I guess the whole point is to understand whether there is a way to know 'in advance' the size of a text (based on string and font) so that it can be positioned accurately in the model at creation.

On my side I did not find a way to do that.

Thanks very much.
I know how to do it via Windows, but if we start programming in Windows SDK instead of Ruby, this is less fun, and more important, additional headaches for deployment.

One suggestion would be a few members of the sites, knowledgeable in system programming (Windows and Mac) group their efforts to build a utility libraries (dll) to complement the Ruby API of Sketchup (menu, events in tools, mouse, etc...). Unless the Sketchup SDK is more powerful!

I was wondering if there is a way to move programmatically the mouse pointer, within a Tool.
This may be useful when you have to open a dialog box and want to reposition the mouse pointer where it was before.

Thanks in advance for your help

It seems that at least you triangulate the faces, from what I could judge with the deer's kees (and the few faces that were not triangular in the original model).
This is a great tool, and judging from the video, the performance are more than "better"! (compared for instance to a simple shear of the model, that would take 40 seconds with a Ruby script). I imagine that you use a compiled version in C++).
I would encourage you to continue on this path, really!
If you need some help for the GUI, just post a description of what you need and I am sure we would find a way to contribute.

I have this one in preparation, for a composite model (i.e. any drawing elements, groups, etc... with texture transfer, etc...) and a curve in 3D space.
Actually, the main difficulty is about managing some topological cases where it is necessary to remove and cut faces that are squizzed in the curvature.

Looks like it does not find the cursor PNG file (cursorshear.png)
It must be in the Plugins directory
I thought I had put some code to check in any case, but apparently this does not work.

@unknownuser said:

Mais c'est bestial comme truc!

Ne serait-il point possible de répéter les dernières entrées paramétrables (celles-ci étant satisfaisantes d'une précédente action) sans passer par l'ouverture du Menu et l'appui de la touche "OK"?

Par un simple clic droit sur l'icône par exemple?
Mais c'est pour finasser sur l'ergonomie et booster la vitesse de modélisation
On doit gagner une seconde peut-être deux

Merci pour l'appreciation. J'espere que cela rendra quelques services.

Je me suis poser la question d'une execution 'rapide', mais on tombe rapidement dans l'indecidable car il y a pas mal de parametres. Sketchup ne permet pas de capturer un clic droit sur l'icone, et en creer un autre conduit a une proliferation des icones deja fort nombreux. En fait, je pense que quand on travaille sur un modele serieusement, on n'utilisera cette operation que rarement, et avec une preparation qui fait que quelques secondes de plus ou de moins ne changeront pas grand chose.

Azuby,

In theory, we could imagine that a user would not call a PDF at each Skecthup session, and if he does, after all, he could wait a little bit.
But your macro is fast enough that this fine too. When we have hundreds of PDF and HTML documenting scripts, we could revisit the problem

All this proves that we have a damned good team of Ruby script programmers on this forum, as it took less than 24 hours to get 3 good versions after the idea just popped up.

This is a great idea and can be done with no major issue (the center of the circle is at the crossing of parallels of segments, at a distance equal to the specified radius). There are of course some tricky cases where the specified radius would be too big and would overlap more that 2 segments, but this can be adjusted.

Great script!
And works fine for me on v6
I gave me some inspiration about what you can do with Sketchup and Ruby, when I redesigned the Bezier curves from the @Last version.

Thanks again

Fredo

TIG,

There is an error, due to a missing closing brace '}' on line 22.

Otherwise, it seems to work, though I may prefer to have the menu / submenu version (one suggestion would be to build the submenus at first call, rather that systematically at load time).

Thanks a lot

Adam,

You are right, we can do much cooler things, but there is still the problem of compatibility with v5.
But I still don't understand the logic of the basic Sketchup Windows-based dialogs for dimensioning combo boxes!

Just to inform you that there is an small update to Bezierspline scripts, available at Crai Ruby Depot http://www.crai.archi.fr/rubylibrarydepot/ruby/fr_Newest_scripts.html

Nothing big, but some bug fixing in the area of inference, as well as the support of Portuguese translation, with a Quick reference card as well (many thanks to Edson).

Fredo

PS: I am still expecting some gentle souls more accomplished in the art of geometry than me to publish new curve extensions

Didier,

Very nice idea.
It seems to have problems with the subfolders of the Plugins directory however. For instance with the PDF of Bezierspline which are in subfolder BZ_DIR. I think it is because the file name contains a '.' (on top of the extension).

Also, do you understand why the combo boxes in Sketchup Dialogs are never properly sized? I came across this problem too.

Daniel,

You are an artist. Adopted, and will go in the final published version, if OK with you

Thanks again

@cadfather said:

not sure what's happening but on my pc JPP seems to have clashed with beziers, either some icons disappear altogether, or the curves cannot be drawn. - i have noticed this before a couple of times while working on the rubymenus thus think the weakness might have more to do with beziers than jpp

EDIT: or it may be me of course!

I am not sure it is an interference with JPP, but there were a few bugs in Bezierspline regarding inference, and in some situation it could create the impossibility to position the starting point at the origin.

I made some fixes, and also incorporate the Portuguese translation (many thanks to Edson). Didier just published the new version of Bezierspline on his site today.

@cphillips said:

@gata said:

CPhillips was playing around with somethng similar http://www.sketchucation.com/forums/scf/viewtopic.php?f=57&t=2570&start=0&st=0&sk=t&sd=a but SketchyPhysics is a bit distracting. It's good to see this sort of tool revived.

I semi-finished that thing I was working on. Problem was that it was really slow a reasonable size model could take minutes to update. Maybe someday I'll go back an optimize it.

Shear looks like a handy tool to have!

I think your Free Form Deformation ballon is really one big step forward, because I guess it implies to handle the individual deformation of faces and edges, and then to re-arrange the topology of the model to keep meaningful geometrical continuity. In comparison, PlaneShear is just a linear offset of vertices that keep faces as faces and edges as edges, without any individual complexification.

Bon courage! I am quite admirative and impressed by what you can achieve with SketchyPhysics!

@pixero said:

I think it would be a good idea to let the plane shear also have a "deformation box" so that you could use it more like a tweak tool and try different amount of shear (something like the scale tool).
The current version is more suited for use with a known amount. In my humble opinion.
Anyway, thanks again!

It would be very difficult to consider an interactive deformation of the model. Keep in mind that the transformation reconstructs the selected portion of the model, face by face, edge by edge, etc...., as there is no native transformation in Sketchup doing a shear (i.e. a non uniform offset).
However, what I have done is to show interactively the deformation of the selected model in wireframe mode. This is much lighter and performance seems acceptable.

You need to download the 2 attached files nult.rband LibProtractorTool.rb.

version with interactive wireframe deformation

version with interactive wireframe deformation

@urgen said:

...Unfortunately I have former problems...

Could you turn on the Ruby Console and see what it says (or does not say)

Thanks

@greyhead said:

Fredo, I do think that there is something here that is not consistent with "as if you would also extrude the neighbouring face." See this diagram (plan view of an extruded dodecagon). Notice that the two end extrusions of the five contiguous faces are different. (I did these both on the same dodecagon for simplicity but I get the same result if I do them on separate objects so I don't think there is interference between them.)

Scratching my head here but I don't see an obvious pattern.

Bob

[attachment=2:2u7nlb8r]<!-- ia2 -->push-pull.png<!-- ia2 -->[/attachment:2u7nlb8r]

This abnormal behavior is actually almost 'normal'.
Because the polygon has 12 segments, the angle between the normals of 2 contiguous edges is exactly 30 degree, which is the default value of the Angle of Influence. So depending on the rounding up, some outside faces do influence, some others do not.

You should change the value of the Angle of influence to either 29 Degree (to have strict perpendicular borders) or to 31 degrees, to have them all radial.