[Bf-committers] Freestyle branch - request for code review
Dear all,
I would like to ask for code review of the Freestyle branch aiming at a trunk
merger in view of the upcoming 2.67 release. The reference revision for the
code review is 54826.
Since the first round of branch code review, all the comments, suggestions
and requested changes have been addressed as briefly summarized below.
Any support and response concerning this code review request are duly
acknowledged.
With best regards,
T.K.
--
A summary of changes since the first round of branch code review
1. Freestyle Python API
1.1) Most get/set methods of Python API classes were transformed into
properties (attributes) written using PyGetSetDef. Vector get/setters have
been implemented using mathutils Vector callbacks.
1.2) The naming of methods and attributes was fixed to follow the naming
conventions of the Blender Python API (i.e., lower case + underscores for
methods and attributes, and CamelCase for classes).
1.3) Per-coordinate get/setters (e.g., .getX and .setX) were removed in
favor of vector get/setters (e.g., .point_3d). Exceptions are
Interface0D.projected_[xyz], because Interface0D.projected_z is the only
property that gives access to the Z component of a projected point from the
3D camera to 2D image space. The other two component-wise properties are
redundant but kept for consistency.
1.4) Proper handling of keyword arguments has been implemented in all class
constructors and methods.
[Implementation notes: Because of the extensive use of constructor
overloading in the underlying C++ classes, the corresponding Python
wrappers try to parse arguments through multiple calls of
PyArg_ParseTupleAndKeywords() if needed. The downside of this
implementation is that most argument errors result in the same error
message ("invalid argument(s)") without indicating what is wrong. For now
this issue is left for future work.]
1.5) Boolean arguments of class constructors only accept values of boolean
type. Input values of other types are considered as error.
1.6) "__getitem__" methods were removed and the Sequence protocol has been
used instead.
1.7) Fix for wild card import (e.g., "from Freestyle import *") was done.
Now import statements are either without using "from" or with all imported
names explicitly listed.
2. User interface
2.1) The new "Freestyle" panel has been added to the Render properties. The
new panel accommodates global Freestyle options, and the global Freestyle
toggle button is shown in the panel header.
2.2) The new "Render Layers" tab has been created and all per-layer options
have been move from the Render tab to the new tab.
2.3) Old and slow raycasting algorithms were removed. Now only two
variations (culling and non-culling) of the fastest raycasting algorithms
are available, and they are switched by a toggle button.
2.4) Factory settings contain more meaningful initial Freestyle settings
that generate plain strokes by just enabling Freestyle and rendering.
2.5) General placement of UI controls has been improved.
3. Code
3.1) The entire branch has been updated for better coding style conformance.
3.2) A compiler flag for disabling Freestyle has been added.
3.3) The new --debug-freestyle option has been added for enabling debug
information in Freestyle.
--
--
KAJIYAMA, Tamito
___
Bf-committers mailing list
Bf-committers@blender.org
http://lists.blender.org/mailman/listinfo/bf-committers
Re: [Bf-committers] Blender developer meeting notes, 24 February 2013
> replacing GL_SELECT...Could be simply replaced with using back-buffer color codes. Is that the idea where the buffer contains a unique solid color code for each object? So you know what is under the mouse by testing that one color under the mouse pointer? I might be completely misunderstanding this, but that seems like it would work well for simple cases but would not help with our quite unique selection method in Blender. I now click once and it selects that obvious object under the mouse, but click again and it selects the nearby object behind it, click again and it selects the next nearest object that might be completely occluded by the first and second. So what is currently being pointed at isn't an easy question that could be helped with that color buffer. But again, I might be completely misunderstanding it all... Harley ___ Bf-committers mailing list Bf-committers@blender.org http://lists.blender.org/mailman/listinfo/bf-committers
Re: [Bf-committers] Blender developer meeting notes, 24 February 2013
Hi, I would advise against replacing GL_SELECT before swiss-cheese merge. There is enough changes in OpenGL code which makes merging very painful. Recoding GL_SELECT would make it even worse. With Wilkins help, we can get swiss cheese up and running (desktop only) before April. Best, Alex On 2/24/2013 12:13 PM, Ton Roosendaal wrote: > Hi all, > > Here are the notes of today's session in irc.freenode.net #blendercoders > > 1) 2.66 release review > > - Some showstoppers have been found already: > > - A change in how textures use alpha (option moved to Image block), required > a do-version patch which crashes for library linking in cases. > > - Compositor: File Output node missed socket input menu > > - Dynatopo crash in some cases. > > - We will check next sunday again, then do apply fixes on the release branch > and call for an 'a' build... > > - Important note: let's not do cleanups of code now please, except when > approved by all involved module owners! > > 2) Targets for 2.67 > > - Freestyle, no real news, but final review can start soon. > > - Sergey Sharybin worked on (motion tracking) Bundle Adjustment, should be in > svn soon. > > - Proposal for release planning: > > March 10 bcon2 (= targets are defined) > March 31 bcon3 (= targets are in svn) > April 14 bcon4 (= only fixes, test builds and/or RC) > April 21 bcon5 (= release imminent) > > 3) Other projects > > - Antonis Riakiotakis works on subsurf drawing speed improvement. > > - Ton Roosendaal suggested to work on replacing GL_SELECT (OpenGL select > feature, used for object mode). Too many problems are arising with it now. > Could be simply replaced with using back-buffer color codes. Alternative is > to do brute-force raycast tests. Might even work surprisingly fast. > > Thanks, > > -Ton- > > > Ton Roosendaal Blender Foundation t...@blender.orgwww.blender.org > Blender Institute Entrepotdok 57A 1018AD Amsterdam The Netherlands > > ___ > Bf-committers mailing list > Bf-committers@blender.org > http://lists.blender.org/mailman/listinfo/bf-committers ___ Bf-committers mailing list Bf-committers@blender.org http://lists.blender.org/mailman/listinfo/bf-committers
[Bf-committers] Inconsistency between 3D view panning with mouse vs. trackpad
Hi again Ton, and the rest of the gang. This is a follow up to an exchange I had with Ton when I tried to open a bug against 2.66: http://projects.blender.org/tracker/?func=detail&atid=498&aid=34389&group_id=9 I'm trying to demonstrate that the trackpad behavior in 2.66 when it comes to panning in the 3D view is inconsistent with other parts of Blender. Specifically, that it behaves in the exact opposite way to the mouse if you are a default user that has not set Natural scrolling in OS X and there is no configuration to make them consistent. Panning with the mouse goes one way, but panning with the trackpad does the exact opposite. See here: http://www.youtube.com/watch?v=PjtMlucvkFU I would like to volunteer to fix this, but as it is a UI issue, first I would like to have a discussion so we can all understand whether it is a real problem or not. If someone shows me how a user could configure this away, I'm good. If the argument is to make the code easier even if it confuses the users, I'm willing to do the work and manage the hordes of trackpad users until the issue is resolved. But please, please, don't just quickly disregard this as just a single user with his particular way of doing things without at least looking at my video and exchanging some arguments with me. And I'd like to point out that with 2.66 is the first time EVER that I feel like I could operate Blender from a Trackpad without a 3 button mouse. That is huge!! Thanks!! This is but one small tweak towards perfection. Later ___ Bf-committers mailing list Bf-committers@blender.org http://lists.blender.org/mailman/listinfo/bf-committers
Re: [Bf-committers] Proposed default margin increase for baked textures
+1 for the increased default margin as well from me. On Feb 24, 2013, at 3:29 AM, Knapp wrote: > On Sun, Feb 24, 2013 at 1:17 AM, metalliandy...@googlemail.com > wrote: >> Bake panel> Margin #px :) >> >> On 23 Feb 2013, at 21:24, Knapp wrote: >> >>> On Sat, Feb 23, 2013 at 8:30 PM, Sean Olson wrote: +1 from me too - I always bump that level up to a min of 16. I could see less if your UV's were really packed in tight, but 16 or higher usually is what does the trick for killing nasty seams for me. One less thing to worry about for most cases would be great. -Sean >>> >>> Where do you set that? >>> >>> >>> -- >>> Douglas E Knapp > > Ah, thanks. I see my problem. I have been only working in cycles for > some time but this panel only shows up with internal render. PS > someone working the manual might want to point that fact out. > > -- > Douglas E Knapp > > Creative Commons Film Group, Helping people make open source movies > with open source software! > http://douglas.bespin.org/CommonsFilmGroup/phpBB3/index.php > > Massage in Gelsenkirchen-Buer: > http://douglas.bespin.org/tcm/ztab1.htm > Please link to me and trade links with me! > > Open Source Sci-Fi mmoRPG Game project. > http://sf-journey-creations.wikispot.org/Front_Page > http://code.google.com/p/perspectiveproject/ > ___ > Bf-committers mailing list > Bf-committers@blender.org > http://lists.blender.org/mailman/listinfo/bf-committers ___ Bf-committers mailing list Bf-committers@blender.org http://lists.blender.org/mailman/listinfo/bf-committers
[Bf-committers] build documentation incorrect, non-existing link
Hello, according to the "Bug report guidelines" it seems i should post this here, in the freshly downloaded blender-2.66 sources in doc/build_systems/cmake.txt it says that the Dependencies are listed at: http://www.blender.org/cms/Getting_Dependencies.135.0.html But it seems that page does not exist. Surely just a very minor issue,but to my understanding worth changing. Best regards Torsten ___ Bf-committers mailing list Bf-committers@blender.org http://lists.blender.org/mailman/listinfo/bf-committers
Re: [Bf-committers] Blender developer meeting notes, 24 February 2013
On Sun, Feb 24, 2013 at 6:13 PM, Ton Roosendaal wrote: > Hi all, > > Here are the notes of today's session in irc.freenode.net #blendercoders > > 1) 2.66 release review > > - Some showstoppers have been found already: > > - A change in how textures use alpha (option moved to Image block), required > a do-version patch which crashes for library linking in cases. > > - Compositor: File Output node missed socket input menu > > - Dynatopo crash in some cases. > > - We will check next sunday again, then do apply fixes on the release branch > and call for an 'a' build... > > - Important note: let's not do cleanups of code now please, except when > approved by all involved module owners! > > 2) Targets for 2.67 > > - Freestyle, no real news, but final review can start soon. > > - Sergey Sharybin worked on (motion tracking) Bundle Adjustment, should be in > svn soon. > > - Proposal for release planning: > > March 10 bcon2 (= targets are defined) > March 31 bcon3 (= targets are in svn) > April 14 bcon4 (= only fixes, test builds and/or RC) > April 21 bcon5 (= release imminent) > > 3) Other projects > > - Antonis Riakiotakis works on subsurf drawing speed improvement. > > - Ton Roosendaal suggested to work on replacing GL_SELECT (OpenGL select > feature, used for object mode). Too many problems are arising with it now. > Could be simply replaced with using back-buffer color codes. Alternative is > to do brute-force raycast tests. Might even work surprisingly fast. > > Thanks, > > -Ton- > > > Ton Roosendaal Blender Foundation t...@blender.orgwww.blender.org > Blender Institute Entrepotdok 57A 1018AD Amsterdam The Netherlands What is the plan with subsurf, multiress and dynamic topo? They all seem to fight with each other and as a user it is very confusing. I would not want to be a newbie and face all that. Good news is that it does not seem to crash. -- Douglas E Knapp Creative Commons Film Group, Helping people make open source movies with open source software! http://douglas.bespin.org/CommonsFilmGroup/phpBB3/index.php Massage in Gelsenkirchen-Buer: http://douglas.bespin.org/tcm/ztab1.htm Please link to me and trade links with me! Open Source Sci-Fi mmoRPG Game project. http://sf-journey-creations.wikispot.org/Front_Page http://code.google.com/p/perspectiveproject/ ___ Bf-committers mailing list Bf-committers@blender.org http://lists.blender.org/mailman/listinfo/bf-committers
Re: [Bf-committers] 'help'
It's 3 clicks from wiki start page to svn page: http://wiki.blender.org/index.php/Dev:Doc/Tools/SVN_checkout_and_usage Am 23.02.2013 20:57, schrieb Erwin: > At a first glance that wiki page has no information about the subversion > repository. It usually takes me a while of searching too. > > So what is the svn url? > > Sent from my iPhone > > On Feb 23, 2013, at 11:28 AM, Tom M wrote: > >> Follow the directions here >> >> http://wiki.blender.org/index.php/Dev:Doc/Building_Blender >> >> Which will direct you to the method for downloading the source code from SVN. >> >> Good luck, >> >> LetterRip >> ___ >> Bf-committers mailing list >> Bf-committers@blender.org >> http://lists.blender.org/mailman/listinfo/bf-committers > ___ > Bf-committers mailing list > Bf-committers@blender.org > http://lists.blender.org/mailman/listinfo/bf-committers > ___ Bf-committers mailing list Bf-committers@blender.org http://lists.blender.org/mailman/listinfo/bf-committers
[Bf-committers] Blender developer meeting notes, 24 February 2013
Hi all, Here are the notes of today's session in irc.freenode.net #blendercoders 1) 2.66 release review - Some showstoppers have been found already: - A change in how textures use alpha (option moved to Image block), required a do-version patch which crashes for library linking in cases. - Compositor: File Output node missed socket input menu - Dynatopo crash in some cases. - We will check next sunday again, then do apply fixes on the release branch and call for an 'a' build... - Important note: let's not do cleanups of code now please, except when approved by all involved module owners! 2) Targets for 2.67 - Freestyle, no real news, but final review can start soon. - Sergey Sharybin worked on (motion tracking) Bundle Adjustment, should be in svn soon. - Proposal for release planning: March 10 bcon2 (= targets are defined) March 31 bcon3 (= targets are in svn) April 14 bcon4 (= only fixes, test builds and/or RC) April 21 bcon5 (= release imminent) 3) Other projects - Antonis Riakiotakis works on subsurf drawing speed improvement. - Ton Roosendaal suggested to work on replacing GL_SELECT (OpenGL select feature, used for object mode). Too many problems are arising with it now. Could be simply replaced with using back-buffer color codes. Alternative is to do brute-force raycast tests. Might even work surprisingly fast. Thanks, -Ton- Ton Roosendaal Blender Foundation t...@blender.orgwww.blender.org Blender Institute Entrepotdok 57A 1018AD Amsterdam The Netherlands ___ Bf-committers mailing list Bf-committers@blender.org http://lists.blender.org/mailman/listinfo/bf-committers
Re: [Bf-committers] Difference between "locale" and "locale with parent"
HI, This is what would be named "model space" in other software. It is the postion orientation and scale of the bone relative to the armature root. 2013/2/24 Tobias Oelgarte > Hello, > > is there any documentation on the difference between "local" space and > "local with parent" space? All i could find was this short > explanation[1], but without knowing what it is supposed to do exactly, > it tells me nothing. > > [1] > > http://wiki.blender.org/index.php/Doc:2.4/Manual/Constraints/Common_Interface#The_Constraint_Space_.28CSpace.29 > > Greetings from > Tobias Oelgarte > ___ > Bf-committers mailing list > Bf-committers@blender.org > http://lists.blender.org/mailman/listinfo/bf-committers > ___ Bf-committers mailing list Bf-committers@blender.org http://lists.blender.org/mailman/listinfo/bf-committers
Re: [Bf-committers] [Bf-blender-cvs] SVN commit: /data/svn/bf-blender [54806] trunk/blender: fix for fullscreen on X11 (used by the BGE, not blender application),
On Sun, Feb 24, 2013 at 11:09 PM, Thomas Dinges wrote: > Hi, > one of those commits broke Windows compile. > > http://www.pasteall.org/39992 > > Am 24.02.2013 06:05, schrieb Campbell Barton: >> Revision: 54806 >> >> http://projects.blender.org/scm/viewvc.php?view=rev&root=bf-blender&revision=54806 >> Author: campbellbarton >> Date: 2013-02-24 05:05:29 + (Sun, 24 Feb 2013) >> Log Message: >> --- >> fix for fullscreen on X11 (used by the BGE, not blender application), >> changing the screen resolution wasn't still allowed for larger virtual >> desktops. >> >> added an exclusive option to ghost so the fullscreen window is ignored by >> the window manager and we get all events. (common practice for games on X11). >> > -- > Thomas Dinges > Blender Developer, Artist and Musician > > www.dingto.org Defining empty functions for OSX and Windows will resolve the build error. -- - Campbell ___ Bf-committers mailing list Bf-committers@blender.org http://lists.blender.org/mailman/listinfo/bf-committers
[Bf-committers] Difference between "locale" and "locale with parent"
Hello, is there any documentation on the difference between "local" space and "local with parent" space? All i could find was this short explanation[1], but without knowing what it is supposed to do exactly, it tells me nothing. [1] http://wiki.blender.org/index.php/Doc:2.4/Manual/Constraints/Common_Interface#The_Constraint_Space_.28CSpace.29 Greetings from Tobias Oelgarte ___ Bf-committers mailing list Bf-committers@blender.org http://lists.blender.org/mailman/listinfo/bf-committers
Re: [Bf-committers] [Bf-blender-cvs] SVN commit: /data/svn/bf-blender [54806] trunk/blender: fix for fullscreen on X11 (used by the BGE, not blender application),
Hi, one of those commits broke Windows compile. http://www.pasteall.org/39992 Am 24.02.2013 06:05, schrieb Campbell Barton: > Revision: 54806 > > http://projects.blender.org/scm/viewvc.php?view=rev&root=bf-blender&revision=54806 > Author: campbellbarton > Date: 2013-02-24 05:05:29 + (Sun, 24 Feb 2013) > Log Message: > --- > fix for fullscreen on X11 (used by the BGE, not blender application), > changing the screen resolution wasn't still allowed for larger virtual > desktops. > > added an exclusive option to ghost so the fullscreen window is ignored by the > window manager and we get all events. (common practice for games on X11). > -- Thomas Dinges Blender Developer, Artist and Musician www.dingto.org ___ Bf-committers mailing list Bf-committers@blender.org http://lists.blender.org/mailman/listinfo/bf-committers
Re: [Bf-committers] [Bf-blender-cvs] SVN commit: /data/svn/bf-blender [54806] trunk/blender: fix for fullscreen on X11 (used by the BGE, not blender application),
Confirm r54811 fixed my problem with multisample! Thanks Campbell. Bastien On 24/02/2013 11:09, Bastien Montagne wrote: > Hi Campbell, > > This commit completely breaks Blender for me, when using multisamples > (without it, everything is fine): > * No window's decorators (frame, buttons, etc.); > * Keyboard is completely ignored by Blender, only mouse works (thankfully!); > * Blender display itself (it's OpenGL windows) works OK, even when > invoking another window (user preferences), but it prevents anything > else to be drawn (i.e. impossible to go back to desktop or switch to > another app)! > > Debian Wheezy 64, ATI HD5730 mobility, fglrx drivers 12-6 (wheezy's > "official" ones). > > Best regards, > Bastien > > On 24/02/2013 06:05, Campbell Barton wrote: >> Revision: 54806 >> >> http://projects.blender.org/scm/viewvc.php?view=rev&root=bf-blender&revision=54806 >> Author: campbellbarton >> Date: 2013-02-24 05:05:29 + (Sun, 24 Feb 2013) >> Log Message: >> --- >> fix for fullscreen on X11 (used by the BGE, not blender application), >> changing the screen resolution wasn't still allowed for larger virtual >> desktops. >> >> added an exclusive option to ghost so the fullscreen window is ignored by >> the window manager and we get all events. (common practice for games on X11). >> >> Modified Paths: >> -- >> trunk/blender/intern/ghost/GHOST_ISystem.h >> trunk/blender/intern/ghost/GHOST_IWindow.h >> trunk/blender/intern/ghost/intern/GHOST_System.cpp >> trunk/blender/intern/ghost/intern/GHOST_System.h >> trunk/blender/intern/ghost/intern/GHOST_SystemX11.cpp >> trunk/blender/intern/ghost/intern/GHOST_SystemX11.h >> trunk/blender/intern/ghost/intern/GHOST_Window.cpp >> trunk/blender/intern/ghost/intern/GHOST_Window.h >> trunk/blender/intern/ghost/intern/GHOST_WindowManager.cpp >> trunk/blender/intern/ghost/intern/GHOST_WindowX11.cpp >> trunk/blender/intern/ghost/intern/GHOST_WindowX11.h >> trunk/blender/source/gameengine/GamePlayer/ghost/GPG_Application.cpp >> >> [...] > ___ > Bf-committers mailing list > Bf-committers@blender.org > http://lists.blender.org/mailman/listinfo/bf-committers > ___ Bf-committers mailing list Bf-committers@blender.org http://lists.blender.org/mailman/listinfo/bf-committers
Re: [Bf-committers] [Bf-blender-cvs] SVN commit: /data/svn/bf-blender [54806] trunk/blender: fix for fullscreen on X11 (used by the BGE, not blender application),
Hi Campbell, This commit completely breaks Blender for me, when using multisamples (without it, everything is fine): * No window's decorators (frame, buttons, etc.); * Keyboard is completely ignored by Blender, only mouse works (thankfully!); * Blender display itself (it's OpenGL windows) works OK, even when invoking another window (user preferences), but it prevents anything else to be drawn (i.e. impossible to go back to desktop or switch to another app)! Debian Wheezy 64, ATI HD5730 mobility, fglrx drivers 12-6 (wheezy's "official" ones). Best regards, Bastien On 24/02/2013 06:05, Campbell Barton wrote: > Revision: 54806 > > http://projects.blender.org/scm/viewvc.php?view=rev&root=bf-blender&revision=54806 > Author: campbellbarton > Date: 2013-02-24 05:05:29 + (Sun, 24 Feb 2013) > Log Message: > --- > fix for fullscreen on X11 (used by the BGE, not blender application), > changing the screen resolution wasn't still allowed for larger virtual > desktops. > > added an exclusive option to ghost so the fullscreen window is ignored by the > window manager and we get all events. (common practice for games on X11). > > Modified Paths: > -- > trunk/blender/intern/ghost/GHOST_ISystem.h > trunk/blender/intern/ghost/GHOST_IWindow.h > trunk/blender/intern/ghost/intern/GHOST_System.cpp > trunk/blender/intern/ghost/intern/GHOST_System.h > trunk/blender/intern/ghost/intern/GHOST_SystemX11.cpp > trunk/blender/intern/ghost/intern/GHOST_SystemX11.h > trunk/blender/intern/ghost/intern/GHOST_Window.cpp > trunk/blender/intern/ghost/intern/GHOST_Window.h > trunk/blender/intern/ghost/intern/GHOST_WindowManager.cpp > trunk/blender/intern/ghost/intern/GHOST_WindowX11.cpp > trunk/blender/intern/ghost/intern/GHOST_WindowX11.h > trunk/blender/source/gameengine/GamePlayer/ghost/GPG_Application.cpp > > [...] ___ Bf-committers mailing list Bf-committers@blender.org http://lists.blender.org/mailman/listinfo/bf-committers
Re: [Bf-committers] Proposed default margin increase for baked textures
On Sun, Feb 24, 2013 at 1:17 AM, metalliandy...@googlemail.com wrote: > Bake panel> Margin #px :) > > On 23 Feb 2013, at 21:24, Knapp wrote: > >> On Sat, Feb 23, 2013 at 8:30 PM, Sean Olson wrote: >>> +1 from me too - I always bump that level up to a min of 16. I could see >>> less if your UV's were really packed in tight, but 16 or higher usually is >>> what does the trick for killing nasty seams for me. One less thing to >>> worry about for most cases would be great. >>> >>> -Sean >> >> Where do you set that? >> >> >> -- >> Douglas E Knapp Ah, thanks. I see my problem. I have been only working in cycles for some time but this panel only shows up with internal render. PS someone working the manual might want to point that fact out. -- Douglas E Knapp Creative Commons Film Group, Helping people make open source movies with open source software! http://douglas.bespin.org/CommonsFilmGroup/phpBB3/index.php Massage in Gelsenkirchen-Buer: http://douglas.bespin.org/tcm/ztab1.htm Please link to me and trade links with me! Open Source Sci-Fi mmoRPG Game project. http://sf-journey-creations.wikispot.org/Front_Page http://code.google.com/p/perspectiveproject/ ___ Bf-committers mailing list Bf-committers@blender.org http://lists.blender.org/mailman/listinfo/bf-committers
