bug#68899: Treesitter's forward-sexp-function

["João Távora" <joaotavora@gmail.com>]

On Sun, Feb 4, 2024 at 5:35 AM Yuan Fu <casouri@gmail.com> wrote:

> Thanks for looking into this, Joao. IME a very useful characteristic of forward-sexp is that it stays in the same “level” and doesn’t go up automatically when there’s no siblings (when there’s a closing delimiter). Eg, in an Elisp buffer, forward-sexp stops at the closing parenthesis, in C, it should stop at the closing bracket.

I agree.  There are other useful characteristics, but this is one of them.
It allows be to mark regions of text up to points that I'm not even seeing.

> Also you don’t want to check for prev when moving forward, and vice
> versa, ie, we don’t want to check (null next) and (null prev) together.

I get it.  I used those existing results as a proxy to know if we're
in the middle
of a leaf.  I _think_ it's sound (maybe I'm wrong).

> So, how do you like this patch:

It works fine, but as far as I can tell does exactly the same as mine, and
looks to be slightly more difficult to read and uses a further treesit
query to check if this is a leaf node.  But it's absolutley fine really.

One way my patch can be described in plain english is
"if we're not at an inter-thing boundary, we navigate to such boundary"
And then the meaning of checking prev _and_ next becomes obvious
and it isn't necessary to perform the additional check that we're in a
leaf node.

So go ahead and push whichever patch you prefer, and thanks.

João