--- Log opened Wed Feb 20 00:00:26 2008
00:16 < genete> ok, updated the extra or missing spaces.
00:16 < genete> thanks dooglus for point it out
00:18 -!- MangoFusion [n=jamesu@host81-132-205-163.range81-132.btcentralplus.com] has quit ["Not here"]
00:21 < CIA-27> synfig: dooglus * r1775 /synfig-studio/trunk/src/gtkmm/renderer_ducks.cpp: Limit the radius ducks to their quadrant before transforming. Otherwise rotations, negative stretches, etc. can cause problems.
00:21 < CIA-27> synfig: dooglus * r1776 / (synfig-studio/trunk/po/es.po synfig-core/trunk/po/es.po): Apply 1897389: Revised missed or extra spaces on Spanish translation.
00:21 < dooglus> ok, thanks. 
00:21 < dooglus> CIA beat me to it!
00:22 < dooglus> ah, wasn't meaning to commit the duck change yet - was still testing.
00:30 -!- omry [n=omry@bzq-84-108-20-56.cablep.bezeqint.net] has joined #synfig
00:31 < genete> dooglus: a tiny change: can you arrange the Misc. settings tab?
00:31 < genete> it needs the checkboxes to be grouped and the text to be left or right justified instead of centered
00:32 < dooglus> yes
00:32 < dooglus> can you suggest a specific layout?
00:33 < genete> I suggest this order:
00:33 < genete> Time Format
00:33 < genete> Unit system
00:34 < genete> Recent Files
00:34 < genete> BAckup interval
00:34 < genete> Browser command
00:34 < genete> and at the end the checkboxes
00:35 < genete> and left or right justified
00:35 < genete> for the text instead of centered
00:40 -!- Factor- [n=Factor@32.145.131.197] has quit ["mIRGGI meni puis"]
00:53 < genete> dooglus: add a polygon gives me an english word as resumee of the set of actions: "Move"
00:54 < genete> in the history panel
00:54 < dooglus> :)
00:56 < genete> there is not tool tip for metadata panel buttons: Add Metadata and Remove Metadata
00:56 < dooglus> ok
00:59 < genete> Also I obtain this history string when use the fill tool: "Establecer Layer Parameter (Circle168):Color"
01:00 < genete> Layer and Parameter are not trasnlated
01:01 < dooglus> aah, I see
01:02  * pixelgeek going home - catch you later
01:03 < genete> also move a duck with normal tool produces a: Establecer Value Node bla bla bla
01:03 -!- pixelgeek [i=86868805@gateway/web/ajax/mibbit.com/x-ee0eb7f217a9c863] has left #synfig []
01:03 < genete> Value Node is not translated
01:03 < dooglus> genete: did you make update-po?
01:04 < genete> i removed the es.po files
01:04 < dooglus> 			description = strprintf("%s %s", _("Value Node"),
01:04 < dooglus> it should be translated
01:08 < genete> well, too late to me. 
01:08 < genete> see you tomorrow.
01:08 < dooglus> night
01:08 < genete> nice anyway
01:08 < genete> night
01:08 -!- genete [n=Genete@84.122.39.172.dyn.user.ono.com] has quit ["Abandonando"]
02:37 -!- pixelgeek [n=chatzill@c-71-59-140-184.hsd1.or.comcast.net] has joined #synfig
02:47 < CIA-27> synfig: dooglus * r1777 /synfig-studio/trunk/src/gtkmm/dockable.cpp: Uncomment the code that gets add_button() to set the tooltip. The function only seems to still be used for the 'Meta' panel anyway.
02:47 < CIA-27> synfig: dooglus * r1778 /synfig-studio/trunk/src/gtkmm/renderer_ducks.cpp: Fix the displayed radius size while hovering over (or dragging) a radius duck.
02:47 < CIA-27> synfig: dooglus * r1779 /synfig-studio/trunk/src/gtkmm/ (duckmatic.cpp duckmatic.h workarea.cpp state_circle.cpp): Remove the special case code for when the screen is 'flipped'. Arbitrary transformations can be in place, so let's not try to special case them all. Just let the radius ducks go where they are transformed to.
02:47 < CIA-27> synfig: dooglus * r1780 /synfig-studio/trunk/src/gtkmm/dialog_setup.cpp: Reorder the rows so the checkboxes are together at the end.
03:16 < CIA-27> synfig: dooglus * r1781 /synfig-studio/trunk/src/gtkmm/app.cpp: Tidied up the default naming of the file and canvas upon creation.
03:16 < CIA-27> synfig: dooglus * r1782 /synfig-studio/trunk/src/gtkmm/ (instance.cpp instance.h): Moved the check for whether an instance has a real filename into a new method, since it's used in a couple of places.
03:16 < CIA-27> synfig: dooglus * r1783 /synfig-studio/trunk/src/gtkmm/canvasview.cpp: Don't show "Synfig Animation 1" in the title bar twice for unnamed documents. If the document isn't saved yet, only show the canvas's name.
03:17 < pixelgeek> +1 to 1783 commit
03:17 < pixelgeek> Have I been hanging out on digg too much?
03:17 < dooglus> yup
03:18 < pixelgeek> You still awake?
03:18 < pixelgeek> (obviously)
03:19 < pixelgeek> Did I recall you mentioning that you'd found an issue with references being cast incorrectly or something?
03:19 < pixelgeek> Did you make any progress with that one, or did genete distract you?  ;)
03:20 < dooglus> pixelgeek: lots of things distracted me :)
03:20 < dooglus> I have local changes that I've not committed yet, pending tests to see if they really help
03:21 < pixelgeek> Any specifics of where they might help?  
03:21  * pixelgeek looking forward to help test
03:33 -!- crazy_bus [n=philip@CPE-121-217-113-72.nsw.bigpond.net.au] has joined #synfig
03:45 < dooglus> pixelgeek: I was seeing a very repeatable crash from calling get_description on a ValueDesc
03:45 < dooglus> that's something that happens all the time, to format strings in the history panel
03:46 < dooglus> so I don't know why it was crashing for me, but doesn't crash when the history panel does it
03:46 < dooglus> that's what I'll look into before committing
04:39 < pixelgeek> Maybe it only happens randomly when the history panel does it 
04:39  * pixelgeek says hopefully
05:37 -!- pixelgeek [n=chatzill@c-71-59-140-184.hsd1.or.comcast.net] has quit [Read error: 113 (No route to host)]
06:10 -!- crazy_bus [n=philip@CPE-121-217-113-72.nsw.bigpond.net.au] has quit [Remote closed the connection]
06:20 < pabs3> anyone else listen to lugradio? could be cool to get them to review synfig 0.61.08
06:21  * pabs3 puts on todo list
07:50 -!- pabs3 [i=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."]
08:57 -!- pabs3 [i=pabs@d122-105-74-150.per9.wa.optusnet.com.au] has joined #synfig
09:11 -!- factor [n=factor@ip68-14-160-70.ok.ok.cox.net] has joined #synfig
09:33 < dooglus> pabs3: I listened to one of them but found it kind of irritating.
09:34 < pabs3> yeah?
09:37 < factor> no rore feb entry yet?
09:38 < dooglus> dooglus: maybe I should give it another go, but something in the personalities and energy of the presenters just bugged me
09:38 < dooglus> s/dooglus/pabs3/
09:39 < pabs3> perhaps it is the britishness ;)
09:39 < dooglus> I think it is something like that, yes
09:39 < dooglus> a particular type of britishness
09:44 < dooglus> I just started listening again to the episode I downloaded
09:45 < dooglus> it makes me cringe
09:45 < pabs3> what were they talking about?
09:45 < dooglus> "the french don't wear deodorant"
09:45 < factor> heh
09:46 < dooglus> "the guy who sings too loud should be shot"
09:46 < dooglus> something vaguely homophobic about "prancer"
09:46 < pabs3> ugh
09:47 < dooglus> all in the first 90 seconds of S05E07
09:47 < dooglus> maybe I picked a bad one to listen to
09:52 < dooglus> did you see this? http://people.debian.org/~igloo/popcon-graphs/index.php?packages=synfigstudio
09:54 < dooglus> I was wondering whether it's worth contributing to http://debaday.debian.net/ re. studio
09:54 < dooglus> "Contribute
09:54 < dooglus> This blog is a collaborative work. We really need your help: submitting entries about all the great packages you use that would deserve more press."
09:54 -!- xerakko [n=xerakko@lpri.gva.es] has joined #synfig
09:55 < pabs3> dooglus: definitely worth doing that. it goes to planet ubuntu as well IIRC
09:56 < pabs3> I think wait till .08 is out and in sid though
09:56 < factor> ok back off to sleep
09:56 < pabs3> nite factor 
09:56  * pabs3 waves to xerakko 
09:58 < xerakko> hi pabs3 
09:58 < pabs3> :)
09:58 < pabs3> how are things xerakko ?
09:59 < xerakko> nice
09:59 < xerakko> (I'm at work now)
10:00 < pabs3> cool
10:01 < dooglus> pabs3: it's one of the default feeds in liferea too I think
12:06  * AkhIL compiles synfig-studio for puppy linux
12:07 < AkhIL> do you need synfig in .pet packages?
12:09 < pabs3> sure, put them up somewhere and add a link on the download page
12:18 < CIA-27> synfig: dooglus * r1784 /synfig-studio/trunk/src/gtkmm/dialog_setup.cpp: Tidied up the "Misc." tab of the settings dialog.
12:18 < CIA-27> synfig: dooglus * r1785 /synfig-studio/trunk/src/gtkmm/dialog_setup.cpp: Fix the last commit so it doesn't modify the strings. I had added a colon to the end of each label.
12:26 -!- MangoFusion [n=jamesu@host81-132-205-163.range81-132.btcentralplus.com] has joined #synfig
13:17 -!- factor [n=factor@ip68-14-160-70.ok.ok.cox.net] has quit [Read error: 104 (Connection reset by peer)]
13:34 -!- factor [n=factor@ip68-14-160-70.ok.ok.cox.net] has joined #synfig
14:44 -!- xerakko [n=xerakko@lpri.gva.es] has quit ["ChatZilla 0.9.81 [Firefox 2.0.0.12/2008020710]"]
14:47 -!- mwiriadi [n=mwiriadi@203-59-174-181.dyn.iinet.net.au] has joined #synfig
14:55 -!- factor [n=factor@ip68-14-160-70.ok.ok.cox.net] has quit ["Ex-Chat"]
15:01 -!- jcome [n=jcome@124.240.92.81] has joined #synfig
15:01 -!- jcome [n=jcome@124.240.92.81] has left #synfig ["Ex-Chat"]
15:04 -!- Zelgadis [n=zelgadis@87.103.170.132] has joined #synfig
15:23 < Zelgadis> It seems GTK supports transparency without compiz
15:26 < Zelgadis> Take a look: http://zelgadis.profusehost.net/files/gnome_splash.png
15:27 < Zelgadis> This is splash screen displayed on Fedora 7 when loading GNOME
15:28 < Zelgadis> You can enable splash screen in gnome by executing
15:28 < Zelgadis> gconftool-2 --set /apps/gnome-session/options/show_splash_screen --type bool true
15:29 < Zelgadis> (but I'm not sure what GNOME have splash screen with transparency on other distros)
15:34 < Zelgadis> Maybe this would help: http://www.internettablettalk.com/forums/showthread.php?p=145231
15:35 < pabs3> doesn't seem transparent to me Zelgadis
15:36 < pabs3> ohwait
15:37 < Zelgadis> There are many examples on pyGTK for shaped windows
15:38 < Zelgadis> The wheelbarrow.py (http://www.pygtk.org/pygtk2tutorial/examples/wheelbarrow.py) example program displays a full wheelbarrow image on the desktop.
15:38 < Zelgadis> pabs3: did you saw?
15:39 < Zelgadis> More links: http://forum.compiz-fusion.org/showthread.php?t=6294
15:39 < Zelgadis> http://www.pygtk.org/pygtk2tutorial/examples/wheelbarrow.py
15:39 < Zelgadis> http://ubuntuforums.org/showthread.php?t=553183
15:40 < pabs3> looks like it uses a 1-bit mask
15:48 < Zelgadis> pabs3: yeah, you right
15:48 < Zelgadis> "The idea is to change the actual shape of the window. The smooth shapes seen in screenlets are due to some parts of the rectangular window being transparent. Transparency, however, only works in a composited environment, so if there is no compositor, you have to change the actual shape of this rectangular window."
15:48 < Zelgadis> "This makes the pixels of the original rectangle either part of the window or not, so the result is a window with a jagged edge and not the smooth edge of its composited counterpart."
15:48 < Zelgadis> Code:
15:48 < Zelgadis> 	def update_shape(self):
15:48 < Zelgadis> 		"""Update the shape of the window."""
15:48 < Zelgadis> 		if self.window:
15:48 < Zelgadis> 			# Create shape pixmap
15:48 < Zelgadis> 			width, height = self.size_request()
15:49 < Zelgadis> 			shape = gtk.gdk.Pixmap(self.window, width, height, 1)
15:49 < Zelgadis> 			
15:49 < Zelgadis> 			# Create and initialize Cairo context
15:49 < Zelgadis> 			ctx = shape.cairo_create()
15:49 < Zelgadis> 			cairo_clear(ctx)
15:49 < Zelgadis> 			ctx.scale(self.scale, self.scale)
15:49 < Zelgadis> 			
15:49 < Zelgadis> 			# Draw to Cairo context and update shape
15:49 < Zelgadis> 			self.on_draw_shape(ctx)
15:49 < Zelgadis> 			del ctx
15:49 < Zelgadis> 			self.window.input_shape_combine_mask(shape, 0, 0)
15:49 < Zelgadis> 			
15:49 < Zelgadis> 			# Change the shape of the window if there is no composite extension
15:49 < Zelgadis> 			if not self.is_composited():
15:49 < Zelgadis> 				self.window.shape_combine_mask(shape, 0, 0)
15:49 < Zelgadis> 			
15:49 < Zelgadis> 			del shape
15:49 < pabs3> eep, no need to paste in here
15:49 < Zelgadis> sorry
16:25 -!- Zelgadis [n=zelgadis@87.103.170.132] has quit ["????!"]
16:29 -!- factor [n=Factor@32.147.72.164] has joined #synfig
16:46 < CIA-27> synfig: dooglus * r1786 /synfig-studio/trunk/src/gtkmm/duckmatic.cpp: Turning the first point in a polygon 'off' was causing a crash.
16:46 < CIA-27> synfig: dooglus * r1787 /synfig-studio/trunk/src/gtkmm/duckmatic.cpp: Draw in the complete outline of a selected polygon if it is 'looped'. This allows the creation of new points on any existing side.
16:46 < CIA-27> synfig: dooglus * r1788 /synfig-core/trunk/src/synfig/valuenode_dynamiclist.cpp: When creating a list of vectors, create it looped. This lets polygons be closed by default.
16:53 < dooglus> genete: "add a polygon gives me an english word as resumee of the set of actions: "Move"" -- took me a while to reproduce :)
16:54 < dooglus> it's not adding a polygon that makes it say 'move' - it's moving a polygon :)
16:56 < dooglus> genete: "Nudge" was missing, too.
17:00 -!- MangoFusion [n=jamesu@host81-132-205-163.range81-132.btcentralplus.com] has quit []
17:01 < CIA-27> synfig: dooglus * r1789 /synfig-studio/trunk/src/gtkmm/workarea.cpp: Two new words for translation: "Move" and "Nudge".
17:08 < factor> poke poke nudge nudge
17:11 -!- genete [i=d90c1036@gateway/web/ajax/mibbit.com/x-be6f80e36ce828c6] has joined #synfig
17:12 < genete> hi dooglus :)
17:13 < dooglus> hi
17:13 < dooglus> I just checked through the code for other history panel group names that weren't translated.  only 'move' and 'nudge' were there
17:14 < dooglus> is there anything else you suggested last night that's not been done yet?
17:14 < dooglus> I justified & sorted the options dialog
17:14 < genete> Yes I've read it . But, I obtained the "Move" resumee word when create the polygon...
17:14 < dooglus> and added missing tips to the meta buttons
17:14 < rore> hi everyone ! Oh darn, still 30 or 40% of core to translate for me.
17:14 < dooglus> genete: try again - could you have moved it after accidentally?
17:15 < genete> nop, nothing more. I don't want to "distract" you ;)
17:15 < dooglus> heh
17:15 < genete> ok, I'll try when being at home. :)
17:15 < genete> eeeh just kidding. I read the logs during the morning so I catugh pixelgeek joke :)
17:16 < genete> hi rore
17:16 < genete> I cannot avoid speak so much in the IRC :( ...... :)
17:17 < genete> s/catugh/caught
17:17 < dooglus> I'm trying out 'ccache'
17:17 < dooglus> it apparently speeds up compilation by reusing builds that have been done before
17:17 < rore> eeek, someone wake me up, that must be a bad dream http://msdn.microsoft.com/events/hero/sfbio/ (I don't wanna know what kind of stuff they smoke at redmon)
17:17 < genete> rore: I preffer you to work on the feb entry rather than update the translation ;)
17:18 < CIA-27> synfig: dooglus * r1790 /synfig-studio/trunk/src/gtkmm/layeractionmanager.cpp: Simplify the code. There's no need to check whether we're increasing or decreasing the amount; we already know.
17:18  * rore hopes to scribble something during Fosdem, and they vectorise it on synfig :) 
17:19 < dooglus> "Windows Vista Sensei travels from place to place in a quest to help the underprivileged global citizens. He shows up when you least expect him and always when he's most needed"
17:19 < dooglus> jesus
17:20 < rore> yeah, scary isn't it?
17:21 < genete> but is this an official page from MS Windows?
17:21 < rore> yep, looks like
17:21 < dooglus> it's an MS ad campaign by the look of it, yes
17:22 < dooglus> "source FOURce", geddit?
17:22 < dooglus> (because the are - um - seven of them?)
17:22 < rore> And honestly, the figures are just ugly. Like some fat legos
17:22 -!- MangoFusion [n=jamesu@host81-132-205-163.range81-132.btcentralplus.com] has joined #synfig
17:23 < rore> Source Farce is a better name for such a thing
17:24 < dooglus> who is it aimed at?
17:24 < dooglus> it looks like it's for kids, but they're pushing development tools?
17:25 < rore> it's never too early to brainwash people ;)
17:25 -!- MangoFusion [n=jamesu@host81-132-205-163.range81-132.btcentralplus.com] has quit [Client Quit]
17:29 < genete> clear
17:29 < genete> oops I meant /clear
17:31 < factor> ppsft
17:33 < factor> rore working on your feb challege
17:33 < rore> yeah, my brain is working on it ;)
17:35 < genete> rore: our cards are on the table... what's your feb challenge theme? =)
17:36 < rore> erm... abstract, I think :)
17:38 < rore> (I started to think of a city, with a tower roof being the synfig logo, but I'm not good at drawing citys. Then I thought of something with a synfig kite, then synfig flowers, and so on ...)
17:39 < factor> cool
17:39 < genete> I know what you need to add to your abstract thing....
17:39 < factor> kite is neat
17:40 -!- xerakko [n=Miguel@46.pool85-54-226.dynamic.orange.es] has joined #synfig
17:40 < genete> ... you need a kitty :)
17:41 < rore> I already have one, sleeping on my lap ;)
17:44 < dooglus> ccache is great :)  I can go from 'make clean' to a full build in 5 minutes
17:45 < factor> Have you tried distcc
17:45 < dooglus> no, I only have one computer
17:46 < dooglus> well, 2, but the other one is pretty old
17:46 < dooglus> what about 'pch' (pre-compiled headers)?
17:46 < dooglus> I see stuff in the synfig source about them - but how do you use them?
17:46 < factor> heh
17:46 < factor> dunno
17:48 < genete> dooglus: can I use ccache too as a normal user?
17:50 < dooglus> genete: sure
17:50 < dooglus> genete: install package 'ccache'
17:50 < dooglus> and:    export CXX="/usr/bin/ccache /usr/bin/g++-4.2"
17:51 < dooglus> add that line to your .bashrc so it's always set
17:51 < dooglus> then re-do the whole autoreconf & configure thing
17:51 < dooglus> genete: I don't know how much it will help you though
17:52 < dooglus> what ccache does is, when you go to compile something, it runs the 'cpp' (pre-processor) step, looks at the output, and sees if you've built it before
17:52 < dooglus> if you have, it uses the previous build, instead of building it again
17:52 < dooglus> so it's helpful when the timestamp on a file has changed, but the content hasn't
17:52 < dooglus> or if you want to quickly switch between optimized and debug builds
17:53 < dooglus> I was finding that 'git' was changing the modification dates on files that hadn't really changed, causing long rebuilds
17:54 < genete> but I can do that with the trick you told me with the etl.pc.in and synfig.pc.in files modifications ...
17:55 < genete> it is only useful for identical files but different date... but what's identical files for gcc?
17:56 < genete> C++ codes "a++" and "++a" can have the same md5sum value but the meaning is completely different.
17:56 < factor> Nifty catch on git
17:56 < dooglus> genete: ccache doesn't know anything about C++
17:56 < factor> did you tell the git people
17:56 < dooglus> genete: "identical" means "md5sum sum is the same" or some such I guess
17:57 < dooglus> factor: I did, but there's not a lot they can do.  I'm using 'git stash' and 'git stash apply' to hide and retrieve local changes while committing to svn
17:57 -!- genete [i=d90c1036@gateway/web/ajax/mibbit.com/x-be6f80e36ce828c6] has quit ["http://www.mibbit.com ajax IRC Client"]
17:58 < dooglus> factor: both those commands change the files, and so the date needs to be changed.  the files end up the same as they started, but git doesn't know whether I did a build in between the save and apply
17:58 < dooglus> before and after stats of a full build, including rendering the icons from .sif files:
17:58 < dooglus> Wed Feb 20 17:48:55 CET 2008   cache hit/miss:  680/438   called for link: 137   files in cache: 876   cache size: 65.1 Mbytes
17:58 < dooglus> Wed Feb 20 17:53:31 CET 2008   cache hit/miss: 1017/468   called for link: 182   files in cache: 936   cache size: 66.6 Mbytes
18:00 < factor> gotcha
18:01 < dooglus> actually, that was the first time I'd built some of those files, so I guess it'll be quicker a 2nd time
18:01 < factor> git not ccache aware
18:01 < dooglus> git not aware of what builds I do at all, whether using ccache or gcc directly
18:01 < dooglus> and, I guess, nor should it be
18:08 < dooglus> Wed Feb 20 17:59:11 CET 2008   cache hit/miss: 1017/468   called for link: 182   files in cache: 936   cache size: 66.6 Mbytes
18:08 < dooglus> Wed Feb 20 18:03:52 CET 2008   cache hit/miss: 1380/472   called for link: 227   files in cache: 944   cache size: 67.3 Mbytes
18:08 < dooglus> not much difference
18:08 < dooglus> I see just 4 cache misses - I guess that's linking synfig, libsynfig, studio, libsynfigapp
18:10 < dooglus> aah, it has a logging feature
18:10 < factor> how are you compiling synfig on windows
18:11 < dooglus> so I can see what it's having to rebuild
18:11 < dooglus> I can't build on windows - but when I could, I followed the wiki page about it
18:11 < dooglus> building_on_windows or something it's called
18:13 < dooglus> http://synfig.org/Building_on_Windows
18:13 < factor> k
18:14 < dooglus> 4 things it rebuilds each time:
18:14 < dooglus> Placed .libs/libsynfig_la-main.o into cache
18:14 < dooglus> Placed synfig-main.o into cache
18:14 < dooglus> Placed synfigstudio-about.o into cache
18:14 < dooglus> Placed synfigstudio-splash.o into cache
18:14 < factor> wondering if it was built using mingw
18:15 < dooglus> yes, I think so
18:16 < factor> looks like it
18:20 -!- genete [i=d90c1036@gateway/web/ajax/mibbit.com/x-18cb5e22db26b2e4] has joined #synfig
18:20 < genete> sorry, electrical supply was down for a while...
18:27 < factor> storm?
18:34 < genete> nop
18:34 < genete> workers in the building...
18:34 < genete> I'm at the office.
18:35 -!- xerakko [n=Miguel@debian/developer/xerakko] has quit [Read error: 104 (Connection reset by peer)]
18:35 < genete> careless people .... >:(
18:35 < factor> k
18:36 -!- MangoFusion [n=jamesu@host81-132-205-163.range81-132.btcentralplus.com] has joined #synfig
18:44 < dooglus> genete: did you know that the arrow keys 'nudge' the selected ducks one pixel at a time?
18:44 < dooglus> genete: and SHIFT+arrow keys 'nudge' them 10 pixels?
18:44 < dooglus> genete: 'cos the "Keyboard Shortcuts" page doesn't mention it
18:47 < genete> dooglus: nop, nop and let me fix it :)
18:48 < genete> dooglus: do you need to have the normal tool selected to do that?
18:48 < genete> or it is valid in any moment? (I cannot test it now)
18:53 < genete> ttyl
18:53 -!- genete [i=d90c1036@gateway/web/ajax/mibbit.com/x-18cb5e22db26b2e4] has quit ["http://www.mibbit.com ajax IRC Client"]
19:01 < dooglus> it seems to work with any tool, but it's not clear what it does
19:01 < dooglus> ie. I don't know how it decides which duck you are controlling
19:29 < CIA-27> synfig: dooglus * r1791 /synfig-core/trunk/src/tool/main.cpp: Let's not include the time of the build in the binary. The date is enough.
19:29 < CIA-27> synfig: dooglus * r1792 /synfig-studio/trunk/src/gtkmm/ (about.cpp splash.cpp): As before, the date is sufficient. We don't need to keep rebuilding this just to get the current time in the splash screen or about dialog.
19:29 < CIA-27> synfig: dooglus * r1793 /synfig-core/trunk/src/synfig/ (main.cpp version.h): As before, the date is sufficient. Removed get_build_time().
20:41 -!- xerakko [n=Miguel@debian/developer/xerakko] has joined #synfig
20:50 -!- genete [n=Genete@84.122.50.111.dyn.user.ono.com] has joined #synfig
20:52 < genete> dooglus: I've just reached home. And you have already updated the wiki
21:14 < genete> how can I apply a blur only over a selected area of a composition?
21:15 < genete> (for example to simulate look thru a blurred glass
21:18 < genete> In general, how can I apply an effect (blur, noise, twirl, etc.) over a certain region of a composition?
21:20 < dooglus> I don't think you can - but I don't know
21:21 < dooglus> unless you export the composition, use it twice, cropping and bluring one and not the other
21:21 < genete> yes that's the solution I've found. Is there any other?
21:24 < MangoFusion> can't you clip it to an area?
21:25 < genete> I want something like spherize but for a custom shape instead of a circle.
21:26 < genete> MangoFusion: I dont' unsertand your sentence.
21:26 < MangoFusion> clip the blur to a shape?
21:26 < genete> yes
21:27 < genete> blur a background but only over a defined region
21:27 < MangoFusion> not played around with that stuff in a while though so i can't recall any specifics
21:27 < genete> the current solution is what dooglus has explained
21:28 < MangoFusion> ah right
21:28 < genete> but want to know is there is other solution
21:29 < genete> my target is to simulate the waves of the air when it is hot and deform the background sight
21:39 < dooglus> genete: at least you can export and re-use the layers, rather than having to clone them
21:41 < genete> I think that there is not other solution than export and use it twice. One with effect and othre without
21:43 < dooglus> well, there is - don't export, just encapsulate, copy & paste - crop and blur one copy, don't blur the other
21:47 -!- Yoyobuae [n=Yoyobuae@201.224.135.156] has joined #synfig
21:47 < genete> ha ha
21:47 < genete> it is obvious :)
22:07 < dooglus> can you think of a better design?
22:07 -!- Yoyobuae [n=Yoyobuae@201.224.135.156] has quit [Read error: 104 (Connection reset by peer)]
22:07 < genete> I can't
22:08 < dooglus> something like a 'pastecanvas' - but an 'effectcanvas' instead?
22:08 < genete> oh well new features you mean?
22:08 < dooglus> yes
22:08 < dooglus> I know - sound, sound & sound...
22:08 < genete> no, no
22:09 < genete> I'm thinking on a new mlend method
22:09 < genete> blend
22:09 < genete> one very cool is "onto" and straight onto
22:10 < genete> why don't "straight onto" + "onto" ?
22:10 < dooglus> they're not new?
22:10 < dooglus> oh, by the way - did you find out about the lack of translation for 'value node' and 'layer param'?
22:10 < dooglus> did it get fixed yet?
22:10 -!- Yoyobuae [n=Yoyobuae@201.224.135.156] has joined #synfig
22:10 < genete> let me check... (and continue with my idea later)
22:11 < Yoyobuae> hi
22:12 < dooglus> hi
22:12 < genete> hi Yoyobuae 
22:12 < dooglus> I had a bit of a look at the 'list' code today, when fixing the polygon layer
22:12 < genete> dooglus: weird. I've translated Move and Nudge and I still having Move and not "Mover" and I did sudo make install inside po folder
22:13 < Yoyobuae> yeah,  saw several commits related =)
22:13 < genete> I did make update-po before translate too
22:14 < genete> Also still having "Value Node" thing in the history
22:16 < genete> BTW how can I make the word "nudge" appear?
22:18 < genete> ok, back to the blend method I want: 
22:18 < genete> I would like to have a blend method that acts on two steps
22:18 < dooglus> genete: you need to remove the stamp thingy?
22:19 < genete> ok
22:19 < dooglus> genete: you make 'nudge' appear by nudging the ducks :)
22:19 < dooglus> genete: ie. use the cursor keys
22:19 < genete> lol
22:19 < dooglus> you need to have more than 1 duck selected - 'nudge' is the group name, and it doesn't group if there's only 1 duck
22:20 < dooglus> a 'nudge' is a 'move' done with the keyboard
22:20 < dooglus> "rm stamp-po"
22:20 < dooglus> seems to be a problem with the Makefile - it sees the 'stamp-po' file and thinks that means it doesn't need to install the new translations
22:21 < genete> cool!! "nudge" is translated
22:21 < genete> abd move too
22:21 < dooglus> I have:
22:21 < dooglus> -rw-r--r--  1 chris chris  61319 Feb 20 01:02 es.gmo
22:21 < dooglus> -rw-r--r--  1 chris chris 110146 Feb 20 22:15 es.po
22:21 < genete> but Value Node don't
22:21 < dooglus> ie. the gmo file is old, but 'make' doesn't know it should be updated - it just installs the old one
22:22 < dooglus> genete: 'value node' is a core's es.po
22:22 < dooglus> genete: maybe the same problem?  need to remove 'stamp-po'?
22:22 < genete> ok
22:23 < dooglus> genete: I might be completely wrong - maybe it is studio after all
22:23 < genete> don't care
22:23 < dooglus> aah!
22:23 < genete> Value node still there ...
22:24 < genete> after rm stamp-po and sudo make install
22:24 < dooglus> got it - I made a new file synfigapp/value_desc.cpp a month or two ago and didn't add it to the list of files to translate...
22:24 < genete> lol
22:25 < genete> dooglus: I want to explain my new blend method on the discuss area od the Blend Method wiki page, ok?
22:25 < dooglus> ok
22:25 < genete> or maybe in the forum? 
22:25 < dooglus> I just committed a change that will let you translate 'value node', etc.
22:26 < genete> in the forum is best to me, if you don't mind...
22:26 < dooglus> I think the wiki is better - it keeps easy to find
22:26 < dooglus> whatever :)
22:26 < genete> it doesn't matter
22:26 < genete> wiki then
22:30 < CIA-27> synfig: dooglus * r1794 /synfig-studio/trunk/po/es.po: Commit 1898184: Updated translation for studio/es.po. Thanks genete.
22:30 < CIA-27> synfig: dooglus * r1795 /synfig-studio/trunk/po/POTFILES.in: Add value_desc.cpp to the list of files to translate.
22:51 < genete> dooglus: http://synfig.org/Talk:Blend_Method_Parameter
22:51 < genete> please have a look
22:53 < dooglus> k
22:53 < dooglus> I'm trying to remember how I caused the crash I told pixelgeek about a few days ago...
22:53 < genete> ok
22:54 < dooglus> the verb is 'to affect' with an 'a'
22:54 < dooglus> so 'affected'
22:56 < dooglus> genete: it looks like a 'fork' in the layer tree to me
22:56 < dooglus> like a pastecanvas is
22:57 < dooglus> for a pastecanvas, there's a bunch of layers 'under' it, and a bunch of layers it's pasting
22:57 < genete> explain it a little please
22:57 < dooglus> for your 'method x', there's a bunch of layers defining the background, and another bunch defining the cropping area
22:57 < dooglus> isn't it analogous
22:57 < genete> right
22:58 < genete> so what do you suggest?
22:58 < dooglus> maybe something like a pastecanvas
22:58 < dooglus> so the cropping area is an inline canvas
22:59 < Yoyobuae> i had an idea about this also
22:59 < dooglus> each 'effect layer' gets a new parameter type: 'canvas', name 'cropping'
22:59 < Yoyobuae> gonna use the discussion page
22:59 < dooglus> k
23:00 < dooglus> I'll go back to my old bug for a while...
23:00 < genete> dooglus: following your idea: what would be the default value for that new parameter (cropping canvas)?
23:02 < dooglus> genete: nil
23:02 < dooglus> ie. don't crop
23:02 < genete> glups
23:03 < dooglus> effectively a solidcolor then
23:07 < dooglus> genete: I wonder, don't you have 3 groups of layers?  1) background 2) crop 3) effects
23:07 < dooglus> genete: my suggestion only gives you one effect layer
23:08 < genete> but anyway you need the cropping canvas to exists
23:09 < genete> and to be exported that can be annoying sometimes
23:09 -!- Yoyobuae [n=Yoyobuae@201.224.135.156] has quit [Read error: 104 (Connection reset by peer)]
23:11 -!- zotz [n=zotz@24.244.163.157] has joined #synfig
23:12 < dooglus> hi zotz
23:13 < zotz> hey dooglus 
23:13 < dooglus> genete: not to be exported?  why?
23:13 < zotz> how are you today?
23:13 < dooglus> zotz: I'm weird today.  Strangely jumpy.  How are you?
23:13 < zotz> doing fine mostly
23:13 < zotz> enjoying the rpm08 challenge this month
23:13 < genete> dooglus: because if it is a canvas to choose from it should be exported (like the canvas parameter of a paste canvas layer)
23:15 -!- Yoyobuae [n=Yoyobuae@201.224.135.156] has joined #synfig
23:15 < dooglus> genete: I don't think I'm understanding you
23:16 < dooglus> genete: I might want to re-use a cropping canvas in lots of places, so being able to have it exported like that is a good thing, isn't it?  what's annoying for you?
23:16 < genete> maybe I don't understand your proposal
23:16 < dooglus> I don't understand it clearly either.  I think Yoyobuae is going to make it clearer :)
23:16 < Yoyobuae> done
23:16 < genete> beacuse exported canvases cannot be unexported :(
23:17 < Yoyobuae> i posted my idea: http://www.synfig.org/Talk:Blend_Method_Parameter
23:18 < genete> Yoyobuae: I didn't understand a word :)
23:18 < Yoyobuae> i was afraid of that XD
23:19 < genete> can you tell it in a more "normal synfig user" language?
23:19 < Yoyobuae> actually its a concept used in programming
23:20 < Yoyobuae> i not sure how to explain it in simpler terms =/
23:21  * genete needs a select inverse ducks to allow select the non selected ducks of a layer and unselect the current selection
23:21 < factor> i hate when i get my ducks inversed
23:22 < factor> :)
23:22 < Yoyobuae> im gonna try explaining a simpler example
23:25 < dooglus> Yoyobuae: I don't see what the compose layer gives us that we don't already have
23:25 < dooglus> we can already apply a bunch of layers over another bunch of layers by just putting the 1st group over the 2nd
23:27 < Yoyobuae> the layers INSIDE an exported canvas cannot be put on top of some other layers
23:27 < dooglus> context(x) = a(b(c(x)));   inline(x) = d(e(f(x)));   compose(context,inline) = inline(context) = a(b(c(d(e(f(x))))))
23:27 < dooglus> Yoyobuae: no, ok, they can't
23:27 < Yoyobuae> inline(context) is impossible in synfig right now
23:27 < dooglus> Yoyobuae: what I don't see is how your suggestion answers genete's request to be able to blur part of a composition
23:27 < Yoyobuae> unless you copy over all the layers
23:28 < dooglus> yes - inline canvases are generally used exactly for that property - that the composition process stops at their end
23:28 < Yoyobuae> compose(context,blurpart())
23:29 < Yoyobuae> in fact compose(context,compose(blur,crop))
23:29 < dooglus> how is that different than copying over all the layers of the inline canvases?
23:29 < Yoyobuae> there is no copy =)
23:29 < Yoyobuae> blur is a single blur layer
23:29 < dooglus> I mean:
23:30 < dooglus> currently I can achieve the effect of your proposed 'compose' by copying the layers out one by one
23:31 < dooglus> is that right?
23:31 < dooglus> you said inline(context) was impossible unless you copy the layers out
23:31 -!- Yoyobuae [n=Yoyobuae@201.224.135.156] has quit [Read error: 104 (Connection reset by peer)]
23:31 -!- Yoyobuae_ [n=Yoyobuae@201.224.135.156] has joined #synfig
23:31 -!- Yoyobuae_ is now known as Yoyobuae
23:32 < Yoyobuae> the use is that when a user wants to apply the same effect on many places, its easier to have an "apply effect to x" layer
23:32 < Yoyobuae> where x can be anything
23:32 < dooglus> here's my confusion
23:33 < dooglus> (1) you stated that the compose effect can currently be achieved by copying layers out of the inline canvas
23:33 < dooglus> (2) the effect genete is looking for cannot currently be achieved without duplicating the background layer
23:33 < dooglus> (1)+(2)--> the compose effect doesn't solve genete's problem?
23:35 < Yoyobuae> ahh i understand
23:35 < dooglus> maybe a concrete example would help.  we have a png image, and want to blur a circular area of it
23:35 < dooglus> how do we go about that, given that we have your 'compose' layer?
23:35 < Yoyobuae> we also need compose2 =)
23:36 < Yoyobuae> one sec, trying to get my thoughs straight XD
23:36 < dooglus> sure
23:39 < Yoyobuae> compose2(context,inline1,inline2) = blend(inline1(contex),inline2(contex))
23:39 < Yoyobuae> something like that
23:40 < Yoyobuae> the blend method used to compose on and the other canvas would need to be defined inside compose2
23:40 < Yoyobuae> leaving inline2 canvas empty will leave the context unaffected for that side
23:41 < Yoyobuae> something like compose2(context, inline, blank) = blend(inline(context),blank(context))
23:41 < Yoyobuae> assuming blank(context)=context
23:43 < dooglus> if this notation is going to stick, I'd like to switch the order of the parameters to the pastecanvas one
23:43 < dooglus> pastecanvas(inline,context)
23:44 < dooglus> since it composes inline on top of context
23:44 < Yoyobuae> yeah sure
23:44 < dooglus> genete: "context" is what the source code calls "the layers under us"
23:45 < dooglus> genete: so in a pile of layers A,B,C, A's context is B,C; B's context is just C
23:45 < genete> yes I know. Don't follow anyway :)
23:45 < dooglus> ok
23:45 < genete> so hard to my poor mind :(
23:45 < genete> but enjoy that my suggestion make yours work :)
23:45 < dooglus> so for the example of blurring the circular region of a png:
23:46 < dooglus> compose2(blur,circle,png) = blend(blur(circle(png)),png) is what we want?
23:47 -!- cizra [n=cizra@gw.tdng.ttu.ee] has joined #synfig
23:47 < dooglus> hi cizra
23:47 < cizra> Heh
23:47 < cizra> Hello there.
23:47 < Yoyobuae> genete: actually this was very hard to understand the first time for me.  But i have spend hours and hours playing with, during programming with STL that is
23:47 < dooglus> .ee?  is that estonia?
23:47 < cizra> yes
23:47 < genete> hi cizra 
23:47 < cizra> (=
23:47 < Yoyobuae> dooglus: yeah that seems right
23:48 < cizra> I'm doing an animation that explains how TCP RST attacks work.
23:48 < cizra> I'm almost absolutely new to Synfig.
23:48 < dooglus> cool
23:48 < genete> so you're at the right place :)
23:48 < cizra> Aaaand I'm going to have a helluva lot of stupid questions.
23:48 < dooglus> good good
23:48 < genete> shot
23:48 < dooglus> shoot :)
23:49 < cizra> First
23:49 < genete> shooooot please!!!
23:49 < Yoyobuae> wow nice, i didn't think about using synfig for network explanation. its a very nice idea =). and hi btw =)
23:49 < cizra> Do I need two keyframes to teleport a TCP packet instantly?
23:50 < dooglus> cizra: you never need any keyframes
23:50 < cizra> ??
23:50 < dooglus> cizra: they are a convenience thing, but not necessary
23:50 < dooglus> cizra: all you need are waypoints (the green dots you see in the timetrack)
23:50 < genete> maybe he says waypoints
23:50 < cizra> No, I'm actually moving stuff around in time with keyframes.
23:51 < cizra> http://synfig.org/Animation_Basics
23:52 < dooglus> cizra: you can make something jump immediately using a 'constant' interpolation type - but you'll still need 2 waypoints to specify its 2 positions
23:52 < cizra> Yes, I discovered the 'constant' thing already -- I have 2 boxes appear out of thin air.
23:54 < Yoyobuae> you can also specify positions with 'constant' waypoints
23:54 < dooglus> cizra: if you watch the 'Timetrack' panel (the red 'T' icon) you'll see the waypoint appears when you move the layer
23:54 < Yoyobuae> in Animation mode that is
23:55 -!- xerakko [n=Miguel@debian/developer/xerakko] has quit [Read error: 110 (Connection timed out)]
23:55 < cizra> Second
23:55 < cizra> How do I move rectangles around?
23:55 < cizra> I can "resize" them around, but this feels hackish.
23:55 < Yoyobuae> using a paste layer works
23:56 < cizra> What's that?
23:56 < Yoyobuae> right click the rectangle and click "Encapsulate"
23:56 < Yoyobuae> the rectangle gets put inside a layer with a box icon
23:57 < cizra> OK
23:57 < Yoyobuae> you can move that layer's "Origin" parameter
23:57 < cizra> OK
23:57 < Yoyobuae> you can also encapsulate multiple layers
23:58 < cizra> Can I easily move something's position in ALL time? (it should remain constant anyway)
23:58 < genete> or you can move all the rectangle ducks at the same time. Just CTRL-A to select them all and drag one
23:59 < cizra> Just do it without the "animation mode" on?
23:59 < cizra> Duh. Why didn't I know it?
--- Log closed Thu Feb 21 00:00:14 2008