--- Log opened Fri Aug 05 00:00:46 2005
00:12 < vasi> la la la
00:15 < dmacks> You just volunteered to manage FC and you still have time to sing?
00:16 -!- msachs [~msachs@A17-255-98-107.apple.com] has quit []
00:23 -!- Netsplit zelazny.freenode.net <-> irc.freenode.net quits: JosephSpiros, vasi, pnorman, _nico_, kane-xs, runelind, Feanor, cmeme
00:25 -!- Feanor_ [~astrange@feanor.developer.opendarwin] has joined #fink
00:26 -!- eno-away is now known as eno
00:31 -!- cmeme [~cmeme@216.184.11.2] has joined #fink
00:31 -!- vasi [~vasi@modemcable133.147-70-69.mc.videotron.ca] has joined #fink
00:31 -!- _nico_ [nicolas@server.asta.uni-saarland.de] has joined #fink
00:31 -!- JosephSpiros [~joseph@ip-246-036.oberlin.net] has joined #fink
00:31 -!- runelind [~mattias@inte099185.halls.colostate.edu] has joined #fink
00:31 -!- kane-xs [~kane@coke.xs4all.nl] has joined #fink
00:31 -!- pnorman [~pnorman@d216-232-209-79.bchsia.telus.net] has joined #fink
01:16 -!- pogma [~peter@pogma.developer.opendarwin] has quit [Remote closed the connection]
01:16 -!- pogma [~peter@p4026-ipad32kobeminato.hyogo.ocn.ne.jp] has joined #fink
01:27 -!- cmeme [~cmeme@216.184.11.2] has quit [Remote closed the connection]
01:27 -!- cmeme [~cmeme@216.184.11.2] has joined #fink
01:28 -!- cmeme [~cmeme@216.184.11.2] has quit [Remote closed the connection]
01:28 < dmacks> !lart loopy logbots
01:28  * Melian slaps a compatible dib on loopy logbots's head
01:28 -!- cmeme [~cmeme@216.184.11.2] has joined #fink
01:29 < dmacks> megahal: Like loquacious aliteration?
01:29 < megahal> dmacks: And like, which one is awesome.
01:31 -!- moike [~Mike@nat01-hardenburgh-ext.Rutgers.EDU] has joined #fink
01:32 < moike> Hi i am installing the bundle-kde-ssl and I was wondering if I were to stop it can it be resumed?
01:33 < vasi> moike, sort of
01:33 < dmacks> Each package's compile cannot be resumed, but bundle-kde-ssl is a collection of many different packages.
01:33 < vasi> what he said :-)
01:34 < moike> okay
01:34 < moike> thanks
01:34 < dmacks> For the first time ever, /me has not-crappy net lag.
01:36 < vasi> dmacks, should we just forget about the parentEssential bug and push only the swappy code to 0.24.9?
01:37 < dmacks> Is it damaging anything other than our sanity and cirdan's chances of getting us to resolve other complaints of his?
01:38 < dmacks> I haven't heard of any other bugs uncovered in .8 so far.
01:39 < vasi> well, we have the bugs sitting in the tracker
01:39 < vasi> but the tracker's there so we can ignore them, right?
01:39 < moike> I wish someone made binaries for this.
01:39 < vasi> moike, we're working on it
01:39 < dmacks> I keep meanign to look at a weird thing I noticed in the validator regexps.
01:39 < vasi> (at least those of use who aren't whacko elitists)
01:39 < vasi> dmacks, ?
01:39 -!- shres [~sshreyas@202.144.95.244] has joined #fink
01:41 < dmacks> Validator.pm line 539 (in 0.24.8) I think I screwed up case-sensitivity ($field is always lc right?)
01:42 < dmacks> Also am considering making all validator msgs "Error" to shut up the maintainers who keep saying "it's just a warning, what's the big deal?"
01:43 < vasi> heh, well sometimes it's ok
01:43 < vasi> i mean, some packages can't avoid %n in Description
01:43 < vasi> if the %n is a common word
01:43 < dmacks> Right. But most are actually must-fix problems.
01:44 < vasi> do we really have a lot of maintainers going "feh, just a warning"?
01:44 < dmacks> A few loud ones.
01:45 < dmacks> (for example, calling scrollkeeper-update in postinst but forgetting Depends:scrollkeeper can leave your entire apt (and possibly some dpkg commands) hopelessly wedged)
01:45 < vasi> maybe we can do   my $warning = "ERROR" if grep { $ENV{USER} eq $_ } @bad_bad_maintainers;
01:45 < dmacks> heh yeah.
01:46 < vasi> except that's an inefficient way of doing that, should use a hash ;-)
01:46 < moike> night
01:46 -!- moike [~Mike@nat01-hardenburgh-ext.Rutgers.EDU] has quit ["Leaving"]
01:49 < dmacks> vasi: For a good time, see http://sourceforge.net/tracker/?group_id=17203&atid=414256&func=detail&aid=1215426
01:49 < vasi> yeah, wrong case in that regex
01:50 < vasi> oy
01:52 < vasi> wow....um, that guy really needs to find something to lead
01:52 < vasi> cuz he sure ain't good at following anything
01:54 < dmacks> For someone who claims to know all about programming, he shows a distinct lack of knowledge about...um...programming.
02:00 < vasi> hey, have you noticed any flakiness in the new indexing?
02:04 < dmacks> I have not.
02:07 < vasi> hmmm, a user submitted a .info which does InfoN: 2
02:07 < vasi> and it's not caught by the validator
02:08 < dmacks> This looks like a good candidate to apply http://sourceforge.net/tracker/index.php?func=detail&aid=1029590&group_id=17203&atid=317203
02:08 < dmacks> Which tracker item?
02:09 < vasi> https://sourceforge.net/tracker/?func=detail&atid=414256&aid=1251580&group_id=17203
02:10 < vasi> i liked the putting 0.24.7 on -devel to test for a while before release, i think we should do that with 0.25 as well (especially since it's likely to be much more buggy)
02:11 < dmacks> Yeah.
02:11 < vasi> i'm wondering if we should do something like branching branch_0_2x so that major architectural changes can go into HEAD and smaller bugfixes into 0_2x
02:11 < vasi> er, not bugfixes
02:11 < vasi> features, but smallish ones
02:12 < vasi> that don't require tearing apart whole modules :-)
02:12 < dmacks> ?
02:12 < vasi> well look at the 0.26 goal list
02:13 < vasi> we have relatively big stuff like refactoring the indexer, the dep engine, inherited build deps
02:13 < vasi> and small stuff like cirdan's dist-up patch, location detection, 'use Fink'
02:14 < vasi> those aren't gonna be done on the same time frame!
02:14 < dmacks> If we get 0.24.9 out Real Soon(tm), we can branch _0_25 from HEAD.
02:14 < dmacks> (/me was confused by so many "branch" terms and "x" values in your comment:)
02:14 < vasi> ah
02:15 < vasi> cuz 0.x.y releases have typically been just bugfixes (except for 0.24.7)
02:15 < dmacks> Right. Although we did have one branch_0_x+1 that came off branch_0_x instead of trunk.
02:15 < vasi> (which one?)
02:17  * dmacks can't remember...was before I was on core, drm thought the amount of backported stuff warranted a new minor-version and HEAD had progressed major leaps since the branch from it.
02:17 < vasi> yeah, that's the issue...that between 0.x.0 releases, there's typically 1) a lot of time 2) a ton of small features
02:17 < vasi> and they all get delayed
02:18 < vasi> until the big features are done
02:20 < vasi> anyway i'm not sure at all what we *should* do, i'm just thinking about it
02:20 < dmacks> Yeah.
02:21 < vasi> tbh, the .7 release just went pretty well, maybe that's the way we want to do it
02:21 < vasi> just appoint one or two of the .y releases to be a 'backport release'
02:22 < vasi> so that the small features can get in before we go rip fink apart for the 26th time
02:22 < dmacks> I really hate branches of branches.
02:24 < dmacks> That's a pretty cute bug with handle_infon :/
02:25 -!- eno is now known as eno-away
02:26 < cirdan> vasi: testing bootstrap now
02:26 < vasi> cirdan, sweet :-)
02:26 < cirdan> heh
02:26 < cirdan> i hope the old ncueses packge builds on tiger ;-)
02:26 < cirdan> it should
02:27 < dmacks> I guess handle_infon_block() should return a list (\%properties,$infon_level) instead of passing the N metadata as if it were an actual field.
02:27 < cirdan> ooo
02:27 < cirdan> the auto country stuff is niiice
02:27 < cirdan> :-)
02:28 < dmacks> I have no idea why Apple made that .plist so hard to access.
02:28 < cirdan> you could just use perl's gettext, no/.
02:28 < cirdan> ?
02:29 < dmacks> language != location
02:29 < cirdan> eh
02:29 < cirdan> right
02:29 < vasi> it's not so hard to access...what do you mean?
02:29 < dmacks> OTOH, if you're offering to make fink i18n... :)
02:29  * cirdan ^5s vasi anyway
02:29 < vasi> i dunno, maybe it's just me, that's the place i was expecting it to be :-)
02:29 < cirdan> hehe
02:30 < dmacks> That's one of the few, perhaps only, .plist that you have to access by complete pathname.
02:30 < vasi> dmacks, i think it would be easier to just have the handle_infon_block specifically test for this issue
02:30 < vasi> oh yeah, well i dunno...you can do '-g' for global, but it only does user-global
02:31 < dmacks> Move validator code into Package?
02:31 < vasi> (and it's not complete pathname, it's complete path minus the ".plist" :-)
02:31 < cirdan> heh
02:31 < cirdan> yeah, the plist tool are a real pain like that
02:31 < vasi> dmacks, if a .info has bad syntax, it breaks in Services not in Validator
02:31 < vasi> Validation even
02:32 < dmacks> Validator checks for use of InfoN features beyond their N level.
02:32 < dmacks> That's perfectly functional, though against policy.
02:34 < dmacks> (the problem is that user's .GlobalPrefernces completely replaced the system one instead of allowing fallback)
02:34 < vasi> dmacks, yeah
02:34 -!- assassin [~percussor@assassin.user] has quit [Read error: 110 (Connection timed out)]
02:37 < dmacks> Actually, nm...seems like user's *always* replaces system's. That's a silly way to provide "defaults".
02:41 < vasi> i thought it merged them sometimes
02:41 -!- kane_ [~kane@perl.xs4all.nl] has joined #fink
02:42 -!- muesli [~muesli@mail.muehlhaeuser.de] has joined #fink
02:42 < dmacks> I thought so too. Perhaps at a lower-level interface than system(defaults).
02:42 < cianhughes> hey has anyone ever seen a problem where all the text in kdevelop changes to boxes?
02:42 < kane_> are there any freenode ops here?
02:43 < dmacks> cianhughes: Long ago...was a missing gtk+2 or glib2 pkg, or a need to do one of the font cache fixes. Can't remember exactly:(
02:43 < cianhughes> ok thanks, it's affecting one of my two machines
02:44 < dmacks> kane_: You mean someone with ops for this channel, or a system-wide admin person?
02:44 < cianhughes> unfortunately it's the laptop I have to take away in two hours
02:44 < kane_> dmacks: someone i tell to a bot is spamming users :)
02:45 < dmacks> Check in #freenode
02:47 < cianhughes> dmacks: my gtk+2 and glib2 are hopelessly out of date, thanks for the tip
02:47 < dmacks> lilo is idle; can't remember the incantation to see a list of freenode staff.
02:49 < dmacks> (thanks for jogging my memory to deal with a glib2 bug:)
02:51  * vasi is gonna stomp all over TheSin's module
02:51 < vasi> because it's just gross that way :-)
02:52 < vasi> (it does work though, gotta give him credit for that....and it's not his fault what was done to his module)
02:54 < cianhughes> dmacks: could it by any chance have been sudo fc-cache
02:55 < dmacks> Yeah yeah, he's just another gun or gasoline-additive additive:)
02:55 < dmacks> s/ additive/ manufacturer/
02:56 < dmacks> cianhughes: This was somewhat before Tiger, which is the first time fc-cache became often-needed.
02:56 < kane_> dmacks: cheers,... slow day in #freenode it seems.. must be early :)
02:56 < dmacks> I *think* there may have been multiple causes and/or people tried "lots of things and it went away"; never clearly diagnosed down to specific-cause=>specific-solution.
02:57 < cianhughes> ok, well for me updating libtiff, gtk+2 glib2 and then a sudo fc-cache has worked, thanks for poking me in the right direction of the solution.....
02:57 < dmacks> kane_: Good luck with it:(
02:57 < kane_> dmacks: worst case i wait for my cow-orker to wake up.. he's an oper on just about every network i think.. he's also vast asleep :)
02:58 < dmacks> heh
03:02 < cirdan> dmwaters?
03:02 < cirdan> he's an admin
03:03 < dmacks> http://freenode.net/faq.shtml#helpfromstaff
03:05 < cianhughes> Anyone ever seen a message "Sorry - KDevelop
03:05 < cianhughes> Could not create language plugin for C"
03:05 < cirdan> hmm
03:06 < cirdan> vasi: look in my exp for ncurses5.info
03:06 < cirdan> use that in bootstrap
03:06 < vasi> cirdan, does it work or bug out or what?
03:07 < cirdan> i'm not sure
03:07 < cirdan> it was built, but for some reason dselect linked against system ncurses
03:07 < vasi> oops
03:07  * cirdan tries again
03:07 < cirdan> it looks like the -dev was missing
03:08 < vasi> hmm
03:10 < cirdan> ah-ha
03:10 < cirdan> there is no builddeps in dpkg
03:10 < cirdan> at least in bootstrap
03:10 < cirdan> odd...
03:10 < vasi> ?
03:12 < cirdan> ah, we do whacky stuff
03:12 < cirdan> CompileScript: <<
03:12 < cirdan>  %p/bin/fink -y install gettext-bin gettext-dev gettext-tools libiconv-dev libncurses5
03:12 < cirdan> ugh
03:12 < vasi> ewww
03:13  * cirdan fixes...
03:15 < cirdan> vasi: now that we have sane defaults for mirrors, it'd be nice to have a "pick a random top 5 server for my country" feature
03:16 < cirdan> or even just all primary servers, no random
03:16 < vasi> cirdan, one thing at a time :-)
03:16 < cirdan> :-)
03:16 < vasi> um wait...ncurses shouldn't be rotated anyhow, right?
03:16 < vasi> isn't the ncurses.info in bootstrap fine?
03:17 < cirdan> vasi: it is now
03:17 < vasi> it's libncurses-shlibs.info that's ordered weirdly
03:17 < cirdan> it's where i discovered the problem
03:17 < vasi> er, ncurses-shlibs.info
03:17 < cirdan> i'm fixing up the old .info
03:17 < vasi> hmm...well continue on then :-)
03:19 < dmacks> Do not change the CompileScript's use of %p/bin/fink instead of BDep before checking with drm!
03:21 < vasi> dmacks, he's just testing not committing
03:21 < dmacks> okay.
03:21 < vasi> what *is* the whole use of fink in compilescript about anyhow?
03:22 < dmacks> There's some major reason for doing it that way that drm's explained a few times and it never really stuck in my brane.
03:22 < cirdan> hmm
03:24 < cirdan> i think it has to do with serious brokenness in the dep engine during bootstrap
03:24 < cirdan> but i forget
03:24 < cirdan> ugh
03:24 < cirdan> waaay past bedtime
03:30 -!- dmacks [~dmacks@dmacks.active.supporter.pdpc] has quit ["leaving"]
03:36 -!- vasi [~vasi@modemcable133.147-70-69.mc.videotron.ca] has quit ["Client exiting"]
03:38 -!- shres [~sshreyas@202.144.95.244] has quit ["This computer has gone to sleep"]
03:42 -!- shres [~sshreyas@202.144.95.244] has joined #fink
04:10 -!- Netsplit zelazny.freenode.net <-> irc.freenode.net quits: Melian
04:11 -!- Netsplit over, joins: Melian
04:20 < cianhughes> anyone here use kdevelop
05:06 -!- knghtbrd [quacked@d172-104.uoregon.edu] has quit [Read error: 104 (Connection reset by peer)]
05:19 -!- knghtbrd [quacked@d172-104.uoregon.edu] has joined #fink
05:44 -!- Bart-- [~hideout@213.23.123.11] has joined #fink
06:38 -!- shres [~sshreyas@202.144.95.244] has quit [Read error: 110 (Connection timed out)]
06:47 < Feanor_> nobody has ever used kdevelop
07:47 < cirdan> ever
07:51 -!- cirdan [~chris@pcp04356153pcs.glstrt01.nj.comcast.net] has quit ["leaving"]
07:51 -!- cirdan [~chris@pcp04356153pcs.glstrt01.nj.comcast.net] has joined #fink
07:57 -!- akh [~akhansen@ldx3.psfc.mit.edu] has joined #fink
08:34 -!- You're now known as RangerRick
08:49 < cirdan> mornin'
09:07 -!- tenner [~tenner@p50926C5D.dip.t-dialin.net] has joined #fink
09:07 -!- tenner [~tenner@p50926C5D.dip.t-dialin.net] has left #fink []
09:08 < akh> mornin'
09:08 < RangerRick> hola
09:09 -!- chori [~snag@85.8.1.87] has joined #fink
09:09 < chori> Have anybody compiled fluxbox 0.9?
09:09 -!- geewz [~gregreede@ppp122-142.static.internode.on.net] has joined #fink
09:11 < akh> If they have, they probably haven't bothered to let us know so that we can update the fink package.
09:12 < chori> akh: Ok becouse I've compiled it and that worked but I won't start
09:12 < akh> And you're starting the "normal way"?  (in a .xinitrc or some such)
09:14 < chori> when compiled it puts the bin in /usr/local/bin
09:14 < chori> so I started X and then started fluxbox
09:24 < akh> If you didn't put it your .xinitrc or the system's xinitrc, then you can't run it on top of another window manager.
09:26 -!- Bart-- [~hideout@213.23.123.11] has quit ["Computer going to sleep..."]
09:29 -!- mprentice [~mprentice@cpe-24-59-118-103.twcny.res.rr.com] has joined #fink
09:29 < chori> I think Im starting to get it to work
09:29 < chori> D
09:29 < chori> :D
09:33 < akh> good
09:34 < cirdan> vasi and dmacks were up too late
09:34 < cirdan> they need to get on irc
09:34 < akh> heh
09:35  * akh had to use Windows last night and I couldn't get an outbound IRC connection.
09:42 < cirdan> hah
09:42 < cirdan> sux0rs
09:51 < chori> I've got fluxbox 0.9 to run. Think I would be good to have it in Fink?
09:52 < cirdan> sure
09:56 < chori> cirdan: How?
10:04 -!- drm [~drm@m015f36d0.tmodns.net] has joined #fink
10:05 < drm> akh: do you understand how your wiki works?
10:05 < drm> akh: cause i can't seem to write c++ or g++
10:06 < akh> drm:  yeah.  dmacks mentioned the same thing
10:07 < drm> it wants to underline stuff 
10:07 < drm> and putting backslashes in front doesn't work :(
10:08 < akh> hmmm---I just tried putting 'c++' on the home page and it looked OK.
10:09 < akh> drm:  Can you give me a page to look at?
10:09 < drm> akh: the new 10.4 tree page
10:10 < drm> (was i supposed to do something about exporting or publishing?)
10:10 < akh> Nope--that's automatic.
10:12 < akh> hmmm...I think it may be treating one of the plusses as a mathematical operator
10:12 < drm> possibly
10:16 < pogma> what happened to fingolfin, he created free time magically?
10:16 < pogma> hi drm, akh 
10:16 < akh> drm:  Best solution I found was to use [[g++]] :  that way there's just the annoying prompt to create a new page (which we could even do)
10:17 < drm> hi potma
10:17 < akh> pogma:  I think he graduated.
10:17 < drm> also pogma :)
10:17 < pogma> he can't have graduated, germans don't graduate 'til they're 50 :)
10:17 < akh> heh
10:18 < pogma> where is this wiki again?
10:19 < drm> pogma: he finished a thesis and is getting a degree, but you are certainly correct, he is not finished!
10:19 < drm> pogma: was mentioned in fink-devel not long ago (the URL)
10:20 < pogma> found ir
10:20 < pogma> it
10:20 < akh> Ah--that's right.  
10:20 < akh> Still, after writing a thesis there's probably a lull.
10:22 < pogma> Hmm, wiki looks good
10:22 < pogma> I know there is at least one wiki installed on sancho
10:24 < akh> This one's pretty easy to install.
10:24 < pogma> wondering if people would update fink's docs more often on a wiki
10:24 < pogma> probably not
10:25 < pogma> having things spread around is not optimal though
10:25  * pogma shuts up
10:26 < akh> Right.  I'm of the opinion that the docs _would_ get updated more often, though--it takes much less time/knowledge on a wki.
10:26 < akh> wiki
10:26 < pogma> akh: Well, imo, you're the docs dude, your call
10:27 < drm> anyway, if the people who are actually working on stuff will use the wiki to debate plans, we are better off
10:27 -!- kpaul [~kpaul@70.57.247.118] has joined #fink
10:27 < drm> and it will be easier to convert this stuff to the docs afterwards, too
10:27 < akh> Yup.
10:27 < pogma> it's unanimous, we're wikilovers :)
10:28 < pogma> akh: you might want to chat to bbraun if you want to move the wiki
10:28 < akh> Sure
10:29 < akh> I'm not absolutely wedded to this particular one, either--it was just incredibly easy to get running.
10:29 < akh> (if somebody wants a different setup)
10:29 -!- cmeme [~cmeme@216.184.11.2] has quit ["Client terminated by server"]
10:29 < drm> akh: the great thing from my point of view is that i'm not the guy who has to run it 
10:30 < drm> so do whatever you want :)
10:30 < akh> That's good, too.
10:30 -!- chori [~snag@85.8.1.87] has quit [Read error: 110 (Connection timed out)]
10:30 -!- cmeme [~cmeme@216.184.11.2] has joined #fink
10:30 < drm> pogma, you know C, right?
10:30 < pogma> why?
10:31 < pogma> (yes)
10:31  * drm watches pogma squirm
10:31 < drm> well, some patch involving rearranging structs is needed, and i can either wait for msachs to wake up, or... :)
10:31 < pogma> I've nothing to do for 20 minutes, then I'm off to bed, so it's okay
10:31 < pogma> where?
10:32 < drm> odcctools fails under XCode 2.1 with an error in bytesex.c about struct <anonymous> having no member nemed 'environ'
10:33 < pogma> Ah, one sec
10:33 < pogma> Add -D_MACH_I386_THREAD_STATUS_FPSTATE_LEGACY_FIELD_NAMES_ -D_ARCHITECTUR
10:33 < pogma> E_I386_FPU_FPSTATE_LEGACY_FIELD_NAMES_
10:33 < pogma> to CPPFLAGS
10:34 < drm> really? ok
10:34 < pogma> well, without the line break :)
10:34 < drm> right
10:34 < drm> no affect on deb if it was built nder XCode 2.0 already? or does it need a rev up?
10:35 < pogma> yeah, I am supposedly an odcctools developer, but ssen told me those defines a few months ago
10:35 < pogma> when I whined in #opendarwin that odcctools wouldn't build :-)
10:36 < drm> ok... which version should this be done on? should the current unstable go to stable anyway?
10:36 < pogma> akh: bbraun is alive and chatting in #opendarwin now. Do you want me to ask him to join?
10:37 < akh> pogma:  I'm a bit too busy right now--I'm working on a presentation for a meeting that's in 20 minutes--after that's over, though...
10:37 < pogma> drm: I don't know what's in stable, it shouldn't be needed on tiger anyway, already has cctools-576 or later
10:38 < drm> ok, i'll just move 576 to stable then
10:38 < pogma> I'd put the latest version in stable and add the flags
10:38 < pogma> yeah
10:44 < pogma> Oh, I probably should undo my basefiles changes, since Murr announced in channel that the bug was fixed in 10.4.3 :-)
10:45 < drm> does that mean we *don't* need a basefiles release?
10:45 < pogma> drm: I don't know, 10.4.2 took forever and a day to come out
10:46 < drm> so maybe we should release it first, then let you take the changes out for the next version?
10:46 < pogma> apparently the FALLBACK_PATH thing is fixed in 10.4.3, so...
10:46 < pogma> maybe that's best
10:51 < pogma> so odcctools built for you?
10:52 < drm> yes
10:52 < drm> thanks
10:52 < pogma> thank shantonu :)
10:52 -!- regeya [~shane@adsl-sp3-cdale176.micgi.com] has joined #fink
10:55 -!- dsias [~dsias@adsl-068-153-207-210.sip.bct.bellsouth.net] has joined #fink
10:56 < cirdan> drm
10:57 < drm> wow, that "processor" field in the SF file release page is going to mean something to us someday :)
10:57 < cirdan> with cvs head, bootstrap works when i take the fink call out of dpkg's compile script, add the stuff to builddeps
10:57 < cirdan> and even have an essential splitoff of a non essential package
10:59 < drm> the reason for the fink call is for people who are *updating* the package, not the bootstrapers
11:03 < cirdan> but builddeps would still work
11:04 < drm> not unless we change the code... i can give exapmles of how it doesn't work with current code
11:04 < drm> pogma: dunno if you're still subscribed to fink-core, so i cc'd you on a msg
11:04 < drm> you don't have to be core to commit code, BTW :)
11:05 < cirdan> you don't?
11:05 < cirdan> shucks
11:05 < cirdan> that means i gotta update my packages now
11:05 < cirdan> ;-)
11:06 < drm> bbl
11:06 -!- drm [~drm@m015f36d0.tmodns.net] has quit ["Leaving"]
12:23 -!- kane_ [~kane@perl.xs4all.nl] has quit []
12:44 -!- cptj [~sirq@cpc3-stap3-6-1-cust179.nott.cable.ntl.com] has joined #fink
12:45 < cptj> hi there
12:45 < cptj> can anyone give me some advice on using kde?
12:47 -!- dalibor_mac [~dalibor_m@139.19.131.1] has joined #fink
12:48 < akh> Such as?
12:48 < cirdan> fink info kde
12:48 < cirdan> there ya go
12:48 < cirdan> ;-)
12:48 < akh> No such package
12:48 < akh> fink info bundle-kde
12:49 < cptj> i'm getting error messages when i try to run xdarwin after editing my ~/.xinitrc file
12:50 < akh> paste them up via lisppaste
12:50 < akh> lisppaste: URL?
12:50 < cptj> i'm a bit of a beginner... i'm not sure what lisppaste is...!
12:50 < akh> grr...broken
12:50 < akh> that's what it is
12:51 < cptj> heh :)
12:51 < lisppaste> To use the lisppaste bot, visit http://paste.lisp.org/new/fink and enter your paste.
12:51 < cptj> basically i'm getting 'permission denied' errors when it's trying to create some folders
12:51 < akh> Ah, that's better.
12:52 < lisppaste> cptj pasted "kde errors" at http://paste.lisp.org/display/10524
12:53 < cptj> there we go...
12:53 < cptj> that's my log file.
12:54 < cptj> i start xdarwin, it gives me some errors, and it quits.
12:54 < akh> Run "ls -ld ~/.kde" and see who actually owns the file.
12:54 < akh> (directory, rather)
12:54 < cptj> ahh. looks like root
12:55 < akh> Yup.  That's it.
12:55 < cptj> if i change it to me, will that sort it out then?
12:55 < akh> That should do it.
12:58 < cirdan> sudo chown -R sirq.sirq ~/.kde
12:58 < cirdan> or whichever
12:59 -!- dmacks [~dmacks@netspace.org] has joined #fink
12:59 < cirdan> hey dmacks 
12:59 < cirdan> g'afternoon
12:59 < dmacks> hiya cirdan
12:59 < cirdan> finally awake?
12:59 < cirdan> slacker
12:59 < dmacks> Yeah.
13:00 < cirdan> ok, bootstrap works well
13:00 < dmacks> Gonna be a short day at the ol' $JOB today.
13:00 < cptj> ok, changed, and xdarwin started...
13:00 < cirdan> i even moved the fink call to builddeps and they are golden
13:00 < dmacks> Good! I always thought that was a weirdity.
13:00 < cirdan> drm said the fink call is for people upgrading fink/dpkg
13:00 < cirdan> but i believe it isn't needed anymore
13:00 < cptj> looking good! thank you very much :)
13:01 < dmacks> And he [agrees,disagrees]?
13:01 < cirdan> if we need a specific version, then use a versioned builddep
13:01 < cirdan> dmacks: he had no comment
13:01 -!- cptj [~sirq@cpc3-stap3-6-1-cust179.nott.cable.ntl.com] has quit ["Quit"]
13:01  * dmacks really doesn't understand the problem, so isn't a good place to look for advice or approval.
13:02 < cirdan> right, i'm just providing feedback
13:02 < dmacks> '
13:02 < dmacks> 'k
13:02 < dmacks> akh: Do we know exactly what package causes the root-owned ~/.kde to appear?
13:02 < cirdan> no
13:03 < cirdan> something in the kde build, i believe
13:03 < dmacks> Do we at least know if it's created during the build right into ~, or is installed into %d/~ in the .deb?
13:05 < dmacks> (the former case is solved in HEAD, the latter is more complicated)
13:05 < cirdan> pretty sure it's created during the build
13:06 < cirdan> .qt is affected sometimes too
13:06 < dmacks> Hrm, any packages that take less than a week to build? If HEAD fixes it, I'll backport it to 0.24.9
13:11 < cirdan> no idea
13:12 < cirdan> talk to RangerRick aboui t
13:27 -!- hennker [flullup@dsl-213-023-251-011.arcor-ip.net] has joined #fink
13:27 -!- dmacks [~dmacks@dmacks.active.supporter.pdpc] has quit ["leaving"]
13:46 -!- dk0r [~dk0r@cpe-24-194-171-176.nycap.res.rr.com] has joined #fink
14:01 -!- dk0r [~dk0r@cpe-24-194-171-176.nycap.res.rr.com] has quit []
14:16 < akh> w00t!  No more "this test is designed to die!"
14:19 -!- kpaul [~kpaul@70.57.247.118] has left #fink []
14:21 < akh> heh--found a great bug that I haven't seen on -beginners yet.
14:22 < akh> While building something that needs libgettext3-dev, run "fink selfupdate'.  gettext-dev wants to be installed -> selfupdate bombs.
14:22 < akh> I guess that's why we have build locks now.  ;-)
14:28 < cirdan> heh
14:33 -!- drm [~drm@dirac.cgtp.duke.edu] has joined #fink
14:37 -!- MacinMan [~tron@117.101.207.68.cfl.res.rr.com] has joined #fink
14:37 < MacinMan> how's everyone doing today?
14:38 < drm> hi
14:38 < MacinMan> i seem to have fixed my x windows issues
14:38 < MacinMan> got rid of apple's and installed xorg
14:39 < MacinMan> seems much better and reliable
14:44 < akh> I found it to be so.
14:45 < MacinMan> akh i got dia and gnome to work better with xorg
14:45 < MacinMan> then apples
14:45 < akh> Good
14:45 < MacinMan> although with gnome something is still busted when i try to open the computer icon instead of getting the computer contents it says it's not associated with anything
14:46 < akh> hmm
15:00 < MacinMan> yeah i am thinking if i want gnome use fink to install the dependencies and then just compile the latest source by hand
15:03 -!- dmacks [~dmacks@netspace.org] has joined #fink
15:04 < akh> MacinMan: It may not work.
15:04 -!- MacinMan [~tron@117.101.207.68.cfl.res.rr.com] has quit ["bbl"]
15:05 < akh> heh--left before I could paste the Wiki link
15:05 < dmacks> htodd: we're working on getting GNOME up to current versions...could you check that your glib-pm, gtk2-pm, extutils-pkgconfig-pm are up-to-date (and perhaps variantize them...the Perl Modules section of the packaging manual has examples of the current best-practices)
15:05 -!- nkuttler [~nkuttler@dsl-084-058-194-071.arcor-ip.net] has joined #fink
15:06 < akh> ewww...variants
15:07 < dmacks> Don't matter to me, but seems silly to have only -pm581 on Tiger:)
15:07 < akh> I won't argue with that.
15:08 < dmacks> !seen miga
15:08 < Melian> miga <~miga@easyconnect2121138-139.clients.easynet.fr> was last seen on IRC in channel #fink, 25d 5h 16m 45s ago, saying: 'There seems to be some explanation in info files.'.
15:09 < drm> looks like a vacation
15:09 < akh> Better go via -devel
15:09 < dmacks> Yeah.
15:09 < drm> dmacks: i put some stuff on the wiki
15:09  * dmacks reads........
15:10 -!- MacinMan [~tron@117.101.207.68.cfl.res.rr.com] has joined #fink
15:10 < MacinMan> akh yeah it might not work
15:10 < MacinMan> but fink's package seems to work jusgt as good
15:10 < MacinMan> oops just as good
15:11 < MacinMan> so i've decided to try kde
15:11 < MacinMan> the latest version
15:11 < akh> fair enough
15:11 < akh> drm:  ==c++==
15:11 < akh> That's the escape
15:11 < drm> ah, gee, i was hoping for a b-
15:11 < drm> :)
15:11 < akh> (found it in the GCC page)
15:12 < MacinMan> actually i've heard kde is nicer to deal with then gnome these days
15:12 < dmacks> drm: Shoulda studied harder.
15:12 < akh> MacinMan: I think so--it's a bit slower.
15:12 < MacinMan> yeah but ti seems more bsd compatible
15:12 < MacinMan> it even
15:12 < MacinMan> or at least easier to do cross platform
15:13 < akh> Yeah; too bad my other platform doesn't have 3.4.1 available in binaries yet.
15:13 < MacinMan> actually akh 3.4.1 doesn't seem that slow
15:13 < akh> Fink will probably have them before Debian does.
15:13 < drm> is it a conspiracy, akh?
15:13  * dmacks shakes fish at wacko elititst.
15:13 < dmacks> *fist
15:13 -!- regeya [~shane@adsl-sp3-cdale176.micgi.com] has left #fink ["Leaving"]
15:14 < MacinMan> akh fink has 3.4.1 as source
15:14 < drm> i like the idea of shaking fish
15:14 < MacinMan> what's the binary releaswe
15:14  * dmacks puts tuna back in pocket.
15:14 < MacinMan> oops release
15:14 < akh> MacinMan: Yes.
15:14 < akh> Debian has 3.3.x
15:14  * akh doesn't feel like building the whole thing there too.
15:14 < MacinMan> i seriously don't see all that much edifference akh
15:14 < akh> I don't really either.
15:14 < akh> drm:  Can't blame Ranger* for that one.
15:14 < dmacks> vasi!
15:15 < MacinMan> akh i tried gnome 2.10 on linux i didn't care much for it
15:15 < MacinMan> i think 2.0.8 was nice
15:15 < dmacks> Clef_: When you get a chance, there's a new teeny-version of your pyorbit2-pmXX.
15:15  * akh hasn't had GNOME on my Linux box in a while.
15:16 -!- Murr [~neeri@A17-202-20-71.apple.com] has quit [Excess Flood]
15:16 < MacinMan> well akh I don't see much change in gnome
15:16 < MacinMan> just more bloat and brekage
15:16  * dmacks answers drm's and vasi's concerns about my .la/.pc idea for IBD
15:17 < akh> Cool--/me will read later.
15:17 < MacinMan> akh KDE actually would be in my opinion a good aqua clone for non os x
15:17 < akh> Most people think of it as a Windows clone.  ;-)
15:17 < akh> (except less annoying)
15:17 -!- Murr [~neeri@A17-202-20-71.apple.com] has joined #fink
15:18 < MacinMan> well it is a good alternative to windows
15:18 < MacinMan> i think kde could be it's own os
15:18 < MacinMan> without needing linux if they did work on it
15:18  * dmacks thinks wiki should ask for confirmation before posting as an AC
15:18 < akh> dmacks:  probably:  I can't really change that.
15:19 < dmacks> Yeah, just musing out loud.
15:19 < MacinMan> akh would something like ogle work on os x
15:19 < MacinMan> it's a simple dvd player
15:19 < MacinMan> it uses libdvd read play etc
15:21 < akh> I'm not sure--there's probably a good reason it hasn't hitherto been ported.
15:21 < MacinMan> probably
15:21 < MacinMan> how's qt mac working
15:21 < MacinMan> the native aqua version of qt
15:22  * akh downloaded an app that uses it--seems not bad.
15:22 < MacinMan> that's one thing i notice akh
15:22 < MacinMan> qt the source of kde is aqua native now 
15:22 < drm> dmacks: right, your proposal is somewhat opposite to the current IBD proposal
15:22 < MacinMan> gtk still isn't
15:23 < MacinMan> or is there an aqua gtk now
15:23 < dmacks> drm: Right.
15:23 < akh> MacinMan: I believe there's a primitive one--not nearly as advanced as QT
15:23  * dmacks didn'tr realize how far until talked to vasi about it last nite.
15:24 < MacinMan> akh i have gimp.app
15:24 < drm> dmacks: i guess i didn't just mean "deterministic"... i also meant "humanly determinable"
15:24 < MacinMan> it seems like a nice package
15:24 < MacinMan> it uses x to run
15:24 < akh> Yup--I use it too.
15:24 < MacinMan> bgut it's aqua colored
15:24 < dmacks> drm: Right.
15:24 < drm> if something is going wrong, dmacks, with IBD you should be able to track it down... with automated reads of .la/.pc it would be much harder, i think
15:24 < MacinMan> are the actual gtk libs in the package akh?
15:25 < akh> Probably in some format
15:25 < MacinMan> ok
15:25 < drm> no reason we can't have both, i guess, dmacks
15:25 < dmacks> Yeah.
15:25 < MacinMan> i noticed filmgimp is no longer in fink
15:25 < MacinMan> did it die?
15:25 -!- dalibor_mac [~dalibor_m@139.19.131.1] has left #fink []
15:25 < akh> It was renamed
15:25 < drm> macinman: name changed
15:25 < akh> cinepaint
15:26 < MacinMan> ahh ok
15:26 < MacinMan> does it actually work with video files
15:26 < dmacks> OTOH, IBD as a whole is a bit scary, since a package "loses control" over its BD.
15:26 < akh> Haven't tried it yet.
15:26 < MacinMan> from the description i gathered that
15:26 < MacinMan> ok
15:27 < drm> right...but only over BD, not over RTD
15:27 < dmacks> I'm wondering "package foo doesn't build" could be due to a missing IBD at any level of recursion.
15:28 < dmacks> Due to AutoShlibs, BD effectively sets the RTD too.
15:28 < MacinMan> akh did u ever find out if the typo theory was true with the sndplay brekage
15:29 < akh> MacinMan: I think it was an issue with xcodebuild--it was set up for an earlier XCode.
15:29 < akh> That kind of thing happens a log.
15:29 < akh> lot
15:29 < MacinMan> ahh ok
15:29 < drm> dmacks: true
15:30 < MacinMan> well akh i'm really impressed with fink and my ibook it really performs well
15:30 < MacinMan> and compiles go well usually
15:30 < akh> That's what we aim for.  ;-)
15:31 < MacinMan> hehe
15:31 < akh> *though /me grumbles about oo.org.  I don't really feel like hacking that gigantic .patch file to add part of another gigantic patch file.
15:31 < dmacks> What now?
15:31 < akh> the mozilla thing
15:32 < dmacks> Ah yeah. asari fixed it I think.
15:32 < dmacks> (oidata.h, or whatever it was)
15:32 < akh> Really?  I didn't notice a new %r
15:32 < dmacks> It didn't compile at all, so there was never a previous-%r .deb:)
15:33 < akh> Ah--so he did
15:33 < dmacks> (I guess.../me didn't look closely at the -commit msg)
15:33 < akh> "dded Mozilla oiddata.h patch from Fink's mozilla package. Thanks, Fink-users mailing list"
15:33 < dmacks> cool
15:34 < akh> guess I'd better try it.
15:34 < drm> i just committed a fix for sndplay (due to msachs)
15:34 < dmacks> Wait for Dr. Whatever:)
15:34 < akh> He's on 10.3.9
15:34 < dmacks> Oh:(
15:35 < akh> If /me had $300 to spare I'd buy a bus-powered removable so that I could have a portable Panther drive.
15:35 < akh> (or I could save $50 and build it myself, but there's that time thing)
15:36 < akh> Maybe dual FW/USB2
15:36 < akh> My existing FW drive is just a bit too heavy to haul around.
15:37 < akh> And I'd have to plug it in if I wanted to do stuff on the train.
15:37 < MacinMan> do they give you wireless on the train  akh
15:37 < dmacks> Hope you're not going south of NJ then...Philly commuter line snagged the catenary.
15:38 < akh> MacinMan: :  Nope--even cell phone coverage cuts out in places.
15:38 < dmacks> I guess you could say they were powerless to fix it.
15:38 < MacinMan> ok akh
15:39 < akh> dmacks:  heh
15:39 < MacinMan> well kde still takes forever to compile
15:40 < akh> It's better on Tiger than on past OS X versions.
15:43 < akh> hmmm...wonder where I can find a mini DVI -> VGA adapter that works with my powerbook.  I think the one from the Apple Store website is too new.
15:43 < akh> (mine got "borrowed" or mislaid by me)
15:43 < MacinMan> do u have a new powerbook
15:43 < akh> Not that new.
15:43 < MacinMan> ahh ok
15:44  * akh tried an adapter from someone with a newer powerbook and it didn't fit.
15:44 < MacinMan> sounds like akh is up a creak
15:44 < dmacks> Xacto-knife?
15:44 < akh> Something like that.
15:45 < MacinMan> don't you have an ibook too akh
15:46 < akh> nope
15:46 < MacinMan> ahh ok
15:47 < MacinMan> i wonder if apple uses piezo buzzers as the internal laptop speakers
15:48 < dmacks> If a package source and homepage host don't resolve and the pkg is license:restrictive (==no source mirror) and maintainer=none, should I pull it?
15:49 < drm> sounds like it
15:49 < dmacks> (could someone else try them to make sure it's not my DNS problem...package is "hx")
15:49 < drm> dmacks: have tried many times in the past with "hx"
15:50 < dmacks> Okay, I'll kill it.
15:51 < dmacks> hxd too, huh?
15:51 < drm> so it seems
15:52 -!- nkuttler [~nkuttler@dsl-084-058-194-071.arcor-ip.net] has quit [Connection timed out]
15:57 < akh> w00t!  Found one.
15:59 < cirdan> ?
15:59 < cirdan> hxd isn't restrictive
15:59 < cirdan> iirc
16:02 < dmacks> http://fink.sourceforge.net/pdb/package.php/hxd
16:02 < akh> $ fink dumpinfo -fLicense hx
16:02 < cirdan> so
16:02 < cirdan> it was once gpl, and it links to gpl sw
16:02 < dmacks> Feel free to take over maintainership and re-commit it.
16:05 < akh> Assuming that the source can be found, of course.
16:05 < dmacks> Roight
16:05 -!- drm [~drm@dirac.cgtp.duke.edu] has quit ["Leaving"]
16:08 < akh> What is it with Safari and commerce sites?  This is the second time I've had a zipcode field strip my leading zero.
16:10 < cirdan> hmm
16:10 < cirdan> anyone have a cups server w/a printer on it?
16:10 < cirdan> laser printer
16:11 < akh> LPD here and for my laser at home as well
16:11 < akh> (or rather--each printer has its own IP)
16:11 < cirdan> hmm
16:11 < cirdan> mail.app seems to be sending bad ps
16:11 < cirdan> other apps print
16:11 < cirdan> and it prints if i re-run it through shostscript
16:12 < cirdan> or even if i preview it and print from preview.app
16:12 < akh> interesting
16:12 < dmacks> My 10.3 machine has no printers attached...does that help? (OTOH, my Mail.app is broken)
16:12 < cirdan> heh
16:12  * dmacks guesses "no"
16:16 < akh> Disk space collapsing....can't withstand oo.org build...must go home
16:16 < cirdan> haha
16:17 < akh> And only the xorg community can use my .deb.
16:17 < dmacks> heh
16:17 < akh> neener neener
16:18 < akh> s/can/should/
16:18 < dmacks> Face it, akh...you're still not even half the Wacko Elitist that RangerRick is:)
16:19 < akh> Yeah--that's a very high bar.  :-)
16:19 -!- protoplasm [~protoplas@c-24-18-61-81.hsd1.wa.comcast.net] has joined #fink
16:19 -!- dmacks [~dmacks@dmacks.active.supporter.pdpc] has quit ["leaving"]
16:20 -!- akh [~akhansen@ldx3.psfc.mit.edu] has quit ["runaway train"]
16:28 < cirdan> !seen vasi
16:28 < Melian> vasi <~vasi@modemcable133.147-70-69.mc.videotron.ca> was last seen on IRC in channel #fink, 13h 7m ago, saying: 'what *is* the whole use of fink in compilescript about anyhow?'.
16:28 -!- msachs [~msachs@A17-255-98-107.apple.com] has joined #fink
16:29 -!- RLD_osx [~rldempse@24-178-204-108.dhcp.ftwo.tx.charter.com] has quit ["Mac OS X  - - a better alternative to winblow$"]
17:38 -!- Qram [~nonya@pool-71-107-82-18.lsanca.dsl-w.verizon.net] has joined #fink
17:38 < Qram> how do I get 0.7.2 /
17:42 -!- Qram [~nonya@pool-71-107-82-18.lsanca.dsl-w.verizon.net] has quit ["life is not measured by the number of breaths we take, but by the moments that take away our breath"]
17:43 -!- You're now known as RangerAway
17:45 -!- MacinMan [~tron@117.101.207.68.cfl.res.rr.com] has left #fink []
17:57 -!- assassin [~assassin@assassin.user] has joined #fink
17:59 -!- RLD_osx [~rldempse@24-178-204-108.dhcp.ftwo.tx.charter.com] has joined #fink
18:12 -!- msachs [~msachs@A17-255-98-107.apple.com] has quit []
18:31 -!- vasi [~vasi@modemcable133.147-70-69.mc.videotron.ca] has joined #fink
18:31 < vasi> !seen mcostabel
18:31 < Melian> i haven't seen 'mcostabel', vasi
18:32 < vasi> i thought he used to show up? oh well
18:41 < cirdan> vasi: 
18:41 < cirdan> status report...
18:41 < vasi> cirdan:
18:42 < vasi> oh good
18:42 < vasi> let's hear
18:42 < cirdan> moved fink call to builddep and had libncurses5-shlibs as a essential split, the parent is the dev
18:42 < cirdan> all is well
18:43 < cirdan> drm said the fink call is for dpkg upgrades, i guess to install the latest version first
18:43 < cirdan> really, it's not needed anymore
18:43 < cirdan> if we require a certain version of someting, we make a versioned dep
18:43 < cirdan> (according to me)
18:44 < vasi> ok, great :-)
18:44 < vasi> i'll go backport the patch
18:45 < cirdan> vasi: now that that's outta the way...
18:45 < cirdan> dist-upgrade!
18:45 < cirdan> :-)
18:45 < vasi> hehe
18:45 < cirdan> you laugh
18:46 < cirdan> i want somethign at least in 0.25
18:46 < cirdan> current branch works great, but there are some features i'd like to make it more robust
18:46 < cirdan> and i dunno how to do it on my own
18:47 < vasi> ok, i'll put it on the blockers list
18:48 < cirdan> basic featues are there, but there are some things that would be nice/need to be done by the time we release 10.4 tree
18:52 < vasi> cirdan, 10.4 is kinda soon
18:52 < vasi> probably before 0.25
18:53 < cirdan> vasi: the *whole* point of dist-updrade is to...
18:53 < cirdan> upgrade dists
18:53 < vasi> i know, but drm wants 10.4 out *soon*
18:53 < cirdan> yeah, and we need dist-upgrade so ppl don't get fscked up when mixing trees
18:54 < cirdan> current code can at least do that
18:57 -!- protoplasm [~protoplas@c-24-18-61-81.hsd1.wa.comcast.net] has quit ["Leaving"]
18:58 -!- Murr [~neeri@A17-202-20-71.apple.com] has quit [Read error: 54 (Connection reset by peer)]
18:59 -!- Murr [~neeri@A17-202-20-71.apple.com] has joined #fink
19:26 -!- dsias [~dsias@adsl-068-153-207-210.sip.bct.bellsouth.net] has quit [Read error: 113 (No route to host)]
19:29 -!- zizban [~zizban@24-52-0-219.sbtnvt.adelphia.net] has joined #fink
19:34 -!- nkuttler [~nkuttler@dsl-084-058-194-071.arcor-ip.net] has joined #fink
19:51 -!- nkuttler [~nkuttler@dsl-084-058-194-071.arcor-ip.net] has quit ["leaving"]
19:51 -!- dmacks [~dmacks@netspace.org] has joined #fink
19:56 < dmacks> vasi: I got a huge (approx one per package or .info, not sure though) stream of "use of undef in < comparison" line seven-hundred-something while doing the first full-indexing after injecting the latest HEAD (last injection was a few days ago)
19:57 -!- dmacks [~dmacks@dmacks.active.supporter.pdpc] has quit ["leaving"]
20:26 < cirdan> heh
20:28 -!- Feanor_ [~astrange@feanor.developer.opendarwin] has quit [Read error: 110 (Connection timed out)]
20:33 -!- akh [~akhansen@68-118-244-23.dhcp.oxfr.ma.charter.com] has joined #fink
20:42 -!- Murr [~neeri@A17-202-20-71.apple.com] has quit [Read error: 110 (Connection timed out)]
20:44 -!- akh [~akhansen@68-118-244-23.dhcp.oxfr.ma.charter.com] has quit []
20:51 -!- driftkop [~driftkop@user-0c8hrip.cable.mindspring.com] has joined #fink
20:57 -!- driftkop [~driftkop@user-0c8hrip.cable.mindspring.com] has quit ["Client exiting"]
21:16 -!- dsias [~dsias@69-167-113-11.atlsfl.adelphia.net] has joined #fink
21:37 -!- hennker [flullup@dsl-213-023-251-011.arcor-ip.net] has quit ["leaving"]
22:01 -!- drm [~drm@65.210.36.227] has joined #fink
22:04 < drm> hmmm...vasi, can you explain this 'essential fix' to me a bit? it frankly makes me nervous
22:05 < vasi> dmacks, i'll check it out
22:05 < vasi> drm, ok lemme explain
22:05 < drm> (dmacks isn't here)
22:05 < vasi> uh, good point :-)
22:05 < zizban> oh explain anyway :)
22:05 < vasi> ok, we currently make all non-essential packages implicitly depend on essential packages
22:06 < drm> right
22:06 < vasi> but what if a non-essential package is a splitoff of an essential package?
22:06 < drm> indeed
22:06 < vasi> then it gets mixed up
22:06 < vasi> this stops it from getting mixed up :-)
22:07 < drm> ok, but there was a reason for the previous behavior, which i'm trying to remember
22:07 < vasi> basically says if a package is itself essential, OR is built at the same time as an essential package, then it doesn't need to implicitly depend on them
22:07 < drm> has to do with upgrading essential pkgs, or maybe trying to update a non-essential splitoff
22:08 < vasi> ok i must clarify a bit
22:08 < drm> vasi: concerning the code, is that a per-package deplist that is being calculated?
22:08 < vasi> before, it *worked* if a non-essential was a splitoff of an essential
22:08 < vasi> that's what that code is for
22:08 < vasi> now, it works the other way too
22:08 < vasi> ie: an essential can be a splitoff of a nonessential
22:08 < vasi> so we don't need libgettext3-shlibs as a parent :-)
22:09 < drm> ok, has this been tested when trying to change the version number of an essential pkg?
22:09 < drm> (version or revision)
22:09 < vasi> i don't know, cirdan asked for the fix and he said he tested it
22:10 < vasi> try it :-)
22:10 < drm> before it goes into the branch that is scheduled for release soon, i'd like to hear somebody say that it has been tested in lots iof circumstances like that, including a list of how it has been tested
22:10 < drm> don't have time to try it
22:11 < drm> although that is probably more important than finishing 10.4 tree, so i might change my priorities
22:11 < vasi> btw, cirdan says he talked to you about the dpkg CompileScript calling fink
22:11 < vasi> wanna explain that to me?
22:11  * drm tried modifying this code a couple of years ago, and ended up breaking fink installations right and left
22:13 < drm> vasi: not so easy to explain... let me try to remember how/why it works
22:13 < vasi> (btw, i'm going to be gone from sunday through next monday)
22:14 < drm> what i remember clearly is the night that i tried to update dpkg, and dpkg managed to uninstall some lib it depended on 
22:14 < drm> which cause d EVERYTHING to come to a crashing halt
22:14 < drm> couldn't reinstall debs, cause i had no working dpkg!
22:14 < vasi> it uninstalled an essential package!?
22:14 < vasi> that's really screwed up
22:14 < drm> it tried to update
22:15 < drm> but during the update routine, it uninstalled the old one on the way o to installing the new one
22:15 < drm> these things are all related
22:15 < drm> vasi: ok, well, i may release 0.24.9 while you are gone, and whether i put this code in it will depend on the outcome of my testing
22:16 < vasi> alright
22:16 < drm> if somebody can clearly tell me why cirdan needs this change, i would appreciate it
22:16 < vasi> we don't *need* it
22:16 < vasi> but it's a bug
22:16 < drm> why?
22:17 < vasi> because currently, if you innocently try to make an essential package a splitoff of a non-essential
22:17 < vasi> weird things happen
22:17 < drm> i see it as an artifact of trying to glue a splitoff system onto a dependency engine after the fact 
22:17 < drm> without ever having planned for that in the first place :/
22:17 < vasi> i'm not sure our dep engine was ever "planned", per se
22:18 < vasi> it's more like it just mutated
22:18 < drm> sure, the one chrisp wrote was nicely planned
22:18 < vasi> evolutionary programming :-)
22:18 < drm> but indeed it has mutated...oh god...kill it quick!
22:18 < vasi> in a way i wish it would just break catastrophically so we'd have to get a new one
22:19 < vasi> cuz maintaining this one is ridiculous
22:19 < drm> anyway, my fault for not documenting the weirdness that is currently in the code...not even I can remember exactly what the problem was
22:19 < drm> but i will do my best to break the new code
22:19 < drm> if i can't, it stays :)
22:19 < vasi> anyway, this code should *only* effect the situation where an essential package is a splitoff of a non-essential
22:19 < vasi> and we currently have no packages like that
22:20 < drm> but cirdan wants to remove my fink line from dpkg...i think that might give us one
22:20 < vasi> *shrug* nobody has ever satisfactorily explained that to me, so i can't really know
22:21 < drm> anyway, i am letting this set the agenda for me now...sigh
22:22 < drm> too tired to tackle it tonight, i will think abuot it in the morning
22:22 < vasi> if it's really problematic right now, delay it for 0.25
22:22 -!- drm [~drm@65.210.36.227] has quit ["Leaving"]
22:22 -!- vasi [~vasi@modemcable133.147-70-69.mc.videotron.ca] has quit ["Client exiting"]
22:25 -!- mprentice [~mprentice@cpe-24-59-118-103.twcny.res.rr.com] has quit ["They ken neither the heart of a Hielandman nor the honour of a gentleman."]
22:26 -!- vasi [~vasi@modemcable133.147-70-69.mc.videotron.ca] has joined #fink
22:29 -!- dsias [~dsias@69-167-113-11.atlsfl.adelphia.net] has left #fink ["Leaving"]
22:48 -!- geewz [~gregreede@ppp122-142.static.internode.on.net] has quit ["Leaving"]
22:56 -!- zizban [~zizban@24-52-0-219.sbtnvt.adelphia.net] has quit [Remote closed the connection]
23:10 < vasi> pogma, are you awake?
23:42 -!- moike [~Mike@nat01-hardenburgh-ext.Rutgers.EDU] has joined #fink
23:43 < moike> bundle-kde-ssl has been compiling for one day
--- Log closed Sat Aug 06 00:00:07 2005