shsms / tulisp Goto Github PK

Blocked by #67

Currently variables are bound dynamically, without tracking scope.

In the below example, there are two variable definitions for out,
and printer is declared to use the first out. But when called
from the scope of the second out, it uses the value of the second
out.

(let* ((out 10)
       (printer (lambda () (print (format "out is %s" out)))))
  (funcall printer)
  (let ((out 20))
    (funcall printer)))

This needs to change, and variables need to follow lexical scoping.

https://www.gnu.org/software/emacs/manual/html_node/elisp/List-Variables.html#List-Variables

A fix was added to handle these in: 915009c

This needs unit tests, so it doesn't break again.

Blocked by #58

equal on symbols is currently broken, because the comparison is done by comparing names. When comparing symbols, it should instead fall back to using eq.

Add a load function for loading and evaluating a file, like an import statement.

To identify if a symbol is bound to a value (in the current context): https://www.gnu.org/software/emacs/manual/html_node/elisp/Void-Variables.html#index-boundp

https://www.gnu.org/software/emacs/manual/html_node/elisp/Calling-Functions.html#index-apply

This can be achieved by adding a span to the error at each step, as the error gets propagated out.

There are a lot of functions that need to implemented: https://www.gnu.org/software/emacs/manual/html_node/elisp/Sequence-Functions.html

The currently implemented ones are listed here: https://docs.rs/tulisp/latest/tulisp/builtin/index.html#sequence-functions

This would allow back traces to show the correct files, for traces that span across multiple files.

ref: https://www.gnu.org/software/emacs/manual/html_node/elisp/Nonlocal-Exits.html

Ref: https://www.gnu.org/software/emacs/manual/html_node/elisp/Named-Features.html

Ref: https://www.gnu.org/software/emacs/manual/html_node/elisp/Setting-Generalized-Variables.html

It should be possible to have the same symbol with different spans. Right now, interned symbols get the span of the last occurrence of them in the code.

It could be implemented by giving unique ids to interned symbols, which sit inside the Rc, therefore not mutable. And using those to identify and compare symbols.

(> 40 70 30) should return nil. But it returns t.

<, <=, >= have similar problems as well. These need to be fixed.

These are if-let, when-let and while-let.

ref: https://www.gnu.org/software/emacs/manual/html_node/elisp/Conditionals.html

Because in tulisp it appears to be.

It is currently implemented as a macro that rewrites the let* expression with nested lets. This is not optimal, and leads to unnecessary recursive calls in the interpreter.

ref: https://www.gnu.org/software/emacs/manual/html_node/elisp/Plist-Access.html

In addition to the currently available setq, we need a set implementation (Primarily to be able to assign to dynamically created symbols).

https://www.gnu.org/software/emacs/manual/html_node/eintr/Using-set.html

https://www.gnu.org/software/emacs/manual/html_node/elisp/Mapping-Functions.html#index-mapconcat

They should always evaluate to themselves, and assignments to them should fail.

Ref: https://www.gnu.org/software/emacs/manual/html_node/elisp/Constant-Variables.html

They should take an array of tuples, like the constructor of a HashMap: https://doc.rust-lang.org/std/collections/hash_map/struct.HashMap.html#impl-From%3C%5B(K,+V);+N%5D%3E-for-HashMap%3CK,+V,+RandomState%3E

https://www.gnu.org/software/emacs/manual/html_node/elisp/Records.html
https://www.gnu.org/software/emacs/manual/html_node/cl/Structures.html

Also add a defstruct alias.

ref: https://www.gnu.org/software/emacs/manual/html_node/elisp/Combining-Conditions.html

The pcase implementation: https://www.gnu.org/software/emacs/manual/html_node/elisp/pcase-Macro.html

And then the destructuring methods: https://www.gnu.org/software/emacs/manual/html_node/elisp/Destructuring-with-pcase-Patterns.html