**** BEGIN LOGGING AT Mon Sep 12 02:59:56 2005 Sep 12 03:00:13 schurig: They do share include files - I'm not sure that makes a difference. Sep 12 03:00:56 RP: yeah, but the person that experiments with the CSL stuff could add things into his FILESDIR variable, to fetch include files from packages/gcc as well Sep 12 03:01:28 RP: it was just an idea, this would maybe prevent people into using this experimental compiler by accident. There are other ways for this, thought. Sep 12 03:01:51 schurig: I recently removed its DEAFULT_PREFERENCE to avoid this Sep 12 03:02:42 mmmm... once again a stupid question : what should i erase ? /build/tmp/* ? Sep 12 03:02:45 csl is just another version of gcc Sep 12 03:02:57 the gcc-csl/ dir is a relic Sep 12 03:02:58 alan|home: yes Sep 12 03:03:28 schurig: It can probably be handled by correctly set DEFAULT_PREFERENCEs Sep 12 03:03:38 RP : thanks. Sep 12 03:07:54 03koen 07org.oe.dev * r8decd470... 10/packages/ (18 files in 3 dirs): gcc-csl/, gcc-snapshot/: remove unused gcc cruft. All necessary gcc metadate already is in gcc/ Sep 12 03:09:29 koen: is gcc 4.x ready for ARM/XSCALE ? I mean real-life use, like I use gcc 4.x now on my Debian Box as main compiler for everything... Sep 12 03:10:05 schurig: no idea, I don't have the time to test gcc4 before the next familiar release Sep 12 03:10:13 that's why familiar specifies 3.x Sep 12 03:11:15 last I heard someone said we need 4.02 in OE to be usefull Sep 12 03:12:22 4.0.2 is supposed to be very useful Sep 12 03:12:32 i hope someone will add it @ some point of time Sep 12 03:13:28 mickeyl: according to gcc.gnu.org this compiler isn't yet released Sep 12 03:13:34 yep Sep 12 03:13:51 most likely next week Sep 12 03:16:32 mithro: I think you forgot to CC the list Sep 12 03:18:00 mickeyl: try talking jbowler into adding 4.0.2 Sep 12 03:20:10 mickeyl: are you responsible for lyx in OE ? Sep 12 03:20:40 i hate it when lists don't have reply-to set to the list :/ Sep 12 03:20:51 alan|home: more or less, yes. Sep 12 03:21:02 mickeyl: so : thanks thanks thanks thanks thanks thanks thanks thanks thanks thanks thanks thanks thanks thanks thanks Sep 12 03:21:06 ^^ Sep 12 03:21:14 :)) Sep 12 03:21:31 there is some more work needed on resolving some latex issues, but LyX per se works :) Sep 12 03:21:42 glop glop ^^ Sep 12 03:22:41 haaaa... latex on the zaurus... sssssssmoking d(^^)b !! Sep 12 03:23:42 heh Sep 12 03:24:17 yah, thought we need something to attract all the mathemagicians ;) Sep 12 03:24:51 mickeyl: I just added http://www.abisource.org/release-notes/2.3.6.phtml :) Sep 12 03:26:01 heh Sep 12 03:26:02 cool Sep 12 03:26:28 I din't add those fonts, though Sep 12 03:26:29 does AbiWord handle some of the evil proprietary formats? Sep 12 03:26:40 it should handle .doc Sep 12 03:27:23 it does Sep 12 03:31:07 koen: you're an montone expert - do you know any good way to determine the list of changed files between revisions? Sep 12 03:31:31 suitable for parsing by a script? Sep 12 03:31:45 03mickeyl 07org.oe.dev * rad913302... 10/packages/wlan-ng/ (10 files in 3 dirs): Sep 12 03:31:45 wlan-ng: resolve several issues in our wlan-ng 0.2.2 build Sep 12 03:31:45 patch courtesy Frederic Devernay, thanks! Sep 12 03:34:30 CP|Lapto1: monotone log ? Sep 12 03:38:11 koen: ok - can log list changes over multiple revisions? Sep 12 03:38:31 no idea on that Sep 12 03:54:48 re Sep 12 03:55:17 hi hrw Sep 12 04:00:56 someone remember where is defined that bluez-utils goes into rootfs? Sep 12 04:01:56 handhelds-common? Sep 12 04:03:58 I have to check my copies of it then Sep 12 04:05:06 ~lart zombie support in linux Sep 12 04:05:06 * ibot farts in zombie support in linux's general direction Sep 12 04:05:46 I have 6 firefoxes which died and are not removed - so I cannot start new one ;( Sep 12 04:07:11 kio_http x6 too Sep 12 04:07:25 7410 pts/2 D+ 0:00 | | \_ host oe.handhelds.org Sep 12 04:07:44 when linux will give use possibility to remove that processes..... Sep 12 04:09:05 anything which try to resolve name die.. Sep 12 04:09:19 fscking OS Sep 12 04:09:29 bye - reboot time Sep 12 04:09:34 hrw|work: just add this feature if you need it, it's open-source :-) Sep 12 04:11:51 hail zecke Sep 12 04:15:21 hey Sep 12 04:15:40 ging over the http://www.freebsd.org/donations/wantlist.html Sep 12 04:15:50 porting *BSD to ARM seems popular... Sep 12 04:17:55 tsk Sep 12 04:17:59 who needs freebsd? Sep 12 04:18:06 openbsd already runs on the Z Sep 12 04:18:33 netbsd has arm support since... Sep 12 04:26:01 re again Sep 12 04:53:30 :) Sep 12 04:54:20 hmm.. one attempt more and I will gaveup with any OE build work Sep 12 04:54:27 13:52 hrw@home:0912$ bitbake -b ../../oe/org.openembedded/packages/meta/task-bootstrap.bb Sep 12 04:54:47 and it ALWAYS build it - just like when there are not stamps.. Sep 12 04:55:21 ....? Sep 12 04:56:05 AvengerMoJo: it should build once Sep 12 04:56:24 only one time - next ones should just print "started\ncompleted\n" Sep 12 04:57:20 right ... Sep 12 04:57:35 does it stamp? Sep 12 04:57:42 is that clock problem? Sep 12 04:58:12 FOUND! Sep 12 04:58:18 ~lart sourcepkg.bbclass Sep 12 04:58:18 * ibot squishes sourcepkg.bbclass like a bug Sep 12 04:58:45 ~lart sourcepkg.bbclass once more Sep 12 04:58:46 * ibot rm -rf's sourcepkg.bbclass once more Sep 12 05:13:53 looks like sourcepkg is not usable when someone want to use build for few builds Sep 12 05:14:26 that bad? Sep 12 05:16:19 on my system whole build goes ok when I'm not using that class Sep 12 05:16:34 if I inherit it then it rebuild everything each time Sep 12 05:22:45 .... Sep 12 05:22:46 hi all Sep 12 05:22:58 hi Dirk Sep 12 05:23:12 does it over ride the stamp check? Sep 12 05:23:26 hi dirk Sep 12 05:23:47 hi R, hrw|work Sep 12 05:24:22 AvengerMoJo: looks like Sep 12 05:24:37 03rpurdie 07org.oe.dev * r722a1383... 10/packages/linux/linux-openzaurus_2.6.13-mm1.bb: linuz-oz-2.6: Correct right shift key handling on cxx00. Update patch order to better reflect mainline/pushed status and clean up the .bb a bit Sep 12 05:25:34 hrw|work: .... Sep 12 05:26:27 oh rpurdie .. .. who is that :) I was going to report the backspace is not correct either Sep 12 05:26:28 :) Sep 12 05:27:11 AvengerMoJo: me. I only touch the 2.6 kernels though... Sep 12 05:27:37 AvengerMoJo: on what machine etc? Sep 12 05:29:20 C3000 Sep 12 05:30:08 AvengerMoJo: 2.4 kernel I presume? Sep 12 05:30:11 AvengerMoJo: ever noticed file calles keymap-2.4.map in /etc/? Sep 12 05:31:12 hrw|work: yeah ... I was wondering that is the keymap problem ... but I havn't look at it hehe :) being lazy .. Sep 12 05:31:24 hrw|work: I was looking at my wireless card first hehe Sep 12 05:31:50 mt log --diffs --no-merges = 2.5 mln of lines, 92M file Sep 12 07:28:45 * chouimat|tired is back. Sep 12 07:31:20 ~lart the person which broken libopietooth1 Sep 12 07:31:20 * ibot squeezes the person which broken libopietooth1 till the person which broken libopietooth1 turns blue like papa smurf Sep 12 07:32:25 zecke: monotone add libopietooth1;monotone commit;monotone sync - could you? Sep 12 07:37:27 hrw|work: yeah, once I'm home :} Sep 12 07:37:41 hrw|work: did I delete the complete directory? Sep 12 07:38:14 zecke: no - you forgot to push common file Sep 12 07:38:20 so bt stuff fails Sep 12 07:38:33 opie bt stuff Sep 12 07:38:34 ~fishslap Sep 12 07:38:54 hmm mt diff, looked good when I pushed the change Sep 12 07:39:55 ;) Sep 12 07:40:55 in meantime I will build gpe-image then Sep 12 08:21:40 cu Sep 12 08:21:48 hrw|gone: cya later Sep 12 08:28:18 mickey|meetig: I'm going to pick up the tosa tomorrow Sep 12 08:28:42 hey ar Sep 12 08:34:10 zecke: I'm poking about in fetch.py. It would be nice to make these plugable modules, wouldn't it... Sep 12 08:34:30 zecke: I'm part way through adding git support to bitbake Sep 12 08:34:40 cool Sep 12 08:34:59 Good evening. Sep 12 08:35:16 hi sirfred Sep 12 08:35:29 Hello, RP Sep 12 08:36:06 RP: yes, that would rock Sep 12 08:36:17 RP: we could have a subdir fetchers/ Sep 12 08:36:40 RP: and automatically add each fetcher on startup (or we use __init__.py to do it) Sep 12 08:37:34 zecke: Agreed. I'm not sure my python it up to it though. The adding on startup is what causes me concern :) Sep 12 08:38:16 RP: I would cheat by looking at bittest Sep 12 08:38:22 RP: it loads a module by name Sep 12 08:39:04 zecke: I might have to do that. I don't do much python so finding example code to pinch is more difficult ;-) Sep 12 08:40:17 RP: either we hardcode the names of the fetchers Sep 12 08:40:22 RP: or we do a listdir Sep 12 08:40:36 hardcoding sounds more safe for now though Sep 12 08:40:47 I'd agree Sep 12 08:41:00 RP: I'm slowly heading home, we could do that rework tonite Sep 12 08:41:02 I'm too used to initcalls in kernel modules :) Sep 12 08:41:24 iterating over sections in a executable is like a filesystem ;) Sep 12 08:41:32 zecke: I'm out for most of the evening a least earlier on... Sep 12 08:41:46 for each in ls fetchers/*: register_module Sep 12 08:42:34 Its asking for trouble :) Sep 12 08:43:22 we use a dynamic language, we want trouble Sep 12 08:43:54 zecke: Does anything exist in bitbake to let me execute a command with a partial bitbake enviornment with variables like PATH set? Sep 12 08:44:21 no Sep 12 08:44:26 'export PATH=%s; %s' % (data.expand('${PATH}', d), os.path.join(data.expand('${STAGING_BINDIR}', d), "git")) seems a touch evil :) Sep 12 08:44:29 the closest is a shelpp method Sep 12 08:45:04 RP: we have even more dangerour use of system()... Sep 12 08:45:35 damn I feel depressed... Sep 12 08:45:59 about bitbake? Sep 12 08:46:06 no generally ;) Sep 12 08:46:26 hi Sep 12 08:46:27 ah. I'm not sure I can help you there ;-) Sep 12 08:46:31 hi hrw Sep 12 08:46:48 RP: you aren't the emacs physcist? Sep 12 08:47:22 I am a physicist as it happens but not emacs related... Sep 12 08:49:14 zecke: is it you or mickeyl the one who will update opie in OE to 1.2.1? Sep 12 08:50:27 hrw: my script will Sep 12 08:50:50 o Sep 12 08:50:52 brb Sep 12 08:51:25 CP|Dresden: FAB13 in Dresden? Sep 12 08:55:37 zecke: dont forget about libmailwrapper;) Sep 12 08:57:31 hrw: I hope to have updated my script... Sep 12 09:01:37 zecke: I will look if there will be something missed Sep 12 09:01:55 hrw: better do so Sep 12 09:03:23 zecke: btw - one bitbake related question.. if I add few tasks extra then how to get them work ONLY once? currently when I inherit 'sourcepkg' then any package is rebuild each time - nevermind does it was built or not Sep 12 09:05:04 hrw: check if stamps for these tasks are created Sep 12 09:06:03 they are Sep 12 09:06:12 + Sep 12 09:06:25 ignore that '+' Sep 12 09:06:54 I'm hungry and can not say anything useful ;) Sep 12 09:06:57 No, I want to add someting Sep 12 09:07:05 I refuse to ignore the '+' Sep 12 09:07:15 normally a task can ask to not generate stamps Sep 12 09:07:18 zecke: get something to eat then Sep 12 09:07:35 zecke: taskname[nostamp] ="1" - right? Sep 12 09:07:45 something like that, right Sep 12 09:08:30 cleaning keyboard is something which must be done from time to time Sep 12 09:15:56 hey mickeyl Sep 12 09:17:54 mithro: hey Sep 12 09:18:08 mickeyl: I'm heading home now Sep 12 09:18:54 03mickeyl 07org.oe.dev * r720a8c74... 10/ (6 files in 4 dirs): qt: revert classes and applications back to explicit DEPENDS to prevent circular dependencies Sep 12 09:19:15 mickeyl: bug from me? Sep 12 09:19:30 damn my quality sucks... Sep 12 09:21:09 mickeyl: hmm if a provider depends on itself, is this really a problem (not too nice anyway) Sep 12 09:22:06 zecke: no, that was just a bug I've introduced by wanting to be clever and making everything that inherits qt3x11 automatically depends on qt3x11 Sep 12 09:22:09 hi mithro Sep 12 09:22:12 hmm Sep 12 09:22:41 zecke: did the mailman drop of a large sum of money yet? Sep 12 09:22:57 i could've workarounded that, but in general, it's perhaps better to have less implicit things going on Sep 12 09:24:02 zecke: hmm, I think I forgot to put the 'Make zecke and koen filthy rich' goal in to the mail Sep 12 09:24:11 hey what about me, btw.? Sep 12 09:24:24 i don't care too much about money, but it'd be nice anyway :D Sep 12 09:26:19 mickeyl: I would share with you Sep 12 09:26:39 I've lost my notebook to my mother... so I need to setup my mail client on my desktop... Sep 12 09:26:45 whenever that will happen... Sep 12 09:26:47 cya later Sep 12 09:27:45 mickeyl: what do you think of renaming conf/distro/openzaurus.conf to conf/distro/openzaurus.inc? Sep 12 09:28:18 mickeyl: this would stop people from using DISTRO="openzaurus" (it would fail much sooner) and it would make the intention of that file clear Sep 12 09:30:52 mickeyl: I have a commit ready to go but am waiting for your approval Sep 12 09:31:26 JustinP: go ahead with that thing for now. The reasing behind that is not quite correct, but the effect is probably ok Sep 12 09:32:20 to elaborate: openzaurus.conf is a perfectly valid DISTRO configuration. It is not prone to fail, but - you are right - it depends on the actual state of metadata in OE and is probably misleading users who think it will always work. Sep 12 09:32:48 in the light of consistence I'd be happy if koen would do the same Sep 12 09:33:05 mickeyl: I don't see how my reasoning if off. People have *always* been told to use a versioned DISTRO. The unversioned DISTRO pretty much never works. Sep 12 09:33:38 I suppose we could move a bunch of the stuff in -3.5.4 into the main file....but it just makes sense to me to force people to use a version since this is how we currently work. Sep 12 09:35:04 in fact, it looks like nearly everything in -3.5.4 could be moved into the main file Sep 12 09:35:15 * emte still thinks that the unversioned should point to last release Sep 12 09:35:25 which would also be ok Sep 12 09:35:36 and that app develpers should be instructed to use it Sep 12 09:35:36 *except* that we don't have a 3.5.3 conf Sep 12 09:36:12 app/casual users Sep 12 09:36:21 in general, i don't like adding thousands of safety nets to make things idiot proof - just to wait until someone comes up with even more idioty to break the system and forcing us into adding more nets. Sep 12 09:36:27 but let's leave it here. go ahead with your change Sep 12 09:36:29 and I thought that we were going to just make a snapshot of OE when a release goes out for people to use... Sep 12 09:36:40 ok Sep 12 09:37:21 yeah a snapshot works as well, but what happens when tehy want updates packages to support something? Sep 12 09:37:34 they update to the newest version Sep 12 09:37:36 nice grammer ... Sep 12 09:37:42 heh Sep 12 09:37:43 then the snapshot is invalid Sep 12 09:37:44 yeah... Sep 12 09:38:06 well, we'r enot currently set up to support multiple release with OE Sep 12 09:39:04 we'd need multiple branches, one for each release, and we'd have to make changes to the "stable" one when we find a fix which increases stability Sep 12 09:39:11 and we'd need to "release" new ipks Sep 12 09:39:13 we will never be unless we a) stop development, or b) branch releases off, or c) keep all the stuff around permanently Sep 12 09:39:48 *nod*. branches are on the agenda soon Sep 12 09:40:00 I don't think we have the resources for supporting the old releases, honestly. Sep 12 09:40:13 versions are supposed to be hard set for specific package release tho right? so any new packages after that point would be availible to build Sep 12 09:40:40 i am not seeing the reason for a branch Sep 12 09:40:43 If we can find people who are interested in maintaining the releases...."backporting" changes from the dev tree.... Sep 12 09:41:15 emte: as long as we keep around *every single version* of a package (or at least the version in the last release) then we can do that. Sep 12 09:42:00 emte: the problem is that we're not really set up for people to just update whenever they want. We're in more if a FC mode right now (blech) not Gentoo. Sep 12 09:42:23 i see what your saying, your supposed to be able to rollback and override localy specific packages in monotone, no? Sep 12 09:42:37 We make a release and that's what you get until the next release unless you want to go with dev, which is possibly more unstable by definition. Sep 12 09:42:54 emte: I'm not sure what you're saying. Sep 12 09:43:22 emte: I'm saying that with the way we currently do things, we tend to get rid of old versions of packages in favor of new ones. (which keeps out metadata pretty clean) Sep 12 09:43:36 if i wanted gcc-2.95 i could tell monotone to go back in history and get it can i not? Sep 12 09:43:45 re Sep 12 09:43:57 JustinP: you will break my setup ;( Sep 12 09:43:58 emte: that would be tough, because it never was part of OE :D Sep 12 09:44:12 bad example, bit if it was in it Sep 12 09:44:15 heh Sep 12 09:44:15 but* Sep 12 09:44:15 emte: assuming that was in our repo, yes, you could checkout that revision Sep 12 09:44:23 hrw: what will break your setup? Sep 12 09:44:31 * hrw always use DISTRO="openzaurus" - only versioned builds use releases Sep 12 09:44:48 hrw: DISTRO="openzaurus" is and always has been broken Sep 12 09:44:56 especially right now Sep 12 09:44:57 mickeyl: will we add altboot to collie? Sep 12 09:45:00 so you rollback for that release, and tell monotone to not overwrite those significant packages Sep 12 09:45:07 JustinP: really? it works for me Sep 12 09:45:13 that will build gcc-4.0.0, which will *not* work Sep 12 09:45:26 hrw: when's the last time you updated and remove dyour tmpdir? Sep 12 09:45:34 JustinP: let me check Sep 12 09:46:07 0912 is oz-3.5.4 Sep 12 09:46:18 03justinp 07org.oe.dev * r8c6c13a2... 10/conf/distro/ (openzaurus.conf openzaurus.inc openzaurus-3.5.4.conf): conf/distro/openzaurus: rename openzaurus.conf to openzaurus.inc as it's not a valid DISTRO target. This causes builds to fail *before* anything is built. Sep 12 09:46:41 hrw: because I am 100% certain that the recent CSL compiler de-preferencing killed builds from scratch due to gcc-4.0.0 breaking things. Hence my fixes a while ago. But the fix I added was in -3.5.4 Sep 12 09:46:42 JustinP: my 0830 build is oz based - but it will be obsolete soon Sep 12 09:46:58 would need a script i guess to automate that, but would allow app/users to work from a stable release set Sep 12 09:47:14 JustinP: maybe you could also bump binutils to 2.16? Sep 12 09:47:36 as opposed to a devel set Sep 12 09:48:03 anway i need ro run to my phisio appt Sep 12 09:48:04 hrw: I'm not really in a position to know if that works....does it? Sep 12 09:48:54 JustinP: we need to test it - but after 3.5.4 will be best Sep 12 09:49:13 hrw: yes - although I'm not sure yet if I like its insistence on loop images. Sep 12 09:49:33 hrw: besides, I'm working on a graphical boot manager, but that won't make it for 3.5.4 Sep 12 09:49:39 JustinP: currently my builds are not quite tested yet Sep 12 09:49:47 mickeyl: angstrom-bootsomething? Sep 12 09:49:49 hrw: do you think we should ship collie with altboot as default for 3.5.4? Sep 12 09:49:52 hrw: *nod* Sep 12 09:50:12 mickeyl: thats the question for CoreDump rather Sep 12 09:50:20 hrw: Matthias ? Sep 12 09:50:24 s/hrw// Sep 12 09:50:44 mickeyl: "insistence on loop images" Sep 12 09:50:44 ? Sep 12 09:50:45 mickeyl: he use it in his branch and users found it good. Sep 12 09:50:57 mickeyl: I see the loop images as a feature, not a limitation. Sep 12 09:51:00 mickeyl: and we will kill those 'I want root on sd' users Sep 12 09:51:22 mickeyl: It's pretty simple to make it, say, pivot to a different partition instead Sep 12 09:51:24 JustinP: i find that limiting. i want to boot from a filesystem, not only from a loop image. Sep 12 09:51:34 mickeyl: anyway if altboot then another Q: do we need/should/be provide ext2 images Sep 12 09:51:44 mickeyl: so boot from a filesystem...what stops you? I'm booting from /dev/hda6.... Sep 12 09:52:04 JustinP: so that works already with altboot? Sep 12 09:52:26 hrw: IMHO the default install method for collie should be to install to SD. Our FS images are too large for their poor little flash memory. Sep 12 09:52:29 mickeyl: of course! Sep 12 09:52:42 mickeyl: it alreayd pivots to hda1 for spitz by default Sep 12 09:52:58 mickeyl: it's very simple to add another option for another partition Sep 12 09:53:08 JustinP: hmm... even better. last time i looked it could just boot loops Sep 12 09:53:27 mickeyl: I helped make it work on Spitz *quite* a while ago Sep 12 09:53:37 mickeyl: been using it forever Sep 12 09:53:51 mickeyl: and I love the file image functions for testing new images ^_^ Sep 12 09:53:52 (anyway, I would like it to be a bit more generic. I'd like to make it querying for available boot devices @ runtime, not via configuration files) Sep 12 09:53:59 JustinP: collie is awful. I'm nearly not able to get any usefull images fit in that 13M flash space Sep 12 09:54:05 mickeyl: well, that would also be good :-) Sep 12 09:54:35 hrw: exactly. And we have lots and lots of users fail to update/upgrade due to space issues Sep 12 09:54:36 JustinP: angstrom-boot will do that :) Sep 12 09:54:48 hrw: which is why we should make root on SD the "default" Sep 12 09:55:03 mickeyl: :-) Sep 12 09:55:14 oh.. I reminded one thing... Sep 12 09:55:20 mickeyl: as long as you also support loop images I'll be happy Sep 12 09:55:21 anyway, i need to go home. have been staying in the office for long enough Sep 12 09:55:28 mickeyl: w8 - one thing Sep 12 09:55:44 mickeyl: could you test one small patch for opie? Sep 12 09:56:08 sure, what? Sep 12 09:56:21 mickeyl: http://opie-bugs.oszine.de/view.php?id=1695 - I forgot to apply it before 1.2.1 Sep 12 09:56:52 ah, that one. looks good. we should apply it for all vga models Sep 12 09:56:58 it needs to go in OE though Sep 12 09:57:01 mickeyl: sure Sep 12 09:57:05 i can apply it upstream, but i won't move the tag Sep 12 09:57:14 mickeyl: its applied for >=480 screens now Sep 12 09:57:26 mickeyl: I can apply it upstream too Sep 12 09:57:34 ;) Sep 12 09:57:37 i _think_ know why it doesn't work on small screens, but fixing it would mean to poke into qte which i refuse to do Sep 12 09:57:57 bl Sep 12 09:57:58 bbl Sep 12 09:58:02 bye mickeyl Sep 12 10:02:20 make Sep 12 10:02:24 heh Sep 12 10:02:31 arf... wrong window ^^' Sep 12 10:02:57 .... Sep 12 10:03:03 my touchscreen isn't working.... Sep 12 10:46:55 kergoth: how new Tatu album? Sep 12 10:47:30 imho its quite good. feels a bit more refined than the last one, like those girls are growing up or something. *gasp* :P Sep 12 10:47:53 kergoth: finally reached 20/21 years old Sep 12 10:47:57 hehe Sep 12 10:49:52 they're that old already? Sep 12 10:49:57 damn time passes fast Sep 12 10:50:06 gb2: 84,85 born Sep 12 10:50:28 * gb2 boggles Sep 12 10:50:41 i still think people born in the 80s are weird Sep 12 10:50:51 gb2: how old you are? Sep 12 10:50:54 hahaha Sep 12 10:50:59 turned 26 on saturday Sep 12 10:51:07 born in '79 ;) Sep 12 10:51:15 young one Sep 12 10:51:16 '84 Sep 12 10:51:19 76 Sep 12 10:51:23 heh Sep 12 10:51:26 80 here Sep 12 10:51:33 my gf was born in '81 Sep 12 10:52:06 kergoth: I have a brother at your age Sep 12 10:52:08 my girlfriend was born in '87 Sep 12 10:52:13 03freyther 07org.oe.dev * rf3227622... 10/packages/libopietooth/libopietooth1.inc: Sep 12 10:52:13 libopietooth1.inc: Sep 12 10:52:13 -add missing file for libopietooth1. Thanks to hrw who spotted Sep 12 10:52:13 my error/failure Sep 12 10:52:30 zecke: danke Sep 12 10:58:03 I Sep 12 10:58:11 I am doing the big renaming now Sep 12 11:00:56 cu Sep 12 11:01:00 * mickeyl crosses finger Sep 12 11:01:02 s Sep 12 11:01:03 cu hrw Sep 12 11:01:44 mickeyl: actually I just updatet the script Sep 12 11:05:23 hmm bitkeeper was faster Sep 12 11:06:05 hardly Sep 12 11:06:06 :D Sep 12 11:06:16 OT: has evolution better gnupg support? Sep 12 11:06:22 mickeyl: in the bk mv case Sep 12 11:06:29 zecke: sure it was Sep 12 11:06:31 in every case Sep 12 11:06:35 i was kidding Sep 12 11:06:57 mickeyl: hehe, I do not find mt too bad though Sep 12 11:07:02 I Sep 12 11:07:11 I have ordered a Merkle Tree Companion Chip Sep 12 11:07:18 :D Sep 12 11:07:39 good investment Sep 12 11:08:03 mickeyl: I had to decide between Java or Merkle tree acceleration Sep 12 11:11:23 mickeyl: 'require' or any other name for the new keyword? Sep 12 11:13:03 require Sep 12 11:13:10 hmm perhaps something more explicit stating the relation to include would be good? Sep 12 11:13:12 how about Sep 12 11:13:29 hmm no Sep 12 11:14:13 perhaps go with gnu make syntax. 'include' always fails (in our case raises a parser exception) if the file cant be loaded. '-' in front of any gmake line indicates that you dont care if it fails. Sep 12 11:14:25 yeah, i think that would be best Sep 12 11:14:52 always intended to have that, never got around to it Sep 12 11:15:41 php has include and require Sep 12 11:15:58 and require_once Sep 12 11:16:23 hmmmkay $DISTRO.conf Sep 12 11:16:53 giving that the usual use of include will be include and -include will be more seldomly used, i vote for changing the meaning of include and adding -include instead of a completely new keyword Sep 12 11:16:58 in the past i've done builds without any distro specified at all. either allow that, or set the default DISTRO to generic Sep 12 11:17:27 kergoth: good point Sep 12 11:17:46 goodbye cia Sep 12 11:19:00 mickeyl: http://www.oesf.org/forums/index.php?showtopic=11751&view=findpost&p=75718 <- orange line 'fix' for X on tosa Sep 12 11:19:48 hmm Sep 12 11:19:55 thanks, but... where is the patch? Sep 12 11:19:58 fixed in a zaurus way Sep 12 11:20:01 i only see a bz2 Sep 12 11:20:12 who needs patches? Sep 12 11:20:20 hmm thought you said it was rotting in the kdrive bugtracker Sep 12 11:20:22 must have dreamed that Sep 12 11:20:28 everyones shuffles binaries around in a rom nowadays Sep 12 11:20:36 *cough* Sep 12 11:20:48 03freyther 07org.oe.dev * r7b6e4980... 10/packages/ (556 files in 175 dirs): Sep 12 11:20:48 Opie 1.2.1: Sep 12 11:20:48 The great move and version bumping. I hope my script Sep 12 11:20:50 worked out well... I will start a build now. Sep 12 11:20:52 hrw: mailwrapper was moved Sep 12 11:20:57 mickeyl: yeah, I thought it was kdrive bugzilla, but it turned out to be oesf Sep 12 11:21:03 koen: ah k. Sep 12 11:21:35