[10:06] <keesj> Hi
[10:13] <keesj> I am trying to cross compile upstart, for that I would need the nih-dbus-tools to be compiled for the host target and not the target target. did anybody corss compile 0.6.3?
[14:03] <Hatrix> Keybuk: hello keybuk, i was reading some time ago an email from yours where you explain that actually in septembre upstart will be in a state to act as a supervise daemon, is this right on track? As I have to make some decisions regarding a client and what I am going to use to daemon supervising, as I really HATE the djb supervise tool thingy i would LOVE to use upstart
[14:18] <ion> Upstart can already do that for simple cases.
[14:18] <ion> It always has. It just hasn’t been ready for booting the entire system as Upstart jobs.
[14:19] <Hatrix> i can not use it for daemons which forks into the background  ... ?
[14:22] <ion> 0.6 has somewhat working support for it, although 0.10 will do it much better.
[14:28] <Hatrix> hmmm, how do i use it for this forking daemons? i could not get it to work as as soon as the daemon forked, upstart did not now anything about it (state, pid, ...)
[14:36] <plautrba_> Hatrix: you can use expect stanza with daemon or fork, then proccess should be traced for forks by ptrace() 
[14:41] <Hatrix> hmmmm
[14:41] <Hatrix> i did not know about the expect stanza, where is the docu for the current stanzas anyway, anyone knows? (besides of the source ...)
[14:42] <Keybuk> Hatrix: man 5 upstart
[14:42] <Hatrix> ray@dev-01:~$ man 5 upstart
[14:42] <Hatrix> No manual entry for upstart in section 5
[14:43] <Hatrix> hmm, seams in my debian upstart 06.2.1-1 package are no manpages ... hmmm
[14:46] <Hatrix> nope, section 5 missing ... ok, i'l grab the source, thanks 
[14:47] <Keybuk> err, man 5 init sorry
[14:48] <ion> Which reminds me, i should finish my Upstart manpage changes and post a patch.
[14:49] <sadmac2> Keybuk: you're alive! er.. hi.
[14:49] <Hatrix> got it now ... so expect daemon is if it's running from an init-script, and therefore forks twice and expect fork is a direct call to the daemon, am i right?
[14:50] <Hatrix> and, fearing being kicked of the channel, but, when will the 0.10 be useable, any ideas?
[14:52] <ion> keybuk: Would you mind reformatting all the manpages with your editor (ctrl-meta-z-7-skull-and-crossbones-å-5 or whatever) and committing that change, so i can emulate the proper formatting in my changes?
[14:54] <sadmac2> ion: nobody spells it out like that. C-M-z-7-☠-å-5
[14:54] <ion> Ah
[14:54] <ion> Forgive my lack of emacs-fu
[14:55] <sadmac2> ion: in vim its just :piratedance
[14:56] <ion> One can come up with somewhat obscure things with vim, too. :-P 2006-06-02 17:16:13 < _ion> Vip:norm W"xyE$T_D"xp;gU$
[14:58] <ion> Looking at the IRC log, i typed the line "    normal = Gtk::STATE_NORMAL", duplicated it a few times, replacing normal with active, insensitive, prelight etc. and then wrote that command to replace the Gtk::STATE_* entries with the respective equivalents.
[15:01] <sadmac2> ion: I can't make it work
[15:02] <sadmac2> where did you start?
[15:02] <ion> Vip should “paint” the whole block first, then :norm should do the list of normal-mode commands for each selected line.
[15:04] <sadmac2> ion: that can't be right... maybe I need to be in C++ mode or it doesn't work.
[15:04] <ion> V should start selecting lines, and ip should select the current “paragraph”.
[15:05] <sadmac2> yep
[15:05] <ion> Then :norm applies to that selection
[15:05] <sadmac2> yep
[15:05] <sadmac2> and the results are borked
[15:05] <ion> Did you have the indentation? Without it, you need to drop the W
[15:06] <sadmac2> ah.
[15:07] <sadmac2> ion: this is what macros are for btw.
[15:10] <sadmac2> probably would have just typed out the inital words and then started with '<,'>s/.*/\0 = Gtk::STATE_\0/
[15:10] <ion> Not much difference between :norm and macros. The difference only starts to show when the changes are so complex you need to see them happening while typing the commands.
[15:10] <sadmac2> then gone about uppercasing
[15:10] <sadmac2> ion: for your method I probably would have had a macro to do it to one line then done :norm @q
[15:11] <ion> In fact, you could have uppercased them within the expression. :s/.*/\0 = Gtk::STATE_\U\0/
[15:11] <sadmac2> oh that's right.
[15:12] <ion> But that would have missed the
[15:12] <ion> somethinglong = blah
[15:12] <ion> short         = blah
[15:12] <ion> formatting :-P
[15:13] <sadmac2> not big on that anyway. I end up spending more time on whitespace than code.
[15:39] <sadmac2> ls
[15:40] <sadmac2> ...well its been awhile since that's happened
[15:42] <ion> rm -fr ~
[15:42] <sadmac2> ion: ...you seem to be confused as to how this works
[19:30] <wasabi> byxnet.net
[19:30] <wasabi> Oops