May 2012 - Inkscape Devel - Inkscape Mailman

Recursive Drawing
by Veronika 15 May '12

15 May '12

I just saw the video for a recursive drawing tool and thought you might find it interesting: http://recursivedrawing.com/ It seems like a very natural way for designing L-systems. Veronika

1 0

Fwd: Re : Fw: Menu refresh
by Susan Spencer 14 May '12

14 May '12

It would also be helpful to add command line options: inkscape --preferences=mydefault_preferences.xml --template=mydefault_template.svg - Susan On Mon, May 14, 2012 at 2:46 AM, John Smith <john.smith7545@...36...>wrote: > Re: [Inkscape-devel] Re : Fw: Menu refresh

1 0

Fw: Menu refresh
by John Smith 14 May '12

14 May '12

Any thoughts on changes to the main menu layout ? 1. Move "File->Inkscape Preferences" to "Edit->Preferences" This seems to be the preferred location in the Gnome HIG http://developer.gnome.org/hig-book/3.4/menus-standard.html.en 2. Move "Edit->Clone" to "Object->Clone" Does this make more sense, since we are really cloning objects ? 3. Split View menu (including Show/Hide submenu) into 2 menus - View (UI "decorating" the main view) and Window (dockable dialogs) View->Grid View->Snap View->Commands Bar View->Toolbox View->Rulers View->Scrollbars View->Palette View->Status Bar ... etc Window->Icon Preview Window->Input Devices Window->Memory Window->Messages Window->Swatches Window->Show/Hide dialogs 4. Merge "File->Document Metadata" into "File->Document Properties"

13 14

Active guides proposals
by Valerie 14 May '12

14 May '12

The fact that guide improvement is one of the projects accepted by Google Summer of Code projects is really good news! Guides are particularly helpful features in illustration programs. I thought it'd be a good time to write up some of the possibilities of guides. I doubt anyone can implement any of it anytime soon (certainly not before the current GSOC project is finished), but it might be helpful to keep these in mind to not close off future possibilities in the definitions or the GUI. Write-up here: http://wiki.inkscape.org/wiki/index.php/Active_guide_proposals As you can see, features I think will can be particularly useful are: - Symmetry guides: by simply toggling them on and off, you can enter an improved "mirror" mode. Helpful for everything from complicated technical illustrations to just drawing some vases in some corner. - Convergence points: this could be a mere feature of a more general "guide point". When drawing straight lines, the lines would "snap" to the direction of one of the active convergence points. Anyone who's ever tried to illustrate a building in 2D or 3D should know how helpful this would be (because the vanishing points can be so far away from the rest of the drawing that it's really no fun to look for them every time). - More generally, guide points and guidelines could have various properties activated. Guidelines with "orthogonal" and "parallel" properties activated, for example, would cause straight lines drawn to snap in a direction orthogonal/parallel to said guidelines. Very helpful for geometric drawing. - Then there are kinematic templates (by Richard Fung). I don't know why this still hasn't been ported to any of the major open source image programs, but in Inkscape, they would be great fun when combined with the calligraphy tool. I can work more on the GUI proposal in the meantime, but I'm not sure where to go from here, so comments are welcome.

1 0

Menu and keyboard shortcuts
by John Smith 12 May '12

12 May '12

Is there a reason Inkscape doesn't show the keyboard shortcuts in the main menu on Linux ? Keyboard shortcuts show up in the * Inkscape menus on Windows * Inkscape context menus on Linux * GIMP main menu on Linux ... so i'm guessing its not a GTK issue ?

4 7

Re: [Inkscape-devel] Circular Arrangement
by Valerie 07 May '12

07 May '12

Nice proposal! I just find the use of arrows for "bounding box" confusing. For me, arrows refer to a Direction, so my first impression is that you're choosing the direction for a transformation, except that doesn't make sense here. There's a really nice existing blueprint for anchor points here: https://blueprints.launchpad.net/inkscape/+spec/transformation-anchors For the sake of the future interface consistency, I'm wondering if that arrow-less grid shouldn't be used instead, though maybe in a bigger version. A mouse-over will pop up descriptions like "Use objects' lower-left bound." Another question is... how does this feature handle it if you want all the objects to Just touch the guide edge, but you don't rotate the objects? Basically, for an object on the upper-right, the lower-left bound should be used, while for an object on the lower-left, the upper-right bound should be used (which isn't quite an accurate way of putting it either, since the object doesn't necessarily touch the circle that way). See what I mean? Something like "Objects to touch circle's inner edge" (or just "Touch inner edge") and "Objects to touch circle's outer edge." Then again, if anybody needs that, they could submit it as a feature proposal later. I really look forward to the feature, rotational arrangements are a pain to do right now (I can never remember how to use that tiling interface). So thanks in advance!

3 4

affine/transform question
by mathog 03 May '12

03 May '12

In the emf print virtual methods the current transform is pulled out with: Geom::Affine tf = m_tr_stack.top(); Is the stack supposed to hold the final transform, or just the stack of partial transforms? For instance, for this bug: https://bugs.launchpad.net/inkscape/+bug/965594 the text SVG has a set of nested transforms, and the top only returns the inner one. The problem can be worked around by ungrouping everything, which seems to replace the serial transformations with a single one for each object. Otherwise it seems the proper action is to multiply down the transform stack to find the final complete transform operation. It looks like the stack is "fed" by the ::bind method, but that just moves the question out a bit, as that method has a call argument which is itself a transform. (Unfortunately "bind" is a common string in this code, so grep was not very helpful in locating whatever it is that calls the [virtual] ::bind method.) Thank you, David Mathog mathog@...1176... Manager, Sequence Analysis Facility, Biology Division, Caltech

5 8