**** BEGIN LOGGING AT Mon Feb 07 02:59:56 2011
Feb 07 03:00:09 <xMff> nbd: I just wondered... right now you need a running service which can provide a unix socket, right?
Feb 07 03:00:19 <nbd> yes
Feb 07 03:00:19 <OutBackDingo> xMff: the way i see it is uci is good at reading/writing config info
Feb 07 03:00:57 <cshore> OutBackDingo: there's an ubuntu ppa somewhere
Feb 07 03:00:59 <xMff> nbd: it would be cool if one could initiate an ubus connection via stdin/stdout
Feb 07 03:01:07 <cshore> OutBackDingo: for uci I mean
Feb 07 03:01:09 <nbd> xMff: why?
Feb 07 03:01:25 <cshore> from shell?
Feb 07 03:01:32 <OutBackDingo> cshore: hrmmm wonders what uci is capable of on ubuntu even
Feb 07 03:02:19 <cshore> OutBackDingo: reading/writing config files
Feb 07 03:02:21 <xMff> nbd: e.g. to pipe two clients
Feb 07 03:02:26 <OutBackDingo> cshore: are you referring to the afrimesh work ?
Feb 07 03:02:40 <cshore> OutBackDingo: not sure what it was for
Feb 07 03:02:51 <xMff> nbd: might be useful at some point to be able to use it for programs that are not daemons
Feb 07 03:03:24 <xMff> nbd: but that can be throw into the nice to have corner until an actual use case arises
Feb 07 03:03:38 <xMff> if at all
Feb 07 03:03:51 <nbd> i don't see how stdin/stdout would help there
Feb 07 03:04:05 <nbd> the protocol is not meant to connect clients directly
Feb 07 03:04:12 <xMff> ok
Feb 07 03:04:20 <nbd> clients can register objects and make calls to other objects
Feb 07 03:04:32 <nbd> and they can listen for events and send events
Feb 07 03:04:52 <nbd> that doesn't really fit into a client-to-client via stdout/stdin model
Feb 07 03:05:42 <nbd> clients that want to exchange data directly via stdin/stdout could just use the blobmsg library directly
Feb 07 03:06:06 <nbd> and some glue code to split incoming messages
Feb 07 03:06:15 <xMff> hm true
Feb 07 03:06:47 <nbd> and splitting messages with the blobmsg/blob library code is already trivial
Feb 07 03:06:57 <nbd> you read a struct blob_attr sized buffer
Feb 07 03:07:01 <nbd> check the length
Feb 07 03:07:17 <nbd> receive blob_pad_len(attr) - sizeof(struct blob_attr)
Feb 07 03:07:22 <nbd> and then you have your message
Feb 07 03:07:40 <nbd> that's how ubus does it internally
Feb 07 03:07:52 <nbd> well, with a tiny header in front of it
Feb 07 03:07:57 <nbd> but same principle
Feb 07 03:08:17 <nbd> btw. for the shell interaction i decided to not put a shell api in the ubus cli
Feb 07 03:08:32 <nbd> i think i'll write a separate tool to let the shell handle json
Feb 07 03:08:47 <nbd> that's more unix-like and offers the same flexibility
Feb 07 03:09:38 <nbd> later we could even use that to reimplement ifconfig, brctl, route, etc. as ubus based shell scripts which let netifd do the work ;)
Feb 07 03:10:14 <xMff> indeed
Feb 07 03:10:37 <cshore> is netifd another new daemon?
Feb 07 03:10:44 <xMff> yes
Feb 07 03:10:48 <nbd> cshore: yes, it will replace all the network scripts
Feb 07 03:10:53 <cshore> ah, ok
Feb 07 03:11:02 <OutBackDingo> sounds nice
Feb 07 03:11:05 <OutBackDingo> eta ?
Feb 07 03:11:11 <xMff> it will finally allow for complex stuff
Feb 07 03:11:12 <nbd> 4 weeks or so
Feb 07 03:11:13 <OutBackDingo> we talking weeks or months
Feb 07 03:11:18 <OutBackDingo> really
Feb 07 03:11:24 <OutBackDingo> maybe ill wait
Feb 07 03:11:25 <nbd> got a paying customer, so it gets done faster
Feb 07 03:12:14 <cshore> nbd: does that include iptables?
Feb 07 03:12:19 <nbd> no
Feb 07 03:12:24 <nbd> at least not initially
Feb 07 03:12:35 <xMff> I am thinking about reviving my fwd project
Feb 07 03:12:46 <OutBackDingo> nbd: id bounty up on radiotap_rate so it worked on ath9k
Feb 07 03:12:50 <nbd> xMff: that would be good
Feb 07 03:13:15 <xMff> yeah, I want to implement timed rules
Feb 07 03:13:25 <OutBackDingo> iptables can get convuluted though
Feb 07 03:13:27 <nbd> OutBackDingo: this will be implemented upstream anyway, just have to wait for a bit
Feb 07 03:13:34 <OutBackDingo> yupp i know
Feb 07 03:14:02 <OutBackDingo> i need to nudge the wireless guys
Feb 07 03:14:23 <nbd> since my todo list is quite long, i prefer to work on things that nobody else can or will work on
Feb 07 03:14:35 <OutBackDingo> ive seen some stuff going in but not yet to the point where i can drop maintaining patches
Feb 07 03:14:50 <xMff> OutBackDingo: iptables gets a bit simpler if you can control it via C instead of the awkward frontend
Feb 07 03:15:11 <OutBackDingo> nbd: yeah we need to clone a few of you guys...
Feb 07 03:15:51 <nbd> xMff: can you think of any more useful features that you'd like ubus to handle?
Feb 07 03:16:11 <OutBackDingo> nbd:  rpc capabilities to manage it :)
Feb 07 03:16:11 <xMff> nbd: not yet
Feb 07 03:16:20 <nbd> OutBackDingo: ubus *is* the rpc
Feb 07 03:16:46 <OutBackDingo> nbd: sooo
Feb 07 03:16:48 <OutBackDingo> hrmmm
Feb 07 03:16:53 <nbd> OutBackDingo: it's simply there to let a number of different services talk to each other
Feb 07 03:17:12 <nbd> xMff: i think i'll add something to let it pick up hotplug events
Feb 07 03:17:32 <OutBackDingo> what specifically will ubus capabilities be tied to ?
Feb 07 03:17:34 <xMff> nbd: I'll look into a generic lua api
Feb 07 03:17:58 <nbd> xMff: sounds good. should be easy to do as well
Feb 07 03:18:01 <xMff> will ease prototyping
Feb 07 03:18:12 <nbd> just need to write some generic code to handle lua table <-> blobmsg conversion
Feb 07 03:18:27 <nbd> and then some glue code to call the api functions
Feb 07 03:19:17 <nbd> you can use libblobmsg_json as an example for that
Feb 07 03:19:27 <xMff> OutBackDingo: in the future - netifd, soon hostapd / wpa_supplicant, personally I want bindings for radvd, pppd would probably make sense
Feb 07 03:19:32 <xMff> ok
Feb 07 03:19:47 <nbd> i was also considering moving udhcpc into netifd directly
Feb 07 03:19:51 <xMff> relayd
Feb 07 03:19:58 <xMff> fwd
Feb 07 03:20:06 <xMff> uhttpd :)
Feb 07 03:20:19 <nbd> why uhttpd?
Feb 07 03:20:36 <xMff> it starts with u ;)
Feb 07 03:20:40 <nbd> lol
Feb 07 03:21:00 <xMff> no seriously, I could need a capability to spawn new listeners on demand
Feb 07 03:21:10 <xMff> and have them timeout after a while
Feb 07 03:21:18 <nbd> can't it do that on its own
Feb 07 03:21:18 <nbd> ?
Feb 07 03:21:30 <nbd> simply by monitoring how many listeners are in use
Feb 07 03:21:56 <xMff> more something like dynamic vhosts... hard to explain
Feb 07 03:22:46 <nbd> k
Feb 07 03:24:05 <xMff> however, dnsmasq could be interesting, but there's nothing that could not be done with writing out a config and sending sighup
Feb 07 03:25:14 <xMff> if the most important stuff is rpc-ified we could do TR-061 ;)
Feb 07 03:26:50 <xMff> sorry, 69
Feb 07 03:26:54 <nbd> ;)
Feb 07 03:27:09 <nbd> not that i expect much open source community interest in that
Feb 07 03:27:57 <nbd> probably better to sell it separately ;)
Feb 07 04:22:10 <OutBackDingo> xMff: referring to Technology Brief for Service Providers
Feb 07 04:41:41 <xMff> was that a question?
Feb 07 04:49:20 <builder> build #77 of ppc40x is complete: Success [build successful]  Build details are at http://tksite.gotdns.org:8010/builders/ppc40x/builds/77
Feb 07 05:49:24 <swalker> is anyone around that can commit the PKG_REV changes from http://paste.openwrt.org/m2fad682e?
Feb 07 07:03:31 <OutBackDingo> swalker: uhmm feeds/packages ???
Feb 07 07:03:42 <OutBackDingo> should be packages
Feb 07 07:05:40 <OutBackDingo> since "feeds" doesnt exists until you update them
Feb 07 07:08:53 <OutBackDingo> xMff: uhmm yeah kinda in a way, just verifying your reference to tr-069 was in reference to  Technology Brief for Service Providers
Feb 07 10:43:13 <CIA-84> nbd * r25398 /trunk/package/mac80211/patches/590-mac80211_fix_clone_check.patch: mac80211: fix the check for cloned skbs in the tx path
Feb 07 10:44:48 <CIA-84> nbd * r25399 /branches/backfire/package/mac80211/patches/590-mac80211_fix_clone_check.patch: mac80211: fix the check for cloned skbs in the tx path (backport of r25398)
Feb 07 11:29:03 <builder> build #72 of ppc44x is complete: Failure [failed compile_6]  Build details are at http://tksite.gotdns.org:8010/builders/ppc44x/builds/72
Feb 07 13:03:42 <CIA-84> mirko * r25400 /packages/Xorg/lib/qt4/files/qws-qmake.conf: [packages/qt4] fix passing variables to qmake
Feb 07 13:04:29 <CIA-84> mirko * r25401 /packages/Xorg/lib/qt4/files/qmake.mk: [packages/qt4] qmake.mk: pass $(STAGING_DIR); override Build/Compile/Default
Feb 07 14:27:48 <Acinonyx> Bonjour
Feb 07 15:59:33 <CIA-84> obsy * r25402 /packages/net/transmission/Makefile: [packages] transmission: update to 2.20 final
Feb 07 17:01:33 <CIA-84> mb * r25403 /packages/Xorg/lib/qt4/files/x11-qmake.conf: qt4: Port r25400 to qt-x11
Feb 07 17:13:02 <CIA-84> mb * r25404 /packages/Xorg/lib/qt4/files/qmake.mk: qt4: Default build rule: Honor PKG_JOBS, if present
Feb 07 18:17:07 <CIA-84> nbd * r25405 /trunk/package/libjson-c/ (. Makefile): libjson-c: add to core (instead of packages)
Feb 07 18:17:49 <CIA-84> nbd * r25406 /trunk/package/libubox/ (. Makefile): add libubox, a small utility library containing blob/blobmsg (structured message format libraries), uloop, usock, list.h and the olsr avl tree implementation
Feb 07 18:17:56 <CIA-84> nbd * r25407 /trunk/package/relayd/Makefile: relayd: update to 2011-02-07, uses cmake and libubox now
Feb 07 21:10:11 <EqUaTe> xMff: thanks for earlier re tortib...
Feb 07 21:45:25 <CIA-84> acinonyx * r25408 /packages/admin/syslog-ng3/Makefile: [packages] syslog-ng3: Add missing libpthread and zlib dependencies
Feb 07 21:45:29 <CIA-84> acinonyx * r25409 /packages/libs/glib2/Makefile: [packages] glib2: Add zlib and libpthread build dependencies
Feb 07 21:45:35 <CIA-84> acinonyx * r25410 /packages/net/irssi/Makefile: [packages] irssi: Add missing libpthread dependency
Feb 07 21:45:38 <CIA-84> acinonyx * r25411 /trunk/package/busybox/files/telnet: [package] busybox: Check if an SSH daemon is enabled before disabling telnet access (#8760)
Feb 07 21:48:34 <CIA-84> blogic * r25412 /trunk/package/ (9 files in 8 dirs):
Feb 07 21:48:34 <CIA-84> [voip]
Feb 07 21:48:34 <CIA-84> * drop libtapi and tapi_sip (moved to external git)
Feb 07 21:48:34 <CIA-84> * add build variants to pjsip
Feb 07 21:48:34 <CIA-84> * split tapi_dev audio driver out of patch into src/
Feb 07 21:49:35 <CIA-84> blogic * r25413 /trunk/package/ltq-kpi2udp/ (5 files in 2 dirs):
Feb 07 21:49:35 <CIA-84> [ltq-kpi2udp]
Feb 07 21:49:35 <CIA-84> * adss in-kernel udp redirect plugin for lantiq voice optimisation
Feb 07 22:28:41 <CIA-84> acinonyx * r25414 /trunk/package/hotplug2/patches/170-non_fatal_include.patch: [package] hotplug2: Make missing include files non fatal
Feb 07 22:28:45 <CIA-84> acinonyx * r25415 /trunk/package/base-files/files/etc/hotplug2-common.rules: [package] base-files: Add include for target specific hotplug rules
Feb 07 23:10:54 <tripolar> ping Acinonyx
Feb 07 23:13:40 <Acinonyx> pong
Feb 07 23:14:11 <tripolar> Acinonyx the patch you submitted to check for the pubkey
Feb 07 23:14:22 <Acinonyx> yes
Feb 07 23:14:24 <tripolar> wouldnt be correct if someone changes the root homedir
Feb 07 23:14:26 <tripolar> as i did
Feb 07 23:14:58 <tripolar> maybe you should read the homedir from /etc/passwd
Feb 07 23:15:08 <tripolar> and test it with that dir
Feb 07 23:15:19 <tripolar> to make the patch more generic
Feb 07 23:15:33 <Acinonyx> I guess that would be better
Feb 07 23:15:48 <tripolar> it wouldnt matter for me as i removed telnet completely but for everyone else ...
Feb 07 23:17:48 <Acinonyx> at least it doesn't brick the router the way it is now
Feb 07 23:18:39 <tripolar> etc/init.d/sshd enabled - where is this enabled documented?
Feb 07 23:19:06 <tripolar> ahh etc/rc.common
Feb 07 23:19:23 <tripolar> but there is not output for this command - only for enable not enabled
Feb 07 23:19:42 <Acinonyx> it return an exit code
Feb 07 23:20:05 <tripolar> i know - but it not shown when doing /etc/init.d/sshd for example
Feb 07 23:20:11 <tripolar> only enable shows up
Feb 07 23:20:39 <Acinonyx> /etc/init.d/sshd enabled; echo $?
Feb 07 23:20:51 <tripolar> Available commands:
Feb 07 23:20:51 <tripolar>         start   Start the service
Feb 07 23:20:51 <tripolar>  ...
Feb 07 23:21:00 <tripolar> /etc/init.d/sshd
Feb 07 23:21:05 <tripolar> the output of this command
Feb 07 23:21:13 <Acinonyx> hmm
Feb 07 23:22:41 <Acinonyx> it's undocumented in the usage ouput
Feb 07 23:24:21 <tripolar>         enable  Enable service autostart
Feb 07 23:24:21 <tripolar> +       enabled Returns 0 when the service is enabled
Feb 07 23:24:25 <tripolar> add this to rc.common
Feb 07 23:24:36 <tripolar> should i sent you a patch?
Feb 07 23:25:40 <tripolar> Acinonyx: http://www.file-upload.net/download-3197993/rc.patch.html
Feb 07 23:26:25 <Acinonyx> are there any other undocumented options?
Feb 07 23:26:39 <tripolar> dont know
Feb 07 23:27:11 <tripolar> help
Feb 07 23:27:26 <tripolar> help Shows this help :)
Feb 07 23:27:58 <Acinonyx> maybe it is undocumented because it is not used by the user but only in scripts
Feb 07 23:28:08 <tripolar> yeah maybe
Feb 07 23:28:42 <Acinonyx> there are some other option in rc.common
Feb 07 23:28:50 <Acinonyx> like "depends"
Feb 07 23:29:00 <Acinonyx> does nothing
Feb 07 23:29:25 <tripolar> the other ones are not really important as they return 0 or 1
Feb 07 23:29:57 <tripolar> shutdown but as shutdown calls stop its the same
Feb 07 23:31:28 <Acinonyx> I guess they are used in some scripts
Feb 07 23:31:39 <Acinonyx> they get overriden
Feb 07 23:34:16 <tripolar> Acinonyx: shutdown is used when the system goes down
Feb 07 23:34:21 <tripolar> this patch is from me
Feb 07 23:35:23 <tripolar> its calling stop but its possible to end a service on system shutdown deferently as when its called with stop
Feb 07 23:36:03 <tripolar> Acinonyx: as far as i know only my package (openssh) uses this by now
Feb 07 23:36:13 <Acinonyx> cool
Feb 07 23:36:28 <Acinonyx> nginx uses it too
Feb 07 23:36:32 <tripolar> okay
Feb 07 23:37:01 <tripolar> Acinonyx: thats the commit https://dev.openwrt.org/changeset/24540
Feb 07 23:37:19 <Acinonyx> shutdown does $NGINX_BIN -s quit instaed of $NGINX_BIN -s stop
Feb 07 23:38:08 <tripolar> for openssh its needed - because without it you would have to do a restart in screen otherwise it would kick you out of the system when doing a restart
Feb 07 23:38:24 <tripolar> of the openssh service
Feb 07 23:39:07 <tripolar> i would also kick you out but it would start the service again when doing it inside a screen session ....
Feb 07 23:39:21 <Acinonyx> ahh, yes that annoying bug
Feb 07 23:39:36 <Acinonyx> I've run into it in the past
Feb 07 23:39:56 <tripolar> and on restart or poweroff it will end all open sshd session - to get rif of this enoying stucking terminals
Feb 07 23:40:54 <tripolar> Acinonyx could you do a patch to read the root homedir from /etc/passwd?
Feb 07 23:40:56 <Acinonyx> yes
Feb 07 23:41:50 <tripolar> Acinonyx do you use openssh? i dont think that that much people use it
Feb 07 23:41:52 <Acinonyx> I'll try but it would be more complicated that the parsing of passwd in has_root_pwd()
Feb 07 23:42:27 <Acinonyx> no i don't
Feb 07 23:42:49 <tripolar> maybe im the only one ;) - maintainer and only user :)
Feb 07 23:42:59 <Acinonyx> :P
Feb 07 23:43:14 <tripolar> only=lonely ;)
Feb 07 23:43:23 <Acinonyx> the reason i am not using it is its size
Feb 07 23:43:30 <tripolar> its not that bad
Feb 07 23:43:56 <tripolar> i managed to build a image with everything included
Feb 07 23:44:18 <tripolar> + the pubkeys to log in directly after the image is flashed and the box is restarted
Feb 07 23:44:41 <tripolar> but there are some mods necessary
Feb 07 23:45:04 <tripolar> the get the right permission for this files - otherwise openssh refuses to start
Feb 07 23:45:09 <tripolar> the get=to get
Feb 07 23:45:28 <Acinonyx> it pulls openssl and zlib which ultimately make it use more spce than dropbear
Feb 07 23:45:42 <tripolar> yeah but i removed some busybox stuff
Feb 07 23:45:42 <Acinonyx> hmm
Feb 07 23:45:54 <tripolar> like telnet+telnetd
Feb 07 23:46:21 <tripolar> my image with wifi+openssh+irssi+nano and some other stuff is about 6mb
Feb 07 23:46:51 <tripolar> and i have about 7mb flash so thats fine
Feb 07 23:47:41 <Acinonyx> :)
Feb 07 23:48:22 <tripolar> but i think my mod/hacks wont make it into trunk as they are not that generic - i also dont like the rootfs usb overlay stuff so i still use my old pivot root stuff i posted somewhere in the forum
Feb 07 23:48:34 <Acinonyx> what does the postinst script do in openssh?
Feb 07 23:49:52 <tripolar> it adds the group and user for the openssh daemon
Feb 07 23:50:01 <tripolar> but is was there befor i got the prog
Feb 07 23:50:06 <tripolar> never touched that stuff
Feb 07 23:50:19 <tripolar> befor i became the maintainer
Feb 07 23:50:27 <Acinonyx> it seems to me a bit hackish
Feb 07 23:50:32 <tripolar> yeah it is
Feb 07 23:50:34 <Acinonyx> it checks if it runs in squashfs
Feb 07 23:51:14 <tripolar> and deltes the files and copies them back
Feb 07 23:51:42 <tripolar> no
Feb 07 23:51:48 <tripolar> it checks if it is a asymbolic link
Feb 07 23:52:46 <tripolar> i think this could be removed
Feb 07 23:53:07 <tripolar> or do newer installs still use /etc/group and /etc/passwd as symbolic links?
Feb 07 23:53:12 <tripolar> @ Acinonyx
Feb 07 23:53:56 <Acinonyx> mount |awk '/root/ { print $$5 }'
Feb 07 23:55:06 <Acinonyx> the don't use symbolic links I think
Feb 08 00:00:37 <tripolar> mount |awk '/root/ { print $5 }'|grep -v rootfs would be right
Feb 08 00:00:45 <tripolar> mount |awk '/root/ { print $$5 }'|grep -v rootfs would be right
Feb 08 00:01:00 <tripolar> Acinonyx: this awk statement makes no sense
Feb 08 00:01:25 <tripolar> i think this code can be removed
Feb 08 00:02:23 <Acinonyx> yeah, it shouldn't work
Feb 08 00:03:23 <tripolar> as it return
Feb 08 00:03:28 <tripolar> rootfs
Feb 08 00:03:32 <tripolar> squashfs
Feb 08 00:03:37 <Acinonyx> yes
Feb 08 00:03:40 <tripolar> and that will never be true
Feb 08 00:03:44 <tripolar> for the if statement
Feb 08 00:03:59 <tripolar> i checked the base-files and there is no symlink created
Feb 08 00:04:08 <tripolar> i will remove the lines
Feb 08 00:04:34 <Acinonyx> yes
Feb 08 00:04:43 <Acinonyx> since mini_fo symlinks are not needed
Feb 08 00:06:26 <CIA-84> tripolar * r25416 /packages/net/openssh/Makefile: [openssh]: remove obsolete lines
Feb 08 00:06:51 <tripolar> Acinonyx: thanks for the hint
Feb 08 00:06:59 <Acinonyx> np
Feb 08 00:07:30 <tripolar> Acinonyx so now its your turn ;)
Feb 08 00:07:53 <Acinonyx> :)
Feb 08 00:08:11 <Acinonyx> i am working on it
Feb 08 00:09:07 <tripolar> i cant even test it as i use shadow
Feb 08 00:12:41 <tripolar> Acinonyx: this would work
Feb 08 00:12:43 <tripolar> grep root /etc/passwd |sed -e s/.*\:root\://g -e s/\:.*//g
Feb 08 00:13:17 <Acinonyx> i'd prefer not to use sed
Feb 08 00:17:56 <Acinonyx> http://paste.openwrt.org/d1979c8d6
Feb 08 00:18:06 <Acinonyx> could you try this please?
Feb 08 00:22:05 <tripolar> Acinonyx seems to work
Feb 08 00:22:16 <tripolar> but i was never that good at bash
Feb 08 00:22:21 <tripolar> sh
Feb 08 00:22:34 <tripolar> what does this mean exactly?
Feb 08 00:22:35 <tripolar> homedir="${homedir#*:*:0:0:*:}"
Feb 08 00:22:37 <Acinonyx> thanks for spotting it
Feb 08 00:22:41 <Acinonyx> and testing
Feb 08 00:23:07 <tripolar> thats self explaining homedir=$([ -f "$1" ] && cat "$1")
Feb 08 00:23:08 <Acinonyx> it removes what matches the expression after #
Feb 08 00:25:53 <tripolar> and this?
Feb 08 00:25:54 <tripolar> ${homedir%:*}
Feb 08 00:26:14 <Acinonyx> hmm did I put a single %?
Feb 08 00:26:24 <Acinonyx> no
Feb 08 00:26:25 <Acinonyx> :)
Feb 08 00:26:55 <tripolar> oh it seems irc is eating the second %
Feb 08 00:27:01 <tripolar> or xchat ;)
Feb 08 00:27:03 <Acinonyx> heh
Feb 08 00:27:14 <tripolar> what does it mean?
Feb 08 00:27:38 <tripolar> after homedir="${homedir#*:*:0:0:*:}" homedir = /tmp/roothome:/bin/ash
Feb 08 00:27:40 <tripolar> for me
Feb 08 00:27:51 <tripolar> nobody:*:65534:65534:nobody:/var:/bin/false
Feb 08 00:27:51 <tripolar> daemon:*:65534:65534:daemon:/var:/bin/false
Feb 08 00:27:51 <tripolar> sshd:x:99:99:sshd:/var/empty/.sshd:/bin/false
Feb 08 00:27:54 <tripolar> and this linces
Feb 08 00:27:56 <tripolar> lines
Feb 08 00:29:16 <tripolar> what does the % 2 times mean?
Feb 08 00:29:27 <Acinonyx> well, the second means again to remove the expression that matches after %
Feb 08 00:29:47 <Acinonyx> but in this case it is a suffix and it can match the largest pattern
Feb 08 00:30:06 <Acinonyx> for example
Feb 08 00:30:34 <Acinonyx> it you have a string "testing"
Feb 08 00:31:35 <Acinonyx> and use a single % with "t*g" it will delete the whole word
Feb 08 00:32:01 <Acinonyx> sorry it ate my double % also
Feb 08 00:32:24 <Acinonyx> crap
Feb 08 00:32:31 <tripolar> test="testing";echo "${test%t*g}" - with one
Feb 08 00:32:36 <tripolar> gives me tes
Feb 08 00:32:56 <Acinonyx> yes, said it in reverse
Feb 08 00:33:02 <Acinonyx> doit with %
Feb 08 00:33:04 <Acinonyx> grrrr
Feb 08 00:33:08 <Acinonyx> irc sucks
Feb 08 00:33:13 <Acinonyx> doit with double %
Feb 08 00:33:25 <Acinonyx> it will match the whole word
Feb 08 00:33:36 <tripolar> ohh % two times takes the first t and % one time takes the last t - am i right?
Feb 08 00:33:50 <Acinonyx> it matches the largest pattern with double %
Feb 08 00:33:59 <Acinonyx> and the smallest with single
Feb 08 00:34:02 <tripolar> okay
Feb 08 00:34:04 <tripolar> nice
Feb 08 00:34:06 <tripolar> i understand
Feb 08 00:34:16 <tripolar> is there somewhere a nice tut for stuff like this?
Feb 08 00:34:19 <Acinonyx> I hope I made sense
Feb 08 00:34:38 <tripolar> yeah you explaind it very well
Feb 08 00:34:39 <tripolar> :)
Feb 08 00:35:18 <Acinonyx> the difference between # and % is that the first removes prefixes and latter suffixes
Feb 08 00:36:24 <Acinonyx> this means that "testost" with %t*t will delete the "tost" and leave tes
Feb 08 00:36:47 <Acinonyx> and "#t*t" will delete "test" and leave "ost"
Feb 08 00:36:59 <tripolar> okay
Feb 08 00:37:17 <tripolar> and two times # or % will remove the whole word with t*t?
Feb 08 00:37:45 <tripolar> or is ther only two times %?
Feb 08 00:37:45 <Acinonyx> excaclty in this case
Feb 08 00:37:49 <tripolar> okay nice
Feb 08 00:38:17 <tripolar> Acinonyx: but there is for shure more of this advanced shell stuff?
Feb 08 00:39:26 <Acinonyx> ash man describes all these
Feb 08 00:39:50 <Acinonyx> with some fantasy, you can do many stuff
Feb 08 00:40:07 <Acinonyx> *much
Feb 08 00:40:19 <tripolar> ash man page - where do i find it?
Feb 08 00:40:28 <tripolar> or do you mean the bash man page?
Feb 08 00:40:45 <Acinonyx> http://linux.die.net/man/1/ash
Feb 08 00:40:58 <Acinonyx> no, bash is different
Feb 08 00:41:18 <tripolar> but it works with sh here
Feb 08 00:41:33 <tripolar> sh=bash
Feb 08 00:41:53 <tripolar> thanks
Feb 08 00:42:27 <Acinonyx> it depends on the system
Feb 08 00:43:16 <Acinonyx> we use ash in openwrt
Feb 08 00:43:21 <tripolar> yeah i know
Feb 08 00:43:23 <Acinonyx> busybox
Feb 08 00:43:32 <tripolar> just tested it on my desktop system
Feb 08 00:43:58 <Acinonyx> oh, sorry. I am a bit tired and didn't understand
Feb 08 00:46:19 <tripolar> hmm why doesnt this work?
Feb 08 00:46:20 <tripolar> faa="http://asas.de/gt_rt" ; echo "${faa%h*/}"
Feb 08 00:46:28 <tripolar> i want just gt_rt
Feb 08 00:46:34 <tripolar> i use double %
Feb 08 00:46:46 <tripolar> but this works
Feb 08 00:46:53 <tripolar> faa="http://asas.de/gt_rt" ; echo "${faa##h*/}"
Feb 08 00:46:58 <tripolar> @Acinonyx:
Feb 08 00:47:24 <Acinonyx> yes
Feb 08 00:47:36 <tripolar> but why doesnt it work with the double %
Feb 08 00:47:56 <Acinonyx> % matches suffixes
Feb 08 00:48:27 <Acinonyx> because it is not a suffix
Feb 08 00:48:41 <tripolar> ahh i understand thanks
Feb 08 00:48:56 <Acinonyx> it would work if you wanted to delete "/gt_rt"
Feb 08 00:49:26 <Acinonyx> if you play a little, it all comes clear
Feb 08 00:49:26 <tripolar> so % begins to search from the right side
Feb 08 00:49:35 <Acinonyx> yes
Feb 08 00:49:43 <tripolar> yeah now it clear
Feb 08 00:49:51 <tripolar> thanks
Feb 08 00:49:54 <Acinonyx> np
Feb 08 00:50:21 <tripolar> is there a way to do both thinks at wonce?
Feb 08 00:50:24 <tripolar> once
Feb 08 00:50:32 <tripolar> things
Feb 08 00:50:36 <tripolar> im tired too :)
Feb 08 00:50:48 <tripolar> i mean % and # in one statement
Feb 08 00:51:36 <Acinonyx> no i don't think so
Feb 08 00:53:16 <Acinonyx> it's almost 3 here
Feb 08 00:53:29 <tripolar> here its nearly 2am
Feb 08 00:53:33 <tripolar> where are you from?
Feb 08 00:53:57 <tripolar> Acinonyx: dont forget to commit your code :)
Feb 08 00:54:38 <Acinonyx> i am in Athens, Greece
Feb 08 00:54:43 <Acinonyx> ok i'll doit now
Feb 08 00:54:46 <tripolar> okay
Feb 08 01:01:58 <CIA-84> acinonyx * r25417 /trunk/package/busybox/files/telnet: [package] busybox: get root home dir from /etc/passwd in telnet init script (thanks tripolar)
Feb 08 01:04:55 <Acinonyx> I am going to bed
Feb 08 01:05:06 <Acinonyx> gnight tripolar
Feb 08 01:05:15 <tripolar> Acinonyx good night
Feb 08 01:05:43 <tripolar> have a nice day tomorrow
Feb 08 01:07:00 <Acinonyx> thanks, u too
Feb 08 01:07:00 <Acinonyx> ;)
Feb 08 01:54:38 <stuffa> Hi,  does anyone know what the broadcom "ctf" module does ?
Feb 08 01:54:41 <Weedy> 26
Feb 08 01:56:22 <mazilo> Does anyone know if busybox supports pushd and popd?
**** ENDING LOGGING AT Tue Feb 08 02:59:57 2011