Yeah, at a certain point I realized that both the maintenance and the use of the language became much slicker if unnecessary deviations from Python were minimized. After all, when I'm writing Hy code, I'm usually spending a lot more time referring to the documentation of Python or third-party Python libraries than the documentation of Hy. I felt there were a number of ways Python could be improved upon, but e.g. the old feature that let you spell `True` as `true` in deference to Clojure was just a needless complication.
It is true that Hy really shines in those cases where it adopts an existing Python feature and adds meaningful quality-of-life improvements: anonymous functions without limitations; multiple iteration in for-loops; relaxed character set for identifiers. Things that seem completely obvious, once you have them.
It also demonstrates that elegance in a Lisp-on-Python is reached in a very different way than elegance in a stand-alone language, since it becomes an art of making the best out of what is already there.
You can add all the same type annotations as in Python, but from what I've seen, type-checkers expect Python source text and don't just use standard Python introspection, so you'll need to use `hy2py` first to actually check your program's types.
Yes, such as: metaprogramming via macros and reader macros; arbitrary compile-time computation; removal of restrictions on mixing statements and expressions; and other arities for Python's binary operators. See http://hylang.org/hy/doc/v1.0.0/whyhy#hy-versus-python
Dynamically shadowing global variables is not built-in, but easy to write a macro for if you want it. See e.g. https://stackoverflow.com/a/71618732
Yes, and it's a very nice tutorial! I'm interested in implementation details. Maybe there's no hygiene (and no scope sets etc.) to worry about—that would probably make documentation a little shorter. I'm sure the documentation will grow as people run into edge cases.
(I'm also probably a little spoiled with documentation coming from Racket which has like 4 big chapters dedicated to different aspects of macros scattered around the docs, plus some associated papers. Forgive me—I'm not trying to dunk on Hy; I just like reading docs.)
Admittedly, I've tried not to document the implementation. Yeah, they're pretty much simple dirty Common Lisp macros. Internally, they're functions that are called with the arguments converted to models (via `hy.as-model`), and then the return value is converted to a model. If a macro's first parameter is named `_hy_compiler`, it gets access to the current compiler object; this is undocumented since it's only meant for internal use. Reader macros have no parameters, but can access the current reader object as `&reader`. When it's defined, a reader macro is added to the current reader's dispatch table.
Hy-pothetically, yes, you could take Hy code in and spit Python code out via `hy2py`. I think at one point I considered supporting this officially, but then decided there was really no advantage.
That's how I'm using Hy at my job—I write Hy then hy2py it into Python, lightly polish the compiled Python for human consumption, and then share that with my Python-fluent but Lisp-illiterate coworkers.
