.12/bin/evolution-start.sh > /dev/null "); 01:15:32< shres> the same thing should work in a apple script too 01:15:46< dmacks> A script "open-x11-dont-suck" that first runs X11.app, then sets DISPLAY (if needed), then runs some x11 program (first arg) with parameters (rest of the args) 01:17:01< shres> is that bad, this is the solution i could think of 01:17:12< shres> no idea if thats the right way to do or not 01:17:57< dmacks> D'oh...my dont-suck starts a new x11 world every time! 01:18:16< brendan> you can do it from the shell too, with a little ps | grep; open X11.app; set DISPLAY; sleep; run 01:18:26< brendan> I predict 6 lines :) 01:18:40< vasi> if you put in your .xintirc "echo $DISPLAY > ~/.xdisplay" 01:18:55< vasi> then you can use my 'x11' script 01:19:00< dmacks> I predict 1: download Bill Scott's dotfile stuff:) 01:19:04< brendan> or 1 if you love ; 01:19:10< vasi> which i will paste in a second 01:19:26< lisppaste> vasi pasted "x11 scri[t" at http://paste.lisp.org/display/11882 01:19:31< asian> vasi: what does setting display in the .xdisplay do? 01:19:57< vasi> asian: it makes it possible for my script to find what your DISPLAY should be 01:20:14< vasi> i just do 'x11 kwrite somefile' and it opens somefile in kwrite...opening x11 if necessary 01:21:13< asian> vasi: you are amazing 01:21:18< asian> that code is well done 01:21:23< asian> did you just write that 01:21:29< vasi> not just now :-) 01:21:54< brendan> ok, more than 6 lines :) 01:22:20< dmacks> We could golf it down if necessary:) 01:22:37< vasi> uh oh 01:24:11< asian> vasi, .xinitrc did not previously exist -- will it be used now? 01:24:37< asian> and of course i put it in my ~/ 01:24:43< vasi> uh, you should probably copy the standard .xinitrc to ~/ 01:24:57< asian> where is the standard? /etc/X11 or sth ? 01:25:03< vasi> standard one is in /etc/X11/xinit/xinitrc 01:25:03< vasi> then just add the echo to it 01:25:48< brendan> I'm a little mystified by how .xdisplay is getting set... 01:26:31< vasi> brendan...when you launch X11, it sources your .xinitrc, right? 01:27:00< vasi> and display is already set when your .xinitrc is run 01:27:01< vasi> er, DISPLAY 01:27:16< brendan> oh I missed that line 01:27:24< vasi> :-) 01:27:25< brendan> pardon my ADD 01:28:16< brendan> dumb question number 2: why does it open x11.app twice? 01:28:36< asian> vasi: well, i have to go to bed -- wife calling 01:28:43< vasi> toodles 01:28:45< asian> but there are some problems running the script 01:28:57< asian> x11 starts and then closes by itself 01:29:00< asian> no error messages 01:29:05< asian> not sure why 01:29:06< vasi> asian, well come back another time and i'll try to help you with it 01:29:30< asian> and i couldnt copy the standard xinitrc bc sourcing it would close my connect (ssh) 01:29:47< asian> =/ 01:29:49< asian> anywho 01:29:50-!- dmacks [n=dmacks@pdpc/supporter/active/dmacks] has quit ["leaving"] 01:29:52< asian> check y'all laters 01:29:57< vasi> brendan, the second time is just to bring it to the front 01:30:01-!- asian [n=Sugoi@71-178.69-92-cpe.cableone.net] has quit ["Leaving"] 01:30:03< vasi> by asian 01:30:08< brendan> ah 01:30:11-!- lisppaste [n=lisppast@common-lisp.net] has quit ["Common Lisp IRC library - http://common-lisp.net/project/cl-irc"] 01:30:22-!- lisppaste [n=lisppast@common-lisp.net] has joined #fink 01:32:45-!- vasi [n=vasi@modemcable133.147-70-69.mc.videotron.ca] has quit ["Client exiting"] 01:35:26-!- Bart_ [n=hideout@213.23.123.11] has joined #fink 01:35:29< Bart_> howdy 01:35:42< Bart_> is there a way to set top transperent on the os x aqua desktop? 02:02:11-!- Feanor [n=astrange@opendarwin/developer/feanor] has quit [Read error: 110 (Connection timed out)] 02:02:41-!- Bart_ [n=hideout@213.23.123.11] has quit [] 02:03:33-!- kane_ [n=kane@perl.xs4all.nl] has joined #fink 02:52:42-!- geewz [n=gregreed@ppp122-142.static.internode.on.net] has joined #fink 03:02:23-!- blammo [n=blammo@dsl-197-239.zhonka.net] has joined #fink 03:02:50< blammo> anyone having problems with 10.3, oroborosx and gaim? 03:10:21-!- geewz [n=gregreed@ppp122-142.static.internode.on.net] has quit ["Leaving"] 03:16:08-!- asari [n=ASARI@gw08.ecc.u-tokyo.ac.jp] has joined #Fink 03:16:19-!- ringerc [n=craig@dsl-202-72-144-62.wa.westnet.com.au] has joined #fink 04:00:49-!- blammo [n=blammo@dsl-197-239.zhonka.net] has quit ["Download Gaim: http://gaim.sourceforge.net/"] 04:35:52-!- hennker [i=flullup@dsl-082-083-217-022.arcor-ip.net] has joined #fink 04:36:46-!- Feanor [n=astrange@opendarwin/developer/feanor] has joined #fink 04:57:34-!- asari [n=ASARI@gw08.ecc.u-tokyo.ac.jp] has quit ["Quitting!"] 05:18:58-!- ringerc [n=craig@dsl-202-72-144-62.wa.westnet.com.au] has quit ["Whoops, someone let the magic smoke out!"] 05:30:06-!- lisppaste [n=lisppast@common-lisp.net] has quit [Broken pipe] 05:30:15-!- lisppaste [n=lisppast@common-lisp.net] has joined #fink 05:30:22-!- sQuare [n=sqrmac@62.147.61.66] has joined #fink 05:47:05-!- shreyas [n=sshreyas@202.144.95.243] has joined #fink 05:47:27-!- sQuare [n=sqrmac@62.147.61.66] has quit ["DABASSIST.COM // French Drum & Bass"] 05:48:38-!- sQuare [n=sqrmac@62.147.61.66] has joined #fink 05:56:37-!- shres [n=sshreyas@202.144.95.244] has quit [Read error: 110 (Connection timed out)] 05:57:36-!- shreyas is now known as shres 06:00:14< sQuare> hi, what i have to do when i've downloaded and extract fink (/sw) to run it ? 06:01:43< shres> sQuare: http://fink.sourceforge.net/download/index.php?phpLang=en download , double click install, drag fink commander into applications, double click and u are good to go 06:02:39< sQuare> i've ever done that... but i've not found the way to run it 06:05:02< shres> man fink 06:06:03< sQuare> No manual entry for fink 06:06:28< newmanbe> Did you setup your path? 06:07:07< sQuare> how do you do that 06:07:40< newmanbe> You read the documentation. ;) 06:07:50< sQuare> i have not chosen the /sw path... it's the only one thing i have : this /sw path in my disk 06:08:31< sQuare> pathsetup.sh it not here :( 06:08:49< newmanbe> Try running echo $PATH 06:09:01< newmanbe> Does it have where you put Fink in it? 06:10:41< sQuare> /usr/bin:/bin:/usr/sbin:/sbin:/usr/X11R6/bin 06:10:46< sQuare> /usr/bin:/bin:/usr/sbin:/sbin:/usr/X11R6/bin 06:10:46< sQuare> /usr/bin:/bin:/usr/sbin:/sbin:/usr/X11R6/bin 06:10:46< sQuare> /usr/bin:/bin:/usr/sbin:/sbin:/usr/X11R6/bin 06:10:46< sQuare> /usr/bin:/bin:/usr/sbin:/sbin:/usr/X11R6/bin 06:10:47< sQuare> /usr/bin:/bin:/usr/sbin:/sbin:/usr/X11R6/bin 06:10:49< sQuare> oups 06:10:55< sQuare> sorry 06:11:38< newmanbe> Your path does not appear to be setup. :) 06:11:56< newmanbe> Read the documentation, setup depends on your shell. 06:12:03< sQuare> and how i have to do 06:12:19< sQuare> /bin/sh 06:12:36< shres> export PATH=$PATH:/sw/bin 06:13:30< sQuare> thx 06:13:42< sQuare> it work 06:17:39-!- You're now known as gopherd 06:19:11< sQuare> ok fink is upgraded and i install the Gimp... thx for help shres 06:21:32< shres> no problem 06:38:57-!- shres [n=sshreyas@202.144.95.243] has quit ["I am sleepy"] 06:41:23-!- robval [n=Robert@nat.packetfront.com] has joined #fink 06:41:51-!- shres [n=sshreyas@202.144.95.244] has joined #fink 06:53:51-!- Feanor [n=astrange@opendarwin/developer/feanor] has quit [Read error: 110 (Connection timed out)] 06:57:40-!- robval [n=Robert@nat.packetfront.com] has left #fink [] 07:04:22-!- shres [n=sshreyas@202.144.95.244] has left #fink ["Leaving"] 07:04:27-!- shres [n=sshreyas@202.144.95.244] has joined #fink 07:10:50-!- geewz [n=gregreed@ppp122-142.static.internode.on.net] has joined #fink 07:23:35-!- Uriri [n=asari@p1155-ipbf507marunouchi.tokyo.ocn.ne.jp] has joined #Fink 07:27:47-!- Uriri [n=asari@p1155-ipbf507marunouchi.tokyo.ocn.ne.jp] has quit [Client Quit] 07:31:02-!- asari [n=asari@p1155-ipbf507marunouchi.tokyo.ocn.ne.jp] has joined #Fink 07:41:56-!- sQuare [n=sqrmac@62.147.61.66] has quit ["DABASSIST.COM // French Drum & Bass"] 08:08:02-!- akh [n=akhansen@ldx3.psfc.mit.edu] has joined #fink 08:09:21< akh> [OT] If I use Disk Copy's Restore to copy a bootable volume to an empty formatted volume, should the latter be bootable? 08:09:44< akh> If so, then apparently I've got a hardware problem. 08:14:36 * akh browses for removable bus-powered firewire drives, since I can't afford the 2 weeks turnaround to send my laptop to the shop, nor do I feel confident that I can take it apart and slap a new drive in in a timely manner. 08:14:47-!- rajesh_ [n=rajesh@38.112.8.74] has joined #fink 08:15:57-!- rajesh_ is now known as rajesh 08:16:37 * asari suddenly noticed that tomorrow is a public holiday, and starts to drink sake 08:16:49< akh> Have a cup for me. 08:17:05 * asari passes akh a cup of sake 08:17:11< asari> here ;) 08:17:16< akh> Thanks! 08:17:30< asari> Happy thursday evening! 08:22:26 * akh wishes the US had public holidays on Fridays instead of Mondays. 08:23:52-!- rajesh [n=rajesh@nylug/member/rajesh] has quit ["leaving"] 08:25:02-!- RangerAway is now known as RangerRick 08:26:21< asari> Tommorow is " Autumnal Equinox Day", when day and night have the same length. 08:27:46< akh> That's a good reason for a holiday. ;-) 08:29:16< asari> During the days around the day, it is said that the spirit of the dead come back from the another world. 08:29:56< asari> we greet them, then. 08:30:51< akh> Sounds as good or better than most of the days we get off here... 08:46:10< baba_> asari san, akh san, konbanwa 08:46:28< akh> Good morning/evening baba san 08:47:05< baba_> hi akh, im doing phd at todai from this oct 08:47:32< akh> Nice. 08:47:35< asari> baba_: konbanwa 08:47:58< baba_> so, will have another three years with fink 08:48:00< asari> baba: welcome to todai :) 08:48:01< akh> Having a Ph.D. is overrated, though. :-) 08:48:19< baba_> over... 08:48:55< akh> (viewed as having more value than it really does) 08:49:15< baba_> hehe 08:49:18 * akh is living proof. 08:49:31< akh> ;-) 08:50:25< baba_> that's exactly what i want 08:50:55-!- baba_ [n=baba@YahooBB220041000015.bbtec.net] has quit ["Leaving"] 08:51:05-!- baba [n=baba@YahooBB220041000015.bbtec.net] has joined #fink 08:51:15< baba> oops 08:55:11< asari> btw, are there FinkCommander person here? 08:55:29< baba> hmm Clef knows about it? 08:56:12< asari> usataway wants to make a translation of Japanese 08:56:54< baba> FC needs developers 09:01:07< pogma> If you guys ask, they'll give you commit access to the project, I'm sure 09:01:29-!- hennker_ [i=flullup@dsl-082-083-238-168.arcor-ip.net] has joined #fink 09:01:59< pogma> Clef is benh57 09:02:23< baba> mogpa! 09:02:34< pogma> abab! 09:02:57< baba> !eheh 09:04:19< asari> mogpa sounds extremely cute! :) 09:04:44 * asari likes pa pi pu pe po 09:04:51< baba> jessica mogpa 09:05:12-!- shres [n=sshreyas@202.144.95.244] has quit ["I am sleepy"] 09:18:39< asari> baba: if you have time, shall we review the usata's translation? 09:18:50-!- hennker [i=flullup@dsl-082-083-217-022.arcor-ip.net] has quit [Read error: 110 (Connection timed out)] 09:19:12< asari> he said it was almost done 09:21:13-!- Sandking [n=jacek@ogorek.akron.wroc.pl] has joined #fink 09:25:15< pogma> baba: http://pogma.com/misc/jessica_bw.jpg 09:26:46< asari> she's cute 09:28:28< pogma> we think so, but then, we're crazy parents :) 09:30:13-!- lisppaste [n=lisppast@common-lisp.net] has quit ["Common Lisp IRC library - http://common-lisp.net/project/cl-irc"] 09:30:20-!- lisppaste [n=lisppast@common-lisp.net] has joined #fink 09:31:54-!- geewz [n=gregreed@ppp122-142.static.internode.on.net] has quit ["Leaving"] 09:38:47-!- rajesh [n=rajesh@38.112.8.74] has joined #fink 09:47:18-!- regeya [n=shane@65.171.234.176] has joined #fink 09:48:12-!- Sandking [n=jacek@ogorek.akron.wroc.pl] has quit [] 09:49:14< akh> pogma: Arguably not as crazy as my wife. The only reason that she doesn't have a bunch of pictures of her "children" (our dogs) is that one of them is scared $#!+less of the flash. 09:52:34-!- dmacks [n=dmacks@pool-70-22-31-132.balt.east.verizon.net] has joined #fink 09:52:43< cirdan> haha 09:55:03 * runelind compiles openoffice 09:55:06< runelind> actually I started 8 hours ago 09:55:23< runelind> so maybe it will be done in a couple of days 09:55:46< cirdan> heh 09:55:48< cirdan> maybe 09:56:03< runelind> I don't know why I'm doing it either 09:56:13< runelind> machochistic I guess 09:56:16< akh> runelind: It took me a day and a half. 09:56:18< runelind> however you spell it 09:56:19-!- regeya [n=shane@65.171.234.176] has quit [Read error: 131 (Connection reset by peer)] 09:56:24< runelind> akh: on what kind of a box? 09:56:30< akh> G4/450 09:56:39< runelind> ph3ar my dual 533 09:56:40< akh> (dual, but even so) 09:56:44< akh> So you should be fine. 09:56:56< akh> Unless you run out of HD space in the middle... 09:57:02< runelind> I've been looking at G4 upgrades, but they are so spendy still 09:57:49-!- regeya [n=shane@adsl-sp3-cdale176.micgi.com] has joined #fink 09:57:59< akh> Wait until 2008. Then they'll be cheap. 09:59:25 * runelind wants the dual dual-core intel setup\ 10:00:22< asari> runelind: Thanks for using OOo! 10:01:06 * cirdan wants the dual core dual proc g5 setup 10:01:07< cirdan> :-) 10:01:17< cirdan> intel macs will suck 10:01:17< cirdan> :-p 10:02:11< runelind> cirdan: why? 10:02:26< RangerRick> because they won't have magical ppc-ness! 10:02:46< RangerRick> because people get emotionally attached to how their apple computers are different than anything else, rather than how they use them :) 10:03:10< cirdan> no, i just heard they are using 32 bit modtly on the intels 10:03:17< akh> RangerRick: Spoken like a wacko-elitist cross-platformer 10:03:28< asari> Soon .dmg distribution will come with pretty icons and drag & drop accepting applescript bundle, I think 10:04:00< akh> asari: Nice. 10:04:28< asari> then, fink package is no longer needed :) 10:06:01< runelind> cirdan: they will be using 64 bit intels 10:06:20< cirdan> rumors, i tell u! 10:06:21< cirdan> :-) 10:07:13< RangerRick> yes, rumors are always true 10:07:15< pogma> runelind: ot next year 10:11:17-!- cianhughes [n=cian@cian.ws] has joined #fink 10:11:45< dmacks> Mayor of some Louisiana town, when asked about city officials keeping several truckloads of donated goods for themselves "they're hurricane victims just like everyone else!" 10:11:46 * RangerRick boots up his G5 powerbook and thanks god rumors are always true, even though it's sad intel chips will only be used 32-bit ;) 10:16:31< runelind> it will be interesting to see benchmarks of the intels versus the g5's 10:16:55< runelind> maybe kde will only take a week to compile now ;p 10:17:43< dmacks> Depends how many new versions of stuff Mr. WE releases during that time... 10:17:44< RangerRick> bah, only takes a week now :) 10:17:49< RangerRick> unless you're akh 10:18:16< runelind> does fink work in osx86 yet? :) 10:18:17< dmacks> Yeah... 'if USER==akh rm -rf /' is a great optimizer 10:20:09< runelind> I could get a mac mini for the price of a G4 upgrade 10:24:36< runelind> they need to make a G5 mac mini 10:26:42< runelind> and price it at like 699 10:27:58< runelind> erm 799 10:28:14< dmacks> akh: groklaw has link to mp3 of ma.us/M$ meeting about open-standards. No transcript yet... 10:28:30-!- regeya [n=shane@adsl-sp3-cdale176.micgi.com] has quit ["Leaving"] 10:32:41-!- shres [n=sshreyas@59.92.140.196] has joined #fink 10:42:55-!- shreyas [n=sshreyas@59.92.139.74] has joined #fink 10:43:16-!- regeya [n=shane@adsl-sp3-cdale176.micgi.com] has joined #fink 10:53:30-!- shreyas_ [n=sshreyas@59.92.145.111] has joined #fink 10:55:17-!- shres [n=sshreyas@59.92.140.196] has quit [Read error: 110 (Connection timed out)] 10:58:53-!- cianhughes [n=cian@cian.ws] has quit [Client Quit] 11:00:16-!- shreyas_ is now known as shres 11:05:27-!- shreyas [n=sshreyas@59.92.139.74] has quit [Read error: 110 (Connection timed out)] 11:08:05-!- KraMer [n=mark@adsl-70-240-205-240.dsl.hstntx.swbell.net] has left #fink ["Leaving"] 11:17:17-!- KraMer [n=mark@adsl-70-240-205-240.dsl.hstntx.swbell.net] has joined #fink 11:24:39-!- drm [n=drm@dhcp-152-224.asci.uchicago.edu] has joined #fink 11:25:14< drm> !seen msachs 11:25:15< Melian> msachs was last seen on IRC in channel #fink, 5d 3h 20m 6s ago, saying: 'newmanbe, I don't know anything about they do!'. 11:26:03-!- jtyler__ [n=jtyler@iphost-64-56-130-194.edm.wiband.net] has quit ["Leaving"] 11:30:50< dmacks> drm: Melian's got some !seen memory lapses lately. 11:32:31< drm> that's cause she's offline a lot, i think 11:32:46< drm> can't hold her responsible when she didn't "see" :) 11:33:16 * drm checks his watch 11:33:46< drm> whew, it *is* afternoon in EDT...was getting worried for a sec 11:34:08< dmacks> heh 11:34:23< dmacks> OTOH...uchicago? 11:34:37< drm> yeah, for a couple of days 11:35:03< dmacks> Sabbatical from the sabbatical? 11:35:21< drm> giving a talk and working on a research proj. with one of the guys here 11:36:34< dmacks> Can't you use one of them string-thingies and just transport your ideas? 11:37:29< drm> you mean, like two tin-cans attached with a piece of string? 11:37:48< dmacks> Hm, that'd work:) 11:38:02< akh> Nasty time lag, though. 11:38:51< akh> biab 11:38:54-!- akh [n=akhansen@ldx3.psfc.mit.edu] has quit [] 11:42:45-!- jtyler [n=jtyler@iphost-64-56-130-194.edm.wiband.net] has joined #fink 11:43:00< cirdan> yo 11:43:10< drm> oy 11:43:23< cirdan> oi! 11:43:34< asari> oui 11:43:34< drm> !io 11:43:37< jtyler> yo 11:43:45< cirdan> hoho 11:43:50< drm> hai 11:43:54< jtyler> yolla! 11:43:57< dmacks> .la 11:44:00< cirdan> iie... 11:44:04< jtyler> ;D 11:44:04< cirdan> .so 11:44:34< asari> .dylib 11:44:48< dmacks> .f 11:45:10< jtyler> .c 11:45:12< cirdan> .cobalt 11:45:24< cirdan> .asm 11:46:45 * dmacks can't remember Apple II extensions:( 11:47:44< cirdan> .u! 11:47:46 * cirdan hugs pascal 11:48:04 * drm tries to remember JCL 11:48:05< cirdan> only language where a non ascii char is used in the filenale extention 11:48:10< cirdan> JPL! 11:48:30 * dmacks has flashbacks to object-pascal. Thanks, ass 11:48:39< drm> JCL... ibm's Job Control Language... from system 360, and maybe before 11:48:40< cirdan> haha 11:48:48< cirdan> symantic pascal! 11:49:08< drm> you could run cobol programs with it :) 11:49:27< cirdan> MS basic...on system 6.0.2! 11:49:33< cirdan> w00t 11:49:45 * dmacks tries to find PROFILE.EXEC backup to remember CP. 11:50:45< cirdan> oh, JPL was Jet Propultion Lab 11:50:53< dmacks> heh 11:50:55< cirdan> or somehting spelled like it :-) 11:51:47< dmacks> Could marry them, make JP-JCL and have a rocket-propelled card reader. 11:51:57< cirdan> hehe 11:54:50-!- akh [n=akhansen@ldx3.psfc.mit.edu] has joined #fink 11:55:48 * akh verifies that my Tiger CD images seem also to give a bootable iPod. 12:13:45-!- dmacks is now known as dmacks_away 12:19:51-!- gopherd [n=irclogge@tor/session/x-806de8b241040862] has joined #fink 12:19:51-!- Topic for #fink: Have a question? Check the FAQ: http://fink.sf.net/faq || Latest Installers: 0.6.4 (10.2), 0.7.2 (10.3), 0.8.0 (10.4) || Fink 0.24.10: Cameloparadalis 12:19:51-!- Topic set by akh [] [Thu Aug 25 10:00:59 2005] 12:19:51[Users #fink] 12:19:51[ Airo ] [ cls ] [ gopherd ] [ kane_ ] [ pnorman ] [ swix_ ] 12:19:51[ akh ] [ cmeme ] [ gzl ] [ KraMer ] [ pogma ] [ usataway] 12:19:51[ asari ] [ das_ ] [ Henk_Pol1y ] [ lisppaste] [ rajesh ] [ zorton ] 12:19:51[ baba ] [ dmacks_away] [ hennker_ ] [ mcp ] [ RangerRick] 12:19:51[ BleedAway] [ drm ] [ htodd ] [ mdmonk ] [ regeya ] 12:19:51[ brendan ] [ eno-away ] [ jack- ] [ mee_bot ] [ RLD_osx ] 12:19:51[ cirdan ] [ Erik____ ] [ JosephSpiros] [ megahal ] [ runelind ] 12:19:51[ Clef ] [ gecko2 ] [ jtyler ] [ Melian ] [ shres ] 12:19:51-!- Irssi: #fink: Total of 43 nicks [0 ops, 0 halfops, 0 voices, 43 normal] 12:19:53-!- [freenode-info] Welcome Tor users! Please stop by #tor to compare notes and to report any freenode Tor problems. 12:20:00-!- Channel #fink created Sun Aug 3 17:57:20 2003 12:20:00-!- newmanbe [n=newmanbe@tor/session/x-46516b3b73e3ba9a] has joined #fink 12:20:37-!- Irssi: Join to #fink was synced in 51 secs 12:24:06< akh> mmm...1 message out of ~40 on the moderation queue wasn't spam. 12:24:17< akh> And it went to gnome-core, never to be seen again. 12:25:10-!- zecke [n=ich@83-169-171-130-dynip.superkabel.de] has joined #fink 12:30:39-!- Albie [n=ambs@bl5-165-61.dsl.telepac.pt] has joined #fink 12:32:58-!- dmacks_away [n=dmacks@pdpc/supporter/active/dmacks] has quit [Read error: 110 (Connection timed out)] 12:42:28-!- drm [n=drm@dhcp-152-224.asci.uchicago.edu] has quit ["Leaving"] 12:55:36-!- zecke [n=ich@83-169-171-130-dynip.superkabel.de] has quit [Read error: 104 (Connection reset by peer)] 12:59:07-!- dmacks [n=dmacks@pool-70-22-31-132.balt.east.verizon.net] has joined #fink 12:59:52 * dmacks pokes cvs.sf...wakey wakey! 13:01:11-!- zecke [n=ich@83-169-171-130-dynip.superkabel.de] has joined #fink 13:01:31< akh> dmacks: heh. Like that'll work... 13:02:02< dmacks> Well it's not like I have any better ideas.... 13:02:37< dmacks> And I was gonna clear out some -submissions, too! 13:02:37-!- zecke [n=ich@83-169-171-130-dynip.superkabel.de] has quit [Read error: 104 (Connection reset by peer)] 13:10:05-!- zecke [n=ich@83-169-171-130-dynip.superkabel.de] has joined #fink 13:12:37-!- shres [n=sshreyas@59.92.145.111] has quit ["What can i do? I am just being me"] 13:14:33-!- Albie [n=ambs@bl5-165-61.dsl.telepac.pt] has quit ["Leaving"] 13:20:02< brendan> thanks for committing those two packages, dmacks 13:20:23< dmacks> you're welcome:) 13:20:55< dmacks> Linker flags are a giant mess on OS X:( 13:21:05< akh> dmacks: New packages? That's crazy talk. 13:21:34< dmacks> heh 13:21:36< brendan> thank god for libtool ;) 13:21:55< dmacks> akh: Did you just feel a disturbance in the Force? 13:22:07< akh> brendan: That's a sentence I never thought i'd hear here. 13:22:14< akh> (or read,anyway) 13:22:44< akh> dmacks: As if millions of users cried out in terror, and were suddenly silenced. 13:23:09< brendan> If only it were sincere 13:23:17< akh> heh 13:23:34< dmacks> So be it 13:23:51< akh> lol 13:25:29< cirdan> hehe 13:25:56< dmacks> brendan: Actually it's substantially Apple's (or darwin's or NextStep's) fault for not allowing indirect symbol refs. Often those types of flags really are needed. 13:29:00-!- zecke [n=ich@83-169-171-130-dynip.superkabel.de] has quit ["leaving"] 13:30:14-!- lisppaste [n=lisppast@common-lisp.net] has quit ["Common Lisp IRC library - http://common-lisp.net/project/cl-irc"] 13:30:20-!- lisppaste [n=lisppast@common-lisp.net] has joined #fink 13:30:21< brendan> er, what's an indirect symbol ref? I'm picturing using liba:foo where a:foo calls b:bar, and I thought that worked... 13:31:45< dmacks> Consider if libfoo links to libbar and re-exposes libbar's symbols, then you pass -lfoo and try to access those symbols. 13:32:30< dmacks> You're accessing libbar's symbols *indirectly*, which is forbidden...one must also pass -lbar if one wants to access libbar's symbols. 13:33:10< dmacks> (on many machines, because libfoo was linked with -lbar, that's enough to cause -lfoo to enable access to libbar's symbols) 13:35:05< brendan> hmm, it seems sensible to require -lbar if you use them directly, to keep the namespace clean 13:35:22< dmacks> Yeah, there are plenty of reasons. 13:35:53< dmacks> But it means libfoo always declares "I was linked with -lbar, so you should link to *me* with that flag also just to be safe" 13:35:58< brendan> that is, I shouldn't have to worry about whether a:foo calls b:bar, nor should I assume that if I link to -la I can call b:bar myself 13:37:48< dmacks> If I want to link against a high-level gnome lib, I gotta drag along a whole sh!tload of -l, for every lib that every dependent lib links, etc etc. Makes sense, but makes a mess for package managers like fink that separate runtime code from compile-time headers. 13:39:38< brendan> I don't think I follow. I thought if you didn't use symbols from -lgnome-helper directly, you didn't have to link with it explicitly 13:39:57< brendan> (that was the basis of that shout-py hack) 13:42:56< dmacks> That requires knowing where each symbol comes from and which ones are *actually* used (consider #define some_lib_function some_other_lib_function) 13:43:16-!- broeken [n=broeken@5353014C.cable.casema.nl] has joined #fink 13:43:56< dmacks> (...in some_lib.h; without -lsome_other_lib, we're undefined) 13:44:33 * brendan googles "CPP considered harmful" 13:44:38< dmacks> heh:) 13:45:24 * dmacks doesn't necessarily agree with this approach, but That's How It Is. 13:45:30< brendan> yeah, that could be a nuisance. Not sure why some_lib.h is doing that define, but I suppose it has its reasons 13:47:33< dmacks> error-lib.h: #define WARNING(msg) new_gtk_modal("Warning!",msg) 13:48:22< dmacks> So I need -lgtk-2.0 in order to make use of -lerror-lib 13:48:50< brendan> seems a bit silly, but oh well... 13:49:32< dmacks> yah 13:51:05< dmacks> [that's a pretty bogus example btw; it's not linker indirect ref; but there are dozens of ABIs that do legitimately re-export dependent-lib symbols] 13:54:35-!- asari [n=asari@p1155-ipbf507marunouchi.tokyo.ocn.ne.jp] has quit ["Quitting!"] 13:55:23< broeken> anybody know a quick fix to get growl working with fink? 13:55:48< RangerRick> 1. install growl. 2. install mac-growl-pmXXX 13:55:49< RangerRick> 3. there is no rule 3 13:56:09< broeken> too bad rule 2 fails 13:56:31< RangerRick> ah, so what you meant to say is, "I have a bug report, the mac-growl-pmXXX package will not install" 13:56:41< RangerRick> paste the error to lisppaste 13:56:44< RangerRick> lisppaste: url 13:56:44< lisppaste> To use the lisppaste bot, visit http://paste.lisp.org/new/fink and enter your paste. 13:56:54< dmacks> heh 13:57:08< RangerRick> man, today just needs to end 13:57:21 * RangerRick is dinking with crappy report stuff at work 13:57:26< RangerRick> but all I can think about is I'm getting on a plane to vegas this evening :) 13:57:42< dmacks> s/d/dr/ would be better, no? 13:57:59< dmacks> Ahhh...INteresting... 13:58:26< lisppaste> broeken pasted "Growl without glue" at http://paste.lisp.org/display/11898 13:58:58< RangerRick> ick! 13:59:03< RangerRick> that looks scary :) 13:59:28< broeken> please don't blame the messenger 13:59:38 * dmacks never understood why TheSin has Mac::Glue as only Suggests 13:59:47< RangerRick> dmacks: in theory it's supposed to use osascript if mac::glue isn't there 13:59:56< RangerRick> broeken: no, not blaming you, just it looks scary :) 14:00:19< broeken> care to explain what's so scary? 14:00:21 * RangerRick wonders if there's a newer version 14:00:37< broeken> RangerRick: there isn't 14:00:39< RangerRick> just beyond a simple fix 14:01:07 * dmacks thought the module itself could fall back to osascript, but the test didn't. 14:01:14< RangerRick> ah 14:01:32< dmacks> broeken: Do you have mac-glue-pm586 installed? If not, install it. 14:02:40< RangerRick> well, there is a 0.7.2 sdk, maybe we can make a new growl perl bit out of it 14:02:46< broeken> dmacks: is there a mac-glue-pm586? 14:02:48< dmacks> um....maybe? 14:02:56< broeken> guess again 14:03:34< dmacks> "Thanks for the dep-tree hole, mac-growl!" 14:03:37-!- Albie [n=ambs@bl5-165-61.dsl.telepac.pt] has joined #fink 14:04:02< dmacks> You on Tiger? 14:04:05< broeken> yep 14:04:23< broeken> sounds a bit like "you on crack?" ;) 14:04:45< dmacks> Heh:) /me was gonna say "mac-glue-pm581 might suffice" but that would require installing perl581 14:05:16< broeken> i've talked about this with thesin and that's one thing i shouldn't do 14:05:56< broeken> i'd rather stick with 586 14:05:56< dmacks> vasi (Mr. mac-glue) should be here within a few hrs...get him to add -pm586 of it. 14:06:14< broeken> would be nice, dmacks 14:10:33< broeken> i'll see what vasi comes up with 14:10:39< broeken> bye for now! 14:10:48-!- broeken [n=broeken@5353014C.cable.casema.nl] has quit [] 14:11:40< dmacks> Any Particular Reason for the vegas trip? 14:12:00< RangerRick> because vegas is a blast :) 14:12:00< cirdan> heh 14:12:33< RangerRick> the Particular Reason you might be referring to is coming up soon, most likely, but not in Vegas :) 14:12:39< dmacks> Gotcha. 14:12:59< cirdan> RangerRick: scary 14:15:41< RangerRick> yeah :) 14:21:19< dmacks> *grr*...thanks for the resource leak, mozilla 14:29:45< akh> hmmm...looks like xcircuit upstream is getting ready to wind up the current (3.3) series. I saw a 3.4 tarball in their archives. 14:30:02< dmacks> oooooo 14:30:05< akh> Of course, I'm already 4 revs behind top-of-tree on release. 14:30:30< akh> Easily remedied. 14:31:32< akh> 4 revs in 11 days--that's seems about standard. 14:32:20< dmacks> wow 14:32:56< akh> I only move new versions to stable that have been around for a couple of weeks. 14:33:09< dmacks> "stable" in the temporal sense? 14:34:04< akh> Yeah--they're all stable in terms of building and running--especially after I told upstream "no damn hardcoded /sw and 'optimization' flags that break Tiger builds". 14:37:04-!- ambs_ [n=ambs@bl5-161-137.dsl.telepac.pt] has joined #fink 14:37:33< dmacks> When you finish with them, you can take on gtranslator, which only defines $KILLALL on linux and sunos...so on any other platform it spawns a(nother) copy of the process instead of killing the existing ones. 14:37:44< RangerRick> haha 14:37:48< RangerRick> nice 14:37:57< akh> ouch 14:38:58< akh> Hmmm...why does "fink validate" not even say "Package looks good!" when I'm on Verbose: 0 ? 14:39:26< dmacks> 'cuz that's how it's always been? 14:39:44< akh> And why is that? 14:40:00< dmacks> Oy, always with the questions! 14:40:11< dmacks> Probably for use in scripts or somesuch. 14:40:26< dmacks> Gotta be at maximum I think to get it. 14:40:37< akh> Makes sense. 14:41:01< akh> Ick--cvs is like 77K molasses today. 14:41:02< dmacks> (Verbose:2 doesn't even display that msg) 14:41:07< dmacks> ya 14:41:50< dmacks> coffee! 14:41:54-!- dmacks is now known as dmacks_away 14:45:50< regeya> message to whoever decided to make iTerm's icon dance when there's new text in the terminal: die. 14:46:47< akh> regeya: Yeah, that seems unnecessary, doesn't it? 14:47:58-!- Albie [n=ambs@bl5-165-61.dsl.telepac.pt] has quit [Read error: 110 (Connection timed out)] 14:48:06-!- ambs_ is now known as Albie 14:52:52< RangerRick> must be configurable, mine doesn't bounce 14:54:20< regeya> looked for it, will look for it again. 15:02:08-!- regeya [n=shane@adsl-sp3-cdale176.micgi.com] has quit [Remote closed the connection] 15:06:41-!- regeya [n=shane@adsl-sp3-cdale176.micgi.com] has joined #fink 15:11:32-!- Albie [n=ambs@bl5-161-137.dsl.telepac.pt] has quit ["Leaving"] 15:13:28-!- akh [n=akhansen@ldx3.psfc.mit.edu] has left #fink [] 15:21:59< dmacks_away> but but but...I just *installed* a new itunes two days ago! 15:32:05< newmanbe> Heh, I don't install silly updates like those. 15:48:44< newmanbe> Silly dmacks. 15:49:42-!- Snaggle [n=nieder@128.252.206.171] has joined #fink 15:51:41-!- baba [n=baba@YahooBB220041000015.bbtec.net] has quit ["Leaving"] 15:51:58< Snaggle> can someone check a security update in for me? 15:54:31-!- vasi [n=vasi@modemcable133.147-70-69.mc.videotron.ca] has joined #fink 15:55:07< dmacks_away> Snaggle: same .patch? 15:55:14< Snaggle> yes 15:55:49< newmanbe> It's vasi! 15:56:07< vasi> uh oh 15:56:19< newmanbe> !lart lag 15:56:19 * Melian readies the nuke launcher and fires some rounds at lag 15:56:45-!- sQuare [n=sqrmac@62.147.23.162] has joined #fink 15:57:05< newmanbe> This game of updating .info files is annoying. 15:58:22< newmanbe> Especially when the location of the source changes. 15:58:35 * newmanbe blames RangerRick. 15:59:58< vasi> meddle not in the affairs of Whacko Elitists :-P 16:01:09< dmacks_away> Snaggle: So none of the .dylib versioning has changed? (/me doesn't have excess cycles to compile it now, but it's just a %v++ so I'll commit it anyway) 16:01:37 * newmanbe suggests that fink has a setting to ask if you want change the Source-MD5 for packages in the local tree. 16:01:58-!- Albie [n=ambs@bl5-161-137.dsl.telepac.pt] has joined #fink 16:02:57< Snaggle> no, everything looked the same from 1.0.6 to 1.0.7 16:03:37< dmacks_away> Alrighty...just waiting on cvs.sf............... 16:03:40< newmanbe> Hmm, /me finds another bug in curl. 16:04:15 * newmanbe has found two bugs and declares curl not-safe. 16:04:24-!- rajesh [n=rajesh@nylug/member/rajesh] has quit ["leaving"] 16:04:29 * newmanbe wonders what that makes fink. ;) 16:05:00< dmacks_away> "able to use a variety of downloaders to suit individual tastes and tolerances for bugs"? 16:05:12< newmanbe> Heh. 16:12:15-!- Albie [n=ambs@bl5-161-137.dsl.telepac.pt] has quit ["Leaving"] 16:17:04< dmacks_away> Snaggle: committed 16:17:11< Snaggle> dmacks: thanks 16:17:30< dmacks_away> vasi: broeken was asking about getting mac-glue-pm586... 16:17:41< vasi> dmacks, uh yeah... 16:18:03< vasi> see i have a patch-set for mac-glue which lets it put the glues somewhere not-in-%p 16:18:24-!- dmacks_away is now known as dmacks 16:18:27< vasi> (because %p/lib/.... is a bad place to leave them when mac-glue is uninstalled) 16:18:38< vasi> but it turns out my patch-set is a little wonky 16:19:01< vasi> maybe i'll just do a simpler patch 16:19:38< dmacks> For now can you just add an additional perl Type to the existing pkg? /me doesn't know what the dependency tree looks like though. 16:24:14-!- Snaggle [n=nieder@128.252.206.171] has left #fink ["Leaving"] 16:33:03-!- dmacks is now known as dmacks_away 16:40:20-!- kane_ [n=kane@perl.xs4all.nl] has quit [] 16:53:33-!- dmacks_away [n=dmacks@pdpc/supporter/active/dmacks] has quit [Read error: 110 (Connection timed out)] 17:02:02-!- sQuare [n=sqrmac@62.147.23.162] has quit ["DABASSIST.COM // French Drum & Bass"] 17:06:31-!- RangerRick is now known as RangerAway 17:30:09-!- lisppaste [n=lisppast@common-lisp.net] has quit ["Common Lisp IRC library - http://common-lisp.net/project/cl-irc"] 17:30:14-!- dmacks [n=dmacks@pool-70-22-31-132.balt.east.verizon.net] has joined #fink 17:30:20-!- lisppaste [n=lisppast@common-lisp.net] has joined #fink 17:32:06-!- baba [n=babayosh@YahooBB220041000015.bbtec.net] has joined #fink 17:33:38< dmacks> If I have a shell script A run a shell script B, can B learn the arguments passed to A without A passing them explicitly to B? 17:34:11< brendan> don't think so 17:34:16< newmanbe> cat ~/.bash_history ! 17:34:24< brendan> they are in the environment or the cmdline or nowhere 17:35:14< dmacks> ps seems to be able to figure them out, so there must be *some* low-level API to find out "the argv of a given PID" or somesuch? 17:35:53< brendan> yeah, it's in /proc on linux 17:36:30< brendan> suppose you could just parse ps :) 17:38:27< baba> how can i know OS version (10.4, 10.3, ...) in shell? 17:38:38-!- drm [n=drm@dhcp-152-224.asci.uchicago.edu] has joined #fink 17:38:52< newmanbe> uname will tell you the version of Darwin. 17:39:01< baba> yeah, i know 17:40:18< baba> i just want to set MACOSX_DEPLOYMENT_TARGET 10.3 or 10.4. do i need to judge from Darwin version? 17:40:49< newmanbe> That is what I would do. 17:40:52< newmanbe> *shrugs* 17:41:57< brendan> dmacks: ps -o pid,command | grep $PPID | cut -d ' ' -f 3- 17:45:34< baba> yeah, but i am sure it would cause a prob in the future... 17:45:55-!- jtyler [n=jtyler@iphost-64-56-130-194.edm.wiband.net] has quit ["Leaving"] 17:46:16< brendan> uname maps directly to darwin version 17:46:29< newmanbe> Yes, we've established that. :) 17:47:28< baba> why apple doesn't set an environment variable like MACOSX_VERSION? 17:47:41< baba> why? why? whhhhyyy? 17:47:49< newmanbe> Darwin Major Version - 4 = Mac OS X second level version. 17:48:00< newmanbe> Think that's right anyway. 17:48:06< newmanbe> baba: That would make too much sense. 17:48:15< newmanbe> And we couldn't have that now could we? :) 17:49:08< newmanbe> !lart broken patch files 17:49:08 * Melian puts on a hockey mask and jumps out at broken patch files 17:50:00< brendan> what's the difference? 17:50:19< newmanbe> brendan: Between what? 17:50:56< brendan> between uname -r and MACOSX_VERSION? 17:51:11< brendan> except that uname is a lot more portable 17:51:31< newmanbe> The version of Darwin is not the same as Mac OS X. 17:51:40< newmanbe> Mac OS X 10.2 is in the Darwin 7 family. 17:51:56< newmanbe> 6 Family I mean. 17:52:06< newmanbe> While Mac OS X 10.3 is Darwin 7. 17:52:37< newmanbe> And MACOSX_DEPLOYMENT_TARGET doesn't want the version of Darwin. 17:52:47< baba> i am afraid of the possibility apple will break the rule 17:53:01< brendan> export MACOSX_VERSION=10.$(($(uname -r | cut -b1) - 4)) 17:53:09< brendan> I don't think they will 17:53:24< newmanbe> I don't trust them not to. 17:53:29< dmacks> Careful...there have been x.y.z where z!=0 17:55:41< brendan> I'd be really surprised if the first digit failed to match 17:55:50-!- JesseW [n=chatzill@pdpc/supporter/student/JesseW] has joined #fink 17:56:49< dmacks> Use `system_profiler SPSoftwareDataType` ? 17:56:51< baba> in .info, is it ok to use the export command, or better use with 'fink list macosx'? 17:57:22< dmacks> MACOSX_DEPLOYMENT_TARGET is set for you in .info scripts:) 17:57:33< baba> is it? 17:57:44< baba> since when? 17:58:06< dmacks> Long time I think. 17:58:20< dmacks> Check 'fink dumpinfo -fenv some-package-name' 18:00:43< pogma> dmacks: darth pogma ??? 18:01:05< drm> darth pogma? 18:01:14< baba> hehe, i've been wasting our time 18:01:28< baba> thanks dmacks and newmanbe and brendan 18:01:29< baba> anway 18:01:47 * pogma saw nick hilited in this channel, it was dmacks saying "darth pogma" 18:01:49< dmacks> You're welcome:) 18:02:13< baba> darth pogmar 18:02:18< pogma> 03:23 < dmacks> So be it 18:02:44< dmacks> pogma: Yeah, read the few lines around it for context:) 18:05:16< dmacks> baba: But if you really want it, 'defaults read /System/Library/CoreServices/SystemVersion ProductVersion' is a pretty easy way to get the OS X Version. 18:06:04< baba> oh thanks 18:06:07 * dmacks wonders why Fink Installer manually reads .plist instead of just using defaults 18:07:30-!- baba [n=babayosh@YahooBB220041000015.bbtec.net] has quit ["Leaving"] 18:15:00 * dmacks wonders why dmacks wonders about the installer:/ 18:16:15 * drm looks on in wonder 18:16:39< dmacks> heh 18:18:06 * newmanbe steals all of RangerRick's knowledge on making patch files. 18:19:03< newmanbe> Hmm, to finish removeing KDE, I am installing some KDE packages. 18:19:06 * drm hopes that newmanbe stole a *copy of* the knowledge 18:19:18< newmanbe> But there is a net-loss! 18:19:25< dmacks> Try better cabling? 18:19:42< newmanbe> drm: No, I'm on my to becoming a Wacko Elitist. 18:19:52< newmanbe> dmacks: No, net-loss of KDE packages. 18:20:47< dmacks> *gah* Our whole scrollkeeper usage is bolloxed. 18:20:50 * newmanbe is amazed at how much more fits in a terminal by reducing the font size by .5. 18:21:47< drm> dmacks: whaddya mean? 18:22:57< dmacks> It appears to usually *not* get called during PostInst, which leaves dangling pointers to the .omf files in the package that got removed. 18:24:20< drm> hmmm... maybe we should automate this with an OMFfiles command? 18:25:51< dmacks> Yeah. vasi'n'I brainstormed the ability to offload common Pre/PostInst/Rm tasks into external scripts (scrollkeeper, gconf, perlpod). Hence my question about unwinding the shell caller stack. Want it to be easy, not hafta pass lotsa arguments to these things. 18:26:00< dmacks> oop...dinner 18:26:02-!- dmacks is now known as dmacks_away 18:26:50< vasi> er, catching up with log 18:27:24< vasi> we should probably auto-detect common things like scrollkeeper and POD and such 18:27:30< vasi> InfoDocs too 18:27:36< vasi> er oops, dmacks is gone 18:28:38< vasi> drm, thought about the dist-up work? 18:29:55< drm> vasi: thinking about it... actually, i've almost reached the conclusion that we'll have to wait for xcode 2.2 before making our 10.4 tree 18:30:01< vasi> drm, okie doke 18:30:08< drm> this would actually give us some time, perhaps, to think through dist-upgrade 18:30:13< vasi> any idea when 2.2 will be going final? 18:30:19< vasi> we can try to get dist-up into 0.25 18:30:22< drm> but i may implement a primitive version of it soon, for use by fink-core folks 18:30:39< vasi> keep me updated with 'what you want to happen' 18:30:44< drm> vasi: no idea about 2.2 ... could be tomorrow, for all i know :) 18:30:48< drm> sure 18:31:16< vasi> is there a plan yet? 18:31:27< vasi> ie: for what should happen 18:32:18< drm> no 18:32:46< drm> i mean, there is going to be some technical stuff about how you create the new tree and so on, and that is what i will work on, soon 18:33:02< drm> but the basic question of what kind of updating is done for users is not yet resolved 18:33:12< drm> probably needs an email discussion 18:33:41< vasi> been there, done that :-( 18:33:58< vasi> i think it was cirdan who brought it up on -devel a few times 18:34:27< vasi> basically reached the same conclusion as always: no perfect solution, and every imperfect one seriously bugs at least one person 18:34:37 * drm will read the archives 18:35:18< drm> on another matter, it looks like we might need a quick fink release to handle an error in the gettext upgrade process 18:35:31< drm> not sure yet, still need to try to think of another solution 18:35:45< vasi> er...shouldn't the new gettext just replace the -tools? 18:35:54< drm> basically, gettext-tools used to be a splitoff of gettext, and so depended on a specific version of gettext 18:36:10< drm> gettext-tools is now its own package, and it is no longer updated automatically when you update 18:36:43< drm> so when you run fink self-update, fink definitely wants to update gettext, but then notices that this will violate a stated dependency of the already-installed gettext-tools 18:36:50< drm> (i think that's whats going on, anyway) 18:37:14-!- spundun [n=spundun@mermaid.isi.edu] has joined #fink 18:37:16< vasi> well the thing that has me confused... 18:37:16< spundun> hi all 18:37:19< drm> the only workaround i've thought of is to have fink *remove* gettext-tools if the old one is present, before starting to upgrade 18:37:27< drm> (but i naven't tested this yet...) 18:37:42< vasi> there's an old gettext-tools that works with gettext 0.10.40-19 18:37:47< drm> right 18:37:49< vasi> and a new one that works with libgettext3 18:37:52< drm> right 18:38:09< drm> but the old one is installed, and its dependencies say "=", not ".=" 18:38:12< vasi> why isn't there one that works with the current unstable 'gettext' (vers 0.10.40-*)? 18:38:18< drm> ">=" 18:38:49< vasi> if it's needed, then it should exist 18:38:56< drm> vasi: i re-jiggered the splitoffs, so we don't have to build as much during a bootstrap 18:38:58< vasi> if it's not, then new gettext should Conflict/Replace it 18:39:29< drm> its not a question of conflict/replace on the gettext... the point is, gettext is being updated to a new version, but gettext-tools is not 18:39:34< spundun> Question: am I the only one getting trouble building gnome-spell-1.0.6 on 10.4-transitional? 18:40:08< spundun> 10.4 transitional is still the active branch right? or have people moved on? 18:40:21< drm> spundun: still the active branch 18:41:26< drm> vasi: sorry, gotta run 18:41:34-!- drm [n=drm@dhcp-152-224.asci.uchicago.edu] has quit ["Leaving"] 18:42:18< spundun> ok... anyone reportin success building gnome-spell-1.0.6 on 10.4-transitional? 18:42:40< spundun> I looked at the problem.. and looked like pretty much everyone should have the problem 18:44:51< vasi> i'll take a look 18:45:08< spundun> vasi: you talking to me? 18:45:16< vasi> yah 18:45:19< spundun> vasi: ok... thanks 18:45:49< vasi> i don't have a ton of time right now, so if i don't have any quick ideas i won't go exploring 18:46:25-!- Feanor [n=astrange@opendarwin/developer/feanor] has joined #fink 18:49:35-!- JesseW [n=chatzill@pdpc/supporter/student/JesseW] has quit [Remote closed the connection] 18:52:23< spundun> vasi: I already have solution... can you atleast verify that its broken right now? Just to confirm that its not just me... 18:52:33< vasi> will do 18:52:43< spundun> vasi: I just created a patched and it worked... 18:54:27-!- hennker_ [i=flullup@dsl-082-083-238-168.arcor-ip.net] has quit [Read error: 110 (Connection timed out)] 19:21:33-!- akh [n=akhansen@68-118-244-23.dhcp.oxfr.ma.charter.com] has joined #fink 19:23:21< vasi> ugh, my dock just fubared 19:23:56< vasi> spundun, yeah i see a bug where it tries to run %i/share for some reason 19:24:34< vasi> that what you were expecting? 19:40:08< newmanbe> I don't care that a package is in a very bad inconsisten state anymore! 19:40:12< newmanbe> Just remove yourself. 19:40:21< vasi> huh? 19:40:31< newmanbe> dpkg needs a --really-foce-all option. 19:40:54< newmanbe> I have two packages that have an unmet dependency. 19:41:02< newmanbe> And it doesn't like it when I install that dependency. 19:41:19< newmanbe> Which it wants be to do to remove the packages in a very bad inconsistent state. 19:41:24< newmanbe> (apt that is). 19:41:30< newmanbe> Fink doesn't care. 19:42:32< akh> !lart useless users 19:42:33 * Melian decapitates useless users conan the destroyer style 19:42:46< akh> "I got breakage, but I won't tell you what" 19:42:57< akh> "It's the docs fault" 19:42:59< newmanbe> akh: All the users are useless. 19:43:01< akh> yup 19:43:05< newmanbe> Just develop for yourself. ;) 19:43:33< akh> That would make documenting easier. 19:43:56< akh> PEBKAC 19:44:12< newmanbe> !wtf pebkac 19:44:17< Melian> PEBKAC: problem exists between keyboard and chair 19:44:27< newmanbe> Ah yes. 19:44:41< akh> It's a good acronym. :-) 19:44:55< newmanbe> !lart the webmin package 19:44:56 * Melian urinates on the webmin package 19:45:04 * akh works on restoring my laptop HD. 19:45:05< newmanbe> And its related theme packages. 19:45:16< akh> What'd they do to you? 19:45:26< newmanbe> They don't want to remove themselves. 19:45:53< newmanbe> They want me to install the webmin package to remove it, but it just spews errors at me. 19:46:00< akh> ah 19:46:10< akh> dpkg -r --force-depends 19:46:20< akh> Or even dpkg -r --force-all 19:46:31< newmanbe> I've tried that. 19:46:41< akh> Possibly you might need --force-reinstreq 19:47:34 * akh installs 8f28 19:47:59-!- dmacks_away is now known as dmacks 19:48:03< newmanbe> Or mv /sw /sw-old 19:48:11< akh> There's that, too. 19:48:21 * newmanbe wonders if that would take a long time. 19:48:24< akh> Or dpkg has a forget option, too. 19:48:31< akh> mv is quick. 19:48:56< newmanbe> Done! 19:49:01< akh> As I said. 19:49:03< newmanbe> !download 19:49:32< newmanbe> And KDE has been giving me headaches trying to remove it. So this is probably for the best. :) 19:50:00< akh> Ah. Yeah it doesn't auto-remove as well as the wacko-elitist potty-mouth instructions claim. 19:50:25< newmanbe> Hmm, what happened to Darwin 5? 19:50:37< newmanbe> 10.1 is Fink v4 19:50:42< dmacks> Wot's up with gnome-spell? 19:51:04< newmanbe> It got moved to /sw-old 19:51:10< newmanbe> :) 19:51:33< akh> newmanbe: Fink's distro versions didn't originally try to correspond to the Darwin version. 19:51:42< newmanbe> Ah. 19:52:02< akh> It just happened that Panther came out when Fink was at 0.6.x 19:52:21< akh> So it was decided to roll with it. 19:52:32< dmacks> (and "fink" the program wasn't independent of the Fink distro) 19:53:04< newmanbe> Before my time. :) 19:53:10< dmacks> ...eventually the "fink" package took on a life of its own, leaving Fink with no real rationale for version-numbering. 19:53:25< akh> Yah 19:54:04 * akh just loves the Debian kernel upgrade message. 19:54:05< newmanbe> And your homework for tonight class, is to write a ten page, signle-spaced, 12 point Times New Roman, 2.5cm margin paper on the history of Fink. 19:54:15< newmanbe> Upgrade now! 19:54:35 * akh should consider updating that article I wrote and submitting it. 19:54:36< newmanbe> Or else you will be h@X0r3d. 19:54:58< dmacks> Ah yeah..../me remembers that a while ago akh. Got back-burnered? 19:55:18< akh> Yeah--our experiment started running. 19:55:56< newmanbe> k, so I installed this fink thing. 19:56:04< newmanbe> And there's like this package I want to use. 19:56:13< newmanbe> But it says its in some thing called unstable. 19:56:16< akh> RTFFAQ! 19:56:18< newmanbe> On the website I mean. 19:56:22< newmanbe> What's with that? 19:56:37< newmanbe> Because it totatly doesn't make any sense. 19:56:50< akh> And the sad part is, that's about how the questions come out. 19:57:16< newmanbe> !tell me about unstable 19:57:33< dmacks> Yup. Or they say "what's wrong? what should I do now?" when something crashes, and then quote an error message that gives both technical and practical answers to both. 19:57:42< akh> Yeah, that too. 19:58:04 * akh remembers with shame quoting an error from "execution of make failed" 19:58:37< dmacks> The buildlock "try doing [this] then [this]" is a common head-scratcher. 19:59:01 * newmanbe goes to make a list of packages he needs to install. 19:59:03< akh> Indeed. You should find a way to have fink use "say" to yell those out. 19:59:36< newmanbe> Hmm, zsh doesn't much like being moved while it's running. 20:00:10< dmacks> Maybe it should just set a flag with a timestamp, and then Mail.app can see if it's run immediately and say "I am a jackass" loudly. 20:01:06-!- akh_ [n=akhansen@68-118-244-23.dhcp.oxfr.ma.charter.com] has joined #fink 20:01:45< newmanbe> Melian: Tell me all the fink packages I use regularly. 20:02:04< newmanbe> Hmm, she won't tell me. 20:02:33< dmacks> pr0n-downloader pr0n-downloader-shlibs irc-goofoff2 20:02:45< newmanbe> Hehe. 20:03:18< akh_> gopher 20:03:33< newmanbe> Yes, got that. ;) 20:03:53-!- akh [n=akhansen@68-118-244-23.dhcp.oxfr.ma.charter.com] has quit [Remote closed the connection] 20:04:05< akh_> KDE should have a client--"kopher" or something like that. 20:04:43< dmacks> It's a symlink to kobsolete-krap 20:07:00< dmacks> Does someone wanna tell me what's wrong with gnome-spell? 20:07:33< akh_> What's the symptom? 20:07:49< dmacks> vasi and spundun nodding at each other and saying "yup, broken" 20:08:00< akh_> ah 20:08:10< akh_> So you'd like a substantive answer or something. 20:08:13< vasi> dmacks, it tries to RUN %i/share 20:08:20< vasi> which is clearly a directory 20:08:31< dmacks> Ah that's bad. 20:08:51< vasi> yep! 20:09:03< dmacks> Where? 20:09:56-!- akh [n=akhansen@68-118-244-23.dhcp.oxfr.ma.charter.com] has joined #fink 20:10:31< vasi> in the installscript 20:10:49< vasi> i didn't look at 'why?' because spundun just asked me to verify that it occured 20:11:21< dmacks> Builds cleanly for me on 10.3. Weird. 20:12:36< vasi> one sec, i'll get you a log 20:16:41< akh_> Built for me on 10.4 AFAICT, unless this is nonfatal. 20:16:53-!- akh_ [n=akhansen@68-118-244-23.dhcp.oxfr.ma.charter.com] has quit [] 20:20:58< newmanbe> akh: There is a kio_something another Gopher client for Konqueror. 20:21:03< akh> ah 20:21:53< newmanbe> Fink-0.7.2-Install has no mountable file systems. 20:22:08< akh> 0.7.2? That old thing? 20:22:40< dmacks> heh 20:23:22< newmanbe> That's no good... 20:23:43< dmacks> Works For Me(tm). 20:24:05 * newmanbe goes to download again. 20:24:19< newmanbe> This time from the Creator. :) 20:24:34< dmacks> MD5 (Fink-0.7.2-Installer.dmg) = 6a2f7aaf7ddd5d8daface872779f4d60 20:24:35< newmanbe> The question is, who is the Creator? 8) 20:25:02< dmacks> "hDmp", whatever that means. 20:25:32< newmanbe> 845f9a51d9302cf2bd34fea63cc1c950 20:25:38< newmanbe> No match. 20:26:15< dmacks> SF oughta allow addition of checksum metadata to files in FRS. 20:26:37< vasi> lisppaste: url? 20:26:38< lisppaste> To use the lisppaste bot, visit http://paste.lisp.org/new/fink and enter your paste. 20:26:53< vasi> Sf oughta do a lot of things 20:27:06< dmacks> true true:/ 20:27:12< lisppaste> vasi pasted "gnome-spell suckage" at http://paste.lisp.org/display/11904 20:28:43< cirdan> yo 20:28:54< cirdan> vasi: what's up?? 20:28:59< newmanbe> Yo. 20:29:03< vasi> hiya cirdan 20:29:09< akh> hey 20:29:20< vasi> cirdan, have you talked to drm lately? 20:29:33< dmacks> vasi: that's weird (gnome-spell) 20:29:40< vasi> yah 20:29:53< cirdan> yeah 20:29:53< cirdan> said "hi" 20:29:54< cirdan> :-) 20:30:07< cirdan> just found what u said, about dist-upgrade 20:32:02< cirdan> we can add compiler checks, to enforce xcode 2.2... 20:33:08< cirdan> i have a feeling maybe 10.4.3 and xcode 2.2 will be released the same or almost time 20:33:35< akh> Sounds plausible 20:33:56< akh> Maximize the breakage 20:33:57< dmacks> "never" and "round about never or so"? 20:34:06< cirdan> the day after never :-) 20:34:19< cirdan> ...or the day before 20:34:26< cirdan> i keep forgetting which 20:34:40< akh> Probably the former 20:36:55< cirdan> does it really matter? 20:36:58< cirdan> ;-) 20:38:48< akh> nah--not when you're talking about asymptotics 20:39:09< dmacks> Well we're getting close to talking about asymptotics... 20:39:31< akh> heh 20:41:42< cirdan> is that like asymoptics? 20:41:57< akh> Yeah, or "asymptomatics" 20:42:17< akh> "At first I applied the Bessel functions asymptomatically" 20:42:39< cirdan> asymptomaticdisestablishmentanerinisim 20:42:49< cirdan> whew 20:44:11 * dmacks can't make gnome-spell fail to build either as root or --build-as-nobody, with or without the pkg already being installed. 20:44:28 * dmacks hands off to spundun. 20:44:41 * newmanbe makes it fail. 20:44:48< newmanbe> /sw/bin/fink install gnome-spell 20:44:55< newmanbe> /sw/bin/fink doesn't exist. ;) 20:46:00< newmanbe> /usr/local/bin/download fink really really faster 20:46:22< vasi> i suspect it relates to an env var 20:47:01< vasi> i take that back 20:47:21< dmacks> akh: If openvrml:BuildDepends:mozilla-dev, does it need to be in section:crypto? 20:47:36< vasi> i think so 20:48:35< cirdan> yay 20:48:40< cirdan> security update and ipod update 20:48:53< akh> dmacks: ugh. The unstable version wasn't there, either. 20:49:06< vasi> dmacks, ok i know part of why the gnome-spell thing happens 20:49:21< vasi> in the Makefile, it has: $(mkinstalldirs) $(DESTDIR)$(datadir) 20:49:28< vasi> and mkinstalldirs isn't defined to anyhting 20:49:39< vasi> so it just ends up trying to run $(DESTDIR)$(datadir) 20:49:40< cirdan> and? 20:49:41< cirdan> oh 20:49:47< cirdan> hehe 20:49:58< vasi> dmacks, can you check if it's defined to something in your successful builds? 20:50:01< vasi> and if so, what? 20:50:52-!- Feanor [n=astrange@opendarwin/developer/feanor] has quit [Remote closed the connection] 20:51:55-!- beniamino [n=ben@adsl-64-164-10-189.dsl.snfc21.pacbell.net] has joined #fink 20:51:57-!- Feanor [n=astrange@opendarwin/developer/feanor] has joined #fink 20:52:01< dmacks> vasi: Do you have fileutils installed? 20:52:18< vasi> nope 20:52:23< vasi> hi beniamino 20:52:41< akh> dmacks, I did, most likely, when I built it. 20:52:41< beniamino> 'lo 20:52:51< dmacks> I did too.../me tries without it. 20:52:56< akh> beniamino: did half of the unison-aqua test. 20:54:02< beniamino> akh: yeah, got the email, thanks a lot... 20:54:10< akh> no problem 20:54:36< beniamino> it was hard to tell from bruno's email exactly what worked and what didn't... 20:54:45< beniamino> ....which made me doubt the results 20:55:08< akh> Yeah 20:56:17-!- regeya [n=shane@adsl-sp3-cdale176.micgi.com] has quit [Remote closed the connection] 20:57:32 * akh checks to see if the CVS trees flag is in man fink.conf 20:58:47< newmanbe> The Creator's SF.net mirror is slow. 20:59:53< akh> and this is surprising? 21:00:11< newmanbe> No, I suppose not. 21:01:36-!- asari [n=asari@p4144-ipbf601marunouchi.tokyo.ocn.ne.jp] has joined #Fink 21:01:57< newmanbe> !wb asari 21:01:59< Melian> Welcome back asari, o lonely traveller amongst the TCP/IP packets of chaos. 21:02:19< asari> hi newmanbe 21:02:30< newmanbe> 'ello. 21:03:08< vasi> "The Creator"? 21:03:18< newmanbe> University of Minnesota. 21:03:21< vasi> akh, --trees is on the wiki list of things-that-need-docs 21:03:24< vasi> lol 21:03:59< akh> vasi: ah. I _did_ find it in "man fink.conf", fortunately. 21:04:16< vasi> hmm, did i actually update the docs? 21:04:18< vasi> meh 21:04:37< akh> vasi: not --trees. SelfUpdateCVSTrees 21:04:45< vasi> ooooh 21:04:58< vasi> i thought you meant 'the --trees flag in fink-from-CVS' 21:05:08< vasi> !lart ambiguity 21:05:10 * Melian hereby declares ambiguity a troll 21:05:12< newmanbe> Yeah, about a minute left in download number two. 21:05:19< newmanbe> Kickban it! 21:08:59 * newmanbe submits a patch for the fink Installer.app package. 21:09:32< newmanbe> Remove the ladies and gentlemen should be venerable developers and very bad users. 21:10:16< akh> But we theoretically want people to use it. 21:10:43< akh> We're never going to get a full time GNOME maintainer if we scare people off. 21:12:10< newmanbe> Theoretically, we don't want people having problems. 21:12:19< newmanbe> By extention we don't want people using it. ;) 21:12:32< dmacks> Wow, determination of mkdir_p is messy. I give up. 21:12:35< akh> Right--no user, no problem. 21:12:52< akh> Was that a message from something or an editorial? 21:13:11< dmacks> Hehe:) The latter. 21:13:27< akh> Ah. 21:13:28< newmanbe> "Your startup scripts contain an error. I'm giving up. Bye." 21:13:45< newmanbe> Oops, the path setup script. 21:14:43< newmanbe> That is an odd soudning error message. 21:16:12-!- dmacks is now known as dmacks_away 21:18:52< akh> crap. lisppaste seems to be down and I have a HEAD error. 21:19:28< akh> Failed: Fink::SysState: Could not resolve inconsistent dependencies 21:19:28< akh> ...propagated at /sw/lib/perl5/Fink/PkgVersion.pm line 3941. 21:19:40< vasi> akh, use another pastebot? 21:19:44< vasi> or just /msg me 21:19:51< akh> That's it. After a failed gettext swap-in 21:19:52< newmanbe> #flood 21:20:09< vasi> akh, something about gettext-tools? 21:20:35< akh> No. I'm building something concurrently that requires libgettext3-dev. 21:20:48 * newmanbe sighs. 21:22:03< newmanbe> Now if I were Mac OS X, where would I store $SHELL... 21:22:36< vasi> i thought SHELL was defined by the shell you're using? 21:22:39< newmanbe> Ah yes, in the semi-black hole of NetInfo. 21:22:41< akh> yup 21:22:52< vasi> oh, THAT shell 21:22:53< newmanbe> vasi: I was speaking figurativly. 21:23:21< vasi> akh, http://pastebin.ca/ 21:23:25-!- spundun [n=spundun@mermaid.isi.edu] has quit ["This computer has gone to sleep"] 21:23:38< newmanbe> !comfort newmanbe 21:23:39< Melian> There, there, newmanbe. It's OK. I'm here for you. 21:26:27< newmanbe> !alicekill NetInfo\ Manager.app 21:26:29< Melian> I want to kill NetInfo\ Manager.app, I mean, I wanna, I wanna kill NetInfo\ Manager.app. Kill. I wanna see NetInfo\ Manager.app's blood and gore and guts and have NetInfo\ Manager.app's veins in my teeth. Eat NetInfo\ Manager.app's dead burnt body. I mean kill, Kill, KILL, KILL! and I jump around with newmanbe, yelling "KILL, KILL" 21:27:28< akh> vasi: http://pastebin.ca/23660 21:28:45< newmanbe> Okay, I su to a user that can can use sudo on any command. 21:28:55< newmanbe> But its shell just disappeared. 21:29:07< newmanbe> And NetInfo.app won't let me log in as the admin user. 21:29:11< newmanbe> Any ideas? 21:29:18< vasi> akh, looks like a stale buildlock 21:29:23< vasi> try 'fink cleanup --bl' 21:29:39 * newmanbe rereads the manpage for su and think he saw the answer. 21:29:47< vasi> we should probably do a auto-clean on startup, eh? 21:29:51< akh> No--it's not stale. That's an active build. 21:30:37< newmanbe> Nope, that didn't work. 21:31:41-!- lisppaste [n=lisppast@common-lisp.net] has quit [Connection reset by peer] 21:31:49-!- lisppaste [n=lisppast@common-lisp.net] has joined #fink 21:33:08< vasi> akh, but it didn't start building firefox yet 21:33:45< akh> Sure it is-I'm building firefox with another fink instance. 21:33:57< akh> Like I said--it's an active build. 21:34:34< akh> I know it's going to bomb, I just think that returning the error code will scare the users. ;-) 21:35:11< vasi> er...what would you like to happen 21:35:13< vasi> ? 21:35:22< akh> OK, that I'm not sure about. 21:38:55< vasi> i should probably remove the last bit (about the line the error occurs on) 21:39:00< vasi> that's mostly just for my own use debugging 21:39:47< akh> Ah. That was my only objection. 21:39:50< akh> Mainly because it was new to me. ;-) 21:39:55< vasi> i could detect if the problem relates to a buildlock....and then suggest to the user that they either wait for a build to finish, or do 'fink cleanup --bl' 21:40:36< akh> Yeah, though they'll ignore the message and fire off a cluelessgram to the lists... 21:41:10< akh> OMG--fink doesn't support my Tiger! 21:42:05< akh> Hmm.../me wonders if we need to get the newer CVS to stable for fink-0.25. 21:42:24< akh> (for SelfUpdateCVSTrees 21:43:04< vasi> uh, does that not work with apple's cvs? 21:43:58< newmanbe> s/does that/that does not/ s/cvs\?/\./ 21:44:05< newmanbe> And nix the 's too. 21:44:58< vasi> that was a remarkably clear way of putting things, newmanbe 21:45:10< newmanbe> lol 21:45:44< newmanbe> #sed must be bad influence on me. ;) 21:47:38< akh> vasi: That's right. 21:47:45< akh> It does not. 21:47:54< vasi> hmph 21:48:20< akh> And either it doesn't work for 1.11.20 or I'm doing something wrong. 21:49:18< akh> (probably the latter) 21:51:04-!- beniamino [n=ben@adsl-64-164-10-189.dsl.snfc21.pacbell.net] has quit [Read error: 110 (Connection timed out)] 21:51:08< akh> yeah, it was the latter. 21:51:38-!- beniamino [n=ben@adsl-64-164-10-189.dsl.snfc21.pacbell.net] has joined #fink 21:52:40< newmanbe> Ooh, the public radio show stopped screening its calls. 21:52:53< akh> That's dangerous. 21:53:02 * newmanbe just realizes it is a rebroadcast from this morning. 21:53:25< newmanbe> Half of the four phone lines weren't working. 21:53:27< akh> vasi: It helps if you pick a tree that actually exists. 21:53:37< vasi> heh 21:53:41 * akh was trying to do "10.4" 21:54:12< newmanbe> Getting ahead of yourself now aren't you. :) 21:54:19< akh> yup 21:54:28 * akh is ahead of my time. 21:54:31< akh> :-) 21:54:56< newmanbe> I'm even more ahead. 21:55:06< newmanbe> I use the GNU Subsystem instead of the BSD Subsystem. :-p 21:55:26< newmanbe> Then I'd finally have GNU date... 21:55:44< newmanbe> !lart non-GNU date 21:55:45 * Melian readies the nuke launcher and fires some rounds at non-GNU date 22:02:32< akh> And that's ahead how? 22:03:19< newmanbe> It has the --iso-8601[=blahblah] flag. 22:03:20< newmanbe> :) 22:03:31< akh> ah. 22:03:47< newmanbe> Instead of making me do it myself. :) 22:03:54< akh> Fair enough 22:04:41< akh> hmm...if I move openvrml from main/graphics to crypto, is a %r bump required? 22:05:02< newmanbe> Does it change the binary? 22:05:39< akh> That's the question. If I'm not totally off my rocker the section information does go in the binary. 22:06:03 * newmanbe shrugs. 22:06:12< newmanbe> Better safe than sorry? 22:06:18< akh> Yeah, probably. 22:06:24< newmanbe> The worst that happens is that someone has to rebuild it. 22:06:52< akh> Yeah, and I'm not the maintainer, so I won't have to hear about it. ;-) 22:07:18< newmanbe> Heh. 22:07:46 * newmanbe wonders how something in graphics would have crypto in it/depend on something crypto. 22:08:07< akh> It uses mozilla-dev 22:08:13< newmanbe> Ah. 22:09:03< newmanbe> You might have something to answer for if you bumped the revision for Mozilla... 22:20:16< akh> Yeah--I won't bother. 22:20:58-!- lane [n=lane@c-24-118-139-118.hsd1.mn.comcast.net] has joined #fink 22:21:18-!- shres [n=sshreyas@59.92.142.16] has joined #fink 22:21:39< akh> My laptop feels so sensitive now. 22:21:43< akh> (case, that is) 22:23:05< lane> Hi all. I've hit a compile error trying to do a fink install gnome-python-py23 on fink unstable on Tiger. I've done fink selfupdate and fink update-all, and I'm still hitting the problem. Is this a known issue? Or should I email the fink-users list and attach the error message? 22:23:24< akh> It's a known issue. 22:23:35< akh> The maintainer hasn't had time to do anything about it. 22:24:06< akh> You can download the .deb from the 10.3 tree and use it--this works. 22:25:24< lane> umm... how would I go about doing that? 22:26:11< lane> I've never done a manual download like that 22:26:17< akh> I'm working on it--just a sec 22:26:36< lane> sure 22:26:49< akh> Grab the file at http://bindist.finkmirrors.net/bindist/dists/10.3/release/main/binary-darwin-powerpc/gnome/gnome-python2-py23_2.0.0-16_darwin-powerpc.deb 22:27:26< vasi> did dmacks not commit the new gnome-python? 22:27:42< akh> Then change directories to where you downloaded it and use "sudo dpkg -i gnome-python2-py23_2.0.0-16_darwin-powerpc.deb" 22:27:54< akh> vasi: lemme check 22:28:45< vasi> clearly not 22:28:47< vasi> :-( 22:28:51< akh> No, it's there. 22:29:05< akh> (at least in CVS--can't guarantee rsync) 22:29:27-!- lane [n=lane@c-24-118-139-118.hsd1.mn.comcast.net] has quit [] 22:29:39< akh> wait--no. brain fart. That's his exp 22:29:46< vasi> uh, i don't see it 22:29:46< vasi> yeah 22:29:46-!- lane [n=lane@c-24-118-139-118.hsd1.mn.comcast.net] has joined #fink 22:30:02< akh> So he committed it, just not to anyplace useful. ;-) 22:30:10< vasi> useful to me :-) 22:30:17< lane> somehow got disconnected... I'm back now. So does that mean it's fixed, but not yet available? 22:30:35< akh> It's _possibly_ fixed. 22:30:52< lane> ok 22:31:05< akh> dmacks only has a 10.3 system, IIRC, so it'll need testing on Tiger. 22:31:22< lane> but I should be OK installing the 10.3 file that you pointed me to 22:31:50< lane> or I would be willing to try this new file, since I'm running Tiger 22:31:51< akh> Absolutely--I did it myself. 22:32:06< akh> (the 10.3 binary) 22:32:11< akh> I needed it for meld. 22:32:20< lane> yeah, I need it for gramps 22:32:59< lane> I have a family file with thousands of people, and even with 1.5 GB of RAM, Reunion still chokes when I try to open it. 22:33:05< akh> ouch 22:33:32< lane> gramps, on the other hand, chugs and chugs for a good long while, and then says, OK, here you go 22:34:29< akh> Definitely sounds better. :-) 22:35:57< lane> ah. gramps is compiling now. installing that binary worked 22:36:42< akh> Great. 22:37:06< vasi> akh, i tested it on Tiger for him 22:37:29< vasi> it works fine, at least for meld 22:37:37< vasi> so it might be worth downloading and using 22:38:06< akh> vasi: Ah. Didn't know that. 22:38:50< lane> oh, well, I've got gramps running now 22:39:31< lane> if you need another tester, I can. otherwise, I'll just keep what I've got for now 22:39:39 * akh will probably grab it later. 22:40:04< vasi> lane, testing would be great 22:40:30< lane> do I have to download something manually? 22:40:52< vasi> this: http://cvs.sourceforge.net/viewcvs.py/*checkout*/fink/experimental/dmacks/finkinfo/gnome/gnome-python2-py.info?rev=1.2 22:41:03< vasi> and then just put it in /sw/fink/dists/local/main/finkinfo 22:41:06< lane> I just did what akh suggested and manually installed the 10.3 binary 22:41:21< vasi> lane, what you've got now will work 22:41:33< vasi> i'm not trying to push you into testing if you don't want to 22:41:34-!- dmacks_away is now known as dmacks 22:41:44< vasi> hey dmacks 22:41:50< dmacks> hiy'all 22:42:01< lane> no, I can try it 22:42:11< lane> esp if that'll help get it into the distro sooner 22:42:23< lane> so download the file you linked? 22:43:02< vasi> thanks 22:43:15< vasi> dmacks, we've shanghai'd a tester for your gnome-python 22:43:20< dmacks> New gnome-python2 is not in unstable yet because some of its components require some libs that aren't in unstable yet. I could commit the part we do have, but that would mean the new version does not have all the modules that the old version did. 22:43:30< akh> ah 22:43:34< vasi> oh, what's it missing? 22:43:45< dmacks> (OTOH, given we don't really "have" an old version, not sure how much real harm would be done) 22:43:49< dmacks> RTFWiki 22:43:58< vasi> gotcha 22:44:32< vasi> i'm gonna prettify the tables some more 22:44:57< lane> so is the new version something that it's ok for me to try, given that I'm running unstable? 22:45:00< dmacks> I can compile gnome-menus, but don't know what to do about env vars that miga mentioned or conflicts with kde's menu files. I really don't understand menus... 22:45:20< akh> menus are devices by which you can visualize your options. 22:45:27< akh> But that's not important right now. ;-) 22:45:36< vasi> dmacks...gnome-python-py24 is fine, but the -extras isn't? 22:45:41< dmacks> Looks like I picked the wrong week to stop sniffing glue! 22:46:02< akh> lane: most likely--you can revert to the one you downloaded from 10.3 if you don't delete that .deb 22:46:06< dmacks> Even parts of the former are absent. 22:47:39< lane> if I install the new package, will it be missing anything I need for gramps? 22:47:50< dmacks> Okay, let's decide this now: commit the new, potentially-functionally-deficient version to unstable and upgrade it further as new libs go in, or wait? 22:48:21< lane> I'd say commit - but I'm just a user 22:48:40< lane> after all, it is *unstable* 22:48:50< dmacks> Or maybe commit only to 10.4, where we sorta have nothing at all now? 22:48:57< lane> yeah 22:49:00< lane> that's what I'm on 22:49:17< vasi> dmacks, re: the gtk/cairo stuff...i think i've changed my mind 22:49:35< vasi> (dmacks, yeah commit to 10.4T and leave 10.3 alone for now sounds good) 22:49:39< dmacks> vasi: Hold that thought...let's solve this first, and discuss that while CVS thinks about it.... 22:51:07< dmacks> Okay, will make one final compile so I know what's in and out, and commit it. 22:51:30< dmacks> Meanwhile, somewhere in Egypt..... 22:52:25< vasi> or illinois 22:53:30< dmacks> Good point. Well either way, what's your thought now? 22:53:38< vasi> er, wait 22:53:52< vasi> new gtk IS backwards compat, but not forwards? 22:54:20< vasi> or is it not BW, either? 22:54:29< lane> ... and in other news, KDE won't give me anti-aliased fonts. Every time I fire up KDE I go to the control center and select anti-aliasing, and then when I re-start KDE, I still get no anti-aliasing and the option's not selected in the control center 22:54:31< dmacks> Something that was compiled using the new -dev will not run with the old -shlibs. 22:54:44< dmacks> Compiling with old -dev will run fine with new -shlibs. 22:55:01< akh> lane: Do you have a ~/.qt that isn't owned by you? 22:55:22< dmacks> Unless we can verify that there are no indirect symbol refs or header changes (in which case we can fiddle with .la and .pc, and there will be no problems at all) 22:55:28< lane> oh - why yes I do 22:55:33< lane> it's owned by root 22:55:37-!- xhrl [n=ThomasW@S0106000f3d5d5bed.vs.shawcable.net] has joined #fink 22:55:38< vasi> yeah, i was thinking about the .la munging 22:55:48< akh> lane: That's the usual culprit. 22:56:00< lane> so I just need to change that back to owned by me? 22:56:10< akh> I think so--if that doesn't work you can remove it. 22:56:24< vasi> hmm, so if it's bw compat, the 'right thing to do' is probably to do the rev-bumping, yep 22:57:01< vasi> this would be a really nice time to have Shlibs working, though :-( 22:57:15< vasi> cuz it will just automagically do the right thing in this case 22:57:20< dmacks> ya 22:57:51< vasi> depending on how badly you want to release the new gtk 22:57:58< vasi> and how quickly you think we can get 0.25 out 22:57:59< dmacks> I'm in no hurry. 22:58:06< vasi> you may want to just wait for Shlibs 22:59:12 * akh moved openvrml around. 22:59:19< akh> Now I can go to bed. 22:59:28< vasi> hmm, wrt to the buildlocks cleanup 22:59:29< dmacks> gtk+2 cannot be upgraded to its current (upstream) stable branch due to this, and other patches may be needed for that version anyway, so back-burner. 22:59:38-!- akh [n=akhansen@68-118-244-23.dhcp.oxfr.ma.charter.com] has quit ["Leaving"] 22:59:48< vasi> by akh 22:59:58< vasi> bye 23:00:09< dmacks> OTOH, pango1 is at its upstream-stable branch, which has *optional* cairo, so we can experiment with it without having to fight other verison-upgrade issues at the same time. 23:00:47< lane> akh: that fixed my antialiasing problem. thanks 23:01:00< vasi> yeah, so buildlocks cleanup...i'm thinking of adding to some fink module (Engine? Config?) a new method '$token = $pkg->add_cleanup_sub(sub { ... });' 23:01:20< vasi> so that means no more weird global vars, they're kept in the closure 23:01:26< lane> btw, if you guys think of it, you could mention to RangerRick that I'm going to try to compile digikam 23:01:46< vasi> lane, i bet he'd appreciate if you tell him how it goes 23:01:50< vasi> :-) 23:01:57< lane> I'll definitely do that 23:02:12< lane> I'm following the packaging tutorial :) 23:02:13< vasi> dmacks, and then END {} will deal with any cleanup subs 23:03:54< dmacks> I assume there's a remove_cleanup_sub also? Otherwise could accidentally remove buildlock that was later installed by a concurrent fink process. 23:04:12< vasi> dmacks, that's what the $token is for 23:04:14< vasi> :-) 23:04:53< vasi> so you can either do $token->remove or $pkg->remove($token), depending on whether or not we make it a new class 23:05:21< lane> I know this is a newbie question, but is it possible to display swing windows within fullscreen X? 23:05:30< lane> I would think not 23:05:36< vasi> swing as in java? 23:05:40< lane> yeah 23:05:56< vasi> afaik, there's no non-apple JVM that really works well on OS X 23:06:04< cirdan> i think they'll always run as a quartz window 23:06:06< vasi> so not really 23:06:15< lane> that's too bad 23:06:23< lane> but makes sense 23:06:41< vasi> if you have a fair bit of time, you could always try to hack one of the existing JVMs...there's prolly at least one that would compile pretty well on OS X / X11 23:06:49< dmacks> Stash a list of active-buildlock pkgnames in %Fink::Buildlock::active_locks, then Fink::Buildlock::END removes that list of 'em? 23:07:11< lane> but none of the 3rd party ones have swing implemented, do they? 23:07:18< lane> at least not the open source ones 23:07:33< vasi> dmacks, well i'd like a general technique....so we can use it to re-install BuildConflicts after we remove them, as well 23:08:37< lane> but I may take a look around to see what's out there 23:08:50< vasi> lane, some of them have swing i think 23:09:31< lane> getting digikam and k3b ported are much higher up on my list 23:09:53< dmacks> $token=new Fink::Buildlock($pv), then Fink::Buildlock::DESTROY clears the lock. 23:09:54< vasi> understandable :-) 23:10:07< vasi> dmacks, yeah that works too 23:10:23< vasi> again, i'd want it to be extensible :-) 23:10:38< dmacks> That's very general, since whenever you let go of the token, the lock goes away. Or whatever was removed (for BCoflicts) gets reinstalled. 23:11:09< vasi> ah so you want the Buildlock package to also manage BConf removal? 23:11:22< vasi> that could work, since they kinda happen at about the same time 23:11:31< dmacks> I typed the second sentence 'cuz you say you wanted a generalized mechanism:) 23:13:07< vasi> given the number of bad BL situations we're seeing...i think we should auto-clean stale BLs at the start of real_install 23:14:52< dmacks> How about a Fink::PkgBuilder, which, at this time, has new($pv) set a buildlock and DESTROY remove it. Can eventually move the swappy-code removal into new() as well and swappy-code reinstall into DESTROY. 23:15:16< dmacks> Goal: an object that encapsulates "the build process" 23:15:44< vasi> hmmm...i was thinking of making a ->build() method in PV that would basically do all that 23:15:53< vasi> not sure if it's worth turning into its own class 23:16:04< vasi> (then again, prolly better than dumping more cruft into PV) 23:16:20< dmacks> It's all about having a DESTROY auto-cleanup. 23:16:46-!- lisppaste [n=lisppast@common-lisp.net] has quit [Remote closed the connection] 23:16:57< vasi> yeah, i'm just concerned about whether the DESTROY'd object should be caller visible 23:17:15< vasi> ie: Fink::PkgVersion::Cleanup, which is only ever used by PkgVersion 23:17:31< vasi> or Fink::PkgBuilder which anybody who wants to build a package will have to use 23:18:43< vasi> in a way i'd kinda rather use explicit destruction + END rather than DESTROY, because depending on ref-counting could bite us in the rear end unexpectedly later 23:19:08< vasi> (not sure how likely that is, just spouting ideas) 23:19:47< dmacks> What if we new() this object at the start of PV::phase_unpack, stash its ref in the $pv object, then release it at the end of PV::phase_build_deb (or whatever it's called)? 23:20:07< dmacks> That way it's entirely internal and automatic to the PV build process. 23:20:37< dmacks> (no tokens to hold, no external calls to be made) 23:21:55< vasi> sounds good 23:22:20< vasi> wrap the whole phase_foo(); phase_bar(); thing in a ->build() while we're at it? 23:22:54< vasi> unpack->patch->compile->install->build always go together i believe 23:23:16< dmacks> We don't build during bootstrap until we have deb toolchain. 23:24:23< vasi> well that can be conditional on whether we're in bootstrap or not :-) 23:24:26< dmacks> I could envision users wanting to unpack->patch and stop. 23:24:57< vasi> yeah, so we're not gonna remove the phase_* 23:25:28< dmacks> grep all of xml/fink for phase_ and see what patterns there are. 23:25:36< vasi> just make a global method which 'does everything' so any place we want to build things have a single call 23:25:48-!- lane [n=lane@c-24-118-139-118.hsd1.mn.comcast.net] has quit [Read error: 104 (Connection reset by peer)] 23:26:20< vasi> activate/deactivate/purge aren't involved.... 23:26:33< dmacks> right 23:26:43-!- xhrl [n=ThomasW@S0106000f3d5d5bed.vs.shawcable.net] has quit [Read error: 110 (Connection timed out)] 23:27:34< vasi> yeah, so each of the ones we care about is only used twice 23:27:38< vasi> in Bootstrap, and in Engine 23:27:55< vasi> with the only difference being that only Engine uses _build 23:29:03< vasi> anyway, it would be nice to get rid of (en|dis)able_bootstrap 23:29:04< dmacks> So PV::roll_the_whole_damn_thing($bool_create_deb) ? 23:29:29< dmacks> or $bool_we_are_bootstrapifying ? 23:29:32-!- lisppaste [n=lisppast@common-lisp.net] has joined #fink 23:29:34< vasi> er, PV::build_from_source(bootstrap => $bsbase) 23:29:39< vasi> :-) 23:29:48< cirdan> PV::roll_it_nice_and_tight($bool_create_deb)! 23:30:07< cirdan> :-) 23:30:07< vasi> uh, thanks cirdan 23:30:19< dmacks> It's possible that bootstrap status could affect dependencies or other stuff outside of the actual build. 23:30:21< cirdan> we need more 'fun' code in fink 23:30:54< cirdan> keeps ppl sane instead of wanting to gouge out their eyes :-) 23:31:09< dmacks> drm and I have broken (en|dis)able_bootstrap or unexpected side-effects of it too many times to count. 23:31:35< vasi> me too! i can break it good, just as good as you! :-P 23:31:46< dmacks> nu-uh! 23:32:34< dmacks> Tell ya what...we'll take turns swapping pairs of lines in Bootstrap and/or PV and see who gets killed by a murderous mob first. 23:32:41< vasi> anyway, the point isn't to get rid of *able_bootstrap...just to move it into PV::make_a_deb_dammit 23:33:23< vasi> heh 23:33:52< vasi> i'll just show them the Shlibs code until they run screaming :-P 23:33:55< dmacks> Yeah, looks like it's already a tight wrapper on the build process. I was expecting that not to be the case. 23:34:19< cirdan> vasi: don't be mean 23:34:32< cirdan> or *i'll* make you look at debbugs! 23:34:37< vasi> nooooooooo! 23:34:38 * dmacks responds with a real_install +2 strike. 23:34:39< cirdan> :-) 23:34:42< vasi> lol 23:34:53< cirdan> dmacks: saw debbugs once... 23:35:05< cirdan> wore 2 eye patches for weeks 23:35:07< cirdan> :-) 23:35:08 * vasi defends with a +5 vorpal refactoring 23:36:12< vasi> damn too much caffeine... 23:36:27< vasi> hate when i stand up and *BUZZZZZZ* 23:36:31 * cirdan tosses in a -3 code modularity 23:36:33 * dmacks upgrades all of gnome simultaneously, sending engine into a fork-bomb. 23:36:38< cirdan> hahaha 23:37:07< cirdan> vasi: they make stiff drinks to fix that 23:37:09 * vasi gets apple to 'fix' gcc again :-P 23:38:08 * dmacks adds an additional binary-incompatible provider of x11 23:38:36< cirdan> yay! 23:39:05< cirdan> once we get shlibs in fink, any x11 packages can auto dep on only compat x11 versions 23:39:28< vasi> true, true 23:39:30< dmacks> ...once we get maintainers to fix their damn Shlibs fields.... 23:39:31< cirdan> maybe even someone can add auto-varients ;-) 23:39:53< dmacks> ImplicitType: maintainer (cirdan) 23:40:01< cirdan> anything with a rundep on x11 fink will magically create varients 23:40:05< vasi> okily doke....so we're in agreement, more or less, about how we shall deal with buildlock removal? 23:40:29-!- cirdan is now known as kirdan 23:40:41< kirdan> so who's this unlucky cirdan guy? 23:40:43< kirdan> :-) 23:40:46 * vasi assigns $gnome_core = 'cirdan', since he obviously has too much time on his hands :-P 23:40:51< kirdan> hah 23:40:56< kirdan> so much time 23:41:07 * kirdan assigns fink.inc bills to vasi 23:41:13-!- kirdan is now known as cirdan 23:41:38< dmacks> Yeah. New pkg is... Fink::PV::BackEnd? Fink::PV::Buildtime? Fink::Buildtime? 23:41:41< cirdan> the lawyer wants another $2k retainer :-/ 23:41:47< cirdan> anyone have any donations? 23:42:01< dmacks> Tell 'im to stop sucking his damn thumb and his teeth will come in straighter. 23:42:06< cirdan> hehe 23:42:23< cirdan> we have a discount rate eve, $125/hr iirc 23:42:36< cirdan> he did fill out more irs paperwork and tax stuff 23:43:01< cirdan> speaking of whcich, we need at least 3 officers, pres, vice, and treas. 23:43:50< cirdan> but it's bedtime... 23:44:00< cirdan> night all 23:44:05< dmacks> I'll be the pres of vice. 23:44:15< cirdan> k 23:44:19< vasi> PV::Cleanup? 23:44:24< cirdan> maybe drm'll be in charge of $$ ;-) 23:44:32< dmacks> But it's also Setup 23:44:36< vasi> nite cirdan 23:44:51< vasi> PV::BuildWrapper? 23:45:07< dmacks> That'll work. 23:45:17< vasi> PV::ThingWithNoName 23:45:35< dmacks> PV::HardHat, because it protects the actual builder? 23:45:40< vasi> meh 23:45:59< vasi> how about automatically cleaning up stale buildlocks when we start real_install....any concern over that? 23:46:46< dmacks> By automatic, you mean "prompt", right? No automatic until we get flock/fcntl/whatever protection. 23:47:50< dmacks> ("prompt 23:48:00< dmacks> "==default N) 23:48:06< vasi> uh, yeah first do fcntl 23:48:24< vasi> then do real-non-prompting-automatic bl removal in real_install 23:48:39< dmacks> Roight. 23:50:51< dmacks> gnome-python2-pyXX into 10.4T! 23:50:56< vasi> yay! 23:51:05< dmacks> Enjoy meld:) 23:51:07< vasi> hey, do you have any thoughts about how to deal with the common x11 user vs. sdk error 23:51:30< dmacks> Kill all users, let G-d sort 'em out? 23:51:34< vasi> where someone has x11user but not x11sdk installed, and fink's dep engine pisses on 'em 23:52:16< dmacks> I thought we solve that in unstable, where fink offers "apple's X11" (or some name) as one of the x11 choices? 23:52:25< vasi> Mutually Assured Dependency-hell 23:52:29< vasi> oh does it? 23:54:26< dmacks> At least in 0.24, most system-* virtuals always exist as package objects, STATUS_ABSENT if what they represent isn't there. 23:54:54< dmacks> So $user sees system-xfree86 "virtual package representing Ap $hash->{description} = "[virtual package representing Apas a choice along with xfree86 and 23:55:08< dmacks> "[virtual package representing Ap 23:55:22< dmacks> ple's install of X11]" along with xfree86 and x.org actual pkgs. 23:55:26< vasi> i'm gonna try it and see 23:55:59< dmacks> Then if they pick it, fink aborts with a msg telling 'em to instlal it manually, and URL to TFFAQ 23:57:18 * dmacks -> sleep; lemme know what you find. 23:57:21-!- dmacks [n=dmacks@pdpc/supporter/active/dmacks] has quit ["leaving"] --- Log closed Fri Sep 23 00:00:07 2005 .