In message <5373f27629riscos@xxxxxxxxxxxxx> Richard Ashbery <riscos@xxxxxxxxxxxxx> wrote: > In article <4aceea7353.Dentrassis@mail>, Gavin Crawford >> I've now uploaded a very simple example of what I'm suggesting to: >> http://www.dentrassis.com/gav/temp/AW_PathSelectionExample.zip [snip] > So much clearer - kind of you to spend time by illustrating the > problem. My main confusion I think was the selection method - I think > in terms of selection rectangles not an irregular 'marquee'. This > would take a bit of getting use to - the selection marquee would be a > path, because control points are needed for positioning. The result > would be an irregular shape. I wonder how the software would > differentiate between a path and a selection. Yes, my idea was to use a standard closed shape (path) object, that you would draw in the required position using the standard drawing tools, such as the line, curve, freehand tools. I may have confused things a little by calling it a 'marquee' which suggests that you would draw/drag an irregular selection shape with some sort of lasso selection tool. That is not what I had in mind. A normal shape object would allow all the fine tuning and editing needed - and can still be deleted if no longer needed after the operation. ArtWorks would know when you had a single filled shape selected, and un-fade the new proposed menu entries - in a similar way as it does already with the 'Shape info...' entry. > I'm sure fine-tuning shape selection in the way you have described so > succinctly would be useful - I guess it will depend on whether Martin > can spend the time implementing these features which may only be used > by a few. I agree. As usual, these sorts of requests for new features, come down to how much work they take to implement and how much of the inner workings are already catered for within the code. When I first though about this I did think that the problem would lie with the detection of the shape's boundary (it's path) and that the selection process would need to know how that shape interacted / intersected with other objects - and that it couldn't just take the rectangular bounding box of the shape as the selection source. In my own mind, I thought that it would be so complex to implement, that it wouldn't be worth even suggesting, but there are already some clever processes within ArtWorks in this respect that may just mean that some of the work has already been done. For example compare how Draw and ArtWorks detects the selecting of an irregular shape - in ArtWorks you have to select within the shape, whereas in Draw only within the bounding box, and the pointer can actually be outside the shape's line. There's also the incredibly complex Intersect features that Martin has already introduced some time ago, maybe this code would be related to this new selection proposal. So after some though, I decided to post here to see if there was enough support and discussion about it, that would make Martin even briefly consider the idea. Over the years of using RISC OS for graphic design work, I have perfected quite a habit of annoying software developers with requests for new features, as I'm sure both Martin Wuerthner and David Pilling would testify to. And although some suggestions never make it in, some have done, for which I'm very grateful. For me, it's one of the most amazing elements of the RISC OS world [1] - I can't imagine getting anywhere if you asked Adobe to add some features to their software. [1] Quick, pass me a tissue I'm starting to well-up! > Very neat the way you created the pseudo menu - it looks as though > this change has already been implemented ;-) I wish it was that easy. :-) I guess it only really works if you are using the same window theme as me. Regards Gavin ------------------------------------------------------------ To change, suspend or cancel your subscription go to //www.freelists.org/list/artworks ------------------------------------------------------------