An update. > Imo, static type checking adds little to python. If you want proper type checking, try rust or other typed languages.
That's a hasty verdict. mypy runs off of annotations, so it's unfair to expect mypy to be much help without them! The annotations of the types returned from functions and methods are likely to be the most important. Such annotations cut the Gordian knot of type inferences. Local inferences are straightforward; inferring return types is much harder. > Impatience is the clear enemy here. The work is mundane, but I must not introduce any new bugs! Yesterday I completed the annotations for leoGlobals.py so that mypy now passes launchLeo.py. The work was indeed tedious. Afterward, I thought that a wax-on script, the companion of wax-off, would have saved me much of the tedium. But wait. Adding annotations for returned types is much harder than adding annotations for arguments, and wax-on would not help. Oh my. As I write this, I think I remember that my old make_stub_files ( *msb*) project calculates returned types based on pattern matches of the values of all "return" statements. *Aha!* Time to experiment with msb! *Summary* Aha! My old make_stub_files project may be more valuable than I remembered. I'll delay writing the wax-on script. My focus for Leo 6.4 is to complete Leo's mypy command. I'll also carefully fix any bugs that mypy reveals. Except for leoGlobals.pyi, I will create no new stub files for Leo 6.4. I will *not* add annotations to Leo's source files themselves. Edward -- You received this message because you are subscribed to the Google Groups "leo-editor" group. To unsubscribe from this group and stop receiving emails from it, send an email to [email protected]. To view this discussion on the web visit https://groups.google.com/d/msgid/leo-editor/20c35a7e-d6e1-40c5-ba8c-84561d1969d1n%40googlegroups.com.
