Yes. Late binding allows lots of flexibility.

Thanks for all the help. I found the problem. Apparently if you add some geometry to the scene using ruby and then try to export a mesh of non-component geometry, the normals become unsmoothed. An odd quirk.

I agree. I tried to use the linetool to learn how to make a tool, as the API suggests. I got so confused. It is a pretty complex tool, at least more complex than it should be if it is being pointed to as a good script for beginners to learn from. Chris

Guys, Thanks for your help! Great insight.

Thanks very much TIG - here is a first try at the conduit - so easy.[image: kDoy_pig031.png] I think it has great potential, even maybe going straight from concepts to auto-modeling ... a kind of machine made provocation - 'what's wrong with this?'

Thanks, For now this shade analysis ruby worked for me. http://forums.sketchucation.com/viewtopic.php?f=323&t=18222&p=150950&hilit=shade+analysis+tool#p150950

See BUG report I posted: http://forums.sketchucation.com/viewtopic.php?f=11&t=27534 Which apparently was noticed by others as early as 5 years ago (and it's still NOT fixed!)

Good point. Thanks

ThomThom, Again, thanks for your speedy and gracious response. You've provided some good information. Time permitting, I'll work on this this weekend and hopefully come to a better understanding. Take care!

XML structures are normally specified in a XSD (XML Schema Definition) file. Check the header in the XML file to see if a URL link to a Schema file is specified.

I just checked on an older PPC mac. So to sum up this thread: ` # Sketchup Ruby OSX 10.4 PPC * RUBY_PLATFORM: fat-darwin8.8.0 * RUBY_VERSION: 1.8.5 Sketchup Ruby OSX 10.5 Intel * RUBY_PLATFORM: i686-darwin8.10.1 * RUBY_VERSION: 1.8.5`

I think it is a typo, as it is the ALT key on Mac. tested with a simple tool.

Hopefully, this file will work. It uses the system call ShGetFolderPath or ShGetSpecialFolerPath to retrieve the folder, so maybe the paths will be localized. @thomthom said: But on XP the folder names for sure is localised. I can try later today on my old laptop with Norwegian XP. I'd be interested to see if it works.

There is no scripting API for layout, no.

@thomthom said: Like AJAX solutions often do? This approach is most often used for "inter-frame" communication. If there are 2 frames in a page that are hosted on different domains, the JS in one is not allowed to access the other, but many browsers allow you to change the fragment identifier regardless of security, so it might work in this case. (There are a handful of other mechanisms people have used to get around frame-2-frame security, and any of them might work. Searching on the topic might give us a clever workaround.) Ajax apps will often modify the fragment to maintain state in the URL, so you can still email URLs around and have your friends land on the same "page" inside an Ajax app.

thanks, I'll give that a try. David update: Chris, thanks! that worked perfectly.

This is somewhat troublesome. One would then have to be very careful when creating observer instances. Ideally only create one instance - ever, if possible. And also clean up any references the observer has to other objects when possible, as otherwise they won't be GC'd either. grumble refactoring time.

ooops!

Me stupid

@dan rathbun said: @thomthom said: @dan rathbun said: (1) Jim did not subclass from Sketchup::ShadowInfoObserver No need. Su doesn't care. Just implement the methods required. I knew that Sketchup does't (currently,) check the observer's superclass. But isn't it a matter of good Ruby programming practice? And, what if (in the future,) Google implements some built-in methods, constants, etc. in the superclass that will need to be inherited by your custom observer's subclass(es)? You'd have to go and edit and re-release all your plugins that use observer's. Kind of seem to be by design: http://forums.sketchucation.com/viewtopic.php?f=180&t=17047&start=150#p171636