--- Log opened Fri Apr 29 00:00:46 2005
00:06 < valentin> sed 's,\(^\|[ \t]\)#.*$,,'
00:07 < valentin> whatever that is
00:07 < valentin> what happens with the \( ... \) part ?
00:07 < mnemoc> \1
00:07 < valentin> yes, i know, but there is no \1 in the replacement
00:10 < valentin> i think he just means normal alternative: ( .. \| .. )
00:11 < valentin> that expression looks quite drug'ed
00:12 < valentin> i think the author wanted to delete all lines beginning with # or with \t#
00:13 < mnemoc> yep
00:13 < mnemoc> 'indented' comments
00:14 < valentin> but that is not what the expression says
00:14 < valentin> it would also match this line     # <- if the spaces would be a tab !
00:14 < mnemoc> not directly but yes
00:14 < valentin> why not replacing it with
00:15 < mnemoc> i have to go to pick up my girls....
00:15 < valentin> s,[^\\]?#,,g ?
00:15 < valentin> have fun
00:15 < mnemoc> rocknet was a proof of concept from master-obscure-programming
00:17 < mnemoc> uhm... i'm late
00:17 < mnemoc> cu :)
00:20 < valentin> cu mnemoc 
00:38 -!- sparc-kly [~sparc@65-23-205-193.prtc.net] has quit [Read error: 54 (Connection reset by peer)]
00:39 -!- sparc-kly [~sparc@65-23-205-193.prtc.net] has joined #t2
01:13 < mnemoc> re
01:15 < sparc-kly> hi mnemoc 
01:15 < mnemoc> hi sparc-kly 
01:24 -!- ojh [~omer@71-34-251-181.eugn.qwest.net] has joined #t2
01:25 -!- ojh [~omer@71-34-251-181.eugn.qwest.net] has left #t2 []
01:55 -!- sparc-kly [~sparc@65-23-205-193.prtc.net] has quit ["Leaving"]
01:58 -!- sparc-kly [~sparc@65-23-205-193.prtc.net] has joined #t2
02:04 -!- keinek [~Keinek@201.254.12.224] has quit [Read error: 145 (Connection timed out)]
03:25 < CIA-9> amery * r8566 /branches/2.1/package/java/jboss-as/jboss-as.desc: * changed jboss-as to optional because it depends on a optional package, sun-jdk-150.
03:27 < CIA-9> amery * r8567 /branches/2.1/package/java/jboss-as/jboss-as.conf: * cleaned jboss-as, removed hardcoded version inside .conf
03:37 < CIA-9> amery * r8568 /branches/2.1/package/java/apache-ant/apache-ant.desc: * repriorized apache-ant after our JDKs
04:36 -!- keinek [~Keinek@201.254.8.139] has joined #t2
04:36 < keinek> hi again
04:36 < keinek> i down for hours
04:36 < keinek> problems with dns
04:36 < keinek> i change dns and now be here...
04:36 < keinek> is posible put three dns in resolv?
04:37 < keinek> or only use the firsrt two nameserver?
04:38 < mnemoc> it tries all avaialble
04:39 < keinek> good
04:39 < keinek> so i can add more dns?
04:39 < mnemoc> nameserver, yes
04:39 < keinek> hours offline... always i use one dns.
04:39 < keinek> now i have three
04:39 < keinek> :)
04:40 < mnemoc> :)
05:01 -!- Netsplit tolkien.freenode.net <-> irc.freenode.net quits: jsaw
05:03 -!- Netsplit over, joins: jsaw
07:05 -!- mipe [~mika@dsl10040.japo.fi] has joined #t2
07:05 < CIA-9> amery * r8569 /trunk/ (9 files in 8 dirs): (log message trimmed)
07:05 < CIA-9> * merged 8559 from 2.1 - removed empty patch for dnrd
07:05 < CIA-9> * merged 8560 from 2.1 - fixed sge by not applying a .patch-x removed long time ago
07:05 < CIA-9> * merged 8561 from 2.1 - repriorized gtk-sharp, as r8556 did for gtk-sharp2, to really fix mono-doc.
07:05 < CIA-9> * merged 8564 from 2.1 - improved Build-Tools to not rebuild because of a svn-commit.tmp file
07:05 < CIA-9> * merged 8565 from 2.1 - removed debugging flag from nss_ldap.conf
07:05 < CIA-9> * merged 8563,8568 from 2.1 - repriorized fyre and apache-ant after our JDKs
07:54 < mnemoc> rxr: http://lists.ximian.com/archives/public/openoffice/2005-April/000914.html <--- my problem with ooo, and i guess my problem with many others which don't find libraries. he blames pkgconfig>=0.17   i tried 0.17.2 and 0.17.1 both failing for inkscape... what do you have locally where you got OOo build?
08:04 < rxr> re
08:05 < rxr> no idea - some 0.16 ?
08:05 < rxr> MOIN
08:06 < mnemoc> moin :)
08:06 < mnemoc> i just post on ML about it
08:06 < mnemoc> asking about downdating to 0.16
08:06 < rxr> then is relly fixes your stuff ...
08:06 < rxr> but only if tested
08:07 < mnemoc> http://t2.geeks.cl/regressions/stable/regressions.html
08:07 < mnemoc> i'll force a rsync because it finished
08:09 < mnemoc> OT: can you explain me package/network/psi/psi.conf:25, pleasE?
08:10 < mnemoc> resync'ed
08:16 < rxr> moment
08:16 < CIA-9> amery * r8570 /branches/2.1/package/security/ (5 files in 2 dirs): 
08:16 < CIA-9> * downdated libchipcard (1.9.12 -> 0.9.1) because some packages depend on it
08:16 < CIA-9> * added libchipcard2 (1.9.12beta) for the new chipcard2 needed by some reciently updated packages
08:16 < rxr> hm - interseting codeline
08:16 < rxr> what is your question ?
08:17 < rxr> I guess it installs needed data files ...
08:17 < mnemoc> where?
08:18 < rxr> psi.conf:25,
08:18 < mnemoc> yes yes
08:18 < rxr> what is your exact wondering ?
08:18 < mnemoc> but where is it trying to install each .qm file?
08:19 < rxr> does it work or error out ?
08:19 < mnemoc> error
08:20 < mnemoc> i would never noticed it without an error :)
08:20 < rxr> well due to the quoting the cp get's the wildcards ...
08:20 < rxr> I never saw this and don't think cp will handle those ..
08:20 < rxr> so I assume it never worked ?
08:20 < rxr> does the ROCK tree still contain this line ?
08:21 < rxr> it is not marked with error within the Jun 17 02:20:08 2004 .cache fiel ...
08:21 < mnemoc> i don't have a copy of rock...
08:21 < mnemoc> uhm
08:21 < rxr> I usually visit the rep. with my favourite browser to spy such stuff ..
08:21 < rxr> well - just make it a match_source_file loop into it to fix it ...
08:22 < mnemoc> http://t2.geeks.cl/regressions/stable/diff/psi.diff
08:22 < mnemoc> but is psi_#1.qm a dir?
08:22 < mnemoc> i don't know how to make it work because i don't understand what it tries to achieve
08:24 < mnemoc> http://www.rocklinux.net/svn/rock-linux/trunk/package/stefanp/psi/psi.conf
08:24 < mnemoc> now i do understand it
08:25 < mnemoc> for file in $desc_D ; do ???
08:25 < mnemoc> wasn't desc_D the whole [D] line?
08:26 < rxr> I think when you "iterate" this way you get all fields seperated
08:26 < rxr> that is
08:26 < rxr> chksum
08:26 < rxr> file
08:26 < rxr> url
08:26 < rxr> chksum
08:26 < rxr> file
08:26 < rxr> url
08:26 < rxr> ...
08:26 < rxr> use our match_source_file and all is good
08:26 < mnemoc> aha
08:27 < rxr> whe you are at it
08:27 < rxr> it is SDECFG_PKG_KDE3_CORE_PREFIX 
08:27 < rxr> not 32 as in that file ...
08:28 < rxr> but pkgprefix of kdelibs would be even better ...
08:28 < rxr> so - /me getting rolls to make susan happy ...
08:28 < rxr> cu
08:28 < rxr> after breakfast ...
08:28 < rxr> Error logs from default-2.2.0-alpha-sparc-v8-desktop:
08:28 < rxr> 649 builds total, 168 completed fine, 0 with errors.
08:28 < rxr> == 08:41:09 =[5]=> Building base/slang [1.4.9 2.2.0-alpha].
08:29 < mnemoc> :)
08:29 < rxr> the box is slow - so the remaining 4xx packges will take a lot of more time ..
08:29 < rxr> but god to see we did not caused sparc regressions
08:29 < rxr> only sparc64 is to be fixed
08:29 < rxr> I try today
08:31 < mnemoc> we need a good base of minimal-desktop isos
08:32 < rxr> yep - s.th. like that
08:32 < rxr> I think important is to get a minimal boot/install system fully cross build
08:32 < rxr> so we can deploy any arch - even those we have no hardware for
08:32 < mnemoc> ack
08:32 < rxr> == 04/29/05 08:49:09 =[5]=> Finished building package slang.
08:32 < rxr> puh 8 minutes ...
08:33 < mnemoc> minimal.iso for every known arch
08:33 < rxr> 270Mhz U5 with only 128MB RAM ...
08:33 < mnemoc> that's why cross-building is cool
08:33 < rxr> yep - that is why I promote "cross compile anything" so much
08:33 < rxr> == 08:49:23 =[5]=> Building base/mmv [1.01b 2.2.0-alpha].
08:33 < rxr> == 04/29/05 08:50:08 =[5]=> Finished building package mmv.
08:34 < rxr> so - but now rolls ...
08:36 < CIA-9> amery * r8571 /branches/2.1/package/office/aqhbci/aqhbci.desc: * updated aqhbci (1.0.7 -> 1.0.8)
08:41 < mipe> hmm,linux26 fails on stage 1,kcc error,i think this was on ML but it havent been commited to 2.1 branch?
08:42 < mnemoc> i haven't merged it
08:43 < mipe> could you?
08:44 < mnemoc> what rev?
08:45 < mipe> according to mail, its r8541
08:45 < mnemoc> oh
08:46 < mnemoc> i missed that big one
08:46 < mipe> btw. the new upcoming site looks cool, https://www2.t2-project.org
08:47 < mnemoc> jeru have had troubles with 3rd generation modules for i18n, and 2nd generation modules like the tracker
08:48 < mnemoc> but the site is get very good
08:50 < tilix> why you switched to Plone?
08:51 < mnemoc> extendability and i18n where my reasons, others had another reasons
08:51 < mnemoc> were*
08:51 < CIA-9> amery * r8572 /branches/2.1/package/ (16 files in 16 dirs): * merged 8541 from trunk - use of CC=$KCC on postlinux.conf for stage 1
08:54 < CIA-9> amery * r8573 /branches/2.1/package/network/psi/psi.conf: 
08:54 < CIA-9> * fixed psi to install language modules without errors
08:54 < CIA-9> * improved psi prefix detection (pkgprefix)
08:55 < mipe> btw. what are 2.2 goals?
08:57 < mnemoc> mipe: tools rewrites, good bootdisk, livecd and embeded targets, some graphical interfaces and graph based dependency solving mainly
08:57 < mnemoc> tools rewrites means moving most internal scripts to c++
08:58 < mipe> oh i thought that was for 3
08:58 < mipe> or was it build scripts to c++
08:59 < tilix> i could help with the hardware detection part of the livecd
08:59 < mnemoc> tilix: hotplug++?
09:00 < tilix> yes, and hwsetup/hwinfo
09:00 < mnemoc> mipe: only a few tool rewrites will go on 2.2
09:00 < mnemoc> tilix: talk with rxr and jsaw about if you want to get involved in that project
09:01 < tilix> ok
09:02 -!- mtr [~michael@p54AF88BB.dip0.t-ipconnect.de] has joined #t2
09:02 < mtr> moin
09:02 < mnemoc> mtr: hi!
09:02 < mtr> hi mnemoc 
09:02 < mnemoc> mtr: i had to branch libchipcard to make all packages happy
09:03 < mtr> yes, i read in the logs
09:03 < mtr> and you wanted to pull my ears .... ;)
09:03 < mnemoc> yes :)
09:04 < mnemoc> in fact i wanted to kick you :)
09:05 < mtr> i had not done reference build, but compiled and tested all my updates,
09:05 < mtr> though, which packages failed due to libchipcard?
09:05 < mnemoc> openhbci, aqbanking and aqhbci
09:06 < mtr> :(
09:06 < mnemoc> the last two are multi-packages
09:06 < mnemoc> some need libchipcard1 and some libchipcard2
09:06 < mnemoc> they broke chain was ugly :(
09:08 < mtr> i am pretty sure that i've tested aqbanking and aqhbci with the beta version of libchipcard
09:09 < mnemoc> i guess you had orphan libchipcard libraries on your system
09:09 < mtr> hmm, you are probably right ... :(
09:11 < rxr> re
09:11 < rxr> hehe - my athlon hit the sparc64 error way fastern than the real sparc
09:11 < rxr> minutes instead of hours ...
09:11 < mtr> hi rxr 
09:11 < mnemoc> rxr: i started a fix-rebuild with pkgconfig 0.16 and some merges from trunk
09:11 < mnemoc> rxr: i'll merge my last 3 commit to trunk and go to bed
09:12 -!- tilix [~tilix@212.116.139.141] has quit [Remote closed the connection]
09:12 -!- tilix [~tilix@212.116.139.141] has joined #t2
09:12 < rxr> cu in my evening then ,-)
09:12 < rxr> sleep well
09:12 < mnemoc> this 6 hours windows has good sides
09:12 < mnemoc> window*
09:15 < mnemoc> rxr: if you had a minute can you take a look bug #15 (balsa)? my eyes are not good finding weird C/C++ bug these days (years?)
09:16 < rxr> mnemoc: ok - I schedule it
09:16 < rxr> after fixing glibc for sparc64 ...
09:16 < mnemoc> sure
09:18 < mipe> hmm,how easy it is to change file hierarchy in t2? meaning packages goes to /programs/<app>/*, i wanna test this and see if it helps newcomers to gradually switch to linux,first appdir hierarchy and then switch to original hierarchy. 
09:19 < mipe> and maybe write up a report to school to get badly needed credits
09:19 < mnemoc> on target's parse-config
09:19 < mnemoc> if it's massive
09:19 < mnemoc> or in target/pkg_$foo.conf if it's not
09:19 < mnemoc> prefix=programs/$pkg; prefix_auto=1; set_confopt
09:20 < mipe> cool
09:20 < mnemoc> that may work for most packages, but many have hardcoded paths on their code
09:20 < rxr> mipe: you might need to patch gcc and glibc and co if you do not want /usr at all
09:20 < mnemoc> prefix_auto=0 sorry
09:20 < rxr> some stuff has hardcoded pathes for library lookup of core stuff ...
09:22 < mnemoc> compatibility symlinks may do your life easier
09:25 < CIA-9> amery * r8574 /trunk/package/ (6 files in 4 dirs): 
09:25 < CIA-9> * merged 8570 from 2.1 - downdated libchipcard (1.9.12 -> 0.9.1) and libchipcard2 (1.9.12beta) added
09:25 < CIA-9> * merged 8571 from 2.1 - updated aqhbci (1.0.7 -> 1.0.8)
09:25 < CIA-9> * merged 8573 from 2.1 - fixed and cleaned psi
09:26 < mnemoc> mipe: take a look into DJB's file hierarchy proposal
09:27 < mipe> mnemoc:where can i find it, i once saw it but i forgot to bookmark it for later reviewing. and i cant find it anymore :(
09:28 < mnemoc> mipe: http://cr.yp.to/slash.html
09:28 < mnemoc> mipe: but mainly http://cr.yp.to/slashpackage.html
09:28 < mipe> mnemoc:thanks
09:29 < mnemoc> yw
09:31 < mnemoc> mipe: http://code.dogmap.org/spf/ he had made a big effort to make world buildable for /sp/
09:31 < mnemoc> sp=slash package
09:38 < mnemoc> rxr: -> Compiler Cache Hits while pkg build: 96.57% (4303 hits, 153 misses) <-- after CC=$KCC merge
09:39 < mnemoc> ok, gn8 *
09:39 < rxr> woohoo
09:39 < rxr> glibs continues building ...
09:39 < mnemoc> :D
09:39 < mnemoc> go sparc go
09:39 < mnemoc> sparc64* :)
09:39 < rxr> mnemoc: do you consider the above ccache hits good or too less ?
09:40 < rxr> how to fix sprac64 glibc:
09:40 < rxr> var_append GCC_WRAPPER_INSERT " " "-D ASI_PNF=0x82"
09:40 < rxr> ,-)
09:40 < mnemoc> rxr: i guess ot missed all third-party modules due to the change of the name of the compiler.... good enough for me :)
09:41 < rxr> yep
09:41 < rxr> == 09:52:24 =[5]=> Building base/screen [4.0.2 2.2.0-alpha].
09:41 < rxr> == 04/29/05 09:57:36 =[5]=> Finished building package screen.
09:41 < rxr> ^- native sparc build comming along nicely ...
09:41 < mnemoc> great
09:41 < rxr> sleep well mnemoc 
09:41 < mnemoc> i'll try 
09:41 < mnemoc> cu
09:42 < rxr> cu
09:42 < mtr> cu mnemoc 
09:44 < rxr> ah - it seems in the past linux's asm-sparc64/asi.h included that define
09:45 < rxr> hm - it still does ...
09:55 < mipe> hmm,how do i disable unionfs on kernel26 build process?
10:01 < rxr> disable the unionfs package
10:02 < mipe> yep :)
10:06 < tilix> glibc builded successfully for an hour on an Athlon XP 2600+ 512 RAM Is i normal
10:08 < rxr> yes
10:10 < rxr> tilix: did it finished after an hour - or is it still building ?
10:11 < tilix> it finished. now T2 is building the kernel
10:11 < rxr> yeah - glibc is a h*ll of a package ...
10:11 < rxr> I also find the build time unreasonable ...
10:12 < rxr> (guess how long it builds on my 270Mhz UltraSPARC ...)
10:12 < tilix> :-)
10:13 < rxr> == 20:16:20 =[5]=> Building base/glibc [2.3.5 2.2.0-alpha].
10:13 < rxr> == 04/29/05 00:33:09 =[5]=> Finished building package glibc.
10:14 < CIA-9> rene * r8575 /trunk/package/base/linux-header/linux-header.conf: 
10:14 < CIA-9> * fixed sparc64 regression, caused by modularization of linux-header
10:14 < CIA-9>  for x86-64 support
10:16 < CIA-9> rene * r8576 /branches/2.1/package/base/linux-header/linux-header.conf: 
10:16 < CIA-9> * merged r8575 from trunk: fixed sparc64 regression, caused by
10:16 < CIA-9>  modularization of linux-header
10:23 < mipe> hmm,where should i do symlinks for package binaries,etc... after package build... target/$target/build.sh?
10:24 < mipe> like djb, /commands/* has the binary links
10:29 < rxr> also a parse config
10:29 < rxr> where you adjusted the pathes ...-
10:29 < rxr> hook_add final 9 "my_targets_smart_symlink" or so ...
10:29 < rxr> should work
10:40 < rxr> == 10:06:03 =[1]=> Building base/glibc [2.3.5 2.2.0-alpha].
10:40 < rxr> == 04/29/05 10:34:03 =[1]=> Finished building package glibc.
10:40 < rxr> sparc64 cross build on an AthlonXP 2500+
11:04 < CIA-9> rene * r8577 /trunk/package/network/plptools/ (gcc34.patch plptools.desc): * updated plptools (0.12 -> 0.13)
11:05 < CIA-9> rene * r8578 /trunk/package/multimedia/kdetv/kdetv.desc: * updated kdetv (0.8.0 -> 0.8.7)
11:08 -!- mnemoc [~amery@200.75.27.47] has quit [Read error: 113 (No route to host)]
11:10 -!- mnemoc [~amery@200.75.27.47] has joined #t2
11:12 -!- sepp [~sepp@p213.54.212.249.tisdip.tiscali.de] has quit [Read error: 145 (Connection timed out)]
11:34 < tilix> who is in charge of the website?
11:34 < tilix> of the new one
11:37 < rxr> the new one is mostly done by jeru
11:37 < valentin> moin
11:37 < rxr> also known as Chrisian Wiese
11:37 < rxr> tilix: do you have some problem or fix or so ?
11:37 < rxr> moin valentin 
11:37 < rxr> valentin: sparc64 building now, too
11:38 < tilix> i could translate it to Bulgarian
11:38 < rxr> that would be nice
11:39 < rxr> I think there is one outstanding issue with i18n of some parts of the pages
11:39 < valentin> rxr: cool
12:03 < tilix> why it builds GCC two times?
12:04 < tilix> on a boodisk target
12:05 < mipe> at same stage?
12:06 < mipe> compiler is build on stage 0,1,2,5 i think
12:06 < tilix> where to check for a stage?
12:07 < tilix> oh, [0] [1] etc
12:07 < mipe> yep
12:07 < mipe> 13:01:17 =[0]=> Building base/glibc [2.3.4 2.1.0-rc1]
12:07 < mipe> so this is stage 0 glibc
12:07 < tilix> = 13:04:35 =[1]=> Building base/gcc [3.4.3 2.2.0-alpha].
12:07 < tilix> stage 1
12:08 < rxr> http://wiki.t2-project.org/index.cgi?Toolchain
12:09 < mipe> gcc is build on 0,1,2 and 5, 0 is pseudo compiler and 1 is something,2 is built inside chroot and 5 is something :)
12:10 < mipe> ah 5 is all packages build stage i think
12:11 < mipe> rxr: have you done x86-64 bootdisk cross build recently? wondering if the dietlibc error is fixed
12:12 < mipe> or gcc error
12:15 < tilix> is there a script for hardware configuration in the livecd target, or only hotplug (rockplug)?
12:17 < rxr> tilix: well the 2.1 rock inherited live cd sucked majorly
12:17 < rxr> I'm currently rewriting it for 2.2
12:18 < rxr> it will use a fully modular kernel and udev with a tiny hotplug for hardware configuration
12:18 < rxr> mipe: nope - no x86-64 cross builds recently ...
12:18 < rxr> but I could try some time soon
12:18 < tilix> rxr: do you need help with a hwsetup program?
12:20 < rxr> well - currently it is comming along quite fine
12:20 < rxr> what would you hwsetup do?
12:20 < rxr> e.g. all the PCI, USB, and so on device will get their module loaded and /dev entries created
12:21 < rxr> for desktop use the mostly only part missing is a X configuration
12:21 < rxr> I'll also do then
12:21 < tilix> hmmm.
12:22 < tilix> I`m using a modified version of hwsetup in my distribution, but I don`t see how it would integrate with T2. This program scans for the all hardware and then puts it in /etc/sysconfig Then the knoppix-autoconfig does the job. 
12:25 < rxr> what does hssetup exaclty do?
12:25 < rxr> I'm not familiar with all the knoppix "hacks"
12:25 < tilix> Scans for the hardware and puts everything that it has found in /etc/sysconfig
12:26 < tilix> Then a script called knoppix-autoconfig takes care and loads the modules, setups symlinks, etc
12:26 < tilix> And a script called mkxf86config is called - it generates XF86Config-4
12:27 < tilix> Gets monitor info via ddcxinfo-knoppix
12:28 < rxr> well you do not need the first half of this - it is done by udev/hotplug
12:28 < rxr> only a X config is needed - the most interesting part here is how ddxcinfo-knoppix is implemented ...
12:29 < tilix> http://www.kanotix.com/files/debian/ddcxinfo-kanotix/ddcxinfo-kanotix_0.6-11.tar.gz
12:30 < mipe> rxr:is there a easy way to add flist to use programs/$pkg from target/$target/ files, without touching Build-Pkg and adding flistroot options there
12:32 < tilix> (updated version of ddcxinfo-knoppix)
12:32 < rxr> mipe: you have questions ...
12:32 < rxr> ... challenging they are
12:33 < rxr> mipe: you mean so that the files show up in the flist ?
12:33 < rxr> just extend flistroot in your target
12:34 < rxr> var_append flistroot ' ' 'programs libraries' or so should do
12:35 < rxr> ah - BIOS calls
12:35 < rxr> that is lame and unportable
12:35 < rxr> but could serve as a first implementation until we hacked X (I think) to exprt DDC info obtained from the native X driver ...
12:35 < rxr> tilix: for such stuff the 99-final package was intended
12:36 < rxr> so you can solder such auto-detection stuff together by your target (defining what 99-final) should do
12:36 < tilix> rxr: most of the Knoppix tools are old now. That`s why I want to move on T2
12:36 < rxr> hoever we could package this ddcxinfo package 
12:36 < rxr> tilix: I looked into the boot code of knoppix
12:36 < rxr> it was disgusting
12:37 < rxr> hardcoded module probing to get the root CD mounted ...
12:38 < tilix> it even uses 16-years old BSD code (ash) in the initrd
12:44 < rxr> tilix: what would you like to use there?
12:45 < tilix> in initrd ?
12:46 < tilix> busybox is the best
12:47 < rxr> ouhm - I find busybox quite disgusting
12:47 < rxr> too much uclibc bound ...
12:47 < tilix> what do you plan to use?
12:48 < rxr> and it also only contains "made fit" old ash code, on?
12:48 < tilix> yes
12:48 < rxr> I have not yet decided what shell to use
12:48 < rxr> for devices udev with our upcomming hotplug++ will be used, what shell to inject was still to be decided
12:51 < rxr> http://www.joyoftech.com/joyoftech/index.html
12:58 -!- sparc-kly [~sparc@65-23-205-193.prtc.net] has quit [Remote closed the connection]
13:05 < rxr> http://www.golem.de/0504/37799.html
13:11 < valentin> rxr, you implemented that string class - can you tell me how to extract a substring from memory ?
13:13 < valentin> like src.substring (start, end) ?
13:13 < valentin> or (start, length) ?
13:13 < valentin> ok, i consult the std...
13:13 < rxr> ehm
13:13 < rxr> start end
13:14 < valentin> iterator stuff or index ?
13:14 < rxr> size_type idx = 0, size_type len = npos
13:14 < rxr> hm no - len
13:14 < rxr> sorry
13:14 < rxr> do you want index or iterator ?
13:15 < rxr> yep - "retuns a substr of at most len ..."
13:15 < valentin> i want index
13:15 < valentin> ok
13:16 < valentin> thx
13:18 < rxr> no problem
13:18 < rxr> might you like to implement the std::map fo rme ?
13:20 < valentin> oh
13:20 < valentin> i once implemented a full AVL tree
13:20 < valentin> this was some years ago
13:20 < valentin> before i knew there was an stl :)
13:21 < valentin> so that one had a different api, but the tree code should fit
13:21 < valentin> but i think i will re-implement it
13:21 < valentin> do you have a std::vector ?
13:22 < valentin> so we could create a fake-map using a vector first if you need the map soon
13:22 < valentin> the search times will be O(n) instead of O(log(n)) but who cares for a beta-version ?
13:23 < rxr> I'll do the vector today 
13:23 < rxr> yep
13:23 < valentin> when do you need the map ?
13:23 < rxr> tonight
13:23 < rxr> I first protype the new module matching code using the G++ STL - to see what STL API set I need
13:23 < valentin> then i would rely propose to map the std::map api onto a std::vector <std::pair <KEY, VAL>>
13:23 < rxr> after that I switch over to our Embedded STL an implement the missing bits
13:24 < rxr> yeah - reasonable for first testing
13:24 < rxr> damn - the std::istream >> some int type did not stip whitespaces ...
13:24 < rxr> damn
13:24 < valentin> implementing balanced trees is quite tricky when you want to have deletion ...
13:26 < rxr> well the first version would not need to balance that well ...
13:27 < rxr>      7 /lib/modules/2.6.11.7-dist/modules.ieee1394map
13:27 < rxr>      8 /lib/modules/2.6.11.7-dist/modules.inputmap
13:27 < rxr>    142 /lib/modules/2.6.11.7-dist/modules.isapnpmap
13:27 < rxr>   1309 /lib/modules/2.6.11.7-dist/modules.pcimap
13:27 < rxr>   1221 /lib/modules/2.6.11.7-dist/modules.usbmap
13:27 < rxr> not too many nodes ...
13:27 < rxr> of course I could really just use a vector for it ...
13:29 < valentin> nope - write down a subset of the map api
13:30 < valentin> and just make it private : std::vector first
13:31 < valentin> associative containers are quite important even for embedded systems, so sooner or later we will code them
13:31 < valentin> however we should create a common tree class to be used for map,set,multimap, ...
13:33 < valentin> g++ stl has its own rbtree class for that
13:41 -!- tilix [~tilix@212.116.139.141] has left #t2 ["Konversation terminated!"]
13:41 < rxr> hm - except silo (which I fix some time later) sparc64 finished cross building on my athlon ..
13:41 < rxr> I think I copy it onto the U30 somet time today
13:52 < rxr> ah - there is the prob
13:52 < rxr> istream do not want to read i hex values for ints by default
13:58 < valentin> yep
13:58 < valentin> c++ streams suck
13:59 < valentin> for most data i have to parse / create at work i switched to scanf / printf instead
13:59 < valentin> format strings are ugly but way more comfortable to use :(
14:01 < rxr> hm - as it looks this f*cking G++ STL can only read in decimal values ...
14:03 < valentin> use scanf ,)
14:04 < rxr> how can one implement streams without non decimal interger input ...
14:04 < valentin> are you sure there is no hex input ?
14:04 < valentin> i think there was
14:04 < rxr> I just read the source - it looked so
14:05 < valentin> hm
14:05 < rxr>     operator>>(int& __n)
14:05 < rxr> hm - moment
14:07 < rxr> ah - my indeed
14:07 < rxr> in the locale facets there are differentiations for hex and octal ...
14:07 < valentin> ?
14:07 < rxr> ?
14:13 < valentin> why in the locale ?
14:14 < rxr> because all of those stuff is groupped there
14:14 < valentin> so how do you use the hex parser ?
14:14 < rxr> what I wonder more is, why it works in my tiny test and not in my real app
14:14 < rxr> it actually works out of the box
14:15 < rxr> ah not it does not
14:15 < rxr> cin.setf (std::ios::hex);
14:16 < rxr> hm - no - even that is not sufficient
14:16 < rxr>     cin.setf (std::ios::hex, std::ios::basefield);
14:18 < valentin> i have to take a look after Kerstin now - be back in at most 45 minutes
14:28 -!- sparc-kly|U5 [~mubex@66-50-123-218.prtc.net] has joined #t2
14:34  * rxr huting #ingredients for lunch
14:34 < rxr> == 10:33:18 =[5]=> Building x11/xorg [6.8.2 2.2.0-alpha].
14:34 < rxr> still building (sparc32 on the U5 ...)
14:34 < rxr> cu
14:38 < sparc-kly|U5> Linux 172.16.102.5 2.4.27-2-sparc64 #1 Sun Apr 3 05:21:36 UTC 2005 sparc64 GNU/Linux
14:38 < sparc-kly|U5> my U5 :D
14:38 < sparc-kly|U5>  08:23:10 up 3 days, 13:15,  1 user,  load average: 0.16, 0.08, 0.02
14:41 < valentin> profile name for interface is mandatory ?
14:41 < valentin> hi sparc-kly|U5 
14:42 < valentin> what system is running on your U5 ?
14:42 < valentin> bbl
14:48 < sparc-kly|U5> debian 
14:48 < sparc-kly|U5> huhuhuhu
14:48 < sparc-kly|U5> my u2 running gentoo 
15:08 < rxr> valentin: nope - it is not mandatory
15:16 < rxr> http://www.heise.de/newsticker/meldung/59145
15:31 -!- Cypl0dE [~cypl0dE@10-49-112.adsl.terra.cl] has joined #t2
15:35 < Cypl0dE> QUOTE WHOIS
15:35 -!- sparc-kly|U5 [~mubex@66-50-123-218.prtc.net] has quit [Read error: 145 (Connection timed out)]
15:35 -!- Cypl0dE [~cypl0dE@10-49-112.adsl.terra.cl] has left #t2 ["Abandonando"]
15:41 -!- mnemoc_ [~amery@200.75.27.9] has joined #t2
15:52 -!- mnemoc [~amery@200.75.27.47] has quit [Read error: 110 (Connection timed out)]
15:54 < rxr> == 04/29/05 16:09:16 =[5]=> Finished building package xorg.
15:58 < rxr> module-matching.cc: In function `int main()':
15:58 < rxr> module-matching.cc:31: error: `main()::pci_module' uses local type `main()::pci_module'
15:58 < rxr> module-matching.cc:31: error:   trying to instantiate `template<class _Alloc> class std::allocator'
15:58 < rxr> module-matching.cc:31: error: template argument 2 is invalid
15:58 < rxr> interesting errors one can produce in C++ ,-)
16:13 < valentin> re
16:13 < valentin> rxr: the handbook does not speak about non-profiled interface sections
16:22 -!- sepp [~sepp@p213.54.201.207.tisdip.tiscali.de] has joined #t2
16:22 < sepp> hi
16:24 < jsaw> wiki has been spamed
16:24 < jsaw> http://wiki.t2-project.org/index.cgi?TODO
16:41 -!- rxr_ [~rene@e178191252.adsl.alicedsl.de] has joined #t2
16:41 -!- Topic for #t2:  T2 | 2.1.0-beta4 RELEASED | The next generation of System Development Environments (SDE) | http://www.t2-project.org/
16:41 -!- Topic set by menomc [] [Sat Apr 16 00:33:34 2005]
16:41 [Users #t2]
16:41 [ _Ragnar_] [ jsaw  ] [ mipe   ] [ mtr] [ rxr ] [ sepp    ] 
16:41 [ CIA-9   ] [ keinek] [ mnemoc_] [ nzg] [ rxr_] [ valentin] 
16:41 -!- Irssi: #t2: Total of 12 nicks [0 ops, 0 halfops, 0 voices, 12 normal]
16:41 -!- Channel #t2 created Sun Aug  8 21:15:33 2004
16:41 -!- Irssi: Join to #t2 was synced in 10 secs
16:41 < rxr_> jsaw: again ...
16:41 < valentin> hm :((
16:41 < valentin> damn, it even killed our content
16:42 < valentin> we should get rid of wiki soon ...
16:42 < rxr_> mnemoc_: please revert ,-)
16:42 < rxr_> yep
16:42 < rxr_> but mnemox has a CVS or SVN backend - so he can revert that ...
16:42 < jsaw> I added a help section...
16:42 < rxr_> and even expose the IP of the changee
16:43 < rxr_> altough they can be chinese anonymizer ...
16:43 < jsaw> somebody please check/correct "HelpDesk"
16:48 < rxr_> url ?
16:51 < jsaw> http://wiki.t2-project.org/index.cgi?HelpDesk
16:55 < valentin> is there a module for kwiki that rejects all changes that contain some special words like viagra ?
16:56 -!- rxr [~rene@e178145104.adsl.alicedsl.de] has quit [Read error: 110 (Connection timed out)]
17:03 < valentin> jsaw: what do you mean by 'check' ? spellcheck ?
17:08 < jsaw> if it is understandable :)
17:08 < jsaw> kept it as short as possible
17:39 -!- Cypl0dE [~cypl0dE@10-49-112.adsl.terra.cl] has joined #t2
17:40 < mnemoc_> moin
17:41 -!- Cypl0dE [~cypl0dE@10-49-112.adsl.terra.cl] has quit [Client Quit]
17:43 < jsaw> cu
17:43 < mnemoc_> cu jsaw 
17:47 < rxr_> hi mnemoc_ 
17:47 -!- You're now known as rxr
17:52 < valentin> re - moin mnemoc_ 
17:53 < mnemoc_> hi rxr, valentin 
17:53 < mnemoc_> rxr: 0.16.0 seems to work fine
17:53 < rxr> aha
17:54 < rxr> this days you can not even trust such plain config prog updates ...
17:54 < rxr> maybe I should rewrite it in C++ ,-)
17:54 < mnemoc_> hehe
17:55 < mnemoc_> i just wonder if this is a behaviour expected by pkgconfig people, which will continue on next releases, or just a bug
17:55 < rxr> well
17:55 < rxr> let's let the gnome folks sort it out
17:55 < rxr> 0.16 always worked for us, soo ..
17:56 < keinek> hi
17:56 < mnemoc_> hi keinek 
17:56 < rxr> Error logs from default-2.2.0-alpha-sparc-v8-desktop:
17:56 < mnemoc_> rxr: was the sparc bug related to modularization present on 2.1?
17:56 < rxr>  [5] mail/spamassassin                   
17:56 < rxr> 649 builds total, 214 completed fine, 1 with errors.
17:57 < rxr> mnemoc_: yep - that is why I already merged it ,-)
17:57 < rxr> make: *** [spamc/Makefile] Segmentation fault (core dumped)
17:57 < rxr> ^- also on SPARC
17:58 < mnemoc_> weird
17:58 < rxr> so not a gcc or binutils x86 code generation issue
17:58 < rxr> I looked into it
17:58 < rxr> it happens when perls starts itself
17:58 < mnemoc_> rxr: i want to move postinstall out of etc/ ..... /usr/lib/something/postinstall ?
17:58 < rxr> thru an very wired expression in that configure.PL script
17:59 < mnemoc_> ic
17:59 < rxr> mnemoc_: well - your /var/adm/postinstall was quite ok
18:00 < mipe> hmm,$root is acting wierdly
18:00 < mnemoc_> i'm not convinsed of it at var/adm right now
18:00 < mnemoc_> var/adm is our database, not a place for scripts
18:03 < rxr> hm
18:05 < mipe> ups thats just me messing up, whats the diffrence between $xroot and $root? they look the same
18:05 < mnemoc_> $xroot=$root only inside chroot
18:07 < mipe> thats wierd,they are same,on stage 0
18:09 < mnemoc_> iirc they get mixed in some minute of the build time
18:09 < mnemoc_> but i'm not sure when
18:16 < rxr> I also recently wondered what the xroot was for ...
18:16 < rxr> would need to search for that ...
18:17 < rxr> mipe: what do you need $root for ?
18:20 < mnemoc_> $xpkg and var/adm/chksum may be gone on 2.2 also
18:20 -!- sparc-kly [~sparc@65-23-205-193.prtc.net] has joined #t2
18:22 < valentin> mnemoc_: did you revet that wiki spam ?
18:22 < mnemoc_> not yet
18:22 < mnemoc_> what pages?
18:22 < valentin> TODO
18:23 < mnemoc_> todo in spanish means 'everything' :)
18:23 < mnemoc_> rxr: r8216 | rene | 2005-04-13 02:19:44 -0400 (Wed, 13 Apr 2005) | 2 lines
18:23 < mnemoc_> * updated pkgconfig (0.15.0 -> 0.17.1)
18:25 < valentin> mnemoc_: :)
18:25 < valentin> no wonder the spammers filed their 100s of links it under that item
18:26 < valentin> can you create a black list for words like viagra ?
18:26 < mnemoc_> and removed our TODO
18:26 < mnemoc_> not easily
18:26 < mnemoc_> i'm waiting for the new site to drop the wiki
18:26 < valentin> yep
18:26 < valentin> jsaw created new content today ...
18:27 < valentin> in the wiki that is
18:28 < rxr> but the wiki will go
18:28 < rxr> we should transfer all the content then ...
18:29 < mnemoc_> yes
18:32 < mnemoc_> valentin: TODO reverted
18:33 < mnemoc_> CIA-9: ?
18:35 < mnemoc_> rxr: will you fight 0.17.x on trunk or should i downdate also?
18:35 < rxr> mnemoc_: downdate
18:35 < mnemoc_> rxr: ok
18:36 < mnemoc_> keinek: svn up, rebuild pkgconfig and see if that fix your audacity
18:41 < mipe> rxr mipe: what do you need $root for ? <-- for the symlinks, $root/package/$pkg/bin,etc...
18:42 < rxr> ah - and for the chroot you miss a variable with build/.../ in front ?
18:42 < keinek> ok thanks mnemoc_ 
18:43 < mipe> rxr:nope, i just wondered why $root doesnt have TOOLCHAIN/tools.cross/$prefix for some packages
18:46 < rxr> mipe: root has this ${base}/build/... for cross builds ...
18:48 < rxr> mipe: for which ones does it not have this?
18:48 < rxr> within the sandbox root if of course empty ...
18:49 < mipe> rxr:yep, dietlibc and linux-headers just doesnt have TOOLCHAIN prefix, but i got it working now, now i need to add -I/headers -L/libraries but for somereason they got removed
18:49 < mipe> *they get removed* during gcc build
18:59 < mnemoc_> rxr: did you got nitifications for 8579 and 8580?
19:03 < rxr> yes
19:03 < rxr> 8579 is in my inbox
19:04 < mnemoc_> i just got the three
19:04 < mipe> ./scripts/Create-ErrList -cfg system
19:04 < mipe> Error logs from system-2.1.0-rc1-x86-64-athlon64-desktop:
19:04 < mipe> 153 builds total, 149 completed fine, 0 with errors.
19:04 < mnemoc_> mipe: you may need to resync and rebuild pkgconfig
19:04 < mipe> mnemoc_: ok
19:05 < rxr> Error logs from default-2.2.0-alpha-sparc-v8-desktop:
19:05 < rxr> 649 builds total, 221 completed fine, 1 with errors.
19:05 < mnemoc_> is that a normal minimal-desktop or it use weird fh ?
19:05 -!- keinek_ [~Keinek@201.254.13.197] has joined #t2
19:05 < mipe> normal
19:06 < mnemoc_> so, we will have an x86_64 desktop iso? :D
19:06 < mipe> wierd fh is just for testing purpose. i wanna test if my dad learn linux easier with other fh system
19:07 < mipe> mnemoc_: yep, but bootdisk might fail
19:07 < mipe> unless dietlibc confusing gcc is fixed
19:10 < mnemoc_> oh
19:12 -!- keinek [~Keinek@201.254.8.139] has quit [Read error: 145 (Connection timed out)]
19:16 < rxr> mipe: let's fix that this time ,-)
19:16 < mipe> :)
19:18 < mipe> system build done,next bootdisk target...
19:18 < mnemoc_> :D
19:24 -!- sepp_ [~sepp@p213.54.199.88.tisdip.tiscali.de] has joined #t2
19:25 [Users #t2]
19:25 [ _Ragnar_] [ keinek_] [ mtr] [ sepp     ] [ valentin] 
19:25 [ CIA-9   ] [ mipe   ] [ nzg] [ sepp_    ] 
19:25 [ jsaw    ] [ mnemoc_] [ rxr] [ sparc-kly] 
19:25 -!- Irssi: #t2: Total of 13 nicks [0 ops, 0 halfops, 0 voices, 13 normal]
19:27 -!- sepp [~sepp@p213.54.201.207.tisdip.tiscali.de] has quit [Read error: 110 (Connection timed out)]
19:27 < valentin> i wonder who chose this stupid syntax for interface if(prof) where interface if prof would be easier to parse
19:28 < valentin>     ignore=1 if="${1%(*}"
19:28 < valentin>         local prof="${1#*(}"
19:28 < valentin>         prof="${prof/)}"; prof="${prof//,/ }"
19:28 < valentin> i love bash
19:29 < rxr> if you need a sparc for s.th. 2 are just up over here ...
19:35 < valentin> so
19:35 < valentin> interface eth0
19:35 < valentin> is converted to
19:35 < valentin> interface eth0(default)
19:36 < valentin> during parsing
19:37 < valentin> most funny thing of the 6 line bash code is, that it converts
19:38 < valentin> interface if(bla,blub,blubber)
19:38 < valentin> into
19:38 < valentin> interface if bla blub blubber
19:38 < valentin> as i proposed above :)
19:43 < valentin> is anybody using the multiple profile thing ?
19:43 < valentin> i could easily handle a vector of profiles per interface section in my datastructures, but i do not know how to represent this in the UI ...
19:44 < rxr> valentin: leave it out for now
19:44 < valentin> ok
19:45 < valentin> as a hack i could just copy the same interface object a multiple times, but after saving you will have copies in your config file ...
19:47 < valentin> but for now i will just take it as one profile name even when it contains ','
19:49 < rxr> ack
19:49  * rxr ncp'ing the cross build sparc64 biuld onto the U30
19:49 < rxr> let's see if it runs ,-)
19:50 < rxr> ok chroot ok
20:00 < rxr> mnemoc_: for the bugtracker
20:00 < rxr> fuser should be present after crossbuild
20:22 < mnemoc_> uhm?
20:22 < mnemoc_> don't we use fuser only for download?
20:23 < mnemoc_> valentin: you can change that bash to sed
20:23 < rxr> cleanup, too
20:23 < valentin> what bash ?
20:25 < mnemoc_> valentin: interface if(profile) thing
20:26 < mnemoc_> for complex parsing like that we should never have used bash expressions
20:26 < rxr> well - normally that is way faster
20:26 < rxr> only bad that some changed between bash 2 and 3 ...
20:26 < rxr> I would leave it this way - rocknet is already slow enough ...
20:27 < valentin> yep
20:27 < mnemoc_> what i miss more on rocknet is abort-on-error
20:27 < valentin> if we want to change it we should break with the syntax and convert to interface ethx prof1 prof2
20:28 < valentin> it will be like that in our upcomming network tool
20:28 < mnemoc_> current syntax is more clear
20:28 < mnemoc_> even if parser is complex
20:28 < rxr> valentin: we do not need to change the syntax - the upcomming thing will be C++ or LUA or so ..
20:28 < valentin> but it sucks what terrible wrong behavior you get when typing
20:28 < valentin> interface eth1 (blabal)
20:28 < rxr> so we can parse abreatary complex stuff there
20:29 < valentin> rxr: if we take c++ we can keep the parser i am coding now (hint:)
20:29 -!- mnemoc_ is now known as mnemoc
20:29 < valentin> hi mnemoc :)
20:29 < mnemoc> hi :)
20:30 < mnemoc> parsing on lua seems to be even easier
20:30 < valentin> maybe - i should dig into lua
20:31 < rxr> I have no idea if LUA is that perfect
20:31 < valentin> we should do a unique and generic config concept for all config stuff then
20:31 < rxr> e.g. it does not allow structural grouping
20:31 < rxr> like struct {
20:31 < valentin> no ?
20:31 < rxr> int
20:31 < rxr> bla
20:31 < rxr> }
20:31 < rxr> IIRC
20:31 < valentin> that SUCKS
20:31 < rxr> only some form of custom table stuff
20:31 < rxr> I need to really play with it
20:31 < mnemoc> lua is not perfect, is a limited scripting language, but afaics complete enough for our needs
20:31 < rxr> at this point I would ont decide what to use
20:32 < rxr> we really need to review the alternatives
20:32 < mnemoc> article article article :D
20:32 < rxr> maybe small / embryo is nicer
20:32 < rxr> it has C like syntax - that might make it easier for us C++ adicts ...
20:33 < mnemoc> does embryo compiles to elf or to bytecode?
20:33 < rxr> bytecode
20:33 < rxr> embryo is a small variant ...
20:33 < rxr> for some platforms there even exist s JIT compiler ...
20:35 < mnemoc> uhm... pilot-link (prio:145.900) has java (800.000) dependencies
20:35 -!- minto [~chatzilla@82-217-66-7.cable.quicknet.nl] has joined #t2
20:35 < minto> hi * :-)
20:35 < rxr> hi minto 
20:36 < valentin> hi minto
20:37 < mnemoc> yahoo! ooo finished buidling!
20:37 < mnemoc> on reference
20:37 < rxr> cool!
20:38 < rxr> how many hours? 
20:38 < rxr> I guess due to my flist and dep tuning about 2h less than with the historic rock scripts ? *g*
20:38 < rxr> mnemoc: it also built on PowerPC - just did not start up
20:39 < mnemoc> == 07:02:15 =[5]=> Building office/ooo [1.9-m90 2.1.0-rc1].
20:39 < mnemoc> == 04/29/05 14:19:35 =[5]=> Finished building package ooo.
20:39 < rxr> ,-)
20:39 < rxr> can you scp it on some desktop and try it out ?
20:39 < mnemoc> -rw-r--r--  1 root root 90M 2005-04-29 14:21 ooo-1.9-m90.gem
20:40 < mnemoc> ok, i'll try
20:40 < rxr> only if it fits in your workflow ...
20:40 < mnemoc> yes, don't worry
20:42 < valentin> is there a variant of find_first_of ( ) where you can specify a starting position for the search ?
20:42  * valentin just to lazy to look it up
20:42 < rxr> yes
20:42 -!- mtr_ [~michael@p54AFA861.dip0.t-ipconnect.de] has joined #t2
20:43 < rxr> valentin: in string?
20:43 < valentin> yes
20:43 < rxr> char c, size_type idx
20:43 < rxr> simillar for const char* and const string&
20:43 < valentin> thanx !
20:46 < valentin> is there a way to get a substring from a position to the end
20:46 < valentin> perhaps passing npos as length ?
20:47 < rxr> yes
20:47 < rxr> or just leaving it out
20:47 < rxr> substr(start)
20:47 < valentin> npos is fine
20:47 < rxr> the later, too
20:47 < rxr> in optimized implementations this saves you one argument
20:47 < rxr> however in my embedded one it is just a overload with len = npos as default argument
20:48 < rxr> just substr(start) is pretty fine and better style
20:50 < valentin> ok
20:54 -!- mtr [~michael@p54AF88BB.dip0.t-ipconnect.de] has quit [Read error: 110 (Connection timed out)]
21:00 < mnemoc> rxr: having shared-mime-info writing the database show us which other packages do it also
21:01 < rxr> ?
21:01 < rxr> you mean sharing the database file ?
21:01 < mnemoc> yes
21:01 < rxr> we could touch the file normally created to detect shared files ...
21:02 < mnemoc> good idea
21:02 < rxr> I do later
21:09 < rxr> $ xset m default
21:09 < rxr> puh
21:09 < rxr> f*cking gnome ...
21:19 < mnemoc> # du -sh ccache-x86/
21:19 < mnemoc> 3.2G    ccache-x86/
21:20 < mnemoc> 1561 packages
21:20 < mnemoc> including ooo
21:20 < rxr> ,-)(
21:20 < rxr> but ooo is not yet building completely out of the chache
21:20 < rxr> what was the numbers I posted? only 20-30% chache hits on rebuilds ...
21:20 < rxr> maybe some timestamp in some generated header included in many files or so ..
21:20 < mnemoc> i'm more worried about phppgadmin skipping flist
21:21 < rxr> I have nto yet reviewed that and put out a beer as bounty to fix that on the mailing list
21:21 < mnemoc> :)
21:22 < rxr> mnemoc: how does it install the files ?=
21:23 < mnemoc> direct untar
21:23 < mnemoc> over is symlink
21:23 < rxr> that is known to no work
21:23 < rxr> never did
21:24 < rxr> was it tested to ever fill the flist? e.g. with the untuned scripts?
21:24 < mnemoc> it use to work, i have installed phppgadmin by .gem
21:24 < rxr> I doubt it ever has
21:24 < rxr> ok - I review in detail
21:24 < rxr> put it on my list ..
21:24 < mnemoc> :)
21:24 < rxr> near - eutils and what was the one you wanted?
21:24 < rxr> this morning ...
21:24 < rxr> there was s.th. ...
21:25 < rxr> does anyone have a suse system handy?
21:25 < rxr> a recent one?
21:25 < rxr> with kde 3.4 ...
21:25 < rxr> and best gnome 2.10 ... ?
21:26 < mnemoc> i only have a broken xfce here
21:28 -!- mipe [~mika@dsl10040.japo.fi] has quit ["Leaving"]
21:53 < mnemoc> 1647 builds total, 1605 completed fine, 42 with errors. <-- looks better
22:20 < mnemoc> sepp_: let's try xmule :)
22:21 < sepp_> maybe the unicode crap works in wx2.6, who knows
22:22 < mnemoc> lets make it work on stable first :)
22:35 < mnemoc> 1647 builds total, 1608 completed fine, 39 with errors
22:36 < mnemoc> sepp_: do you have account on www2? can you close the bug you fixed?
22:36 < sepp_> no
22:36 < mnemoc> :(
22:36 < mnemoc> jeru!!
22:37  * sepp_ is confused with all the jdks seems libpano works with suns 1.5
22:40 < rxr> re
22:40 < rxr> mnemoc: link xmule with an older wxwidgets or delete it from 2.1 ...
22:40 < rxr> xmule is one of the least needed packages ...
22:41 < mnemoc> but it's fixed
22:41 < mnemoc> ===[build_this_package:815 (last $?=0)> make CC=gcc CPP=cpp CXX=g++ prefix=/opt/kde3 install
22:41 < mnemoc> make[2]: Entering directory `/TOOLCHAIN/src.kdevelop.1114650956.24440.a8c00201/kdevelop-3.2.0'
22:41 < mnemoc> mv: cannot stat `_SUBDIRS': No such file or directory
22:41 < mnemoc> that is bad
22:42 < rxr> ...
22:43 < mnemoc> # mine -i ooo-1.9-m90.gem
22:43 -!- minto [~chatzilla@82-217-66-7.cable.quicknet.nl] has quit ["Chatzilla 0.9.66 [Mozilla rv:1.7.6/20050318]"]
22:46 < mnemoc> $ oowriter1.9
22:46 < mnemoc> ....
22:46 < mnemoc> no suitable windowing system found, exiting.
22:46 < mnemoc> great
22:49 < mnemoc> 16:52:55 stat64("/opt/mozilla/lib/libnss3.so", 0xbfffd980) = -1 ENOENT (No such file or directory) <0.000019>
22:49 < rxr> ?!?
22:50 < rxr> that is s.th. I have not seen so far ...
22:50 < mnemoc> the no suitable thing?
22:50 < rxr> == 19:25:56 =[9]=> Building base/gcc [3.4.3 2.2.0-alpha].
22:50 -!- mtr [~michael@p54AF9D5F.dip0.t-ipconnect.de] has joined #t2
22:50 < rxr> mnemoc: ack
22:50 < rxr> == 04/29/05 20:53:17 =[9]=> Finished building package gcc.
22:50 < rxr> ^- a full gcc on sparc64 ...
22:51 < mnemoc> :)
22:51 < rxr> now running on the faster U30 ...
22:52  * mnemoc downloading mozilla.gem
22:54 -!- mtr_ [~michael@p54AFA861.dip0.t-ipconnect.de] has quit [Read error: 60 (Operation timed out)]
23:06 < mnemoc> strace -vvfF /opt/ooo/lib/OpenOffice.org/program/soffice 2>&1 | less       hunting runtime dependencies
23:07 < sparc-kly> i have one stupid question . what CPU is more fast . G3 366mhz or PII 366mhz ?
23:08 < rxr> maybe PII
23:08 < valentin> depends
23:08 < rxr> the G3 is very weak in the floating point area ...
23:09 < sparc-kly> hmmm
23:10 < mnemoc> 1647 builds total, 1609 completed fine, 38 with errors.
23:13 < rxr> we have some problem with multilibs and sparc64
23:13 < rxr> it overinstalled the libgcc_s.so with a 32bit version ...
23:14 < mnemoc> outch
23:14 < rxr> resulting in rendering gcc unusable ...
23:14 < rxr> I wonder why it first installed the 64bit version there just to overwrite it later
23:14 < rxr> I'll review tomorrow
23:15  * rxr in bed - cu all
23:15 < rxr> == 23:31:46 =[5]=> Building develop/libsigc++12 [1.2.5 2.2.0-alpha].
23:15 < rxr> Error logs from default-2.2.0-alpha-sparc-v8-desktop:
23:15 < rxr>  [5] mail/spamassassin                   [5] base/mtools
23:15 < rxr> 649 builds total, 250 completed fine, 2 with errors.
23:19 < mnemoc> gn8 rxr 
23:21 < mnemoc> rxr! i got it running!
23:21 < mnemoc> i have ooo-1.9-m90 open on my laptop
23:22 < rxr> !!!!!!!!!!!!!!
23:22 < rxr> cu
23:22 < mnemoc> cu rxr :)
23:26  * mnemoc removed the ugly pre-build ooo-1.1.4 he had installed before
23:42 < valentin> cu rxr
--- Log closed Sat Apr 30 00:00:47 2005