T2 IRC Log: 2005-01-30

This is the log as captured by an IRC bot in the channel. The statements are those of the individual people and might not neccessarily reflect the policy and legal rules as set forth by the T2 SDE Project.

« prev | next »

--- Log opened Sun Jan 30 00:00:46 2005
00:13 < _Ragnar_> rxr_: I'm getting '`TOOLCHAIN/localtime': File exists' after every package build;)
01:34 < Postal|Away> hey all
05:03 -!- Postal|Away [~root@ool-43562131.dyn.optonline.net] has quit [Remote closed the connection]
05:03 -!- Postal|Away [~root@ool-43562131.dyn.optonline.net] has joined #t2
05:06 < Postal|Away> Hey all
06:37 < rxr_> moin
06:37 < rxr_> _Ragnar_: I also get this - and thought it would be some local sparc64 problem
06:37 < rxr_> _Ragnar_: which architecture ?
06:40 -!- You're now known as rxr
06:56 < _Ragnar_> rxr: ix86-32-uclibc
07:01 < _Ragnar_> btw I got a build on a64 running right now
07:16 < rxr> I hope the new glibc23 package does not do too much harm on x86-64
07:16 < rxr> it is still untested
07:16 < rxr> but I will give it a try on sparc64 today
07:17 < _Ragnar_> new glibc? oh
07:17 < rxr> no 32bit libc for stuff like grub and silo build ,-)
07:17 < rxr> or other old binary only software
07:18 < _Ragnar_> ooh
07:18 < _Ragnar_> btw do you recommend to enable gcc multilib?
07:23 < rxr> ah - yes
07:23 < rxr> you should
07:23 < _Ragnar_> kay
07:23 < rxr> but I'll remove the option and hardwire it into the architecture config so you get it when you need it - and do not need to think about it ...
07:24 < _Ragnar_> ^_^ good
09:04 < CIA-8> rene * r5850 /trunk/architecture/ (sparc64/gcc-options x86-64/gcc-options):
09:04 < CIA-8> * fixed sparc64/gcc-options to use the correct ROCKCFG variable
09:04 < CIA-8> * fixed x86-64 to use GCC_WRAPPER_* and not the (to be removed) GCC3_*
09:06 < _Ragnar_> :)
09:06 < rxr> ^- commit from another sparc64 box ... ;-)
09:06 < rxr> soooo koool to have a working svn ;-)
09:06 < rxr> (and ping *g*)
09:07 < _Ragnar_> *lol*
09:18 -!- rxr_ [~rene@p213.54.209.198.tisdip.tiscali.de] has joined #t2
09:18 -!- Topic for #t2: T2 | 2.1.0-beta2 RELEASED | The next generation of System Development Enviroments (SDE) | http://www.t2-project.org/ | CIA, don't ever leave us again!
09:18 -!- Topic set by mnemoc [] [Wed Jan 12 03:04:43 2005]
09:18 [Users #t2]
09:18 [ _Ragnar_] [ daja77] [ nzg ] [ Postal|Away] [ rxr ] [ valentin]
09:18 [ CIA-8 ] [ jsaw ] [ ooo_] [ praenti ] [ rxr_]
09:18 -!- Irssi: #t2: Total of 11 nicks [0 ops, 0 halfops, 0 voices, 11 normal]
09:18 -!- [freenode-info] please register your nickname...don't forget to auto-identify! http://freenode.net/faq.shtml#nicksetup
09:18 -!- Channel #t2 created Sun Aug 8 21:15:33 2004
09:19 -!- Irssi: Join to #t2 was synced in 11 secs
09:21 -!- rxr [~rene@p213.54.197.164.tisdip.tiscali.de] has quit [Nick collision from services.]
09:21 -!- You're now known as rxr
09:26 < _Ragnar_> and xorg still doesn't compile *grr*
09:26 < _Ragnar_> (for uclibc)
09:28 < rxr> :-(
09:29 < rxr> I bet you'll soon have it working ...
09:29 < rxr> _Ragnar_: btw commit anytime soon ?
09:30 < _Ragnar_> yeah
09:31 < _Ragnar_> should I commit each of the changes on its own, or all in one?
09:34 < rxr> well - depending if you want to cheat in this graph or not:
09:34 < rxr> http://svn.exactcode.de/
09:34 < _Ragnar_> how do I do each anyway?
09:34 < rxr> I think all normal patches in one chunk should be fine ...
09:34 < rxr> svn commit one-file-to-commit
09:35 < _Ragnar_> then the log message?
09:36 < rxr> svn will fire up the $EDITOR
09:36 < _Ragnar_> yeah
09:36 < rxr> http://www.t2-project.org/svn-commit-tutorial.html
09:37 < rxr> please use this layout for the message:
09:37 < rxr> * the changes ...
09:37 < rxr> * the next change you have done ...
09:37 < rxr> when you have more than one change per message
09:39 < _Ragnar_> okay
09:39 < rxr> if you need inspiration:
09:39 < rxr> http://svn.exactcode.de/ChangeLog-t2
09:39 < _Ragnar_> ;)
09:39 < rxr> ^- automatically prettified from the svn log ...
09:39 < _Ragnar_> how do I tell it to not try the commit as root?:)
09:39 < rxr> hit enter at the passwd promt
09:39 < rxr> it will ask for the user then
09:40 < _Ragnar_> ah okay
09:40 < _Ragnar_> ;) done
09:40 < CIA-8> ragnar * r5851 /trunk/package/x11/xorg/xorg.conf: * added calls to prepatch...postmake to xorg build script
09:41 < rxr> svn status is a nice help to check if forgotten files (read not added) are lingering around
09:46 < CIA-8> ragnar * r5852 /trunk/package/base/uclibc/add_sinf_cosf.patch:
09:46 < CIA-8> * uclibc lacks sinf/cosf calls which some programs (notably xorg) need.
09:46 < CIA-8> Add wrapper functions to call respective sin/cos(double).
09:50 < _Ragnar_> can I do svn commit package/base/uclibc/patches/* ?
09:54 < rxr> svn commit package/base/uclibc/patches/
09:54 < rxr> ?
09:54 < rxr> possibly with a
09:54 < rxr> svn add package/base/uclibc/patches/*
09:54 < rxr> before ...
09:55 < _Ragnar_> okay;)
09:56 < CIA-8> ragnar * r5853 /trunk/package/base/uclibc/uclibc.conf: * Turn on some more features in uclibc to make more packages happy.
09:59 < CIA-8> ragnar * r5854 /trunk/package/base/uclibc/parse-config-9:
09:59 < CIA-8> * Add linux-gnu -> linux-uclibc sed fixes for some more packages.
09:59 < CIA-8> * prevent shared file conflict about /usr/share/locale/locale.alias
10:00 < CIA-8> ragnar * r5855 /trunk/package/base/uclibc/patches/ (34 files): * add/modify patches for packages to compile with uclibc
10:01 < _Ragnar_> okay so far :)
10:02 < rxr> ;-) cool
10:57 < _Ragnar_> okay I'm off to bed
10:57 < _Ragnar_> goodnight
11:26 < rxr> sleep well _Ragnar_
13:04 < rxr> cu later
15:17 -!- mnemoc [~amery@200.75.27.8] has joined #t2
15:27 < rxr> mnemoc: ?
15:33 -!- Postal|Away [~root@ool-43562131.dyn.optonline.net] has quit [Read error: 110 (Connection timed out)]
16:15 -!- ooo [~o@dsl-olugw3p33.dial.inet.fi] has joined #t2
16:15 -!- ooo_ [~o@dsl-olugw3p33.dial.inet.fi] has quit [Read error: 54 (Connection reset by peer)]
16:36 < CIA-8> rene * r5856 /tags/2.1.0-beta3/package/graphic/libmng/libmng.desc: * updated libmng (1.0.5 -> 1.0.9)
16:41 < CIA-8> rene * r5857 /trunk/package/graphic/libmng/libmng.desc: * merged r5856 from tags/2.1.0-beta3: updated libmng (1.0.5 -> 1.0.9)
17:28 -!- maddin [~martin@pD9E724DD.dip.t-dialin.net] has joined #t2
17:28 < maddin> hi again
17:37 < rxr> hi maddin
17:37 < CIA-8> rene * r5858 /tags/2.1.0-beta3/package/base/dietlibc/dietlibc.conf:
17:37 < CIA-8> * fixed dietlibc for powerpc (well, ...)
17:37 < CIA-8> * implemented native (hostlibc) diet - untested - but soon will
17:44 < rxr> GNU GRUB 0.96
17:44 < CIA-8> rene * r5859 /trunk/package/graphic/inkscape/inkscape.desc:
17:44 < CIA-8> * updated inkscape (0.40 -> 0.40-20050130) - this is an arbitrary
17:44 < CIA-8> snapshot but it honours the hidden state of layers while exporting
17:44 < CIA-8> ;-) However it has one new visible oddity and regressions for one
17:44 < CIA-8> of my script usages ... :-(
17:46 < CIA-8> rene * r5860 /trunk/package/x86/grub/grub.desc: * updated grub (0.95 -> 0.96)
17:55 < maddin> hi rxr
18:02 < rxr> hi mnemoc, how are you ?
18:08 < jsaw> re
18:08 < rxr> hi jsaw !!!
18:08 < jsaw> hi rxr :)
18:09 < jsaw> mnemoc was not here, still?
18:10 < rxr> no - and I even sent him a "All ok?" mail with no response yet ... :-(((
18:11 < jsaw> hope that we misunderstood him and he meant this weekend...
18:11 < rxr> my mail was sent "01/29/2005 10:33 AM"
18:11 < jsaw> guess I should have committed my nvidia work before I left...
18:11 < rxr> yes!
18:12 < jsaw> damn...
18:12 < rxr> release often release early ;-)
18:13 < rxr> how war your weekend ?
18:13 < jsaw> was at my mom's place with marlene... I worked, they played ;)
18:16 < rxr> ;-)
18:17 < jsaw> only one minus (nvidia): patches do not apply to the ia64 package...
18:18 < rxr> .oO
18:18 < rxr> jsaw: have fun implementing some more patch apply code then ...
18:18 < rxr> jsaw: you have ia64 access now ?
18:22 < jsaw> we have a fresh altix machine with 8 ia64 on it running redhat/sgi.
18:23 < jsaw> 8Gig RAM...
18:23 < jsaw> nice.
18:23 < rxr> and you build t2 on it ?
18:23 < jsaw> sadly no root access...
18:23 < jsaw> It's a production (compute) server...
18:24 < jsaw> Once I finished my PhD I'll ask if I may compile T2 on it...
18:24 < valentin> hi - everybody votes against replacing current graphviz maintainer (nazgul) by valentin ?
18:24 < valentin> s/everybody/anybody
18:25 < rxr> no - go ahead
18:33 < jsaw> dinner - cu later
18:33 < rxr> cu
18:35 < CIA-8> rene * r5861 /tags/2.1.0-beta3/package/base/dietlibc/hostlibc.diff: * one should better not forget to svn add files: dietlibc/hostlibc.diff
18:40 < nzg> valentin: do it
18:41 < nzg> i have no time to maintain it atm
18:41 < rxr> oh - nazgul was you?
18:41 < nzg> yep
18:41 < rxr> .oO
18:42 < rxr> sorry :-(
18:42 < rxr> did not match in my brain ... :-(((
18:42 < nzg> what for?
18:42 < rxr> nazgul ...
18:42 < maddin> hi, I need a little help with grep
18:42 < rxr> if you actively use it the maintainer must of course not be changed
18:42 < rxr> maddin: go ahead ;-)
18:43 < maddin> I will grep to *.zip *.pdf and more.
18:43 < nzg> i will use it again when i get the time to set up my laptop....
18:44 < maddin> rxr, there are two new hentries: http://stunk.tfh-berlin.de/apfelmus/
18:44 < rxr> maddin: grep some file for *.pdf and co ?
18:45 < maddin> grep String1|String2|String3 |=OR
18:46 < maddin> I know, I can use [ABC] for characters, but what can I use for strings?
18:50 < rxr> sorry - read your site ...
18:51 < maddin> It's okay. I have some minutes ...
18:52 < rxr> well - | should wok
18:52 < rxr> the quick way is grep -e string1 -e string2 -e string3
18:52 < rxr> I lookup how the other syntax was
18:52 < rxr> what do you need this for ?
18:52 < rxr> syp?
18:54 < maddin> No, for a klausuren(?)-CD
18:54 < rxr> exams
19:03 < CIA-8> valentin * r5862 /trunk/package/scientific/graphviz/graphviz.desc: * updated graphviz (1.16 -> 2.2)
19:03 < valentin> maddin: use "abc\|def"
19:04 < rxr> huh
19:04 < valentin> huh what ?
19:04 < rxr> why does it need quoting ? of the pipe ?
19:04 < valentin> because | and + have to be backslashed in grep
19:04 < rxr> annoying ..
19:05 < valentin> whereas . and * can be used without
19:05 < rxr> just read the info page and did not found why my | did not work out as I remembered
19:06 < valentin> the most annoying thing is: every program supporting some kind of regexp has other oddities in the syntax
19:06 < rxr> yes
19:06 < rxr> screams for a ExactOS
19:06 < rxr> unified regex everywhere - and nothing else ..
19:07 < rxr> except maybe file globbing - but maybe it is better replaced by full regex anyway
19:07 < rxr> some days++ ago I also tried s.th. with file globbing that was not possible
19:07 < valentin> you mean wildcards like "*.txt
19:07 < valentin> " ?
19:07 < rxr> yep
19:07 < maddin> hi valentin, thanks. It works
19:08 < valentin> rxr: a problem with filenames is the common use of the '.'
19:08 < valentin> so you just cannot use '.' as a character match
19:08 < rxr> wel - get common to quote it ;-)
19:09 < rxr> +l
19:09 < valentin> lol
19:10 * rxr reading diff linux-2.6.8.1 linux-2.6.9
19:10 < valentin> the diff or the changelog ?
19:10 < rxr> between those swapping on sparc64 stopped to work :-(
19:10 < rxr> diff
19:11 < rxr> the f*cking sparc does not seem to boot without keyboard
19:11 < rxr> at least the reboot did not brought my university sparc back online ...
19:11 < rxr> maddin: btw. I heared you have take a u5 home ?
19:13 < rxr> taken of course ...
19:13 < CIA-8> rene * r5863 /tags/2.1.0-beta3/package/base/dietlibc/dietlibc.conf: * fixed the hostlibc diet
19:15 < maddin> rxr: Yes, for a long long time.
19:16 < CIA-8> rene * r5864 /trunk/package/base/dietlibc/ (dietlibc.conf hostlibc.diff):
19:16 < CIA-8> * merged r5858:HEAD dietlibc changes from tags/2.1.0-beta3 into trunk:
19:16 < CIA-8> * powerpc fixes
19:16 < CIA-8> * hostlibc diet build and fixes
19:19 < rxr> long long ?
19:19 < rxr> I miss one for two weeks or so ...
19:20 < rxr> how much RAM is plugged in your one?
19:20 < rxr> do you actually have it running at all ?
19:23 < maddin> Dont' know, how much RAM is in it. I have it since 2 month or so.
19:23 < rxr> hm
19:23 < rxr> how long has this "ramin" (or so) the sparc?
19:24 < rxr> two weeks ago one vanished - and last week sven concluded it could be the one you have - but this does not seem to be the case ...
19:26 < maddin> I don't know, who has take the u5 for two weeks.
19:26 < rxr> actually I would not care if it would not be the one where sven placed all the RAM into - and now there are two (or three?) without RAM at all ..
19:27 < maddin> Must I stand up now and look in it?
19:27 < rxr> no
19:28 < rxr> it is not yours ...
19:28 < maddin> ok
19:28 < rxr> I just wanted to express my frust about that ...
19:28 < rxr> maddin: do you run s.th. on it ? FreeBSD I guess ?
19:29 < maddin> yes, but i will have t2. but i have no time ...
19:29 < rxr> soon are holidays ;-)
19:29 < maddin> yes :-)
19:30 < rxr> .oO
19:30 < rxr> forgot that the just running kernel on the sparc has this scsi defect ...
19:30 < rxr> no wonder the HD is crunching like crazzy ...
19:30 < maddin> What are good/needed options to mkisofs for create a normal ISO-File from a dir?
19:30 < rxr> -jt
19:30 < maddin> :-(
19:30 < rxr> err
19:30 < rxr> -jr
19:32 < maddin> This enough? I have a example: mkisofs -v -J -r -d -D -f -l -L -N -o /tmp/xxx.iso /dir/path/
19:32 < rxr> jaja - no
19:32 < maddin> Okay, I will read the manpage
19:32 < rxr> haha even
19:32 < rxr> -Jr is good ;-)
19:32 < rxr> never used more ...
19:32 < maddin> okay :-)
19:33 < maddin> -l Allow full 31 character filenames. ? I need this ...
19:34 < rxr> ignore the rest ...
19:34 < rxr> fuck
19:34 < rxr> damn crap
19:34 < rxr> why does the whole linux kernel change between each release
19:34 < rxr> just reverting the mm stuff to see it this braks sparc64 is and endless task
19:34 < rxr> due to API changes all over the place ...
19:35 < maddin> I see, we need a lovely bot for rxr :-)
19:36 < rxr> bot ?
19:37 < rxr> what for? fixing the ugly linux kernel ?
19:39 < maddin> No, to give you the right answers. One, how understand you every time.
19:45 -!- ooo_ [~o@dsl-olugw3p33.dial.inet.fi] has joined #t2
19:45 -!- ooo [~o@dsl-olugw3p33.dial.inet.fi] has quit [Read error: 104 (Connection reset by peer)]
19:45 < rxr> ?
19:46 < _Ragnar_> moin
19:53 < rxr> moin!
19:55 < rxr> maybe I'm soon into bed ...
19:59 < _Ragnar_> ;)
19:59 < _Ragnar_> xorg builds!:)
20:01 < rxr> congrats!
20:01 < _Ragnar_> =)
20:01 < rxr> big patch ?
20:02 < _Ragnar_> no
20:02 < _Ragnar_> already included with the last ones
20:04 < _Ragnar_> but breakfast first :) bbl
20:08 < rxr> only the xorg.conf ?
20:08 < _Ragnar_> yes
20:09 < _Ragnar_> and package/base/uclibc/add_sinf_cosf.patch
20:20 < rxr> the interesting part is whether it also runs ;-)
20:23 < rxr> ok - n8 all
20:24 < maddin> gn8 rxr
20:34 -!- maddin [~martin@pD9E724DD.dip.t-dialin.net] has quit ["Verlassend"]
20:48 < _Ragnar_> gn8 rxr ;)
21:21 * mnemoc finally got internet connection again!!!!
21:21 < mnemoc> lot of email to read :\
21:21 < _Ragnar_> w00t ;) wb
21:21 < mnemoc> hi _Ragnar_
21:23 < valentin> hi mnemoc
21:23 < mnemoc> hi valentin
21:24 < valentin> rxr feared you were gone forever ...
21:27 < mnemoc> i was near ;)
21:27 < _Ragnar_> what happened?
21:27 < mnemoc> money :\
21:27 < _Ragnar_> oh :/
21:27 < mnemoc> banks and related crap
21:28 < mnemoc> but health is good :)
21:29 < _Ragnar_> good :)
21:32 < valentin> .oO
21:32 < mnemoc> "since 2.1.0-beta3 got branched off and we should start..." it seems i was dead for too long
21:33 < _Ragnar_> *g*
21:33 < mnemoc> brb
21:38 < mnemoc> re
21:52 < jsaw> wee, mnemoc's back...
21:55 < mnemoc> :D
21:59 < jsaw> hi all
21:59 < mnemoc> hi jsaw
21:59 < jsaw> hola mnemoc
22:01 < jsaw> relocating self to lab, cu in a few seconds.
22:07 < CIA-8> mnemoc * r5865 /trunk/package/network/openldap/openldap.desc: * updated openldap (2.2.20 -> 2.2.23)
22:07 < mnemoc> :)
22:13 < CIA-8> mnemoc * r5866 /trunk/package/security/keepalived/keepalived.desc: * updated keepalived (1.1.7 -> 1.1.8)
22:16 < jsaw> re
22:17 < jsaw> what would I have to do if I want to create a cross compiler, say for hc12... in other words, what do you think would be the best approach, create a gcc-cross package eg.?
22:21 < mnemoc> i miss multi-copy packages from loooong time ago
22:23 < _Ragnar_> multi-copy packages?
22:24 < _Ragnar_> btw shouldn't we just fix ./*/config.sub and ./configure and ./*/configure in general for uclibc?
22:24 < mnemoc> uhm
22:25 < mnemoc> i'm not sure how safe can that be
22:25 < _Ragnar_> do you know one package where it breaks?
22:25 < mnemoc> by multi-copy i mean to be able to build and install different copies of the 'same' package, in this case gcc, built with diferent options
22:26 < jsaw> *rxr*: ./scripts/Create-ErrList -newdelete
22:26 < jsaw> ./scripts/Create-ErrList: line 1: pkgcksum: command not found
22:26 < mnemoc> oh
22:26 < mnemoc> _Ragnar_: have you commited or send your uclibc patches these days?
22:26 < _Ragnar_> oh okay
22:26 < _Ragnar_> yes
22:27 < mnemoc> commit or send?
22:27 < _Ragnar_> commited
22:27 < mnemoc> :D
22:28 < mnemoc> some configure are not autoconf configures
22:28 < _Ragnar_> xorg builds now, too
22:28 < mnemoc> xorg with uclibc??
22:28 * _Ragnar_ nods
22:28 * mnemoc falls down
22:28 < CIA-8> jsaw * r5867 /trunk/scripts/Create-ErrList: * fix typo (pkgc_h_ksum)
22:28 < jsaw> hehe, mnemoc missed quite some good stuff
22:28 < _Ragnar_> if they're not autoconf, chances are there's no linux-gnu in it, too
22:28 < jsaw> hi _Ragnar_
22:28 < _Ragnar_> hey jsaw ;)
22:29 < mnemoc> _Ragnar_: true
22:29 < mnemoc> jsaw: i noticed
22:29 < jsaw> we need multi-copy back.... hmmm.
22:30 < jsaw> I need it if I want to succeed with the DistCC-LiveCD...
22:30 < mnemoc> back... back what?
22:30 < mnemoc> afaik we have never had multi-copy, did we?
22:30 < jsaw> don't know exactly when rxr branched.
22:31 < mnemoc> you mean 'pkgfork'?
22:31 < jsaw> <- limited memory.. things not used get lost after a month or less in my brain
22:31 < jsaw> mnemoc: yep
22:32 < _Ragnar_> for gcc: doesn't that already support building in different locations for diff. architectures?
22:32 < jsaw> gcc itself yes.
22:32 < mnemoc> jsaw: pjgfork means is there for different versions of the same package, example gcc2.95,gcc3.3,gcc3.4,gcc-3.5
22:32 < jsaw> oh, that only.
22:33 < mnemoc> jsaw: but not different options like gcc-native, gcc-${some_cross_arch}, ...
22:34 < mnemoc> or linux26-udev and linux26-devfs
22:35 < jsaw> linux26-rtai, etc.
22:35 < jsaw> yep.
22:35 < mnemoc> :)
22:35 < mnemoc> i doubt config/$config/packages looping makes that possible
22:37 < jsaw> What about a special package naming scheme, like "package/base/gcc-%", for a package that supports conditioned copies?
22:37 < mnemoc> _Ragnar_: don't you want to start a uclibc-reference build with a more wider auto_patcher?
22:38 < _Ragnar_> mnemoc: I'm currently preparing for it :)
22:38 < mnemoc> jsaw: rock use '=' as delimited
22:38 < mnemoc> _Ragnar_: :D
22:38 < jsaw> mnemoc: I meant in the file name already.
22:39 < mnemoc> jsaw: package that supports foo has foo flag in current schema
22:39 < jsaw> the gcc=gcc-sth. I know. Here, I meant, gcc-hc12 would match the package "gcc-%" but not the original "gcc" one
22:39 < jsaw> ? foo flag ?
22:39 < mnemoc> uhm
22:39 < mnemoc> .oO
22:40 < mnemoc> how to configure a build to include gcc-hc12?
22:40 < mnemoc> how to expand cofnig/$config/packages accordingly
22:40 < jsaw> everything but "target=hc12-unknown-none" unchanged
22:41 < jsaw> oh, that u mean.
22:41 < jsaw> unsure.
22:42 < jsaw> With the "-%" extension, matching would be quite easy imo.
22:42 < mnemoc> $config/packages may get expanded accordingly using pkgsel and/or a pkg${clone} command but i'm not use how to menu that on Config
22:43 < mnemoc> how can user know what $pkg support for that '%'
22:43 < jsaw> .oO(no config - expert wisdom)
22:44 < _Ragnar_> *lol*
22:45 < jsaw> hehe
22:45 < mnemoc> .oO
22:46 < mnemoc> .oO
22:46 < jsaw> real answer: no idea :(
22:48 < mnemoc> i guess i can do Config stuff if you do '%' processing :)
22:48 < jsaw> good deal!
22:49 < jsaw> brb (small chat with colleague)
22:49 < mnemoc> parsing $pkg on .conf is easy
22:50 < mnemoc> now i only doubt about how to implement $pkgfork's .desc conditionality [V]s and [D]s
22:53 -!- freddy [~junix@pD9E76D36.dip.t-dialin.net] has joined #t2
22:53 < freddy> hi
22:57 < mnemoc> hi freddy
22:57 < freddy> hi mnemoc
22:58 < mnemoc> jsaw, rxr, _Ragnar_: i'll change LIBC's USE_IT with a single ROCKCFG_LIBC= option
22:59 < mnemoc> `-- comments?
22:59 < _Ragnar_> which libc are you talking about?
23:00 < jsaw> hi freddy
23:00 -!- praenti [~praenti@82.139.198.99] has quit [Read error: 104 (Connection reset by peer)]
23:00 < mnemoc> _Ragnar_: bool ROCKCFG_{UCLIBC,DIETLIBC}_USE_IT ---> ROCKCFG_LIBC={glibc,uclibc,dietlibc}
23:00 -!- praenti [~praenti@82.139.198.99] has joined #t2
23:01 < jsaw> mnemoc: good simplification
23:01 < freddy> hi jsaw
23:01 < _Ragnar_> dietlibc isn't going to work for most of the system afaik
23:01 < mnemoc> but dropping it is not good PR
23:01 < mnemoc> rene loves dietlibc too
23:01 < _Ragnar_> no one talks about dropping it
23:01 < daja77> hi mnemoc
23:01 < mnemoc> hi daja77
23:02 < _Ragnar_> just leave it as an addon
23:02 < jsaw> hi daja77
23:02 < daja77> hi jsaw
23:03 < jsaw> mnemoc: but ain't the semantics a bit messed right now? USEIT for dietlibc also handles the cases where some progs are linked to while the system libc is still glibc, or?
23:03 < mnemoc> _Ragnar_: rene even rejected to have an official target based on 'a dirty and unclearly-licensed' package like uclibc
23:03 < _Ragnar_> *lol*
23:03 < _Ragnar_> uclibc isn't dirty
23:03 < mnemoc> it is to rene
23:04 < jsaw> hmm. technicians...
23:04 < _Ragnar_> what does he think about glibc then?
23:04 < jsaw> don't ask ...
23:04 < mnemoc> :)
23:04 < daja77> gpl
23:04 < _Ragnar_> *hrhr*
23:06 < _Ragnar_> anyway, uclibc is gnu licensed
23:06 < mnemoc> jsaw: i'll see what can i do to keep dietlibc working exactly as now with the new var
23:06 < mnemoc> _Ragnar_: he dislikes the copy&paste done on uclibc
23:07 < jsaw> _Ragnar_: http://dl.exactcode.de/irclogs/t2/t2-2005-01-25.html
23:09 < _Ragnar_> jsaw: yes, I know ;)
23:09 < jsaw> mnemoc: so change it
23:09 < _Ragnar_> at least uclibc is a smaller target, and most patches to packages are quite superficial
23:14 < mnemoc> 118 more commits to read :\
23:14 < _Ragnar_> =)
23:14 < _Ragnar_> were there so many?
23:16 < mnemoc> yep
23:17 < jsaw> mnemoc: the pkgfork code seems to exist in t2...
23:18 < mnemoc> it has been removed by parts
23:18 < jsaw> by parts... ic
23:19 < mnemoc> is rene building beta-3 isos?
23:26 < jsaw> he's still merging fixes to beta3.
23:27 < mnemoc> nice
23:30 < mnemoc> jsaw: it's not hard to use ${pkg}-% to set well known options to packages, but what do you think about different versions? ($pkg=$xpkg sucks)
23:37 < jsaw> this would also match %. E.g. like the target thing (machine-host-os-etc...), we could use condition(-version)?
23:38 < jsaw> otoh maybe that's too much. Just think about gcc-(machine)-(version)... how many [D] tags u could have...
23:39 < jsaw> then we would have to have conditonal .desc files...
23:42 < jsaw> I'll write a proposal to the ML a bit later.
23:44 < mnemoc> ok
23:44 < mnemoc> $pkgclone needs to know $ver when clonning cofngi/$config/packages line
23:45 < mnemoc> and i guess removing stages <3 from the clone
--- Log closed Mon Jan 31 00:00:46 2005