**** BEGIN LOGGING AT Fri Oct 11 02:59:58 2013 Oct 11 09:12:39 build #396 of cobalt is complete: Failure [failed compile_5] Build details are at http://buildbot.openwrt.org:8010/builders/cobalt/builds/396 Oct 11 10:16:30 build #375 of rb532 is complete: Failure [failed shell_12] Build details are at http://buildbot.openwrt.org:8010/builders/rb532/builds/375 Oct 11 10:18:22 build #375 of ppc44x is complete: Failure [failed shell_12] Build details are at http://buildbot.openwrt.org:8010/builders/ppc44x/builds/375 Oct 11 13:12:33 build #353 of sibyte is complete: Failure [failed shell_12] Build details are at http://buildbot.openwrt.org:8010/builders/sibyte/builds/353 Oct 11 13:21:43 kaloz r38363 packages/net/coova-chilli/ Config.in Makefile * [coova-chilli]: add some compile time options (more to come) Oct 11 13:51:12 jow_laptop: is luci.json meant to handle \u00xx escapes? or is that a known problem? http://pastebay.net/1320260 Oct 11 13:52:03 maybe a bug Oct 11 13:57:47 with \xc8\x92 instead of \u0212 (the utf8 repr of \u0212) it doesn't strip the character, it just fails to parse outright) Oct 11 14:05:57 let me take a look Oct 11 14:07:37 (that web page has Ȓ which is html escaping a character that did get printed to my screen, and in the text box before I pressed submit) Oct 11 14:08:12 well the parser currently does not handle codepoints > 0xFF Oct 11 14:08:39 I thought about replacing luci-lib-json with a binding to json-c anyway Oct 11 14:20:27 seeing as json-c is required anyway, that would seem to be an excellent choice :) Oct 11 14:23:11 sure about > 0xff? it doesnt' deal with \u00e6 either, Oct 11 14:24:11 probably > 0x7f Oct 11 14:24:25 yeah, 0x80 is where it switches to 2 bytes Oct 11 14:44:44 karlp: http://luci.subsignal.org/~jow/luci-json-unicode-decoder.patch Oct 11 14:44:57 # lua -e 'lj=require "luci.json"; print(lj.decode([[{"e":"ddd\u0212www"}]]).e)' Oct 11 14:45:00 dddȒ0212www Oct 11 14:54:50 karlp: oops, forgot a sub(5), patch is updated at the same location and now works peroply Oct 11 14:55:08 build #408 of orion is complete: Failure [failed compile_8] Build details are at http://buildbot.openwrt.org:8010/builders/orion/builds/408 Oct 11 15:42:58 kaloz r38364 trunk/target/linux/ar71xx/image/Makefile * [ar71xx]: make the generated images actually flashable on the Unifi devices. Oct 11 16:08:42 build #363 of avr32 is complete: Failure [failed compile_5] Build details are at http://buildbot.openwrt.org:8010/builders/avr32/builds/363 Oct 11 16:10:41 build #398 of lantiq is complete: Failure [failed compile_5] Build details are at http://buildbot.openwrt.org:8010/builders/lantiq/builds/398 Oct 11 16:36:38 I hate when someone does complain about package breakage, but he/she does not provide configuration for it. Especially when it does work for me. :-/ Oct 11 16:54:56 jow_laptop: that looks a lot better, thanks :) (was that really easier than just deferring everything to json-c?) Oct 11 17:44:54 karlp: yes Oct 11 17:48:12 jow_laptop: I see that a lot of packages uses "depends on" in /config macros. Wiki is not clear about that. A few examples shows "depends" with "on" and without it. I suppose it should be just "depends", right? Oct 11 17:48:33 slachta: 'depends on' is the right syntax Oct 11 17:51:20 nbd: okay, I was a little bit confused (I know, I can look at the source) - http://wiki.openwrt.org/doc/devel/dependencies this says "depends" and this says both in examples http://wiki.openwrt.org/doc/devel/packages Oct 11 17:51:58 ok, this page needs to be fixed then Oct 11 17:52:51 karlp: deferring to json-c is not that hard either but needs a bit more code, packageing and testing Oct 11 18:01:56 jow_laptop: did you have a chance to see the DHCP broadcast reply fix I submitted on patchworks? 3963 I think it was, a couple of months back... Oct 11 18:02:26 fairly trivial patch... 6 or 7 lines or so. Oct 11 18:04:59 also, Adam's patch looked really good but I don't think that's been picked up either. Oct 11 18:08:10 nbd: did the eglibc vpath issue ever get figured out? Oct 11 18:13:25 philipp64: can't find any reference to the "--dhcp-broadcast" flag Oct 11 18:13:50 philipp64: what issue was that exactly? Oct 11 18:14:19 How can I read dhcp options in openwrt? Oct 11 18:15:02 nbd: http://patchwork.openwrt.org/patch/3963/ Oct 11 18:15:11 imho it is a bit arbritary Oct 11 18:15:48 i meant the 'vpath issue' Oct 11 18:34:31 jow r38365 trunk/package/network/ services/dnsmasq/files/dnsmasq.init services/dnsmasq/Makefile * dnsmasq: add option broadcast to host sections Oct 11 18:41:32 build #388 of uml is complete: Failure [failed compile_5] Build details are at http://buildbot.openwrt.org:8010/builders/uml/builds/388 Oct 12 02:17:26 build #376 of rb532 is complete: Failure [failed shell_12] Build details are at http://buildbot.openwrt.org:8010/builders/rb532/builds/376 Oct 12 02:17:43 build #376 of ppc44x is complete: Failure [failed shell_12] Build details are at http://buildbot.openwrt.org:8010/builders/ppc44x/builds/376 **** ENDING LOGGING AT Sat Oct 12 02:59:59 2013