Turns out it's a stdcall which has no mangling. I can debug the c extension in CBuilder. Very fast. Now I need to experiment with arguments etc. That will be the acid test for stdcall If anyone is interested I can post the files and the steps. Once I get a real good handle on this I'll try it out with XE7 which is a compiler that compiles Win32, Win64 and OSX. I will also test this on my lap top where Borland memory manager (a dll) doesn't exist. It will tell me if I need to include any of these files or not.

fantastic as always. Thank you slbaumgartner and sdmitch. The flip all x works perfectly! The sdmitch code already was great. This forum is always great. I will make a simple free plug-in to help road build, and share here when it is ready.

thanks John, already did that though (screenshot above is a watermark). the problem with this system is that you are forcing a new style on the user: if the user has a sketchy style on, you cannot have the bars and keep the sketchy style. (in other words we cannot load watermarks without loading a style as well). that's why i mentioned about another route: find user current style > save it as variable > load the 'grid' style > restore user style from variable when finished. alas, no way to get the previous style back in. whether from the 'in model' style, or trying to save it to a temp directory and loading it back from there. (at least i found no info at all anywhere). PS least of all the api docs!

Thank you Thom.

For my own extensions that I write now I usually drop support for anything older than SU2014 - due to Ruby 2.0 and other improvements since then.

I've run into issues with taking the fist or second Mac Address out of ifconfig or ipconfig /all. I've changed the code to look at the entire file and then to step through it line by line and treat it as a bit of a state machine. Thanks to pgarmyn and Driven I think I've got some code that will work. They both helped with files as have a handful of other people. With windows I've tested files in English, French, Russian and some with VMWare set up. With OSX I've just tested in English. Mac Address

@thomthom said: The BoundingBox object you get when you query the instance will always be oriented to world space. To get the bounding box as drawn in the UI you need to get the bounds of the definition - fetch all the points from its corners and transform them using the current edit_transform and the selected instance transform. That extension I linked to does all that. Ah! Thanks for the explaination ThomThom! I didn't realise that's what happened with instances' bounding boxes...very confusing for a novice! I did take the time to check out your extension and was able to understand what you're doing and apply that to my own code... thanks again sir!

@jolran said: As a continuation of Thomthoms advice. Should you ever want to create a C-extension easily for distance between Points the code is already written and available https://bitbucket.org/thomthom/sketchup-ruby-c-extension/src/e2fe0f2e43e43ebae568535b6d7a98e160b4624c/src/Example%20-%20Basics/SXBasics.c?at=default Note: That example is OLD! You can refer to it for how I resolve the distance calculation, but had I done that today I would have done it in C++ using Visual Studio and Xcode: https://github.com/SketchUp/ruby-c-extension-examples

@archidave said: but in SU 2015; > 2.m.abs gives the NoMethodError This is pretty bad (it has broken my plugin for SU 2015) and means I now need to go hunting through the code for all occurrences of <Length>.abs replacing them with <Length>.to_f.abs and building-in extra validation where the class is not intrinsically known, just in case there could be a Length instead of a Float. The Length class used to be a subclass of Float. However this isn't really possible in Ruby, but from the Ruby C API this was hacked together a way to do so. When we added 64bit support in SU2015 this broke because the data structures where different. So we had to re-implement Length by forwarding the methods to Float. Initially this used the Ruby Standard Library, but due to another bug in Ruby where the StdLib would not always load if there was unicode characters in the user's username the StdLib would fail and the Length class would not properly initialize. There was also a bug in SU at some point where the StdLib would fail to load if the user started SketchUp by double clicking an SKP file. We later corrected the new Length implementation to not use StdLib so it would not fail if the StdLib failed to load. And we also fixed the bug regarding starting SU via SKP files. But those fixes should be in SU2015M2 - so I don't understand why you'd experience such problems. I am not able to reproduce your error. Can you try without any extensions enabled - just so we can eliminate that possibility?

@unknownuser said: There is no "external" ruby pipeline. Phew, I was worried for a moment.

If anyone is interested, I finally solved my imported group's origin problem fairly simply with this... ga = entities.grep(Sketchup::Group) ga.each { |e| n=e.name; e.model.entities.add_group(e); e.parent.instances[0].name=n} Basically it just takes all the groups after import, adds them into a new group and names the new group the same name. This resets the origin, and allows the swap code to work properly.

@driven said: well, they have always said not to rely on them... but this works on my mac, but isn't in the list... Sketchup.send_action "select3dTextTool;" john it works also in windows 7 it was not documented thank you

Note that if only close the currently open instance. If you are deeply nested you need to loop until you are back in the root - (if that's what you desire). And before SU2014 the method was bugged and could mess up geometry if the close was undone. (The method actually creates an undo item)

@sdmitch said: @lbsswu said: Hi everyone, Suppose I have opened the Sketchup software, I want to know whether a 3D point is visible or not in current view. For example, the 3D point may locate on the back of a person, thus we can't see it when the person facing to us. This should be done by Sketchup ruby code.Could anyone help me? I placed a construction point, the little black dot just below the horizon, and "Nancy" in the model. I then used the following code to test if the point was visible. mod = Sketchup.active_model > ent = mod.active_entities > sel = mod.selection > vue = mod.active_view > eye = vue.camera.eye > cp = ent.grep(Sketchup;;ConstructionPoint)[0] > pt = cp.position > vec = pt.vector_to(eye) > hit = mod.raytest([pt,vec]) > if hit > puts "pt is not visible" > else > puts "pt is visible" > end > Hi, sdmitch. Thank you very much. It works!

@tig said: view.screen_coords returns the point in the Model as X,Y coords - screen value. But you must ensure that the exported image used the same size as the window, otherwise there is no direct relationship between the point and the image's pixels... Got it! Thanks for your help!

I just thought I'd post this snippet in case it will help someone with a similar problem. def set_origin(compdef, compinst, pt0, pt1, pt3) tr1 = compinst.transformation rotvector = pt1-pt0 originvector = Geom;;Vector3d.new(1,0,0) org = pt3 angle = rotvector.angle_between originvector if angle < 90.degrees angle2 = -angle elsif angle > 90.degrees angle2 = angle end #if if angle == Math;;PI/2 angle2 = -90.degrees end if pt0.y > pt1.y && angle == Math;;PI/2 #Check if it's drawn at 90 degrees but in opposite direction angle2 = 90.degrees end compinst.transform! tr1.inverse tr2 = Geom;;Transformation.rotation(org, Z_AXIS, angle2) tr3 = Geom;;Transformation.new(org) compdef.entities.transform_entities(tr2, compdef.entities.to_a) compdef.entities.transform_entities(tr3.inverse, compdef.entities.to_a) compinst.transformation = compinst.transformation * tr3 compinst.transform! tr1 * tr2.inverse compdef.invalidate_bounds end #def

There are a few things that I had to deal with. I found it easier to put the dynamic attributes in a nested group as opposed to the outside group. This way you can rotate the entire door and jamb to fit it into the door opening and you don't have to keep changing the dynamic rotz. These doors can be copied, flipped and or rotated while preserving the desired angles of rotation. 4 doors.skp [image: COJv_4doors.PNG]

@cadfather said: Thanks Guys, i'll have a look at Jan's table. can't believe it's so messy working with the keyboard. Yes, messy - but the mess originates in the OS, not SketchUp!

Got it, thanks Thom. I realised that I could have optimised it to recurse fewer times, but at some point with increasingly complex surfaces it will run into a stack overflow again. I had a little read elsewhere and found that there are workarounds (ulimit and altering the interpreter, sounds like a bit of a dodgy hack to me..), but eventually it's always got an upper limit. I very much like the idea of a Geometry walker traversing geometry, and ended up using a solution very similar to yours. Cheers! -CJ

i guess one workaround is to make a webdialog tiny enough not to intrude but able to 'catch' the keystrokes repeating...