The checkoff list has been winnowing down to its last elements. Left to do is:
Matrix math.
Properties.
Channels.
Classes.
Also clean up tasks:
Multitasking in back end and associated procedures.
Full debug of vectors and other elements.
I also think I will go ahead and tackle the following:
Addition of unicode UCS-16 and UTF-8 modes.
Cardinals.
These are not complex, and will add needed functionality.
The list of to dos went down rapidly, but that was partially because I tackled the easiest ones first. These last items are in fact challenging. After full Pascaline functionality happens, I expect an extended period of time while I construct a Pascaline acceptance test from the IP Pascal acceptance test. That test is a subset of all Pascaline.
After the Pascaline language is fully functional, it needs a lot of testing to become hardened and relatively error free. I know for a fact it has slipped in error handling since mid year, since I have seen it break many times on bad program input. That needs cleaning up.
After that it will need reorientation for the code generator. I have plans for a better framing system that matches generated assembly code better than the legacy Pascal-P system.
Thus, though I expected to be functionally complete by year end, it is actually close.
Happy new year!
If you would like to refer to this comment somewhere else in this project, copy and paste the following link:
The checkoff list has been winnowing down to its last elements. Left to do is:
Also clean up tasks:
I also think I will go ahead and tackle the following:
These are not complex, and will add needed functionality.
The list of to dos went down rapidly, but that was partially because I tackled the easiest ones first. These last items are in fact challenging. After full Pascaline functionality happens, I expect an extended period of time while I construct a Pascaline acceptance test from the IP Pascal acceptance test. That test is a subset of all Pascaline.
After the Pascaline language is fully functional, it needs a lot of testing to become hardened and relatively error free. I know for a fact it has slipped in error handling since mid year, since I have seen it break many times on bad program input. That needs cleaning up.
After that it will need reorientation for the code generator. I have plans for a better framing system that matches generated assembly code better than the legacy Pascal-P system.
Thus, though I expected to be functionally complete by year end, it is actually close.
Happy new year!