Each model is stored in a separate directory, along with the version
of the spec on which it was last run. And models can be locked.
However, I don't see the use of essentially keeping a separate copy of
each procedure in a model since, if testing the entire algorithm
reveals that a procedure needs to be changed, you'd want to change
that copy as well.
Anyway, all programming languages have some form of procedure. If no
one has thought it important enough to add such a feature to their
debugging tools, I have no reason to believe that it will be useful
enough to warrant adding it to the Toolbox. Especially when there are
so many other things that we'd like to do if we had the time.
On Friday, April 10, 2015 at 7:24:18 AM UTC-7, fl wrote:
3 - Suppose I want to develop an algorithm with pcal under toolbox . There are several procedures. I want to test each procedurewith a model before testing the main algorithm. There is no simple way to have models that do that I think. I mean we can't havea model that tests a procedure, another model that tests another procedure and eventually a model that tests the main algorithm.An idea that might solve this problem. The models are kept in their own subdirectories. And a copy of the thaplus specificationis copied in these directories every time the model is activated.Well suppose some models can be declared as frozen. No fresh copy of the specification would replace the old one. Thatway one might test a procedure (calling it in the main algorithm) and then proceed further with the specification andonce every procedure is developped with its own specification in its frozen model subdirectorie, we might develop themain algorithm with its own (non-frozen) models.--FL