Yes, I agree with thomas on the idea of collaborating on basic constructs. I have an example that might be a good to illustrate. I have found that many models are easily created using an extruded profile. The operation involves creating a 'sketch', which is a 2d set of closed curves, and the sweeps it along anoher curve to create a solid.
This operation is of cousre possible diectly with occ, but it would bee too much code. In this example, there are a few abstactions that are useful: the idea of a sketch, its semantic relationship with the curve to create an object. Basically, I would like to be able to relate these abstractions one time and have the kernel rebuild the solid when they change. A useful and important aspect of these abstractions is tha they should encapsulate building behaviors, while exposing properties allowing them to be parametricly modified. On 12/22/10, Thomas Paviot <tpav...@gmail.com> wrote: > Hello Julien, > > I was not aware of your project, although I try to stay up-to-date with > pythonocc related project. Internet is still a forest where google can get > lost. > > Your pycado is a really interesting project. I don't know exactly what are > your final objectives with it, but the beginnings seem promising. I will > test it as soon as possible. > > Your approach of developing a scripting engine can be classified in the > "Macro-Parametric solutions" category. As far as I understand your code, you > developed a kind of scripting language aiming at map the feature tree > available in most CAD softwares. That's indeed something we discussed here a > few days ago, and it looks like there is here a few fans of such a > technology. > > However, I'm still not convinced by this approach. In my opinion, developing > a scripting engine (with a parser, lexer etc.) with limited features on top > of a dynamic programming language like python is a kind of inconsistency. It > makes me think about something I'd call BASIC for CAD (I love BASIC, I > learnt programming with this language!-but it was a long time ago). > Furthermore, I do think that the modeling of 3D shapes is something much > more complex than a simple temporal sequence of basic operations (you might > call them "features"). Anyway, it's only my current opinion, and I could > fall in love with your project and change my mind. > > Whatever the technical solution you chose to perform CAD operations > (scripting or not), I'm more interested in the semantics you defined for CAD > data handling/creation/etc. At some point, it's a topic on which we could > collaborate (we've been here discussing the development of a high level API > for pythonOCC for a long time). > > Best Regards, > > Thomas > > 2010/12/22 julien blanchard <julien...@yahoo.fr> > >> Hello, >> >> I have just read the thread of Dave Cowden and I think that a project >> we're >> working on might interest someone. >> >> We have started the pycado >> project http://julienbld.github.com/pycado/ which seems quite similar to >> what >> you're looking for and to http://openscad.org/. >> It is based upon pythonocc and still in an early stage but it is working >> whith a >> restricted number of features. >> >> You'll find more information into the web page and feel free to download >> the sources and ask us anything about the project. >> >> We are now two developers and started the project on spring 2010 but we >> have not >> been active for the last months due to lack of time. We would be >> very insteresting in continuing the development with other people or to >> merge >> with a similar project. >> >> Julien Blanchard. >> >> >> >> >> _______________________________________________ >> Pythonocc-users mailing list >> Pythonocc-users@gna.org >> https://mail.gna.org/listinfo/pythonocc-users >> > -- Sent from my mobile device _______________________________________________ Pythonocc-users mailing list Pythonocc-users@gna.org https://mail.gna.org/listinfo/pythonocc-users
