| 00:04.55 | Neil | ha ha, new app written |
| 00:04.59 | Neil | total password awareness |
| 00:05.03 | Neil | now to package and build with OE |
| 00:08.39 | gb2 | hrm. |
| 00:09.17 | gb2 | I wonder how i can tell from within anonymous python in a .oe if i'm the top level OE, or if i'm being included. |
| 00:09.40 | pb_ | I don't think you can, at the moment. |
| 00:09.55 | gb2 | that means my plan fails unfortunately |
| 00:10.16 | pb_ | It probably wouldn't be too hard to arrange for you to be able to do "if ${FILE} != ${TOP_FILE}:", or something |
| 00:10.32 | gb2 | I have an idea which might work. |
| 00:14.54 | gb2 | ERROR: executing anonymous function: 'module' object has no attribute 'SkipPackage' |
| 00:14.55 | gb2 | odd |
| 00:16.43 | pb_ | it's oe.parse.SkipPackage |
| 00:18.04 | gb2 | yeah. |
| 00:18.10 | gb2 | i did the wrong machine trick again |
| 00:18.18 | pb_ | ah, heh |
| 00:19.32 | CIA-5 | 03mickeyl 07 * r1.550.1.1 10oe/classes/package.oeclass: add automatic package dependencies based on dangling symlinks - patch courtesy pb_.\ |
| 00:19.33 | CIA-5 | 03mickeyl 07 * r1.552 10oe/: |
| 00:19.33 | CIA-5 | Merge http://openembedded@openembedded.bkbits.net/oe\ |
| 00:19.33 | CIA-5 | into gandalf.tm.informatik.uni-frankfurt.de:/usr/local/projects/oe\ |
| 00:19.34 | pb_ | gb2: that patch I just checked in should give you automatic dependencies of -dev packages on the corresponding library, with a bit of luck. |
| 00:20.04 | gb2 | cool |
| 00:20.26 | reenoo_ | ...for tomorrow |
| 00:20.28 | reenoo_ | doh |
| 00:20.31 | pb_ | heh |
| 00:20.50 | CIA-5 | 03gbritton 07 * r1.553 10oe/conf/oe.conf: package libdir/.o files too\ |
| 00:23.09 | reenoo_ | 'night all |
| 00:24.31 | pb_ | night reenoo_ |
| 00:24.34 | gb2 | foo |
| 00:24.44 | treke|home | gb2: we getting 3.4 any time soon? |
| 00:24.49 | gb2 | i want a logical xor |
| 00:24.56 | gb2 | treke: tonight if i'm lucky |
| 00:25.07 | treke|home | sweet |
| 00:25.17 | pb_ | gb2: isn't that just "!="? |
| 00:25.21 | gb2 | you can't compare booleans in python with != ? |
| 00:25.27 | gb2 | pb: i figured it was |
| 00:25.30 | treke|home | remind me to bring you some scotch next time i'm in mass :) |
| 00:25.31 | gb2 | but it didn't seem to work |
| 00:25.32 | gb2 | oh |
| 00:25.33 | gb2 | i know why |
| 00:26.25 | pb_ | wow, what crazy stuff is pcmcia-cs trying to do in its configure script? |
| 00:26.29 | pb_ | chmod: changing permissions of `/etc/CORBA': Operation not permitted |
| 00:26.30 | pb_ | chmod: changing permissions of `/etc/Muttrc': Operation not permitted |
| 00:27.02 | reenoo_ | heh |
| 00:27.37 | gb2 | <PROTECTED> |
| 00:27.37 | gb2 | <PROTECTED> |
| 00:27.43 | gb2 | EXCEPT (gcc-uclibc, linux-uclibc) |
| 00:27.54 | gb2 | the indentation is actually correct, it pasted odd |
| 00:28.19 | pb_ | gb2: re.match returns a Match object, not a bool |
| 00:28.24 | gb2 | oh? |
| 00:28.27 | gb2 | grr. |
| 00:28.45 | gb2 | object or None |
| 00:28.48 | pb_ | you'll need to test against None, or otherwise force it to a boolean |
| 00:29.00 | gb2 | so i want logical xor |
| 00:29.01 | gb2 | ;) |
| 00:29.28 | pb_ | heh |
| 00:29.40 | gb2 | invalid syntax |
| 00:29.41 | gb2 | odd |
| 00:31.31 | CIA-5 | 03mickeyl 07 * r1.552.1.1 10oe/classes/base.oeclass: respect existing value for RDEPENDS_pkg - patch courtesy pb_.\ |
| 00:31.33 | CIA-5 | 03mickeyl 07 * r1.554 10oe/: |
| 00:31.33 | CIA-5 | Merge http://openembedded@openembedded.bkbits.net/oe\ |
| 00:31.33 | CIA-5 | into gandalf.tm.informatik.uni-frankfurt.de:/usr/local/projects/oe\ |
| 00:32.00 | CIA-5 | 03mickeyl 07 * r1.555 10oe/classes/package.oeclass: also write out ALLOW_EMPTY into subpackage metadata file - patch courtesy pb_.\ |
| 00:33.04 | treke|home | is this the gateway at work? |
| 00:33.10 | pb_ | yup |
| 00:33.22 | treke|home | should add a seperate user for the gateway |
| 00:33.37 | treke|home | I'll poke mickey|zzZZzz about that tomorrow |
| 00:33.41 | pb_ | good idea |
| 00:34.06 | gb2 | NOTE: package patcher-native-1.0-r0: task do_unpack started |
| 00:34.06 | gb2 | NOTE: exceptions.KeyboardInterrupt: while evaluating: |
| 00:34.06 | gb2 | os.popen('uname -m').read().strip() |
| 00:34.06 | gb2 | NOTE: package patcher-native-1.0-r0: task do_unpack completed |
| 00:34.06 | gb2 | NOTE: package patcher-native-1.0-r0: task do_patch started |
| 00:34.07 | gb2 | heh |
| 00:34.18 | pb_ | heh |
| 00:35.09 | pb_ | I wonder what's eating that exception. |
| 00:36.32 | gb2 | something that changed today |
| 00:36.38 | gb2 | i didn't have this yesterday |
| 00:37.17 | pb_ | odd |
| 00:40.13 | gb2 | cool. my idea works. |
| 00:40.19 | gb2 | i can check PN |
| 00:40.26 | pb_ | ah, cool |
| 00:40.26 | gb2 | successfully |
| 00:40.34 | gb2 | <PROTECTED> |
| 00:40.34 | gb2 | <PROTECTED> |
| 00:40.34 | gb2 | <PROTECTED> |
| 00:40.38 | gb2 | raise oe.parse.SkipPackage... |
| 00:41.25 | pb_ | ingenious. which file does that go in? |
| 00:41.37 | gb2 | gcc_3.4.0.oe |
| 00:41.44 | gb2 | so it works for all the gcc sub-oe's |
| 00:41.49 | pb_ | right, I see. |
| 00:41.53 | pb_ | sounds cool |
| 00:43.20 | pb_ | the only case that might lose would be if someone wanted to do a canadian cross of i386-linux x arm-linux x arm-linux-uclibc, or something. |
| 00:43.43 | gb2 | yeah. |
| 00:44.04 | pb_ | though actually, I guess even that should be ok. |
| 00:44.14 | pb_ | TARGET_OS would still be uclibc in that case, it's just HOST_OS that would be different |
| 00:44.33 | gb2 | # |
| 00:44.33 | gb2 | # TODO: For shared setups with both uclibc and glibc, libstdc++ and the |
| 00:44.33 | gb2 | # fortran (libg2c, libfrtbegin) will need to be moved into separate |
| 00:44.33 | gb2 | # per-compiler paths. Shared libraries will also need some |
| 00:44.33 | gb2 | # separation. This will also require changes to the dynamic linkers # for the two libc's to support selection of appropriate libraries. |
| 00:44.35 | gb2 | # |
| 00:44.37 | gb2 | # For now, we will skip building of a gcc package if it is a uclibc one |
| 00:44.39 | gb2 | # and our build is not a uclibc one, and we skip a glibc one if our build |
| 00:44.41 | gb2 | # is a uclibc build. |
| 00:44.43 | gb2 | # |
| 00:44.45 | gb2 | that's what i've got in here |
| 00:44.50 | pb_ | right |
| 00:47.03 | *** join/#oe woglinde (woglinde@pD95D1C09.dip.t-dialin.net) |
| 00:49.42 | *** join/#oe _stigger (~jeff@203.41.111.76) |
| 00:49.50 | pb_ | eh, pcmcia-cs.oe really does have "chmod ug+rw /etc/*" in its do_configure() |
| 00:49.54 | pb_ | what joker added that? |
| 00:50.20 | woglinde | hm because some scripts nee the rights |
| 00:50.31 | woglinde | but thats bullshit anyway |
| 00:51.00 | *** join/#oe badalex (~badalex@dialup-4.228.60.189.Dial1.Denver1.Level3.net) |
| 00:53.41 | pb_ | heh. well, it's history. |
| 00:53.56 | treke|home | hmmm |
| 00:54.03 | CIA-5 | 03mickeyl 07 * r1.1209.1.1 10packages/pcmcia-cs/pcmcia-cs_3.2.7.oe: remove spurious chmod call from do_configure - patch courtesy pb_.\ |
| 00:54.03 | CIA-5 | 03mickeyl 07 * r1.1212 10packages/: |
| 00:54.03 | CIA-5 | Merge bk://openembedded@openembedded.bkbits.net/packages\ |
| 00:54.03 | CIA-5 | into gandalf.tm.informatik.uni-frankfurt.de:/usr/local/projects/packages\ |
| 00:56.45 | treke|home | pb, oddly enough there is no log for pcmcia-cs |
| 00:57.41 | gb2 | pb: heh, i was wrong about grub |
| 00:57.42 | treke|home | oh wait |
| 00:57.51 | treke|home | the bkbits web interface was just too complicated for me |
| 00:57.54 | gb2 | since it is "...-linux" that'll match "...-linux-uclibc" |
| 00:58.05 | gb2 | since it isn't anchored to the end of string |
| 00:58.06 | pb_ | gb2: aha, right |
| 00:58.37 | treke|home | pb_: mickey did it :p |
| 01:01.00 | gb2 | to be a bit more consistant |
| 01:01.20 | pb_ | NOTE: package task-bootstrap-1.0-r0: task do_package_ipk started |
| 01:01.21 | pb_ | *** Error: CONTROL/control is missing field Source |
| 01:01.22 | pb_ | doh |