--- Log opened Mon Aug 01 00:00:27 2005
00:27 -!- eno-away is now known as eno
00:42 -!- JesseW [~chatzilla@JesseW.student.supporter.pdpc] has joined #fink
00:48 -!- tgpo [~tgpo@pcp427696pcs.naugus01.ga.comcast.net] has joined #fink
00:49 < tgpo> howdy everyone
00:52 -!- tgpo [~tgpo@pcp427696pcs.naugus01.ga.comcast.net] has quit [Client Quit]
01:06 -!- xhrl [~ThomasW@S0106000f3d5d5bed.vs.shawcable.net] has joined #fink
01:11 -!- shres [~sshreyas@59.92.134.80] has joined #fink
01:39 -!- xhrl [~ThomasW@S0106000f3d5d5bed.vs.shawcable.net] has quit [Read error: 60 (Operation timed out)]
01:45 < asari> Does fink have libmspack?
02:07 -!- dmalloc [~whatever@chello080108111128.5.11.univie.teleweb.at] has joined #fink
02:07 < dmalloc> hello
02:16 -!- eno is now known as eno-zzz
02:23 -!- dmalloc [~whatever@chello080108111128.5.11.univie.teleweb.at] has left #fink []
02:36 -!- xhrl [~ThomasW@S0106000f3d5d5bed.vs.shawcable.net] has joined #fink
02:37 -!- xhrl [~ThomasW@S0106000f3d5d5bed.vs.shawcable.net] has quit [Client Quit]
02:43 -!- sshreyas_ [~sshreyas@59.92.135.131] has joined #fink
02:54 -!- dmalloc [~whatever@chello080108111128.5.11.univie.teleweb.at] has joined #fink
02:54 -!- shres [~sshreyas@59.92.134.80] has quit [Read error: 110 (Connection timed out)]
03:09 -!- sshreyas_ is now known as shres
03:17 -!- dmalloc [~whatever@chello080108111128.5.11.univie.teleweb.at] has left #fink []
03:21 -!- shadowarts [~shadowart@adsl-68-251-227-65.dsl.sfldmi.ameritech.net] has joined #fink
03:21 < shadowarts> svn-client-ssl will not compile
03:30 -!- chris01 [~chris01@212.126.165.246] has joined #fink
03:30 -!- chris01 [~chris01@212.126.165.246] has quit [Client Quit]
03:31 -!- chris01 [~chris01@212.126.165.246] has joined #fink
03:31 -!- kane_ [~kane@perl.xs4all.nl] has joined #fink
03:34 < chris01> shadowarts: how is svn-client-ssl failing to compile?
03:34 < chris01> lisppaste: url?
03:34 < lisppaste> To use the lisppaste bot, visit http://paste.lisp.org/new/fink and enter your paste.
03:53 < JosephSpiros> Hmm
03:53 < JosephSpiros> Either I was dreaming
03:53 < JosephSpiros> or it exists
03:53 < JosephSpiros> command line tool that provides system repair and administration tasks
03:54 < JosephSpiros> for easy maintainence of Mac OS X via remote shelling
03:54 < JosephSpiros> anyone know of anything like what I'm talking about?
03:55 < chris01> JosephSpiros: man diskutil
03:56 < JosephSpiros> bah
03:56 < JosephSpiros> but if I recall correctly
03:56 < JosephSpiros> it also wrapped software update
03:56 < JosephSpiros> which, before you say it
03:56 < JosephSpiros> is "softwareupdate"
03:56 < chris01> :)
03:56 < JosephSpiros> yes, I'm aware of pretty much every built in tool on Mac OS X
03:56 < JosephSpiros> but this was like a neat little thing that provided a menu
03:56 < chris01> aha, ok.
03:56 < JosephSpiros> that did all sorts of things
03:56 < JosephSpiros> I think I read about it on one of my RSS feeds
03:57 < JosephSpiros> but I don't remember and I searched the ones that were likely
03:57 < JosephSpiros> and can't find it
03:57 < JosephSpiros> :(
03:58 < JosephSpiros> The problem is
03:58 < JosephSpiros> if it was a dream, I won't know if it was
03:58 < JosephSpiros> and I want whatever it is I remember
03:58 < JosephSpiros> so I'll write it myself
03:58 < JosephSpiros> but then when I'm done, it'll appear out of nowhere
04:38 -!- JesseW [~chatzilla@JesseW.student.supporter.pdpc] has left #fink []
04:40 -!- asari [~ASARI@gw02.ecc.u-tokyo.ac.jp] has quit ["Quitting!"]
04:56 -!- sshreyas_ [~sshreyas@59.92.128.119] has joined #fink
05:09 -!- shres [~sshreyas@59.92.135.131] has quit [Read error: 110 (Connection timed out)]
05:09 -!- sshreyas_ is now known as shres
05:22 -!- sshreyas_ [~sshreyas@59.92.137.60] has joined #fink
05:31 -!- marc7 [~marc@CPE0011246b4139-CM001371879a48.cpe.net.cable.rogers.com] has quit []
05:36 -!- shres [~sshreyas@59.92.128.119] has quit [Read error: 110 (Connection timed out)]
05:54 -!- hennker [flullup@dsl-082-082-235-171.arcor-ip.net] has joined #fink
06:01 -!- sshreyas__ [~sshreyas@59.92.139.172] has joined #fink
06:02 -!- sshreyas__ is now known as shres
06:12 -!- sshreyas_ [~sshreyas@59.92.137.60] has quit [Read error: 110 (Connection timed out)]
06:19 -!- _mcp [~hightower@wolk-project.de] has joined #fink
06:21 -!- mcp [~hightower@wolk-project.de] has quit [Read error: 104 (Connection reset by peer)]
07:21 -!- ddfreyne [~ddfreyne@dD5E06610.access.telenet.be] has joined #fink
07:21 -!- shadowarts [~shadowart@adsl-68-251-227-65.dsl.sfldmi.ameritech.net] has quit ["This computer has gone to sleep"]
07:27 < ddfreyne> For some reason, when I start gnome-session the gnome settings daemon doesn't activate
07:28 < ddfreyne> and the entire screen stays gray... any idea what's going wrong here?
08:06 -!- geewz [~gregreede@ppp122-142.static.internode.on.net] has joined #fink
08:10 -!- geewz [~gregreede@ppp122-142.static.internode.on.net] has quit [Client Quit]
08:13 -!- akh [~akhansen@ldx3.psfc.mit.edu] has joined #fink
08:26 < akh> !lart freetype
08:26 * Melian installs a bad bootloader on freetype and turns freetype into a brick
08:28 < akh> Hmmm...since it's an include issue, that's kind of like a bad loader already. ;-)
08:29 < akh> No points for Melian on that one.
08:29 -!- akh [~akhansen@ldx3.psfc.mit.edu] has quit []
08:38 -!- shres [~sshreyas@59.92.139.172] has quit [Read error: 60 (Operation timed out)]
08:38 -!- shres [~sshreyas@59.92.136.54] has joined #fink
08:42 -!- akh [~akhansen@ldx3.psfc.mit.edu] has joined #fink
08:47 -!- ddfreyne [~ddfreyne@dD5E06610.access.telenet.be] has quit []
09:05 -!- pogma [~peter@p4026-ipad32kobeminato.hyogo.ocn.ne.jp] has quit ["Lost terminal"]
09:13 -!- sshreyas_ [~sshreyas@59.92.144.210] has joined #fink
09:15 -!- sshreyas_ [~sshreyas@59.92.144.210] has left #fink ["Leaving"]
09:16 -!- _mcp is now known as mcp
09:19 -!- ddfreyne [~ddfreyne@dD5E06610.access.telenet.be] has joined #fink
09:22 -!- shres [~sshreyas@59.92.136.54] has quit [Read error: 110 (Connection timed out)]
09:52 -!- pogma [~peter@p4026-ipad32kobeminato.hyogo.ocn.ne.jp] has joined #fink
09:53 -!- hennker [flullup@dsl-082-082-235-171.arcor-ip.net] has quit ["leaving"]
10:29 -!- regeya [~shane@adsl-sp3-cdale176.micgi.com] has joined #fink
10:43 -!- zizban [~zizban@24-52-0-219.sbtnvt.adelphia.net] has joined #fink
10:44 -!- chris01 [~chris01@212.126.165.246] has quit ["bye"]
10:58 -!- eno-zzz is now known as eno-away
10:59 -!- ddfreyne [~ddfreyne@dD5E06610.access.telenet.be] has quit []
11:00 < cirdan> morning all
11:00 < zizban> hey cirdan
11:02 < pogma> hi cirdan
11:06 -!- marc7 [~marc@CPE0011246b4139-CM001371879a48.cpe.net.cable.rogers.com] has joined #fink
11:07 < akh> morning
11:07 < zizban> hey akh
11:11 -!- xhrl [~ThomasW@S0106000f3d5d5bed.vs.shawcable.net] has joined #fink
11:35 -!- xhrl [~ThomasW@S0106000f3d5d5bed.vs.shawcable.net] has quit [Read error: 110 (Connection timed out)]
11:39 -!- zizban [~zizban@24-52-0-219.sbtnvt.adelphia.net] has quit ["Leaving"]
11:39 -!- zizban [~zizban@24-52-0-219.sbtnvt.adelphia.net] has joined #fink
11:53 -!- slyrus [~slyrus@adsl-68-121-172-169.dsl.pltn13.pacbell.net] has joined #fink
11:59 * akh can't remember if CVS fink has additional compiler-selection stuff above and beyond what's in the release version.
12:17 -!- TheSin [~TheSin@iphost-64-56-130-194.edm.wiband.net] has joined #fink
12:21 < slyrus> any suggestions on what to do about: /usr/bin/ld: Undefined symbols: _crc32 _inflateReset _deflate ... in trying to build pilot-link9?
12:34 -!- zizban [~zizban@24-52-0-219.sbtnvt.adelphia.net] has quit ["Leaving"]
12:54 -!- ringerc [~craig@dsl-202-72-144-62.wa.westnet.com.au] has joined #fink
13:08 -!- dmacks [~dmacks@dmacks.active.supporter.pdpc] has joined #fink
13:09 < dmacks> pmd
13:24 < slyrus> answering my own question... it seems that rebuilding libpng3 fixes that problem
13:26 -!- RLD_osx [~rldempse@24-178-204-108.dhcp.ftwo.tx.charter.com] has quit ["Mac OS X - - a better alternative to winblow$"]
13:26 -!- RLD_osx [~rldempse@24-178-204-108.dhcp.ftwo.tx.charter.com] has joined #fink
13:30 < akh> pmd?
13:49 < dmacks> Tis a command that monitors my incoming mailboxes (ProcMail Display, or somesuch), that I perpetually type in the wrong window.
13:50 < akh> ah
14:02 -!- hennker [flullup@dsl-082-082-235-171.arcor-ip.net] has joined #fink
14:10 < dmacks> pmpmd
14:11 < dmacks> (befor eyou ask, that's a typo of i because my connectivity is sucking today)
14:17 -!- Albie [~ambs@bl6-36-212.dsl.telepac.pt] has joined #fink
14:37 < akh> dmacks: I wasn't going to ask.
14:38 -!- Netsplit zelazny.freenode.net <-> irc.freenode.net quits: regeya
14:43 -!- Albie [~ambs@bl6-36-212.dsl.telepac.pt] has quit ["Leaving"]
14:45 -!- Netsplit over, joins: regeya
14:45 < dmacks> Is gmane hosed today?
14:49 -!- gpaci [~gpaci@158.59.200.115] has joined #fink
14:50 < gpaci> Hey, all. I have a question about hosed X.
14:51 < gpaci> Specifically, I think the easiest way to fix it is to upgrade from 10.3 to 10.4. Anything I should be careful of?
14:56 < dmacks> Dunno what "hosed X" is. Dunno what's in need of fixing in your case. You should obviously do a backup of anything you care about before doing a major os upgrade. The installer's 'archive and install' is much better than 'upgrade' method.
14:59 -!- pawz [~pawz@catpa.ws] has joined #fink
15:02 < akh> Especially because that flushes out your old X11 stuff.
15:04 < dmacks> ...and old XCode files that tend to confuse new XCode installs.
15:04 -!- Feanor [~astrange@dsl-80-44-94-252.access.as9105.com] has joined #fink
15:04 < dmacks> "Installer.app--is there anything it can't do...poorly?"
15:06 < akh> It can't install Fink in an arbitrary tree.
15:07 < akh> It can't uninstall stuff.
15:11 -!- megahal [~netking@70.85.113.106] has quit [Remote closed the connection]
15:13 -!- megahal [~netking@70.85.113.106] has joined #fink
15:17 < dmacks> Hrm, good point.
15:19 -!- ringerc [~craig@dsl-202-72-144-62.wa.westnet.com.au] has quit ["zzzz"]
15:21 -!- megahal [~netking@70.85.113.106] has quit [Remote closed the connection]
15:21 -!- megahal [~netking@70.85.113.106] has joined #fink
15:24 -!- TheSin [~TheSin@iphost-64-56-130-194.edm.wiband.net] has quit ["Client exiting"]
15:25 -!- shadowarts [~shadowart@adsl-68-251-227-65.dsl.sfldmi.ameritech.net] has joined #fink
15:26 -!- shadowarts [~shadowart@adsl-68-251-227-65.dsl.sfldmi.ameritech.net] has left #fink ["Leaving"]
15:40 < akh> hmm...spent three days proving something that should have been obvious.
15:40 < akh> Hate it when that happens.
15:41 -!- pawz_ [~pawz@catpa.ws] has joined #fink
15:42 -!- msachs [~msachs@A17-255-98-107.apple.com] has joined #fink
15:50 -!- kpaul [~kpaul@70.57.247.118] has joined #fink
15:50 < dmacks> A few weeks in lab can easily save a 10-minute trip to the library.
15:51 < akh> heh--/me was analzying data, and the damn equation was sitting there in front of me in the report I was writing up.
15:52 < akh> Luckily I found something else to use as the X-axis.
15:52 < akh> And I did confirm that my numerical calculations weren't just totally bogus.
15:52 * akh masochistically starts again on oo.org
15:53 < dmacks> akh: It looks *much* better now
15:53 < dmacks> (and led to two fink-doc improvements too)
15:54 < akh> Yeah, that was the gist I got from your conversation with asari and the resulting commits.
15:54 * dmacks didn't have the heart to use the exact oo.o intermediate .info as the example of a forbidden package layout.
15:54 < akh> heh
15:54 < akh> That would have been a bit harsh.
15:56 < akh> Unfortunately my old tarball went *poof* in v1 of 'cleanup --source' so I have to download that again.
15:56 < akh> If only I had gigabit--100Mbit is just too slow.
15:56 < dmacks> Sorry 'bout the cleanup:/
15:57 < akh> That's OK--I'm going to need every bit of disk space that I can muster.
15:58 < akh> Maybe OO.org should Replace mozilla
15:59 -!- vasi [~vasi@modemcable133.147-70-69.mc.videotron.ca] has joined #fink
16:00 < dmacks> I don't know why I still use moz, since development prety much stopped in favor of ffox.
16:00 < akh> I haven't used it in a while, myself.
16:01 * dmacks used to use chatzilla, but not any more...just need browser.
16:01 < akh> Which is ironic because I frequently wind up opening both firefox and thunderbird.
16:01 * akh used chatzilla on Windows, but now that I don't use Windows...
16:01 < akh> (much)
16:02 < akh> vasi: Did you get my message about having enabled a password on my wiki?
16:03 < dmacks> I do most mail/news using remote clients (mutt/slrn), so I don't even consider thurderbird (or mail.app)
16:03 < vasi> akh, nope
16:03 < vasi> dmacks, there's a chatzilla for ffox
16:04 -!- pawz [~pawz@catpa.ws] has quit [Read error: 110 (Connection timed out)]
16:04 < dmacks> chatzilla pissed me off once too many times and doesn't have a text-mode form, so it's no longer an option.
16:05 < vasi> i just use it when i need irc on a friend's windows box
16:05 < vasi> easy way to install a "program" that doesn't look like a program is via a ffox extension :-)
16:06 < akh> nice
16:11 < vasi> dmacks, i think i found a bug in Shlibs
16:11 < dmacks> vasi: I've never looked at that code.
16:11 < akh> ooh--a 0.25-blocker?
16:12 < vasi> akh, yeah that page reminded me to check out the bug
16:13 < akh> ah
16:13 -!- akh is now known as akh_gone
16:13 -!- Albie [~ambs@bl6-36-212.dsl.telepac.pt] has joined #fink
16:14 < dmacks> "thanks wiki, for making it hard to type things like 'cleanup --bl' and 'AutoShlibs'
16:14 < vasi> heh
16:14 < vasi> so what do you think about Info3?
16:14 < dmacks> The new space handler is *very* nice.
16:14 < akh_gone> dmacks: you have to escape them
16:15 < vasi> dmacks, you can use == as an escape mechanism :-)
16:15 < vasi> for example, i needed ==g++ or c++==
16:17 < dmacks> 'k
16:17 < vasi> i'd like to deprecate RFC whatever
16:17 < vasi> 822?
16:17 < vasi> er, not deprecate, more like remove support
16:17 < dmacks> Fine with me.
16:18 < vasi> that way the spaces-vs-tabs distinction becomes less important
16:19 < vasi> final thing, how do we intend InfoN to be used?
16:19 < dmacks> ?
16:19 < vasi> i mean....let's say there's a minor incompatible change, that still allows the .info file to be parsed
16:19 < vasi> like adding comments in Depends
16:19 < vasi> but will make older finks get confused at a post-parsing stage
16:20 < vasi> how do we deal with that? N++? InfoMinorVersion like thesin and cirdan want? some other way...?
16:22 < vasi> (on a somewhat related note, maybe we should allow a per-tree default-N? so we can say at a certain point "everything in 10.4R is assumed to use Info2, since you really shouldn't have an old fink and 10.4")
16:27 < dmacks> That last bit promotes not-backporting of packages.
16:27 < dmacks> Any clever ways to include .info snippets in a wiki page?
16:28 < vasi>
16:28 < dmacks> (for example, to make the long-range plans page not look like crap)
16:28 < vasi> er,
16:28 < vasi> and then end with
16:30 < vasi> dmacks, while we're refactoring the source/patch issue, it might be nice to deal with the 'patch not noticing if sed fails' issue
16:31 < vasi> Maybe PatchReplace:
16:31 -!- regeya [~shane@adsl-sp3-cdale176.micgi.com] has left #fink ["Leaving"]
16:31 < dmacks> Yeah. Permissions problems will be solved automatically (MD5 requires being able to read it:)
16:31 < vasi> that too
16:32 < dmacks> PatchReplace is nice!
16:32 < vasi> would be especially nice to be able to have more than one patch, but i'm not sure how we could easily fit that into our directory structure
16:32 < dmacks> Patch alrady allows multiple filenames.
16:33 < dmacks> (but we have no good structure other than "lots of files in %a")
16:34 < vasi> what about a patches subdir, and then auto-prepend %{Ni}
16:34 < vasi> and append .patch
16:34 < vasi> er, no
16:35 < vasi> anyway, something to figure out later
16:35 < vasi> but you didn't answer my infoN question earlier ;-P
16:35 < dmacks> Yeah. I haven't heard much complaint about "just strings all the patches into a single file" (except from cirdan:)
16:35 * dmacks didn't have a good answer:)
16:36 < vasi> ok, thoughts then?
16:36 < vasi> :-)
16:37 < dmacks> Does BuildConflicts:fink<<0.x not solve this one?
16:39 < dmacks> (i.e., what's the harm caused by the misparsing?)
16:39 -!- zizban [~zizban@24-52-0-219.sbtnvt.adelphia.net] has joined #fink
16:40 < dmacks> Take your time...I'm off to get a haircut then looooong staff-meeting.
16:40 -!- dmacks is now known as dmacks_away
16:50 -!- zizban [~zizban@24-52-0-219.sbtnvt.adelphia.net] has quit ["Leaving"]
16:50 -!- zizban [~zizban@24-52-0-219.sbtnvt.adelphia.net] has joined #fink
16:55 -!- Albie [~ambs@bl6-36-212.dsl.telepac.pt] has quit ["Leaving"]
17:01 -!- kane_ [~kane@perl.xs4all.nl] has quit []
17:03 -!- slyrus [~slyrus@adsl-68-121-172-169.dsl.pltn13.pacbell.net] has quit [Read error: 110 (Connection timed out)]
17:08 -!- slyrus [~slyrus@flypaper.BDGP.Berkeley.EDU] has joined #fink
17:27 -!- drm [~drm@user-152-3-154-137.wireless.duke.edu] has joined #fink
17:27 < drm> pogma: awake yet?
17:37 < drm> msachs, you around?
17:40 < msachs> hey drm
17:40 < msachs> How's it going?
17:40 < drm> hi msachs
17:40 < drm> i've got a weird problem in my latest 10.4 run
17:40 < msachs> Yeah?
17:40 < drm> (using gcc-4.0 and g++-4.0)
17:41 < drm> yeah, i've got a package which at link time is trying to use both /usr/lib/gcc/../4.0.0/libgcc.a and
17:41 -!- Feanor_ [~astrange@dsl-80-44-79-47.access.as9105.com] has joined #fink
17:41 < drm> /usr/lib/gcc/.../4.0.0/libgcc_s.dylib
17:41 < drm> and there are duplicate symbols
17:42 < drm> (.. was powerpc-apple-darwin8)
17:42 < msachs> Which package?
17:42 < drm> its libgtkhtml2
17:42 < drm> my build strategy is different than yours:
17:43 < drm> i build stuff from existing .info files
17:43 < drm> so some of the dependent libs might have been compiled with gcc 3.3
17:43 < msachs> Okay, well I do see that same problem in my build.
17:43 < drm> ah
17:43 < drm> must be a libtool thing, then
17:44 < drm> because on the command line there is no evidence of pulling in libgcc at all
17:44 < msachs> Yeah.
17:44 < msachs> Maybe when it linked some of the .a that it's linking in there, it said "oh, static archive, I'd better link in libgcc.a" ?
17:44 < drm> well, maybe, but that strikes me as a HUGE linker bug :)
17:45 < msachs> What's -all_load do?
17:45 < msachs> "Loads all members of static archive libraries."
17:45 < drm> according to "man ld", it loads all symbols from static libs
17:45 < drm> msachs is too fast
17:45 < drm> :/
17:46 < msachs> Let's see, when it makes those .a files...
17:46 < drm> well, in a sense it is good to know that your build has the same problem
17:46 < msachs> It doesn't look like it's doing anything suspicious.
17:46 < msachs> Is there a reason it's being built with -flat_namespace ?
17:46 < drm> history, probably
17:47 < drm> i.e., was needed one but never removed
17:47 < msachs> Taking that out might fix it.
17:47 * drm will try
17:47 < msachs> If that doesn't work, I can ping my coworkers.
17:48 < drm> it built for the bindist
17:48 < vasi> drm, this looks like it could be the repair-perms bug
17:48 < drm> which is weird, since the only change now should be the change of g++ compilers
17:48 < drm> vasi: what's that?
17:48 < vasi> fixed in 0.24.8
17:48 < msachs> vasi: I saw it on my build, and I've never run repair permissions on that machine :)
17:49 < vasi> msachs, ok then nm
17:49 < drm> vasi: my build uses 0.24.8
17:49 < msachs> Doesn't work with my 4.0.1 build either, which had a nice fresh compiler install.
17:49 < vasi> that's just a symptom of the repair perms issue, multiple symbols from libgcc_s and libgcc :-)
17:49 < vasi> so i thought, but i'm wrong
17:49 < drm> oh
17:49 < drm> why is that a symptom of repair-perms, and what is/was the bug?
17:50 < msachs> "Repair Permissions" makes some symlinks under /usr/lib/gcc which are supposed to be present when 3.3 is selected but *not* when 4.0 is selected. The symlinks confuse 4.0.
17:50 < vasi> drm, 1) i dunno, i just fixed it i didn't analyze it 2) bug was that repair-perms tries to "restore" some symlinks that are part of gcc 3.3
17:50 < vasi> and didn't respect gcc_select
17:51 < vasi> so fink just checks "is gcc_select 4? then run gcc_select --force 4.0"
17:51 < vasi> which fixes that bug
17:51 < drm> well, my gcc_select is 4.0
17:52 < vasi> yeah, so then at the top of your build you should have seen some notice about "Default compiler has been set to:"
17:52 < vasi> which is the fix happening
17:52 < drm> i get that
17:52 < vasi> clearly the multiple symbols is symptomatic of multiple bugs then :-)
17:53 < drm> is libgcc_s.dylib supposed to be a symlink? it is on my system
17:54 < msachs> For me it's libgcc_s -> libgcc_s.1 -> libgcc_s.1.0
17:54 < drm> same
17:55 < drm> although there is a directory change
17:55 < msachs> Mine are all in /usr/lib...
17:56 < drm> i have libgcc_s.dylib in /usr/lib/gcc/powercrap/4.0.0
17:56 < msachs> Well and then there's -- yeah, that one symlinks to the /usr/lib one
17:56 < drm> but the things it links to are in /usr/lib
17:56 < vasi> why's it using all_load for a dylib?
17:56 < drm> vasi: there are some static there, no?
17:56 < vasi> i thought it only works on executables though?
17:57 < vasi> may be different now
17:57 < vasi> or i may be confused
17:57 < drm> man gcc refers one to man ld
17:57 < drm> its a valid option for gcc
17:58 < drm> it must be done by libtool? cause its not in the info file
17:58 < vasi> yeah, libtool
17:58 < vasi> didn't all_load used to load all linked in dylibs at startup?
17:58 < drm> which is why i started by pinging pogma :)
17:58 < drm> i don't know
17:58 < vasi> or is that something else?
17:59 -!- Feanor [~astrange@feanor.developer.opendarwin] has quit [Connection timed out]
17:59 < vasi> oh, i'm thinking of bind_at_load
17:59 < vasi> uh yeah, so forget what i said :-)
18:01 < drm> all_load is being put before the stuff that .la expands to, which i think is correct
18:02 < drm> these are the currently-static symbols that we are about to group together into the dylib
18:04 < vasi> http://lists.apple.com/archives/Unix-porting/2005/May/msg00005.html
18:05 < vasi> google is wonderful :-)
18:06 < drm> cool...don't even have to wait for pogma to get out of bed!
18:06 < drm> thanks!
18:07 < vasi> np
18:08 < vasi> i knew i had seen that bug before somewhere
18:11 < vasi> sheesh, montreal wants the 2016 olympics!? we haven't even finished paying for the one 30 years ago
18:11 < drm> really? wow
18:12 < vasi> they built this huge stadium in east montreal, to help stimulate development
18:12 < vasi> with a retractible roof and all this stuff that was cool at the time :-)
18:12 < drm> that's where the expos played, right?
18:12 < vasi> yeah....it never got fully built, the roof never worked, nowadays nobody uses it and they're trying to raise $500 million to knock it down
18:13 < vasi> it's officially the "Big O" (for olympic) but everyone here writes it "Big Owe"
18:14 < drm> man, puns must be difficult when everything is supposed to be bilingual
18:14 < vasi> nah they're easier, twice the opportunity
18:14 < vasi> aha, found a bug in Shlibs!
18:14 < vasi> *tsk tsk*
18:15 < drm> Le Grand Lieu?
18:15 < vasi> not an easy bug to fix either, good thing it TheSin's problem
18:15 < drm> just doesn't have the same ring
18:15 < vasi> heh, it kinda works though
18:16 < drm> did i ever tell you my story about speaking french in montreal?
18:16 < vasi> nope!
18:16 < drm> only visited there once... was a bit frustrated, because my french ain't great but i get by in france...but i was having a hard time understanding the quebec version
18:16 < drm> however, the taxi ride back to the airport was great
18:17 < drm> got into a conversation with the driver, felt i was finally understanding canadian french
18:17 < drm> until he told me that he emigrated from paris after wwii !
18:18 < msachs> heheh
18:18 < msachs> Two of my flatmates last year, one spoke Parisian French, the other would always piss her off by speaking Acadian French.
18:20 < vasi> heh
18:20 < vasi> did ya know they actually subtitle quebecois movies in france?
18:20 < drm> i'm not surprised
18:21 < vasi> my french teachers have always been admirers of "french french", so i have trouble with quebecois accents too
18:21 < drm> there are certainly parts of england (not to mention the rest of the uk) where the english as spoken is almost unintelligible to a north american
18:21 < vasi> my gf's mom, i can't understand a thing she says :-/
18:21 < drm> vasi: that has certain advantages :)
18:21 < vasi> hehe, yeah that's true!
18:21 < msachs> "Why yes, I think I'm awesome too! I'm glad we agree!"
18:23 -!- kpaul [~kpaul@70.57.247.118] has left #fink []
18:25 < drm> ok, battery dying, bbl
18:25 < msachs> *waves*
18:25 -!- drm [~drm@user-152-3-154-137.wireless.duke.edu] has quit ["Leaving"]
18:35 < vasi> msachs, wasn't the libssl/libcrypto thing fixed in 10.4.2?
18:36 < msachs> vasi: I thought it was supposed to be.
18:36 < vasi> do i have to rebuild a package for the fix to work?
18:36 < msachs> ISTR we fixed a radar that was supposed to fix it, I don't recall seeing anyone say "hey, I updated to .2, it actually works now."...
18:37 < msachs> Oh.
18:37 < vasi> hmm
18:37 < msachs> Ah, no, that one wasn't fixed in .2
18:39 < msachs> If I'm looking at the right one
18:40 < msachs> dyld does not load libraries correctly if matching install_name in /usr/lib
18:40 < msachs> I think that's the one, yeah.
18:40 < vasi> looks right
18:44 < pogma> that one will be fixed "sometime"
18:44 < pogma> morning :)
18:45 < vasi> hi pogma
18:45 < vasi> sorry to see you're leaving core
18:45 < msachs> Have they given you your gold pocketwatch yet?
18:45 < vasi> (and i'm praying you keep working on libtool, cuz i don't wanna touch that monster ;-) )
18:47 < pogma> no pocketwatch, and yes, I am still a gnu libtool maintainer
18:47 < vasi> thank you for that!!
18:48 < pogma> If you've completed copyright assignment with the FSF I'd be happy to give you libtool commit after approval access though :-)
18:51 -!- thomasjo [~thomasjo@203-206-22-223.dyn.iinet.net.au] has joined #fink
18:54 -!- citizen_0 [~user@va-sterling-u1-c5c-a-146.chvlva.adelphia.net] has joined #fink
19:00 -!- citizen_0 [~user@va-sterling-u1-c5c-a-146.chvlva.adelphia.net] has quit ["Leaving"]
19:03 -!- citizen_0 [~user@va-sterling-u1-c5c-a-146.chvlva.adelphia.net] has joined #fink
19:04 -!- KraMer [~mark@adsl-70-240-196-20.dsl.hstntx.swbell.net] has quit [Remote closed the connection]
19:06 -!- gpaci [~gpaci@158.59.200.115] has quit [Remote closed the connection]
19:14 -!- KraMer [~mark@adsl-70-240-196-20.dsl.hstntx.swbell.net] has joined #fink
19:29 -!- citizen_0 [~user@va-sterling-u1-c5c-a-146.chvlva.adelphia.net] has quit ["Leaving"]
19:30 -!- citizen_0 [~user@va-sterling-u1-c5c-a-146.chvlva.adelphia.net] has joined #fink
19:52 * pogma reads gcc list and giggles
19:53 < msachs> What are those crazy cats doing this time?
19:53 < pogma> our friend jack howarth is trying to get geoff keating to make fsf gcc accept -bundle as the frist argument
19:54 < msachs> heheh, undle
19:54 < pogma> geoff said "submit a patch", jack "I haven't done the paperwork"
19:57 < msachs> OS X should have undle support.
20:04 < pogma> :)
20:06 < pogma> love how autoconf adds -DHAVE_CONFIG_H even if the config file is named foobar.h ... NOT
20:07 < msachs> Fun.
20:11 < runelind> wtf, I can't export iMovies into mp4's or anything like that?
20:11 < vasi> uh, i dunno runelind, i'm sure there's a way
20:11 < msachs> mencoder?
20:11 < runelind> this is my first time using iMovie
20:12 < msachs> Mm. Haven't gotten around to trying that one.
20:12 < runelind> I have an mp4 video file taken with my smartphone
20:12 < runelind> and only quicktime seems to want to play it
20:23 < zizban> vlc
20:24 -!- msachs [~msachs@A17-255-98-107.apple.com] has quit []
20:29 < pogma> are you sure that it is mp4? My phone does not take .mp4 it does .3gpp
20:33 -!- marc7 [~marc@CPE0011246b4139-CM001371879a48.cpe.net.cable.rogers.com] has quit []
20:33 < vasi> ah, you want to import a mp4 into iMovie?
20:33 < vasi> maybe import it into quicktime, then export it as something else, then import into imovie?
20:44 -!- RLD_osx [~rldempse@24-178-204-108.dhcp.ftwo.tx.charter.com] has quit [Read error: 104 (Connection reset by peer)]
20:44 -!- RLD_osx [~rldempse@24-178-204-108.dhcp.ftwo.tx.charter.com] has joined #fink
20:47 -!- hennker [flullup@dsl-082-082-235-171.arcor-ip.net] has quit ["leaving"]
20:56 -!- asari [~ASARI@gw08.ecc.u-tokyo.ac.jp] has joined #Fink
20:59 -!- vasi [~vasi@modemcable133.147-70-69.mc.videotron.ca] has quit []
21:05 -!- slyrus [~slyrus@flypaper.BDGP.Berkeley.EDU] has quit [Read error: 110 (Connection timed out)]
21:08 -!- regeya [~shane@adsl-sp3-cdale176.micgi.com] has joined #fink
21:14 -!- dk0r [~dk0r@cpe-24-194-171-176.nycap.res.rr.com] has joined #fink
21:17 -!- dmacks_away is now known as dmacks
21:17 < dmacks> asari: BuildConflicts is now documented:)
21:20 < asari> Thanks :)
21:23 -!- akh [~akhansen@68-118-244-23.dhcp.oxfr.ma.charter.com] has joined #fink
21:25 -!- dk0r [~dk0r@cpe-24-194-171-176.nycap.res.rr.com] has quit []
21:27 < zizban> hey Profressor akh
21:27 < zizban> are you a professor?
21:27 < akh> Not as such.
21:27 < zizban> ah
21:28 < zizban> I didn't know, just a guess :)
21:29 < akh> I'm on research staff--it's similar pay but no teaching and no tenure
21:29 < zizban> ahhhh
21:29 < zizban> I see
21:29 -!- vasi [~vasi@modemcable133.147-70-69.mc.videotron.ca] has joined #fink
21:31 < vasi> hey all
21:31 -!- Feanor [~astrange@feanor.developer.opendarwin] has joined #fink
21:33 < akh> hey, vasi
21:36 < akh> Anybody see the message on -beginners about the 3 day new install?
21:38 < akh> (no builds--that's just from the .pkg)
21:39 -!- slyrus [~slyrus@adsl-68-121-172-169.dsl.pltn13.pacbell.net] has joined #fink
21:39 * dmacks didn't see nothin'...gmane's pulling as SF.
21:39 < akh> heh
21:40 < akh> my thought was Installer.app silliness.
21:41 * dmacks visits lists.sf
21:41 -!- akh_gone [~akhansen@ldx3.psfc.mit.edu] has quit [Remote closed the connection]
21:42 * akh explores my oo.org build failure to see if it's the same as the post on -users from today. That way I can just say "me, too, and I have the following setup"
21:42 < dmacks> installer.app silliness yup-yup
21:43 -!- Feanor_ [~astrange@feanor.developer.opendarwin] has quit [Client Quit]
21:43 < akh> I usually force quit if nothing happens after 3 _minutes_.
21:43 < dmacks> (could be a missing BildConflicts)
21:44 < akh> yeah, maybe.
21:48 < vasi> dmacks, BuildDepends: fink (>= whatever) is not sufficient
21:48 < vasi> because fink uses Depends even for not-building tasks
21:49 < vasi> though it will at least force the user to upgrade fink
21:49 < dmacks> howzat?
21:49 < vasi> well if you do 'fink install foo', and foo is already built, fink doesn't use the Depends in the .deb
21:49 < vasi> but it uses the depends in the .info file
21:50 < vasi> so if someone does something stupid like downgrading fink (or shares debs), fink will get mixed up
21:50 < vasi> on another somewhat related note, that's what the bug i found in AddShlibDeps is about
21:50 < dmacks> I thought that was (at least partially) changed as a result of AutoShlibDeps.
21:51 < vasi> see: http://sourceforge.net/tracker/index.php?func=detail&aid=1249916&group_id=17203&atid=117203
21:51 < dmacks> (since hwat matters is the Depends encoded while the pkg was built, not the ones fink would determine at a later installtime.
21:51 < vasi> all that AddShlibDeps does is add to the Depends in the .deb
21:52 < vasi> based on the Shlibs fields that exist in fink, and the binaries in the .deb
21:52 < dmacks> Ah, so I'm basically telling you what you previously said:/
21:52 < vasi> i'm not sure what you're telling me :-P
21:53 < vasi> i'm not sure how to corral thesin to make him fix this bug
21:53 < vasi> because it's Not Easy
21:53 < vasi> hmm, well it could be easier actuall
21:53 < vasi> y
21:54 < vasi> just check if the .deb is present and read the depends line with uh...
21:54 < dmacks> I *think* the last-paragraph issue is a non-issue (or it's just a special case of the more general one in the tracker item)
21:55 < vasi> dmacks, not really
21:55 < vasi> it's not such a big problem because if you're building a package, you're guaranteed to have the shlibdeps installed
21:57 < vasi> but it's vulnerable to changes in the dep engine, if we ever make it not leave all the builddeps sitting around after a build
21:58 < dmacks> ?
21:59 < vasi> ie: if you do 'fink install foo pkg-that-conflicts-with-foos-builddeps' it should be doable
21:59 < vasi> (in some better engine than what we have)
22:00 < vasi> but if foo uses AddShlibDeps, we can't tell which of the builddeps and their dependencies can be removed so that pkg-that-conflicts... can be installed
22:00 < dmacks> Maybe phase_activate should read the Depends from the .deb and fire up a mini dep-engine check?
22:01 < dmacks> vasi: Good point. /me hates the whole pre-check thing even more than before.
22:01 < vasi> the solution to the tracker bug is for 1) all methods to use the same API for getting the Depends list 2) that method should check if is_present(), and in that case use 'dpkg-deb' to get the depends list from the deb
22:02 < dmacks> I had thought that was the long-ago agreed-upon process to use.
22:03 < vasi> must have been long ago enough that i never heard of it :-)
22:06 < vasi> i think the only place left that gets depends directly is resolve_depends?
22:06 < vasi> ah, also dumpinfo
22:08 < vasi> wow, do we really have a variable called "$struct"
22:08 < vasi> ?
22:08 < dmacks> Please talk to TheSin about this. I'm prety sure he knows about the need to read the .deb control-file.
22:09 < dmacks> heh. I guess so:)
22:09 < vasi> yeah, i'm gonna talk to him whenever he un-vacations himself
22:10 < dmacks> If the rest of the ancientish core is any guide, $struct is probably a boolean.
22:10 < vasi> lol
22:10 -!- drm [~drm@65.210.36.227] has joined #fink
22:10 < dmacks> No, $lol is a ref to a scalar.
22:10 < drm> what a sentence to enter on :)
22:11 < dmacks> drm: actually, it was meant exactly as you probably read it.
22:11 < drm> $lol
22:12 < vasi> SCALAR(0x180d974)
22:12 < drm> $rofl
22:13 < drm> so, i was about to backport the so-called swappy code to branch_0_24... any objections? does it need more work or testing?
22:14 < drm> or maybe somebody already did this?
22:16 -!- regeya [~shane@adsl-sp3-cdale176.micgi.com] has quit [Client Quit]
22:17 < vasi> go ahead drm
22:18 < drm> by the way, we need to release fink-mirrors pretty soon... things get a little weird during bootstrap when the fink-mirorrs pkg is old
22:20 -!- zizban [~zizban@24-52-0-219.sbtnvt.adelphia.net] has quit ["Leaving"]
22:21 < vasi> oh yeah
22:23 -!- akh is now known as akh_gone
22:26 < vasi> are there any pkglist fields that aren't in a .deb?
22:26 < drm> pkglist?
22:27 < vasi> fields that are lists of package specs, and which you access with the methods $pv->pkglist and friends
22:27 < drm> ah... not too familiar with what $pv->pkglist accesses
22:28 < vasi> it just gets param($field), where $field could be Depends or something
22:28 < vasi> but it does funky things
22:28 < dmacks> vasi: Build* aren't in .deb
22:28 < vasi> like filtering conditionals
22:29 < vasi> dmacks, oh yeah
22:29 < vasi> *brain rot*
22:29 < drm> conditionals *are* in the deb
22:29 < vasi> drm, are you sure?
22:29 < dmacks> no.
22:29 < vasi> i don't think dpkg knows how to deal with those
22:29 < dmacks> Conditionals are either made explicit or removed completely long before the .deb is written.
22:29 < drm> no? oh, a terminology confusion
22:30 < drm> i'm talking about "foo (>= 0.1.0)"
22:30 < drm> and even "foo | bar"
22:30 < drm> but that's prolly not what you guys meant :/
22:30 < dmacks> Bingo.
22:32 < vasi> yeah, we're talking about the variant-conditions
22:32 < dmacks> The $pv->pkglist methods handle the the *before*-the-pkgname if/then conditionals, not the dpkg-ish versioned dependnencies or alternatives.
22:32 < vasi> ( %type_pkg[-foo] = foo) bar
22:33 < Knghtbrd> fink has no openal..
22:33 < Knghtbrd> hmm
22:33 < vasi> port it!
22:34 < drm> dmacks, vasi: got it
22:35 < dmacks> I say backport the swappy-code if it already isn't (/me has not done so, dunno about y'all).
22:35 < drm> just backported it
22:35 < dmacks> Cool.
22:35 < dmacks> (no gmane, so I can't see -commits right now)
22:36 < vasi> dmacks, why do we do the to_be_rebuilt checking in the PACKAGELOOP rather than in the preflight?
22:36 < Knghtbrd> meh, it may not be necessary, Apple ships the framework with Tiger and provides an installer for older versions of MacOS X
22:36 < dmacks> Sweet...<24 hrs for someone to send a null msg about fink-0.24.8
22:36 * drm is actually subscribed to commits
22:36 < drm> dmacks: null means good IMHO
22:36 < vasi> drm, me too but i do a lot of filtering
22:37 < Knghtbrd> and um, Fink doesn't even provide a framework version of SDL currently =(
22:37 < dmacks> I think I might be also, but "filter"=="set to NOMAIL" :)
22:37 < vasi> knightbrd, that's correct and by design
22:37 < drm> Knghtbrd: fink doesn't do frameworks
22:37 < drm> at least, not at the moment
22:37 < Knghtbrd> I realise that.
22:37 < vasi> except when we do
22:37 < dmacks> vasi: I have no idea 1) what that means, 2) why it is, or 3) almost anything else about the engine design.
22:37 < Knghtbrd> lol vasi
22:38 < vasi> dmacks, oh :-(
22:38 < vasi> well somebody must know
22:38 < vasi> where somebody is whoever wrote that code?
22:38 < drm> vasi: that is a big assumption :)
22:38 < drm> when code is written and rewritten and rewritten again, does anybody really know?
22:38 < vasi> i dunno?
22:39 < dmacks> me neither
22:39 < drm> does anybody really know what time it is? does anybody really care?
22:39 * drm supplies musical accompaniment
22:39 < dmacks> Do you know the way to San Jose?
22:39 < drm> actually i do
22:39 < drm> even to Cupertino :)
22:40 < dmacks> 1) Get on a plane to [city], 2) meet Burt Bacharach, 3) ???, 4) profit!
22:40 < vasi> hmm, it's fingolfin's code
22:41 < drm> vasi: seriously, the problem with the current engine "design" is that there was a clean design which did significantly less than the current engine, which has been overlayed with jury-rigged additions several times
22:41 < vasi> drm, this i recognize
22:41 < drm> and each time, it gets harder to understand and harder to modify
22:41 < vasi> a serious deficiency of the current engine is that it tries to figure out everything at the start
22:42 < vasi> and then later fixes anything that it incorrectly figured out at the start
22:42 < drm> well, that was a good feature of the original engine
22:42 < vasi> yes yes, but it makes it impossible to modify effectively
22:42 < drm> sorry, only the first part was a good feature
22:42 < drm> no, what it really means is that we need to figure things out *better* at the beginning
22:43 < vasi> preflighting is *bad*....a fink build can go on for days, who knows what's changed in the meantime?
22:43 < drm> one fink build at a time, right?
22:43 < drm> can we make that a design assumption?
22:43 < vasi> drm, and users of course always follow that "rule"
22:43 < vasi> and never touch dpkg or apt
22:43 < vasi> or remove .debs
22:44 < vasi> and shlibdeps don't exist anymore
22:44 < drm> well, do they remove .debs while fink is building?
22:44 < vasi> drm, i'm sure somebody does :-)
22:44 < drm> vasi: clearly, building should be done in a chroot jail or something
22:44 -!- vasi [~vasi@modemcable133.147-70-69.mc.videotron.ca] has quit [Remote closed the connection]
22:45 < drm> no, not you!
22:45 < drm> the program!
22:45 -!- vasi [~vasi@modemcable133.147-70-69.mc.videotron.ca] has joined #fink
22:45 < vasi> grr, damn irc client
22:45 * drm wasn't planning to send *vasi* to jail :)
22:45 < vasi> anyhow, fink should just figure out what the user wants, and then do-whatever-is-necessary-to-make-that-happen
22:46 < dmacks> A Buildconflicts could cause a pkg that is installed but has not .deb to be removed but later needed by a BuildDepends.
22:46 < drm> well, all software should do that
22:46 < drm> "do what i meant, not what i said"
22:46 < drm> the goal of all programming
22:46 < vasi> drm, but right now fink is really bad at it
22:46 < dmacks> drm: Right, that's what vasi meant, he just didn't say it well:)
22:46 < vasi> i mean, why is real_install not-very-recursive?
22:47 < drm> i agree that Buildconflicts and the swappy code are both jury-rigged, half-assed solutions to real problems
22:47 < vasi> i *really* wish we could easily plug into a real dep engine
22:47 < vasi> but apt has a shitty perl API
22:47 < vasi> smartpm is 1) in python 2) broken
22:47 < vasi> and i don't know of any other good option
22:47 < drm> yum?
22:48 < drm> doesn't somebody have yum working for dpkg?
22:48 < drm> i mean, debs
22:48 < vasi> yum is broken too i think
22:48 < drm> yum is maintained by a guy in my building (although a guy who is leaving soon...dunno what will happen when he changes jobs)
22:48 < vasi> but it might be worth checking out
22:49 < dmacks> How about we completely scrap the current pre-determination loop and just have dry-run and real modes for a compeltely recursive processing loop?
22:50 < drm> dmacks: this still doesn't solve the problem of the environment changing as you compile
22:50 < dmacks> Make a copy of the status file and actually twiddle it.
22:50 < vasi> dmacks, that's more or less what i'd do if i had time to rewrite and debug a whole farking engine tonite
22:50 < vasi> drm, how do you mean?
22:50 < vasi> at every step, you just do "satisfy all the deps and conflicts i need", recursively
22:51 < drm> well, i'm assuming he wants to do a dry-run of the whole thing, and then run it?
22:51 < drm> or do user choices have to be made all the way through? even in the middle of the night?
22:51 < vasi> er, the way i've considered it, the dry-run would just be for meaningful user-choices
22:51 < vasi> which would then be considered advisory only
22:52 < drm> so what happens when there is a choice to be made? do we interrupt and wait for user input?
22:52 < vasi> (in the sense that if fink sees "oh, well the opposite of the choice is currently installed", it won't swap it)
22:52 < vasi> drm, if it showed up in the dry-run, we obey the user....otherwise we just pick
22:52 < vasi> apt-get works pretty well for people, and it has no user-choices whatsoever! :-)
22:52 * dmacks considered a literal dry-run of the entire process. That way every choice will be made, every missing .deb will get noted, .
22:53 < vasi> dmacks, but that sucks if things change
22:53 < drm> dmacks: right... and that will be fine provided that nothing changes while you are running
22:53 < vasi> also it asks too many questions for me
22:53 < vasi> if the user has to make a choice to decide what should be installed, that's fine
22:54 < drm> but maybe an interrupt of the form "you have to make a choice, most likely becuase you changed something outside of fink while i was running" might do it
22:54 < vasi> but for things that are only to build something? who cares?
22:54 < dmacks> If a user does something jack-assy while fink is building something, it's his own damn fault. Fink *should* either crash outright or stop and ask for help.
22:54 < drm> vasi: i'll tell you the worst choices that exist right now... foo vs. foo-ssl
22:55 < drm> vasi: these are bad for several reasons, among which are the fact that the -shlibs are not really interchangeable but they are being treated as if they were
22:55 < dmacks> And if that causes fink to have to build something it didn't expect, which causes it to need bdeps it didn't expect and downlaod tarballs it didn't initally want, it's his own damn fault.
22:55 < vasi> drm, i was bugging pogma about that issue
22:55 < vasi> they should have different install names if they're different!
22:55 < drm> vasi: what we really should have is variants for -ssl, which depend only on the -ssl guys
22:56 < drm> lots of -ssl variants
22:56 < vasi> drm, can't we just use apple's openssl for the vast majority of things, and then not need non-ssl variants?
22:56 < drm> but with the current system, they are a pain to code, since our separation of crypto relies in part on keeping separate info files
22:57 < vasi> we don't have to separate crypto for things that apple distributes :-)
22:57 < drm> if we could make "turning off crypto" into a software thing, we'd be better off
22:57 < drm> vasi: there are objections to using apple's openssl, because they don't keep it very up to date
22:57 < drm> makes us too dependent on them, particularly for something subject to security probs
22:57 < dmacks> Perhaps drm would like to document exactly what the problem is with foo-shlibs|foo-ssl-shlibs on our new wiki? I'm tired of fighting this battle from scratch every few months.
22:58 * drm needs an account on the wiki
22:58 < drm> also, is it pointed to from anywhere yet?
22:58 < dmacks> akh_gone: Come back!
22:58 < dmacks> Un-_gone thyself!
22:58 < drm> (i'll catch him tomorrow)
22:58 < vasi> drm, currently it's not accessible (even for reading) without the pw
23:04 -!- dmacks [~dmacks@dmacks.active.supporter.pdpc] has quit ["leaving"]
23:14 -!- drm [~drm@65.210.36.227] has quit ["Leaving"]
23:19 -!- cmeme [~cmeme@216.184.11.2] has quit [Read error: 104 (Connection reset by peer)]
23:20 -!- cmeme [~cmeme@216.184.11.2] has joined #fink
23:20 -!- cmeme [~cmeme@216.184.11.2] has quit [Remote closed the connection]
23:21 -!- cmeme [~cmeme@216.184.11.2] has joined #fink
23:40 < asari> freetype219 has /sw/lib/freetype219/lib/pkgconfig/freetype2.pc: How can I use this file via pkg-config?
23:47 -!- xhrl [~ThomasW@S0106000f3d5d5bed.vs.shawcable.net] has joined #fink
23:48 * asari is reading `man pkg-config`
23:51 < asari> SetPKG_CONFIG_PATH: /sw/lib/freetype219/lib/pkgconfig/ ?
23:53 < vasi> er, i don't think SetXXX works for arbitrary stuff
23:54 < vasi> but you can do in your compilescript PKG_CONFIG_PATH= ./configure
23:54 * asari read packaging reference
23:55 < asari> Okay, thanks! I'll try it.
23:56 < vasi> you may need to do something so pkg-config uses both the standard and non-standard paths
--- Log closed Tue Aug 02 00:00:29 2005