--- Log opened Thu Mar 06 00:00:00 2008 00:00 < dooglus> a parameter can be one of two things: a value, or a valuenode 00:00 < dooglus> a valuenode can be one of 3 things: constant, animated, or linkable 00:00 < dooglus> constant (has blue line, but no waypoints, and isn't expandable) 00:00 < dooglus> animated: has waypoints 00:01 < dooglus> linkable: has a little triangle next to it which expands it to show you its links 00:02 < genete> so it was a valuenode constant 00:02 < dooglus> yes 00:02 < dooglus> and when you disconnected it, it because just a value 00:04 < genete> Good Night, it is time to go bed 00:05 < dooglus> night 00:05 < genete> And I make it fast with one more thing... 00:05 < dooglus> I'm just going to check in code to stop it linking offsets when only one layer is created 00:05 < dooglus> in the bline tool 00:05 < dooglus> playing the honky tonk again? 00:05 < genete> we're the sultans of synfig =) 00:05 < genete> ha hah ha ha haç 00:06 < genete> never knew to translate "honky tonk" :( 00:06 < dooglus> lol 00:06 < dooglus> I don't know what it means 00:06 < dooglus> the rolling stones sung about a honky tonk woman, right? 00:07 < dooglus> who gave him the honky tonk blues 00:07 < dooglus> http://en.wikipedia.org/wiki/Honky_tonk 00:08 < dooglus> "the term came from the sound of geese" 00:08 -!- Elk [n=_Elk@ti0110a340-0355.bb.online.no] has joined #synfig 00:08 < genete> wikipedia is absolutely amazing :) 00:08 < dooglus> maybe we need to rename our ducks? 00:08 < factor> heh 00:11 < genete> think on that just make me goosepumps 00:11 -!- xerakko [n=Miguel@debian/developer/xerakko] has joined #synfig 00:11 < genete> good night then :) 00:12 -!- genete [n=Genete@84.122.32.253.dyn.user.ono.com] has quit ["Abandonando"] 00:18 -!- MangoFusion [n=jamesu@host86-147-212-148.range86-147.btcentralplus.com] has quit ["Not here"] 00:18 -!- genete [n=Genete@84.122.32.253.dyn.user.ono.com] has joined #synfig 00:18 * dooglus didn't need that image of genete with his goosepump :) 00:19 < dooglus> I just checked in the change to not link 1 layer's offsets to itself :) 00:19 < genete> dooglus: I cannot sleep if I don't tell you a new idea ;) 00:22 < genete> Mirror Layer 00:23 -!- _Elk [n=_Elk@ti0110a340-0355.bb.online.no] has quit [Read error: 110 (Connection timed out)] 00:23 < genete> with some ability you can make things like that: http://i85.photobucket.com/albums/k74/Genete/Mekhnes_Place_El-Hedine_Mosaique3.jpg 00:26 < genete> sleep now 00:26 -!- genete [n=Genete@84.122.32.253.dyn.user.ono.com] has quit ["Abandonando"] 00:28 -!- Elk [n=_Elk@ti0110a340-0355.bb.online.no] has quit [Read error: 110 (Connection timed out)] 00:30 < CIA-35> synfig: dooglus * r1860 /synfig-studio/trunk/src/gtkmm/state_bline.cpp: Removed old comment - plants do have offsets now. 00:30 < CIA-35> synfig: dooglus * r1861 /synfig-studio/trunk/src/gtkmm/state_bline.cpp: The "Link Offsets" option in the BLine tool now only links offsets if we're creating more than one layer at a time. 00:38 * dooglus can't see how to rotate a "BLine Tangent" by 90 degrees 00:40 < factor> out back later 00:40 -!- factor [n=Factor@32.146.9.61] has quit ["mIRGGI meni puis"] 01:19 -!- pabs3 [n=pabs@d122-105-74-150.per9.wa.optusnet.com.au] has joined #synfig 01:37 -!- pabspabspabs [n=pabs@d122-105-74-150.per9.wa.optusnet.com.au] has quit [Read error: 110 (Connection timed out)] 01:46 < CIA-35> synfig: dooglus * r1862 /synfig-core/trunk/src/synfig/valuenode_blinecalctangent.cpp: Allow Real values to be converted to "BLine Tangent". The ValueNode returns the length of the tangent at the selected point. 01:58 -!- pabspabspabs [n=pabs@d122-105-74-150.per9.wa.optusnet.com.au] has joined #synfig 02:03 < CIA-35> synfig: dooglus * r1863 /synfig-core/trunk/src/synfig/ (4 files): Added 3 new parameters to the "BLine Tangent" ValueNode, and incremented the canvas version number to allow checking for their presence in .sif files created from now on. 02:16 -!- pabs3 [n=pabs@d122-105-74-150.per9.wa.optusnet.com.au] has quit [Read error: 110 (Connection timed out)] 02:19 -!- pabs3 [n=pabs@d122-105-74-150.per9.wa.optusnet.com.au] has joined #synfig 02:37 -!- pabspabspabs [n=pabs@d122-105-74-150.per9.wa.optusnet.com.au] has quit [Read error: 110 (Connection timed out)] 03:20 -!- Yoyobuae [n=Yoyobuae@201.224.135.156] has joined #synfig 04:17 < pabs3> dooglus: Experience of One Mentor from gsoc 2005 http://shlang.com/writing/soc2005.html 04:17 < pabs3> dooglus: http://code.google.com/p/google-summer-of-code/wiki/AdviceforMentors 04:21 < pabs3> dooglus: the icons in that thin-blue-line image on the tabs look like they are opaque rather than alpha-blended 04:53 -!- Netsplit kornbluth.freenode.net <-> irc.freenode.net quits: Bombe, timonator, CIA-35, xerakko 04:54 -!- Netsplit over, joins: xerakko, Bombe, timonator, CIA-35 05:12 -!- factor [n=factor@ip68-14-160-70.ok.ok.cox.net] has joined #synfig 05:13 -!- AkhIL [n=AkhIL@90.188.201.182] has quit [Read error: 104 (Connection reset by peer)] 05:14 -!- AkhIL [n=AkhIL@90.188.201.182] has joined #synfig 05:45 -!- pabspabspabs [n=pabs@d122-105-74-150.per9.wa.optusnet.com.au] has joined #synfig 05:47 -!- pabs3 [n=pabs@d122-105-74-150.per9.wa.optusnet.com.au] has quit [Read error: 110 (Connection timed out)] 06:03 -!- storbeck [n=storbeck@cpe-65-189-190-215.columbus.res.rr.com] has joined #synfig 06:04 < storbeck> Is there anyway to make the .sif to a .swf? 06:15 -!- storbeck [n=storbeck@cpe-65-189-190-215.columbus.res.rr.com] has left #synfig ["Quack Quack"] 06:31 -!- Yoyobuae [n=Yoyobuae@201.224.135.156] has quit [Read error: 110 (Connection timed out)] 06:40 -!- pixelgeek [n=chatzill@c-71-59-140-184.hsd1.or.comcast.net] has joined #synfig 06:41 < pixelgeek> pabspabspabs: genete and I were talking about "March is Monster Month!" for the Challenge 06:49 -!- pabspabspabs is now known as pabs3 06:50 < pabs3> pixelgeek: sounds cool :) 07:03 < factor> GRRR! 07:03 < factor> monster month 07:05 < factor> ok off to sleep 07:26 -!- pabs3 [n=pabs@d122-105-74-150.per9.wa.optusnet.com.au] has quit ["brb"] 07:35 -!- pabs3 [n=pabs@d122-105-74-150.per9.wa.optusnet.com.au] has joined #synfig 07:58 -!- pixelgeek [n=chatzill@c-71-59-140-184.hsd1.or.comcast.net] has quit [Read error: 113 (No route to host)] 09:42 -!- factor [n=factor@ip68-14-160-70.ok.ok.cox.net] has quit [Connection timed out] 09:43 -!- factor [n=factor@ip68-14-160-70.ok.ok.cox.net] has joined #synfig 10:31 -!- mwiriadi [n=mwiriadi@124-169-123-199.dyn.iinet.net.au] has joined #synfig 10:31 < mwiriadi> 'lo 10:31 < dooglus> 'ello 10:37 < pabs3> hi mwiriadi 10:37 < mwiriadi> hey 10:37 < mwiriadi> how's release coming along 10:37 < mwiriadi> I'm getting pinged 10:37 < pabs3> mwiriadi: was just about to comment on the bug; new release out a few days ago 10:37 < mwiriadi> it's out already? 10:37 < pabs3> sorry, forgot to let you know about it 10:38 < pabs3> yup http://synfig.org/Download 10:38 < mwiriadi> yep no worries 10:38 < mwiriadi> I'm in rawhide at the moment 10:38 < mwiriadi> so I'll be compiling against gcc4.3 10:39 < pabs3> yeah, we fixed all the 4.3 bugs we could find a few weeks before the release IIRC 10:40 < mwiriadi> yeah I posted that 10:45 < mwiriadi> Building the package now 10:45 < mwiriadi> I'll upload it later on in about 20 mins so I'll wait for the review :) 10:46 < pabs3> cool 10:48 < dooglus> as I remember it, a few of the packages we depend upon don't build cleanly using gcc4.3 10:48 < mwiriadi> like what dooglus ? 10:49 < mwiriadi> as long as synfig and etl do I'm happy :) 10:50 < dooglus> the 3 that gave me problems were pangomm, glibmm, sigc++ 10:51 < dooglus> I remember now 10:51 < dooglus> it wasn't building those packages that was the problem (I didn't try building them; I use binary packages of them), it was the headers they provide 10:51 < dooglus> their headers give warnings when used, and we build synfig with -Werror (meaning 'treat warnings as errors') 10:52 < pabs3> thats only if you do --enable-warnings=hardcore isn't it? 10:53 < dooglus> oh, maybe 10:54 < dooglus> I'll check 10:55 < dooglus> I don't use that, and I was having problems building against those packages - I'll see if I can find out why 10:55 < mwiriadi> Just so you know I've packaged the new etl-devel 10:56 < mwiriadi> I've uploaded it blah blah waiting on the review atm 10:56 < mwiriadi> I'll do synfig now 10:56 < dooglus> example error: /usr/include/sigc++-2.0/sigc++/signal.h:1675: error: declaration of 'typedef struct sigc::slot_list > sigc::signal0::slot_list' 10:58 < mwiriadi> pabs3, https://bugzilla.redhat.com/show_bug.cgi?id=428568 can you add yourself to this 10:58 < dooglus> my build line for that error: 10:58 < dooglus> /usr/bin/g++-4.3 -DHAVE_CONFIG_H -I[...] -D[...] -D_DEBUG -g -O0 -W -Wall -MT libsynfig_la-layer_bitmap.lo -MD -MP -MF .deps/libsynfig_la-layer_bitmap.Tpo -c layer_bitmap.cpp -fPIC -DPIC -o .libs/libsynfig_la-layer_bitmap.o 10:59 < dooglus> no -Werror in there 11:02 < pabs3> looks like the debian sigc++ package has patches to fix that dooglus: see the resolved bugs: http://bugs.debian.org/cgi-bin/pkgreport.cgi?src=libsigc%2B%2B-2.0 11:03 < dooglus> pabs3: this is debian sid 11:03 < mwiriadi> does synfig or synfig or both require imagemagick? 11:03 < mwiriadi> synfig-devel 11:03 < pabs3> libsynfig does 11:03 < dooglus> it's optional, but allows synfig to render nice animated gifs 11:03 < mwiriadi> to build as well pabs3 ? 11:04 < mwiriadi> dooglus, no optional in fedora so it's a requires 11:04 < pabs3> mwiriadi: yeah, for the new libmagic++ target 11:04 < mwiriadi> thats one thing I like in Fedora no such thing as optional you either take it or leave it :) 11:04 < mwiriadi> k 11:05 < dooglus> pabs3: I see there's a new sigc++-dev when I go to update sid... 11:05 < dooglus> pabs3: I'll check the other 2 packages too then, see if they've also been updated since my last build 11:11 < mwiriadi> pabs3, You are probably missing libtool-ltdl-devel as BuildRrequires, but you cannot 11:11 < mwiriadi> install 11:11 < mwiriadi> %{_libdir}/libltdl.a 11:11 < mwiriadi> %{_includedir}/ltdl.h 11:11 < mwiriadi> as theses files will conflict with libtool-ltdl-devel 11:11 < mwiriadi> how do I deal with that one remove it? 11:12 < pabs3> IIRC we changed it so libltdl is compiled in, rather than installed 11:12 < mwiriadi> cool 11:12 < mwiriadi> It's compiling now so it'll generate an error 11:13 < pabs3> well, not really, ideally it could use the external libltdl when available and an internal copy otherwise, but the libtool autotools macros don't make that possible IIRC 11:13 < pabs3> (not really cool I mean) 11:13 < mwiriadi> could you do it in the configure script? 11:13 < mwiriadi> or would that cause mass messing around? 11:16 < pabs3> fedora could very simply patch configure.ac to use the external libltdl, but you'd need to rerun autoreconf at build time though 11:16 < pabs3> that is what I'll be doing for Debian 11:16 < mwiriadi> Can you link me to the patch? 11:17 < dooglus> it seems the rest are mostly just warnings, like: 11:17 < dooglus> /usr/include/glibmm-2.4/glibmm/helperlist.h:113: warning: suggest a space before ';' or explicit braces around empty body in 'for' statement 11:17 < dooglus> /usr/include/pangomm-1.4/pangomm/attributes.h:219: warning: type qualifiers ignored on function return type 11:19 < mwiriadi> Have you packaged it already so it would be under packages.debian.org? 11:19 < pabs3> mwiriadi: no patch or package yet, but it would be sed -i -e s/AC_LIBLTDL_CONVENIENCE/AC_LIBLTDL_INSTALLABLE/ configure.ac 11:19 < pabs3> and the build-dep on libtool-ltdl-devel 11:19 < pabs3> the/then 11:21 < pabs3> ideally there would be a AC_LIBLTDL_USE_SYSTEM_THEN_CONVENIENCE, but there isn't :( 11:21 < mwiriadi> yep 11:21 < mwiriadi> I've added that in 11:22 < mwiriadi> m4/gettext.m4:442: warning: macro `AM_LANGINFO_CODESET' not found in library 11:22 < mwiriadi> acinclude.m4:6680: the serial number must appear before any macro definition 11:22 < mwiriadi> configure.ac:16: error: possibly undefined macro: AC_LIBLTDL_INSTALLABLE 11:22 < mwiriadi> If this token and others are legitimate, please use m4_pattern_allow. 11:22 < mwiriadi> See the Autoconf documentation. 11:22 < mwiriadi> configure.ac:80: error: possibly undefined macro: AC_LIBTOOL_WIN32_DLL 11:22 < mwiriadi> configure.ac:81: error: possibly undefined macro: AC_LIBTOOL_DLOPEN 11:22 < mwiriadi> configure.ac:82: error: possibly undefined macro: AC_DISABLE_STATIC 11:22 < mwiriadi> configure.ac:83: error: possibly undefined macro: AC_ENABLE_SHARED 11:22 < mwiriadi> configure.ac:84: error: possibly undefined macro: AC_PROG_LIBTOOL 11:22 < mwiriadi> configure.ac:336: error: possibly undefined macro: AC_LIB_LTDL 11:22 < mwiriadi> autoreconf: /usr/bin/autoconf failed with exit status: 1 11:22 < mwiriadi> error: Bad exit status from /var/tmp/rpm-tmp.4048 (%build) 11:22 < mwiriadi> excuse spam 11:22 < mwiriadi> after running autoreconf 11:23 -!- pabs3 [n=pabs@d122-105-74-150.per9.wa.optusnet.com.au] has quit [Read error: 104 (Connection reset by peer)] 11:24 -!- pabs3 [n=pabs@d122-105-74-150.per9.wa.optusnet.com.au] has joined #synfig 11:24 < pabs3> damn modems 11:27 < dooglus> mwiriadi: the libtool package should define those things 11:27 < mwiriadi> so running autoreconf fails 11:27 < dooglus> /usr/share/libtool/libtool.m4 has lines like "AC_PROVIDE_IFELSE([AC_LIBTOOL_WIN32_DLL]," - does yours? 11:32 < mwiriadi> helps if I have libtool installed :( 11:32 < dooglus> right :) 11:32 < mwiriadi> it's working now 11:33 < dooglus> http://synfig.org/Build_instructions has a list of dependencies 11:34 < dooglus> synfig 11:34 < dooglus> Requires: ETL, libxml++, libsigc++, libltdl, libtool, gettext, cvs 11:34 < dooglus> cvs? apparently so 11:34 < pabs3> yep, thats a dep of autopoint 11:35 < pabs3> only needed if running autoreconf 11:35 < mwiriadi> you have to remember that those dep's are all debian deps 11:35 < mwiriadi> there are equivalent packages for fedora but yeah 11:35 < pabs3> mwiriadi: feel free to list the fedora ones on that page 11:36 < mwiriadi> So trying to figure what the equivalents are, are annoying 11:36 < mwiriadi> Yeah it'll be a copy and paste situation 11:36 < mwiriadi> What I might do is give you guys the spec file 11:36 < mwiriadi> and then just include that in your next release 11:36 < mwiriadi> using spec.in 11:36 < mwiriadi> if you want 11:37 < mwiriadi> I do that with upstream for mediatomb 11:37 < mwiriadi> I do the spec file and any patches etc he takes them and includes them for the spec.in file 11:37 < pabs3> I tend to think the packaging belongs in the distros 11:41 < mwiriadi> Fair enough 11:42 < dooglus> what about other redhat-like distros? they would find your spec file useful surely? 11:42 < mwiriadi> Yep 11:43 < mwiriadi> it helps in that way 11:43 < dooglus> I guess they'd know where to find it though on a fedora site somewhere presumably 11:43 < mwiriadi> I think so 11:43 < mwiriadi> just depends if they can be bothered tbh 11:43 < mwiriadi> I know I don't look at the other rpm distro's 11:44 < dooglus> I suppose synfig doesn't want to end up with hundreds of distro-specific files 11:44 < dooglus> do we really want the .xyz file from poodle-linux? 11:45 < mwiriadi> Fair enough 11:45 < mwiriadi> I know that they cater for the majority being debian and rpm 11:45 < dooglus> there really is a poodle linux? heh. 11:46 < mwiriadi> I know 11:46 < mwiriadi> never used it 11:46 < mwiriadi> I just know the majority is rpm and deb 11:56 < pabs3> dooglus: so basically, I get the impression that gsoc mentors guide the student through their project, help them out when they have problems and evaluate their final contribution 11:56 < pabs3> dooglus: also found this: http://www.gnome.org/~federico/docs/summer-of-code-mentoring-howto/index.html 12:03 < mwiriadi> synfig is done and packaged waiting for the review 12:03 < mwiriadi> speaking to the reviewer now 12:09 -!- MangoFusion [n=jamesu@host86-147-212-148.range86-147.btcentralplus.com] has joined #synfig 12:16 -!- pabs3 [n=pabs@d122-105-74-150.per9.wa.optusnet.com.au] has quit [Read error: 110 (Connection timed out)] 12:20 -!- pabs3 [n=pabs@d122-105-74-150.per9.wa.optusnet.com.au] has joined #synfig 12:22 < pabs3> mwiriadi: whats this about the .so files? 12:22 < mwiriadi> He wants the *.so file in the main synfig package and the *.so.* in the devel package 12:22 < mwiriadi> yet I'm getting rpmlint errors 12:23 < pabs3> that sounds wrong 12:23 < pabs3> in debian we have .so symlinks in libsynfig-dev, *.so.* in libsynfig and /usr/bin/synfig in the synfig package 12:24 < pabs3> IIRC 12:24 < mwiriadi> Yep 12:24 < mwiriadi> Thats what I proposed 12:24 < pabs3> yep http://packages.debian.org/sid/i386/libsynfig-dev/filelist 12:25 < mwiriadi> is synfig included in ubuntu? 12:26 < pabs3> yep, they copy the debian package, like most of their stuff 12:26 < mwiriadi> yep 12:26 < mwiriadi> Fedora is having a huge bitch about ubuntu 12:27 < mwiriadi> claiming stuff is theirs when they just copied Fedora stuff 12:30 < pabs3> uh, I thought Fedora used RPM? 12:30 < pabs3> and Ubuntu used .debs 12:37 < mwiriadi> pabs3, they do but what they do is take Fedora source and package it in deb and then claim that they created it. 12:37 < mwiriadi> not the deb but other stuff. 12:37 < mwiriadi> Like system-config-selinux 12:37 < mwiriadi> things like that 12:39 < pabs3> so, they are packaging stuff fedora wrote? sounds like what any distro does with upstreams 12:40 < pabs3> except the claim that they created it, slap a GPL violation on their ass if they really did (and you use the GPL for that stuff) 12:44 < mwiriadi> Yeah 12:44 < mwiriadi> but Fedora doesn't because there is no alteration to the source 12:44 < mwiriadi> and the Ubuntu users eat it up. 12:44 < mwiriadi> Debian uses a heap of Fedora stuff but they don't say it's debian created 12:44 < mwiriadi> We (Fedora) has a real good relationship with debian 12:45 < mwiriadi> I speak to deb dev's all the time 12:47 < pabs3> what Fedora stuff does Debian use? I know we have rpm, mach/mock, yum and stuff, didn't know there was much more 12:55 < mwiriadi> system-config stuff 12:55 < mwiriadi> rpm isn't fedora's anymore it's been split off into it's indepedant which is a great idea 12:56 < pabs3> ah 13:17 -!- MangoFusion [n=jamesu@host86-147-212-148.range86-147.btcentralplus.com] has quit ["Not here"] 13:23 -!- pabspabspabs [n=pabs@d122-105-74-150.per9.wa.optusnet.com.au] has joined #synfig 13:25 -!- pabs3 [n=pabs@d122-105-74-150.per9.wa.optusnet.com.au] has quit [Read error: 110 (Connection timed out)] 13:37 -!- factor [n=factor@ip68-14-160-70.ok.ok.cox.net] has quit [Read error: 104 (Connection reset by peer)] 13:44 -!- pabspabspabs is now known as pabs3 13:44 < mwiriadi> pabs3, ping 13:45 < pabs3> yo 13:45 < mwiriadi> does libsynfig call the binaries for imagemagick 13:45 < pabs3> there are two plugins that use imagemagick, one uses $PATH/convert and the other uses the C++ library 13:46 < mwiriadi> cheers 13:46 < mwiriadi> just trying to fix those problems 13:46 < pabs3> you want to build both, since one the imagemagick allows import, and the other one produces very nice .gif images with 7.3 13:46 < mwiriadi> or the packaging problems 13:46 < pabs3> I replied to the bug about some of the modules 13:46 < mwiriadi> k 13:47 < pabs3> the docs are just doxygen output, not user docs (those are on the site) 13:47 < mwiriadi> yep so they should go in the devel package 13:48 < pabs3> in debian we have a -doc package so the buildds don't need to install the docs to build synfigstudio 13:50 < mwiriadi> Yeah 13:50 < mwiriadi> Similar 13:50 < mwiriadi> Interesting I have GraphicsMagick-c++-devel installed 13:51 < mwiriadi> and yet it didn't pick it up 13:51 < AkhIL> hey. I just got and compiled svn ETL and it returns $ ETL-config --version 13:52 < AkhIL> ETL 0.4.10 13:52 < pabs3> GraphicsMagick-c++-devel != imagemagick 13:52 < AkhIL> but synfig-core requires 11 13:52 < pabs3> woops! 13:53 < pabs3> AkhIL: try rerunning autoconf 13:53 < pabs3> autoreconf rather 13:54 -!- factor [n=factor@ip68-14-160-70.ok.ok.cox.net] has joined #synfig 13:54 < AkhIL> ok.. thanks 14:00 < pabs3> mwiriadi: GraphicsMagick is a fork of ImageMagick, which synfig doesn't use 14:00 < mwiriadi> k 14:01 < mwiriadi> I'm still not getting mng output 14:01 < pabs3> got libmng.h? 14:01 < mwiriadi> nm 14:01 < mwiriadi> nope stupid devel file 14:02 < mwiriadi> What is gradient supplied by then? 14:02 < mwiriadi> and Using OpenEXR's "half" type ------> no 14:03 < pabs3> gradient is a bunch of layer types and should be always built 14:03 < pabs3> leave OpenEXR half off, it makes things slower apparently (comment on the build instructions page) 14:10 < mwiriadi> Interesting I'm still getting Magick++ -------------------------> no 14:11 < dooglus> mwiriadi: it checks for the existance of Magick++.h 14:11 < dooglus> I have it in /usr/include/Magick++.h 14:12 < pabs3> in debian it is in libmagick++9-dev 14:16 < mwiriadi> /usr/include/GraphicsMagick/Magick++.h 14:28 < mwiriadi> Is there an option to specify where that library is? 14:29 < pabs3> try ImageMagick-c++-devel instead of GraphicsMagick-c++-devel 14:31 < mwiriadi> trying now 14:32 < mwiriadi> Yep that worked cheers 14:35 < mwiriadi> What is synfig_modules.cfg 14:35 < mwiriadi> for? 14:35 < pabs3> libsynfig uses it to determine which modules to load at runtime 14:36 < mwiriadi> so it's required for synfig-devel or synfig? 14:36 < mwiriadi> synfig 14:46 < dooglus> libsynfig I would think 14:46 < mwiriadi> Does synfig_modules.cfg need to be overwridden each time there is an update 14:47 < mwiriadi> as in it has a list generated of what modules to load? 14:47 < mwiriadi> Cause we are trying to figure whether the cfg needs to be updated after each build? 14:49 < mwiriadi> dooglus, 14:49 < pabs3> not sure what you mean 14:49 < mwiriadi> What does the cfg file do it loads modules 14:49 < mwiriadi> if you add more modules in the future 14:49 < mwiriadi> how will it know what modules to load? 14:50 < mwiriadi> If the config file isn't replaced 14:50 < pabs3> we'll add them to synfig_modules.cfg.in and it will be installed 14:50 < mwiriadi> Yes but we have an option called (noreplace) 14:51 < mwiriadi> now I'll alter it 14:51 < mwiriadi> so it does replace it 14:53 -!- pabspabspabs [n=pabs@d122-105-74-150.per9.wa.optusnet.com.au] has joined #synfig 14:55 < pabspabspabs> what does noreplace do? 14:55 < mwiriadi> It doesn't replace the config files 14:55 < mwiriadi> so it creates a file called x.conf.rpmnew 14:56 < mwiriadi> and it's up to you to check it 14:56 < mwiriadi> this causes breakages on updates sometimes 14:56 < mwiriadi> the other thing is to break upgrades with replace modified config files 15:00 -!- factor [n=factor@ip68-14-160-70.ok.ok.cox.net] has quit ["Ex-Chat"] 15:01 < pabspabspabs> hmm, ok. in debian we have mechanisms to upgrade when the config is not user modified, or prompt if it is 15:02 -!- pabs3 [n=pabs@d122-105-74-150.per9.wa.optusnet.com.au] has quit [Nick collision from services.] 15:02 -!- pabspabspabs is now known as pabs3 15:03 < mwiriadi> Yeah I just asked that 15:03 < mwiriadi> I think we do as well I'm going to double check I think it's an md5 macro 15:16 -!- pabspabspabs [n=pabs@d122-105-74-150.per9.wa.optusnet.com.au] has joined #synfig 15:18 -!- pabs3 [n=pabs@d122-105-74-150.per9.wa.optusnet.com.au] has quit [Nick collision from services.] 15:18 -!- pabspabspabs is now known as pabs3 15:22 -!- _Elk [n=_Elk@ti0110a340-0355.bb.online.no] has joined #synfig 15:37 -!- pabspabspabs [n=pabs@d122-105-74-150.per9.wa.optusnet.com.au] has joined #synfig 15:37 -!- pabs3 [n=pabs@d122-105-74-150.per9.wa.optusnet.com.au] has quit [Nick collision from services.] 15:37 -!- pabspabspabs is now known as pabs3 15:47 -!- pabs3 [n=pabs@d122-105-74-150.per9.wa.optusnet.com.au] has quit ["Don't rest until all the world is paved in moss and greenery."] 15:47 -!- pabs3 [n=pabs@d122-105-74-150.per9.wa.optusnet.com.au] has joined #synfig 16:17 -!- genete [n=Genete@84.122.52.245.dyn.user.ono.com] has joined #synfig 16:19 -!- genete [n=Genete@84.122.52.245.dyn.user.ono.com] has quit [Client Quit] 17:07 -!- pabspabspabs [n=pabs@d122-105-74-150.per9.wa.optusnet.com.au] has joined #synfig 17:10 -!- pabs3 [n=pabs@d122-105-74-150.per9.wa.optusnet.com.au] has quit [Nick collision from services.] 17:10 -!- pabspabspabs is now known as pabs3 17:33 -!- pabspabspabs [n=pabs@d122-105-74-150.per9.wa.optusnet.com.au] has joined #synfig 17:51 -!- pabs3 [n=pabs@d122-105-74-150.per9.wa.optusnet.com.au] has quit [Read error: 110 (Connection timed out)] 18:00 -!- _Elk [n=_Elk@ti0110a340-0355.bb.online.no] has quit [] 18:08 -!- pabspabspabs is now known as pabs3 18:14 * dooglus is working on implementing the first suggestion in http://synfig.org/Linking_to_Blines 18:36 -!- pabspabspabs [n=pabs@d122-105-74-150.per9.wa.optusnet.com.au] has joined #synfig 18:40 -!- pabspabspabs [n=pabs@d122-105-74-150.per9.wa.optusnet.com.au] has quit [Client Quit] 18:53 -!- pabs3 [n=pabs@d122-105-74-150.per9.wa.optusnet.com.au] has quit [Read error: 110 (Connection timed out)] 20:57 < AkhIL> http://pastebin.ca/930698 - compiling falls because of splash 21:01 < rore> baaah don't hurt me, not my fault /o\ 21:04 * AkhIL bugfinder 8) 21:09 -!- genete [n=Genete@84.122.51.163.dyn.user.ono.com] has joined #synfig 21:11 < genete> hi 21:11 < genete> dooglus: can you give some details on what you're working on? 21:11 -!- _Elk [n=_Elk@ti0110a340-0355.bb.online.no] has joined #synfig 21:12 * genete rubs his hands together... 21:15 < AkhIL> genete: can you help me ? http://pastebin.ca/930698 21:15 < genete> hi AkhIL 21:16 < AkhIL> Yah... hello! 21:16 < genete> rore's splash seems to look for the logo.sif file 21:16 < genete> I don't have that error 21:16 < genete> it is weird 21:17 < genete> If I touch the splash maybe I can repeat the compilation and show you the messages? 21:18 < genete> let me update to the current svn and make a test for you... 21:28 < dooglus> genete: I'm trying to get the "select ducks then right-click on bline and "link to bline"" to work 21:29 < genete> dooglus: nice!! 21:29 < dooglus> genete: it's mostly working - but if I draw 2 lines, and link the end of one to the middle of the other, and then link the end of the other to the middle of the 1st - it crashes :) 21:29 < dooglus> genete: the weird thing is, it crashes in a way that doesn't let me debug it 21:30 < dooglus> no stack trace, no valgrind error - just a crash 21:30 < genete> circular references? 21:30 < dooglus> AkhIL: I introduced that bug yesterday, sorry. 21:31 < dooglus> AkhIL: I thought all gradients had widths after canvas version 0.1 - but I guess I was wrong! 21:31 < genete> so the problem is in the logo.sif not rore's splash directly 21:32 < rore> yay!, not my fault, I told him :D 21:32 < dooglus> genete: yes, it is. logo.sif's fault :) 21:33 < dooglus> genete: looks like some monkey edited the logo.sif (r1429) canvas version without really saving it as a 0.3 canvas... 21:34 < genete> dooglus: when you introduced the bug you didn't notice it? maybe you did not render splashs again and AkhIL yes. 21:34 < genete> monkey? 21:36 < dooglus> genete: I didn't re-render the images/ folder - only the examples/ folder 21:36 < dooglus> AkhIL: quick fix - edit images/logo.sif and change the genete: ok, it was me :) 21:37 < AkhIL> thanks i'll wait next commint. 21:38 < dooglus> AkhIL: committing now 21:38 < AkhIL> commit* 21:38 < dooglus> ok, done 21:38 < dooglus> there were a few older commits queued up that I had forgotten about 21:39 < dooglus> genete: are you interested to see the 'Link to BLine' thing? I can check it what I have if you like 21:40 < genete> dooglus: what a quaestion! sure! 21:40 < genete> question* 21:43 < genete> rore: how can I make a big circle in GImp? 21:44 < rore> with the "elipse selection" tool, draw a circle (and hold the shift key if you want it to be a perfect circle), then fill with whatever color you like :) 21:45 < genete> I've tried it but seems to select nothing if there is only a white background 21:45 < genete> should be there other layer? 21:46 < rore> no, the "default" background should be enough 21:47 < rore> isn't there a kind of symbol on the mouse cursor, like a "+" or "-" or "n" ? (in addition to the cursor cross and "circle" for the elipse) 21:47 < genete> Mmm the default ellipse selection mode is sustractive 21:47 < genete> it is not ntuiive 21:47 < genete> intuitive 21:48 < genete> thanks rore! you roks! 21:48 < rore> no, the default is the 1st mode, the simple selection, if it's substractive it's that it remembered some manipulations you did before :) 21:49 < rore> I know gimp is not really intuitive, but not *that much* :D 21:50 < CIA-35> synfig: dooglus * r1864 /synfig-studio/trunk/src/synfigapp/actions/ (16 files): Tidy up some is_candidate() methods. 21:50 < CIA-35> synfig: dooglus * r1865 /synfig-core/trunk/src/modules/ (example/filledrect.cpp mod_geometry/circle.cpp): Don't allow negative feather parameters for circles (or for the example 'filled rect' layer). 21:50 < CIA-35> synfig: dooglus * r1866 /synfig-studio/trunk/images/logo.sif: Change the canvas version back to 0.1. It is a 0.1 version canvas - in 0.3 the ValueNode would have a width. 21:52 < genete> dooglus: please ping me when you commit the link to bline thing 22:06 < dooglus> genete: I need to split my changes into groups 22:06 < dooglus> it's all mixed up at the moment 22:06 < genete> ok no hurry 22:09 -!- AkhIL [n=AkhIL@90.188.201.182] has quit [Read error: 104 (Connection reset by peer)] 22:10 < dooglus> genete: it's only for vertices at the moment too - no tangent or width converts 22:10 < dooglus> genete: and linking multiple ducks to a bline doesn't connect their amounts, etc. 22:10 < dooglus> it's just the basic operation at the moment 22:10 < genete> for sure 22:11 < genete> but I guess you need other monkey to make some tests ;) 22:11 -!- AkhIL [n=AkhIL@90.188.201.182] has joined #synfig 22:17 < dooglus> yup 22:17 < dooglus> I'm ready - except I've not tried building my commits yet 22:19 < dooglus> how should "Link to BLine" be written in the menu - just like that? Or just "Link"? (maybe confusing) - isn't it best to have it slightly different than the existing "Link" entry? 22:20 < genete> yes. sometimes the context menu is so confusing because it has two Disconnect things 22:20 < genete> oh! 22:20 < genete> that New link can be undone? 22:21 < genete> "BLine Link" 22:25 < genete> I guess you just select the vertex and convert it to composite 22:28 < dooglus> "BLine Link" sounds like a noun 22:28 < dooglus> I like my menu entries to be verbs 22:29 < genete> "Link to Bline" 22:29 < dooglus> committed, by the way 22:29 < dooglus> 1871 22:29 < genete> core too? 22:29 < dooglus> shouldn't be 22:30 < dooglus> I made core changes about 24h ago - adding 3 new params to 'bline tangent' ready for this stuff 22:30 < dooglus> you can have 'fixed length' tangents now, and add offset angles or multiple by a scale 22:30 < genete> I was up to date to that 22:30 < dooglus> ok 22:30 < genete> but didn't tried it =) 22:31 < genete> your cimmit on that seems so interesting 22:31 < dooglus> I was surprised to find that I couldn't add 90 degrees to a tangent easily 22:31 < dooglus> using existing ValueNode types 22:31 < dooglus> can you see how? 22:31 < genete> maybe you need to convert a tangent to radial composite and offset the angle 22:32 < genete> then it becomes not maipulable by the duck 22:32 < dooglus> it's more like I need to convert it *from* radial composite, isn't it? 22:32 < dooglus> I mean a "BLine Tangent" 22:32 < dooglus> I want to link to a point on a BLine, and have the link be at 90 degrees to the bline 22:32 < genete> yes *from* I still my brain worng on that =) 22:33 < dooglus> the "BLine Tangent" can give me a vector or an angle 22:33 < dooglus> (and now a Real for the length, too) 22:34 < dooglus> so I convert my BLinePoint to Radial Composite, getting 'Theta' and 'Radius' params. I convert 'Theta' to an "Add", and connect LHS to 90 and RHS to "BLine Tangent", but what about the "Radius" of the Radial Composite? 22:34 < dooglus> I didn't see any way of getting that to be the length of the tangent 22:35 < CIA-35> synfig: dooglus * r1867 /synfig-core/trunk/src/synfig/interpolation.h: Number the interpolation types in comments. 22:35 < CIA-35> synfig: dooglus * r1868 /synfig-studio/trunk/images/synfig_icon.sif: Replace "./././logo.sif#" with "logo.sif#". 22:35 < CIA-35> synfig: dooglus * r1869 /synfig-studio/trunk/images/sif_icon.sif: Replace "././././logo.sif#" with "logo.sif#". 22:35 < CIA-35> synfig: dooglus * r1870 /synfig-studio/trunk/images/about_icon.sif: Replace "./././logo.sif#" with "logo.sif#" 22:35 < CIA-35> synfig: dooglus * r1871 /synfig-studio/trunk/src/ (10 files in 3 dirs): Made a start at implementing the ideas in http://synfig.org/Linking_to_Blines . Doesn't work particularly well yet. 22:36 < genete> I see, so that's the reason you made the reals be able to be converted from a tangent lenght? 22:36 < dooglus> yes 22:36 < dooglus> just for completeness reall 22:36 < dooglus> y 22:37 < dooglus> I think missing converts Vector -> Real (length) and Vector -> Angle (atan2(y,x)) would be useful - don't you? 22:37 < dooglus> also Vector -> Real (vector.x) and Vector -> Real (vector.y) ? 22:38 < genete> yes for the first and don't understand the seconds 22:39 < genete> ah yes 22:39 < genete> wrap the x coordinate 22:39 < genete> or the y coordinate 22:39 < dooglus> just to let us resolve a vector into its x and y components 22:39 < genete> I think the most important are the two first 22:40 < dooglus> it's like "Composite" but in reverse 22:40 < dooglus> "Composite" lets us specify an X and a Y to make a Vector 22:40 < dooglus> these 2 let us specify a Vector to make an X or a Y (as Real values) 22:41 < dooglus> (they're like 'Cos' and 'Sin', but working on a Vector, rather than an Angle) 22:42 < genete> let me remember you some of my initial request on convert talk page: http://synfig.org/Talk:Convert 22:42 < dooglus> given a Vector, there are 4 things you might want to know about it: length, angle, x, and y. those are the 4 converts I'm talking about. 22:42 < dooglus> oh, yes. 22:42 < dooglus> how do you make your screencasts? 22:42 < dooglus> I forgot the program name 22:42 < genete> and also this one: http://synfig.org/Talk:Convert#Angle_Vector_Dot 22:42 < genete> don't change the theme =) 22:43 < genete> those were my initial convert request!!!! ;) 22:43 < genete> gtk-recordMyDesktop 22:43 < dooglus> how about deleting the sections of that page which have been done already, like Range? 22:43 < dooglus> (Interval) 22:43 < genete> for sure 22:44 < genete> and switch 22:54 < genete> dooglus: cool 22:54 < genete> beautiful 22:55 < genete> and beautiful crash too 8) 22:55 -!- factor [n=Factor@32.147.206.60] has joined #synfig 22:57 < dooglus> genete: I just tried a cycle of 5 lines: link a->b->c->d->e->a 22:57 < genete> dooglus: it also crash if you want to Link to Bline a vertex to the same Bline 22:57 < dooglus> it only crashes when you link e back to a (completing the loop) 22:57 < dooglus> http://dooglus.rincevent.net/random/out.ogg 22:57 < dooglus> genete: does it crash if you do the same thing the "old fashioned" way? 22:58 < genete> I'll try 23:00 < genete> yup 23:00 < genete> if you export the Bline and convert a vertex to Bline Vertex and then connect the Bline param to the exported one it crash 23:00 < genete> obvious 23:00 < genete> it is a circular reference 23:02 < dooglus> right 23:02 < dooglus> so I've not added a new bug 23:03 < dooglus> I've just made it a lot easier to trigger the old bug 23:03 < genete> just discovered it 23:03 < dooglus> I think the only reason it's crashing is because it's trying to work out what to put in the history dialog :) 23:04 < dooglus> the description is infinitely long :) 23:04 < genete> Mmm 23:04 < genete> sure? 23:04 < genete> I believe that there is a bline that have a bline inside a bline and .... 23:05 < genete> what makes it crash too 23:05 -!- catseye [i=3f57d51e@gateway/web/ajax/mibbit.com/x-2b6b53ab8cdcda33] has joined #synfig 23:05 < genete> hi catseye 23:05 < catseye> hi 23:05 < dooglus> genete: I think it's running out of stack due to infinite recursion 23:06 < genete> dooglus: yes but not only the history but the references too 23:06 < dooglus> genete: I don't think the reference counting will be a problem - other than that the circular reference might mean these lines can never been deleted from memory 23:06 < catseye> dose anyone know if there is a sound tutorial for synfig 23:07 < genete> dooglus: also link a end point to another bline crashes 23:07 < dooglus> catseye: there's no sound in synfig yet, unfortunately 23:07 < genete> there is not sound support in synfig yet :( 23:08 -!- mwiriadi [n=mwiriadi@124-169-123-199.dyn.iinet.net.au] has quit [Read error: 113 (No route to host)] 23:09 < catseye> oh then how do I out in my animation 23:09 < dooglus> genete: any idea how I find out which function is crashing? 23:10 < dooglus> catseye: out? 23:10 < catseye> put 23:10 < catseye> sorry 23:10 < dooglus> catseye: I guess you use some other video editing software afterwards 23:11 < genete> dooglus: no idea. I only can suggest to start adding console outputs for every step done to see when it crash 23:12 < genete> if debugger doesn't say you where is the stack overflow 23:12 < catseye> I guess it would be better to have a program more specific to fror that 23:12 < dooglus> genete: the stack overflow seems to trash the stack 23:12 -!- skrylar [n=jcearley@adsl-71-145-160-133.dsl.austtx.sbcglobal.net] has joined #synfig 23:12 < dooglus> but hitting a quick control-c caught it: 23:12 < dooglus> #420 0xb7d1ef29 in synfig::ValueNode::on_changed (this=0x9058518) at valuenode.cpp:292 23:12 < dooglus> #421 0xb7cf6d69 in synfig::Node::changed (this=0x9058518) at node.cpp:201 23:12 < dooglus> #422 0xb7cf6ddc in synfig::Node::on_changed (this=0x905c000) at node.cpp:302 23:12 < dooglus> #423 0xb7d1ef29 in synfig::ValueNode::on_changed (this=0x905c000) at valuenode.cpp:292 23:13 < genete> hi skrylar 23:13 < skrylar> Do the commit bots actualy help that much? :( 23:13 < skrylar> Hi 23:14 < genete> catseye: what's your platform? 23:14 < dooglus> skrylar: they let people know when there's stuff to update and rebuild 23:14 < skrylar> ah. 23:14 < catseye> xp 23:14 < skrylar> I've only ever seen them go crazy and spam channels. 23:14 < skrylar> But I think the guy in lightfeather broke his at the time 23:15 < dooglus> skrylar: but CIA-xx is pretty slow - so I usually end up manually saying when I've committed something 23:15 < genete> catseye: I'm on linux. Only wanted to recommned you some good edition programs to do that :/ 23:15 < dooglus> "editing" 23:15 < genete> thanks dooglus :) 23:15 < catseye> I wish I knew some 23:16 < skrylar> For video? 23:16 < dooglus> this stack track is still printing... it's on frame 10,000 now 23:16 < dooglus> catseye: doesn't Windows Movie Maker let you add sound to a silent movie? 23:16 < dooglus> catseye: I'm sure it must do 23:17 < skrylar> It does, but it wont export to any format except for wmv. 23:17 < dooglus> aah, I see 23:17 < catseye> I haven't thought of that 23:17 < catseye> I'll try it 23:17 < catseye> thanks 23:18 < skrylar> On windows you could probably use VDub to put a soundtrack on a movie. 23:18 < genete> dooglus: I think you cannot solve looped blines problem. Only detect it and avoid linking it. 23:19 < genete> bline A and bline B. bline A has a vertex that has an internal paramter that points to bline B 23:19 < genete> and bline B has a vertex that has a parameter that points to bline A 23:19 < skrylar> Oh noes circular reference :< 23:19 < genete> there is a circular reference that would produce a stack overflow for sure 23:19 < skrylar> There's one way to handle that, methinks. 23:20 < catseye> if movie maker dosen't wok I try VDub 23:20 < dooglus> genete: I think the draw tool lets me link a line to itself doesn't it? 23:20 < genete> but not in that sense 23:20 < skrylar> A references a coordinate of B, and B references a coordinate of A, but you could instead create an overarching item C that provides those coordinates? 23:20 < genete> you link vertex not converted vertex 23:20 < skrylar> That didn't make as much sense as I hoped it would ._. 23:21 < catseye> thanks again 23:21 -!- catseye [i=3f57d51e@gateway/web/ajax/mibbit.com/x-2b6b53ab8cdcda33] has left #synfig [] 23:21 < dooglus> skrylar: the problem is when line A references line B, and line B references line A 23:22 < skrylar> If it detects you trying to make a reciprocal link, it would create a Link item which was subscribed to an event updated when A or B are modified 23:22 < skrylar> So A and B become loosely coupled to C 23:22 < skrylar> Which removes the circular reference 23:22 < dooglus> also A -> B -> C -> A, etc. also crashes 23:23 * skrylar doesn't see how it would do that... 23:23 < skrylar> I don't know if you have the event system in place to do it though. 23:23 < dooglus> http://dooglus.rincevent.net/random/out.ogg shows how 23:23 < skrylar> Are you trying to join two shapes together? 23:23 < skrylar> Or just link vertices 23:23 < dooglus> trying to join a series of lines together 23:23 < skrylar> Ah, it wouldn't work for that then. 23:24 < skrylar> i was thinking if you were trying to link two points together 23:24 < dooglus> joining vertices onto arbitrary positions on blines, without having to create a new vertex on the bline 23:24 < genete> skrylar: we are trying to link a vertex to a bline without an speciphic vertex on it 23:24 < skrylar> Hrmm. 23:24 < skrylar> Well, you can prevent stack overflows by just remembering entities you've already read, or you should be able to anyway. 23:26 < dooglus> http://dooglus.rincevent.net/random/linked-with-loops.png is fine - no crash there 23:27 < dooglus> but that's linking vertices together - not using the "BLine Vertex" convert 23:27 < genete> that's what I said =) 23:27 < dooglus> yes 23:27 * skrylar thinks. 23:28 < dooglus> skrylar: we can keep a flag with each node for whether it's been "visited" in this sweep or not 23:28 < dooglus> but we sweep the tree of nodes each time a value is changed, so it should be fast 23:28 < skrylar> Trying to figure out why that wouldn't work 23:29 < skrylar> Theoretically you wouldn't have to do the mark/sweep thing unless you were combining two shapes 23:29 < skrylar> Once it's a single shape, you can just draw the shape 23:29 < dooglus> the thing is, one vertex's position depends on a different bline 23:29 < dooglus> so whenever that bline is modified, we need to update the vertex position 23:30 < dooglus> which updates the bline that vertex is on, which triggers an update of the vertices which depend on that bline, etc. 23:31 < dooglus> the sweep itself is relatively small, generally, but I don't want to have to clear every node's "visited" flag before each sweep - that's the expensive part 23:31 < skrylar> I was actually thinking you would have a local visited vector 23:31 < skrylar> when it passed over an item, it pushed a reference on to the vector 23:32 < skrylar> At the end of the operation you just dump the vector 23:32 < dooglus> ok 23:32 < dooglus> works for small sweeps, but gets slow O(n^2)? 23:32 < skrylar> I also figured you'd only have to do it when combining things ;) 23:32 < dooglus> no, we need to do it whenever changing any parameter, really 23:33 < skrylar> I don't think I understand the situation, sadly :( 23:33 < skrylar> My dealings with vector graphics on the code side is pretty weak. 23:33 < dooglus> any parameter can affect any other, due to the "convert" menu 23:33 < dooglus> it's a synfig thing really - do you know synfig much? 23:33 < genete> this is not the first time we deal with circular references 23:34 < skrylar> I've seen it a couple times, just downloaded it last night. 23:34 < skrylar> Looks a lot like a tool I poked with a while back called moho :D 23:34 < genete> eh 23:34 < genete> synfig is better ;) 23:35 < skrylar> Except it crashes making circular shapes ;) 23:35 * skrylar goes to eat dinner. 23:35 < genete> because it can link vertices to blines =) 23:35 < dooglus> hah 23:37 < dooglus> all nodes already have a "mutable int time_last_changed_" field 23:37 < dooglus> that seems to be the solution... 23:37 < genete> cool 23:48 -!- kwizart [n=kwizart@fedora/kwizart] has joined #synfig 23:49 < kwizart> hi all ! 23:49 < kwizart> salut à tous ! 23:49 < kwizart> https://bugzilla.redhat.com/show_bug.cgi?id=428567 23:49 < kwizart> ^^review request for Fedora ! (i'm the reviewer 23:49 < kwizart> ) 23:50 < dooglus> hi :) 23:50 < genete> hi kwizart 23:50 < kwizart> well i really don't understand what's the aim of having ETL as a separate package! 23:51 < kwizart> why don't you just install headers in some place and ask for this place from the dependants apps ? 23:53 < dooglus> I don't know why its a separate package. It could just as easily be part of the 'synfig' package. 23:53 < kwizart> because using pkg-config for just some headers and no libs seems strange... 23:53 < dooglus> pabs3 would be the one to ask, but he's on Australia time 23:54 < dooglus> it's a template library - that's the idea 23:54 < kwizart> but do you know many project using it ? 23:54 < dooglus> maybe the intention was that other projects could use it - it offers stand-alone functionality 23:54 < dooglus> I suspect a grand total of 1 23:55 < dooglus> or 2 if you could synfig-core and synfig-studio as two 23:55 < dooglus> if you *count* 23:56 < kwizart> then i think it would be better to have it built within synfig, and then provide a separate package (if needed ) 23:56 < kwizart> so i will state this on the review, because i don't like how pkgconfig is tweaked... 23:57 < dooglus> the "WORDS_BIGENDIAN" mentioned in the review is used in the synfig source code 23:57 < dooglus> synfig-core/trunk/src/modules/mod_bmp/trgt_bmp.cpp:#ifdef WORDS_BIGENDIAN 23:57 < dooglus> for example 23:58 < kwizart> dooglus, that why this value should be set on the synfig config.h, and not search for another package... 23:58 < kwizart> package are not supposed to export a config.h --- Log closed Fri Mar 07 00:00:00 2008