--- Log opened Tue Apr 10 00:00:57 2007
01:14 -!- igli [n=igli@unaffiliated/igli] has quit [Remote closed the connection]
01:33 -!- igli [n=igli@unaffiliated/igli] has joined #synfig
03:14 -!- pxegeek [n=chatzill@c-71-59-140-184.hsd1.or.comcast.net] has joined #synfig
03:16 < dooglus> hey
03:16 < dooglus> I was just looking for you
03:18 < dooglus> remember the problem I was having with one of your icons where the corners were being cut off?
03:19 < dooglus> well, I fixed the bug recently - this one: http://dooglus.rincevent.net/synfig/sharp.html
03:19 < dooglus> but now, 'sharp cusps' are only drawn for verticies which have split tangents
03:21  * pxegeek looks around
03:21 < pxegeek> Who me?
03:21 < pxegeek> Ah yes
03:21 < pxegeek> Very nice!
03:21 < pxegeek> Was it just cases that weren't considered, or were there bugs in the code?
03:22 < pxegeek> Chanserv seems a little sluggish today - or maybe the IRC server I'm attached to is slow
03:22 < dooglus> there were several bugs in the code
03:23 < dooglus> one was particularly odd
03:23 < dooglus> but the upshot is that the corners are now cut off all over the place unless you split the tangents
03:23 -!- igli [n=igli@unaffiliated/igli] has quit [Remote closed the connection]
03:23 < dooglus> I think it's good to give the user this per-corner choice
03:23 < dooglus> but it does mess up existing files a little
03:24 < dooglus> so I'm not sure what to do about it
03:25 < pxegeek> How bad does it look - have you got a screenshot of one of the examples?
03:26 < dooglus> I'll make one.
03:26 < dooglus> the fix was in r445 for future reference
03:26 < pxegeek> I guess technically they're bugs in the sif files that are now exposed.... ;)
03:26 < dooglus> um, I mean 446
03:27 < pxegeek> There's nothing to say we can't update the examples that ship with synfig, right?
03:28  * pxegeek wonders if we need a compatibility mode for things like this and gamma adjustments
03:30 -!- igli [n=igli@unaffiliated/igli] has joined #synfig
03:30 < pxegeek> hey Igli
03:31 < igli> hey pxegeek :)
03:32 < pxegeek> half past 2 in the morning?
03:34 < dooglus> http://dooglus.rincevent.net/synfig/1176168745.png
03:35 < pxegeek> :)  That looks familiar
03:35 < dooglus> the top half is with the fixed code; the bottom half is with the fixed code *and* split tangents, to get the corners re-drawn
03:36 < pxegeek> OK. I can certainly fix up the icons.
03:37 < dooglus> the code before was saying 'whenever both ends of an outline have zero-length tangents, don't draw cusps at the start end"
03:37 < dooglus> which seems a bit arbitrary to me
03:37 < dooglus> which end is the 'start end'?  you're unlikely to know or care as the artist - it's a loop after all
03:37 < pxegeek> They're probably the most dependent on outlines and vertices without tangents
03:38 < dooglus> I think it's good to be able to have 'blunt corners' if you want them
03:38 < pxegeek> Yes.
03:38 < dooglus> that wasn't possible before with zero tangents
03:38 < dooglus> you had to use a fiddle, like making a 0.001 tangent or something
03:39 < pxegeek> sneaky
03:43 < dooglus> so now I can make shapes like this: http://dooglus.rincevent.net/synfig/square.png
03:43 < dooglus> 2 sharp corners, 2 blunt
03:44 < pxegeek> Looks like the blunt starts from vertex
03:45 < pxegeek> i.e. it doesn't look like it would if you just took a ribbon and folded it at right angles
03:47 < dooglus> no, it draws a line parallel to a line joining the 2 verticies, but offset by half the width
03:47 < dooglus> so yes, it stops at the vertex
03:55 < pxegeek> Makes sense
03:55 < dooglus> anyway, I'm off.
03:55 < dooglus> ttyl
03:55 < pxegeek> later - have fun
03:56 < dooglus> I wouldn't worry about fixing up the icons - the corners aren't noticeable at the size they'll be used
03:56 < pxegeek> hmmm...  I guess pabs is busy packing/unpacking
03:56 < pxegeek> Ah- but you're talking to a Virgo....
03:56 < pxegeek> ;)
04:54 -!- igli [n=igli@unaffiliated/igli] has quit [Remote closed the connection]
04:59 -!- crazy_bus [n=philip@139.168.106.228] has joined #synfig
05:00 -!- igli [n=igli@unaffiliated/igli] has joined #synfig
05:38 -!- crazy_bus [n=philip@139.168.106.228] has quit [Read error: 104 (Connection reset by peer)]
05:38 -!- igli_ [n=igli@82.153.8.78] has joined #synfig
05:39 -!- igli_ [n=igli@82.153.8.78] has quit [Read error: 104 (Connection reset by peer)]
05:40 -!- igli_ [n=igli@82.153.8.78] has joined #synfig
05:41 -!- igli_ [n=igli@82.153.8.78] has quit [Read error: 104 (Connection reset by peer)]
05:41 -!- igli_ [n=igli@82.153.8.78] has joined #synfig
05:42 -!- igli_ [n=igli@82.153.8.78] has quit [Read error: 104 (Connection reset by peer)]
05:43 -!- igli_ [n=igli@82.153.8.78] has joined #synfig
05:54 -!- igli [n=igli@unaffiliated/igli] has quit [Read error: 110 (Connection timed out)]
07:05 -!- pxegeek [n=chatzill@c-71-59-140-184.hsd1.or.comcast.net] has quit [Read error: 113 (No route to host)]
07:09 -!- igli_ [n=igli@82.153.8.78] has quit [Remote closed the connection]
08:23 -!- igli [n=igli@unaffiliated/igli] has joined #synfig
12:20 -!- crazy_bus [n=philip@139.168.106.228] has joined #synfig
12:33 -!- zipola [n=zipola@cable-vrk-fe5cdd00-177.dhcp.inet.fi] has joined #synfig
12:42 -!- crazy_bus [n=philip@139.168.106.228] has quit [Read error: 104 (Connection reset by peer)]
14:09 -!- mememeandme [n=mememean@jabber.hst.ru] has joined #synfig
14:10 < mememeandme> hi
14:12 < mememeandme> I have some problems installing synfig and can't figure out how to correct them
14:12 < mememeandme> could anybody help?
14:17 -!- tokyo [n=tokyo@c-67-175-134-199.hsd1.il.comcast.net] has quit [Read error: 110 (Connection timed out)]
14:19 -!- zotz [n=zotz@24.244.163.157] has joined #synfig
14:22 < dooglus> mememeandme: hopefully
14:22 < dooglus> you still there?
14:22 < mememeandme> yeah
14:23 < dooglus> what's the problem?
14:24 < mememeandme> I have the newest autoconf and automake avaliable in my linux distribution but still have that kind of error:
14:24 < mememeandme> + libtoolize -c -f --ltdl
14:24 < mememeandme> You should add the contents of `/usr/share/libtool-1.5/aclocal/libtool.m4' to `aclocal.m4'.
14:24 < mememeandme> You should add the contents of `/usr/share/aclocal/ltdl.m4' to `aclocal.m4'.
14:24 < mememeandme> Putting files in AC_CONFIG_AUX_DIR, `config'.
14:24 < mememeandme> + aclocal -I config
14:24 < mememeandme> config/pkgconfig.m4:5: warning: underquoted definition of PKG_CHECK_MODULES
14:24 < mememeandme>   run info '(automake)Extending aclocal'
14:24 < mememeandme>   or see http://sources.redhat.com/automake/automake.html#Extending-aclocal
14:24 < mememeandme> config/libxml.m4:10: warning: underquoted definition of AM_PATH_XML
14:24 < mememeandme> config/libxml.m4:205: warning: underquoted definition of AM_PATH_XML2
14:24 < mememeandme> config/libxml++.m4:3: warning: underquoted definition of AM_LIBXMLPP
14:24 < mememeandme> config/freetype2.m4:7: warning: underquoted definition of AC_CHECK_FT2
14:24 < mememeandme> + autoheader
14:24 < mememeandme> + autoconf -o configure
14:24 < mememeandme> configure.in:306: error: possibly undefined macro: AC_LIB_LTDL
14:24 < mememeandme>       If this token and others are legitimate, please use m4_pattern_allow.
14:24 < mememeandme>       See the Autoconf documentation.
14:24 < mememeandme> + set +x
14:24 < mememeandme> bootstrap: Failure.
14:24 < mememeandme> bootstrap: Cleaning up...
14:24 < dooglus> you're building from the svn sources?
14:25 < mememeandme> yes
14:25 < dooglus> I don't know much about the automake/autoconf stuff
14:26 < dooglus> have you tried with automake1.9 and autoconf2.59?
14:26 < dooglus> (I only ask because that's what worked for me)
14:26 < mememeandme> that's what I have
14:26 < dooglus> is this in the ETL directory?
14:26 < dooglus> or did you do ETL already?
14:27 < dooglus> ah - lunch is served...  :)  I'll be back in a while.  hopefully you can stick around.
14:27 < mememeandme> ok
14:29 < mememeandme> ETL is okay... I have stable version in my repository. or should I rebuild it from svn too?
14:29 < dooglus> svn, ideally
14:30 < mememeandme> ok, I'll try this.
14:30 < dooglus> I don't think it will fix this problem,
14:30 < dooglus> but there are changes to ETL in svn which you may as well have
14:32 < mememeandme> ok, thanks :)
14:49 < dooglus> does the bootstrap step work for ETL?
14:52 < mememeandme> yes
15:02 < dooglus> and synfig-core?
15:04 < mememeandme> I don't know yet... I'm trying to tell it to use correct etl from repository...
15:08 < dooglus> export PKG_CONFIG_PATH=/usr/local/lib/pkgconfig
15:08 < dooglus> or some such
15:08 < dooglus> tells it where to look for the .pc file
15:12 < mememeandme> same error with synfig-core :(
15:16 < dooglus> can you pastebin the whole output?
15:16 < dooglus> ie. go to http://pastebin.com/ and paste it there
15:18 < mememeandme> whole bootstrap output?
15:19 < dooglus> please, if you can
15:20 < dooglus> oh, this is different from what I see:
15:20 < dooglus> 14:24 < mememeandme> configure.in:306: error: possibly undefined macro: AC_LIB_LTDL
15:22 < dooglus> did you install package 'libltdl3-dev'?
15:25 < mememeandme> I need libltdl-devel or libltdl-devel-static too?
15:25 < dooglus> I don't know - I've only ever built on debian and ubuntu
15:26 < dooglus> we only have these 2:
15:26 < dooglus> libltdl3 - A system independent dlopen wrapper for GNU libtool
15:26 < dooglus> libltdl3-dev - A system independent dlopen wrapper for GNU libtool
15:26 < dooglus> nothing about 'static'
15:27 < dooglus> do you have a file /usr/share/aclocal/ltdl.m4 ?
15:27 < mememeandme> static is static libraries for libitdl
15:28 < dooglus> on my system, that ltdl.m4 is the file which defines the "possibly undefined macro: AC_LIB_LTDL" you're seeing
15:33 < dooglus> which distro are you using?
15:33 < dooglus> it would be good to get the build instructions wiki page updated with a list of packages needed for it
15:34 < mememeandme> i'm making rpm package for ALTLinux...
15:34 < mememeandme> oh, it's okay now...
15:35 < mememeandme> hope there will be no other errors..
15:36 < dooglus> did you need that 'static' version?
15:36 < dooglus> or just libltdl-devel ?
15:38 < mememeandme> I've tried with both... 
15:39 < mememeandme> next time I'll use only static, i think..
15:39 -!- zotz [n=zotz@24.244.163.157] has quit ["Leaving"]
15:40 < mememeandme> oh... new error.. this time it's distro-related, i think..
15:42 < mememeandme> no, it's problem with ffmpeg
15:42 < mememeandme> i'll fix it and send a patch
15:47 < mememeandme> we have ffmpeg too new for synfig :(
15:49 < dooglus> ok
16:29 < mememeandme> patch done
16:34 < igli> yay!
16:34 -!- igli [n=igli@unaffiliated/igli] has quit [Read error: 104 (Connection reset by peer)]
16:36 -!- igli [n=igli@unaffiliated/igli] has joined #synfig
16:37 -!- igli [n=igli@unaffiliated/igli] has quit [Read error: 104 (Connection reset by peer)]
16:37 < dooglus> where?  in the sf.net tracker?
16:39 < mememeandme> I'll submit it soon.... I just have to finish registration there..
16:45 < dooglus> ok
16:51 < mememeandme> now it's in the sf.net tracker
16:56 < dooglus> shouldn't you be mex4?
16:57 < mememeandme> btw, libltdl-devel without libltdl-devel-static is okay
16:58 < dooglus> ok
16:58 < mememeandme> mex3 is a short variant of mex3andme ;)
16:58 < dooglus> heh
16:58 < dooglus> what's this __STDC_CONSTANT_MACROS macro about?
16:58 < mememeandme> if it's mex4 - then i dunno where should be "and" %)
17:01 < mememeandme> it fixes problem of undefined INT64_C and it's new ffmpeg feature
17:03 -!- omry [n=omry@l85-130-137-160.broadband.actcom.net.il] has quit [Read error: 104 (Connection reset by peer)]
17:04 < mememeandme> there is no such declaration in libavutil/common.h
17:13 < dooglus> this sounds like a bug with ffmpeg doesn't it?
17:14 < dooglus> sound every app that uses it have to define a new macro?
17:22 < mememeandme> yeah... sort of.. but it's sort of they don't want to fix it or something like that.. ffmpeg packager advised me to send patches not to ffmpeg team but to developers of projects who use ffmpeg...
17:26 -!- omry [n=omry@l85-130-137-160.broadband.actcom.net.il] has joined #synfig
17:28 < dooglus> ok.
17:28 < dooglus> I guess it's better to put the flag into a Makefile somewhere, rather than lots of separate source files.
17:30 < dooglus> I just fixed a couple more synfig bugs in svn.  'home' and 'end' work now for jumping to the start and end of an animation
17:46 -!- omry [n=omry@l85-130-137-160.broadband.actcom.net.il] has quit [Read error: 104 (Connection reset by peer)]
17:50 < mememeandme> btw,
17:50 < mememeandme> make[2]: *** [installer_logo.png] Segmentation fault
17:50 < mememeandme> make[2]: Leaving directory `/usr/src/RPM/BUILD/synfig-studio/images'
17:50 < mememeandme> make[1]: *** [all-recursive] Error 1
17:50 < mememeandme> I mean,
17:50 < mememeandme> synfig -q installer_logo.sif -o installer_logo.png --time 0
17:50 < mememeandme> make[2]: *** [installer_logo.png] Segmentation fault
17:50 < mememeandme> make[2]: Leaving directory `/usr/src/RPM/BUILD/synfig-studio/images'
17:50 < mememeandme> make[1]: *** [all-recursive] Error 1
18:06 < dooglus> hmm
18:06 < dooglus> it works here
18:06 < dooglus> it creates this file: http://dooglus.rincevent.net/synfig/installer_logo.png
18:07 < dooglus> can you generate a backtrace?
18:08 < dooglus> ie.    cd synfig-studio/images;  gdb synfig
18:08 < dooglus> then:    run -q installer_logo.sif
18:15 -!- omry [n=omry@l85-130-137-160.broadband.actcom.net.il] has joined #synfig
18:24 -!- omry [n=omry@l85-130-137-160.broadband.actcom.net.il] has quit [Read error: 104 (Connection reset by peer)]
18:32 < mememeandme> I guess it is not that easy because I build package using our special tool that apsolutely unsuitable for debug :(
18:40 < mememeandme> *absolutely..
18:46 -!- omry [n=omry@l85-130-137-160.broadband.actcom.net.il] has joined #synfig
19:16 < dooglus> oops!
19:17 < dooglus> you can configure with --enable-optimization to get it suitable for debugging
19:17 -!- omry [n=omry@l85-130-137-160.broadband.actcom.net.il] has quit [Read error: 104 (Connection reset by peer)]
19:17 < dooglus> ./configure --help to check
19:21 < mememeandme> that's okay... i've almost compile it special for debug - it's faster anyway =)
19:22 < mememeandme> emmmm.... magic %)  now everything is ok...
19:23 -!- omry [n=omry@l85-130-142-163.broadband.actcom.net.il] has joined #synfig
19:25 < mememeandme> may be that was caused by some packages that are newer than needed..
19:26 < mememeandme> I'm not sure if all packages I used this time was the newest version..
19:37 -!- omry [n=omry@l85-130-142-163.broadband.actcom.net.il] has quit [Read error: 104 (Connection reset by peer)]
19:39 -!- omry [n=omry@l85-130-142-163.broadband.actcom.net.il] has joined #synfig
19:42 < dooglus> sometimes compiling for debugging will make bugs 'disappear'
19:48 < mememeandme> hmm.. the problem sure is related to version of BuildRequires. I've got the same error :) I'll try to figure out what could it be..
19:59 < dooglus> valgrind is good at finding this kind of problem
20:00 < dooglus> here's the ~/.valgrind suppression list I use: http://dooglus.rincevent.net/synfig/.valgrind
20:19 -!- omry [n=omry@l85-130-142-163.broadband.actcom.net.il] has quit [Read error: 104 (Connection reset by peer)]
20:39 -!- omry [n=omry@l85-130-142-163.broadband.actcom.net.il] has joined #synfig
22:34 < mememeandme> about segfault - seems, it's our distro problem or some other package bug..
22:34 < mememeandme> hope it would be ok :)
23:27 -!- igli [n=igli@unaffiliated/igli] has joined #synfig
--- Log closed Wed Apr 11 00:00:57 2007