[00:00] <th> hmm
[00:00] <th> so you'd prefer this to removing some single options?
[00:03] <th> at least this would obey our policy "build as the author intended"
[00:07] <th> == 11/20/06 00:11:09 =[5]=> Finished building package mplayer.
[00:07] <th> seems like mplayer wants to have a -O4
[00:07] <th> i removed -Os and added -O4
[00:08] <th> O4 was the option mplayer chose when building without rock
[00:14] <stf^rocklinux> mplayer might even use different optimizations for different cpus and parts of the code. So yes, I think it's better to let mplayer choose the options. Anything else is not even supported by the authors, see above.
[00:14] <th> yea
[00:17] <th> var_append confopt " " "--enable-linux-devfs
[00:17] <th> ugh
[00:17] <th> so we may not even add -Isomething to CFLAGS?
[00:18] <th> well this should be CPPFLAGS anyways, no?
[00:29] <stf^rocklinux> The question is: do we need -l in CFLAGS/CPPFLAGS? If not both should probably be empty (and LDFLAGS as well).
[01:13] blindcoder (i=blindcod@gateway/gpg-tor/key-0x384243C9) left irc: Remote closed the connection
[01:16] blindcoder (i=blindcod@gateway/gpg-tor/key-0x384243C9) joined #rocklinux.
[07:21] <blindcoder> moin
[07:35] rocklogbot (n=rocklogb@pallas.crash-override.net) left irc: Remote closed the connection
[07:35] rocklogbot (n=rocklogb@pallas.crash-override.net) joined #rocklinux.
[08:14] blindcoder (i=blindcod@gateway/gpg-tor/key-0x384243C9) left irc: Remote closed the connection
[08:18] blindcoder (i=blindcod@gateway/gpg-tor/key-0x384243C9) joined #rocklinux.
[08:26] blindcoder (i=blindcod@gateway/gpg-tor/key-0x384243C9) left irc: Remote closed the connection
[08:49] blindcoder (i=blindcod@gateway/gpg-tor/key-0x384243C9) joined #rocklinux.
[10:21] rocklogbot (n=rocklogb@pallas.crash-override.net) left irc: Remote closed the connection
[10:22] rocklogbot (n=rocklogb@pallas.crash-override.net) joined #rocklinux.
[10:44] rocklogbot (n=rocklogb@pallas.crash-override.net) left irc: Remote closed the connection
[10:44] rocklogbot (n=rocklogb@pallas.crash-override.net) joined #rocklinux.
[10:45] rocklogbot (n=rocklogb@pallas.crash-override.net) left irc: Remote closed the connection
[10:45] rocklogbot (n=rocklogb@pallas.crash-override.net) joined #rocklinux.
[10:45] <blindcoder> hmm
[10:46] rocklogbot (n=rocklogb@pallas.crash-override.net) left irc: Remote closed the connection
[10:53] rocklogbot (n=rocklogb@pallas.crash-override.net) joined #rocklinux.
[12:25] <owl> moin
[13:42] <stf^rocklinux> moin moin
[13:43] <blindcoder> moin stf^rocklinux 
[14:02] ubijtsa (n=ubijtsa@karlsson.force9.co.uk) left irc: Remote closed the connection
[14:03] ubijtsa (n=ubijtsa@karlsson.force9.co.uk) joined #rocklinux.
[14:04] <esden> blindcoder: thanks ... heise.de feed works fine now
[14:04] <blindcoder> esden: sure does
[14:04] <blindcoder> esden: I use it too, so it underwent apologetic splicing
[14:04] <blindcoder> esden: err, considerate testing
[14:04] <esden> hehe ;)
[14:11] <th> [SRC] is confusing
[14:11] <th>   Format: [SRC] tarball1 tarball2
[14:11] <th>   This will enable build_this_package to build the content of more than
[14:11] <th>   one tarball.
[14:11] <th> so will it do a ./configure for EACH tarball?
[14:11] <th> or extract both and do it once?
[14:11] <th> i guess it's the first
[14:11] <blindcoder> interesting question
[14:11] <th> so how do i do the 2nd?
[14:12] <blindcoder> preconf hook, I'd say
[14:13] <blindcoder> it doesn't seem like such a common case
[14:13] <blindcoder> but we could expand the srctar variable for that
[14:14] <th> which makes things even more complicated
[14:14] <blindcoder> hardly
[14:14] <blindcoder> for x in $srctar ; do
[14:14] <blindcoder> should be fine, no?
[14:14] <th> for xsrctar in $xsourceballs; do
[14:14] <th> we do this already
[14:15] <blindcoder> then srctar="foo.tar.bz2 bar.tar.bz2" sohuld be fine, no?
[14:15] <th> sourceballs=$( echo "$desc_D" | tr ' ' '\t' | tr -s '\t' | cut -f2 | sed 's,.\(t\?\)\(gz\|Z\)$,.\1bz2,' )
[14:15] <blindcoder> even better
[14:16] <th> no - this loops through each [D] and tries to build every dir on its own
[14:17] <blindcoder> so back to autoextract=0 and a preconf hook
[14:17] <th> seems so
[14:17] <th> prepatch would be more correct
[14:18] <th> but that's only evaluated when $autopatch = 1
[14:26] <stf^rocklinux> btw. [SRC] does a 'configure ; make ; make install' for each source archive given.
[14:28] <th> yea - i figured.
[14:30] <stf^rocklinux> I think it's rarely used because you can often split a package with the [SRC] tag into several smaller ones, which is better wrt. to modularity, and also makes for simpler .conf files (the old openh323 package is a good example).
[14:32] <th> well subversion-deps is only a "addon" it has no makefile or configure
[14:33] <th> it's used by the main package
[14:37] <stf^rocklinux> ic. then preconf seems appropriate (afaik no need to set autoextract to 0 if the main archive is the first in the desc file)
[14:37] <th> true - there is a "head -n 1"
[14:38] <th> stf^rocklinux: could you propose the "CFLAGS=" alternative for mplayer as sm patch? i send the one which replaces Os with O4
[14:39] <stf^rocklinux> or use prepatch, as you suggested (autopatch=1 in subversion.conf, isn't it).
[14:39] <stf^rocklinux> ok, I'll test-build mplayer with this and send a patch.
[14:40] <th> thanks
[14:41] <th> stf^rocklinux: perhaps you can use the update to rc1 which my patch includes...
[14:43] <th> == 11/20/06 14:42:28 =[9]=> Finished building package subversion.
[14:43] <th> (without apache)
[14:47] blindcod1r (i=blindcod@gateway/gpg-tor/key-0x384243C9) joined #rocklinux.
[14:48] blindcoder (i=blindcod@gateway/gpg-tor/key-0x384243C9) left irc: Nick collision from services.
[14:48] Nick change: blindcod1r -> blindcoder
[16:40] blindcoder (i=blindcod@gateway/gpg-tor/key-0x384243C9) left irc: Remote closed the connection
[16:42] blindcoder (i=blindcod@gateway/gpg-tor/key-0x384243C9) joined #rocklinux.
[19:16] <th> started another build: r8068 + journal "2006-11-20 th"
[22:31] kasc_ (n=kasc@dslb-084-060-111-242.pools.arcor-ip.net) joined #rocklinux.
[22:39] ija_ (n=ija@84.19.217.60) joined #rocklinux.
[22:41] kasc (n=kasc@dslb-084-060-115-133.pools.arcor-ip.net) left irc: Read error: 110 (Connection timed out)
[22:41] Nick change: kasc_ -> kasc
[22:44] ija (n=ija@84.19.222.230) left irc: Nick collision from services.
[22:44] Nick change: ija_ -> ija
[23:36] aFlag (n=rafael@201.78.110.155) joined #rocklinux.
[23:39] aFlag (n=rafael@201.78.110.155) left #rocklinux.
[00:00] --- Tue Nov 21 2006