Tree-sitter became more widespread and Emacs took notice and included a bunch of -ts-mode as alternatives to -mode into the core. This is good news and a welcome change, but I have some concerns about the approach.
When I first saw the Tree-sitter talk by Max Brunsfeld I was concerned that the language highlighting “fix” they’re talking about is too much.
The problem is monochromacity. The font lock rules in a lot of the TS modes are… not always thought through. Ramp up python-ts-mode’s font lock level to the maximum and most of everything is dyed the same colour. It’s frustrating and I have had to muck around with treesit-font-lock-feature-list to selectively unpick the chaff from the grist. And it’s not exactly easy changing it either: if you’re new to elisp, or not fluent in the extremely awkward interface for changing this stuff, you’ll have to look up how someone else has done it.
It’s not even customizable, so you can’t just click around and change it either.
The font lock rules in a lot of the TS modes are… not always thought through. Ramp up python-ts-mode’s font lock level to the maximum and most of everything is dyed the same colour.
And then on the other side, people come asking why even when TS is used, highlighting is not “up to par” with VS Code and etc. Then hear about treesit-font-lock-level, ramp it up to the max, and apparently live happily after.
That’s a misinterpretation of my argument. I said monochromacity: that at the highest level, many disparate node types are font locked with the same face. Try it in Python. Everything is the same shade of whatever your variable name face is.
So your complaint is about the non-default behavior? The level 3 was chosen as default explicitly to avoid the abundance of color where it’s not really needed.
I said monochromacity: that at the highest level, many disparate node types are font locked with the same face. Try it in Python. Everything is the same shade of whatever your variable name face is.
I don’t see that “everything is the same shade”, even with level 4. There are problems - like the variable matches in particular (which is not useful IMO, but should at least use a different face). The rest use different faces, but since most of those faces are new, it’s up to the theme authors to differentiate them.
Here’s a patch fixing the one problem I found (try it), and below is a screenshot with this patch applied along with the new faces customized to be distinct.
So your complaint is about the non-default behavior? The level 3 was chosen as default explicitly to avoid the abundance of color where it’s not really needed.
That’s not what I wrote.
I don’t see that “everything is the same shade”, even with level 4. There are problems - like the variable matcher in particular (which is not useful IMO, but should at least use a different face).
The variable matcher, indeed, is the one I was talking about…
The problem is monochromacity. The font lock rules in a lot of the TS modes are… not always thought through. Ramp up python-ts-mode’s font lock level to the maximum and most of everything is dyed the same colour. It’s frustrating and I have had to muck around with
treesit-font-lock-feature-listto selectively unpick the chaff from the grist. And it’s not exactly easy changing it either: if you’re new to elisp, or not fluent in the extremely awkward interface for changing this stuff, you’ll have to look up how someone else has done it.It’s not even customizable, so you can’t just click around and change it either.
And then on the other side, people come asking why even when TS is used, highlighting is not “up to par” with VS Code and etc. Then hear about
treesit-font-lock-level, ramp it up to the max, and apparently live happily after.That’s a misinterpretation of my argument. I said monochromacity: that at the highest level, many disparate node types are font locked with the same face. Try it in Python. Everything is the same shade of whatever your variable name face is.
So your complaint is about the non-default behavior? The level 3 was chosen as default explicitly to avoid the abundance of color where it’s not really needed.
I don’t see that “everything is the same shade”, even with level 4. There are problems - like the
variablematches in particular (which is not useful IMO, but should at least use a different face). The rest use different faces, but since most of those faces are new, it’s up to the theme authors to differentiate them.Here’s a patch fixing the one problem I found (try it), and below is a screenshot with this patch applied along with the new faces customized to be distinct.
diff --git a/lisp/progmodes/python.el b/lisp/progmodes/python.el index d3cb5a77e22..9f9344e0eb4 100644 --- a/lisp/progmodes/python.el +++ b/lisp/progmodes/python.el @@ -1235,7 +1235,7 @@ python--treesit-fontify-variable (when (python--treesit-variable-p node) (treesit-fontify-with-override (treesit-node-start node) (treesit-node-end node) - 'font-lock-variable-name-face override start end))) + 'font-lock-variable-use-face override start end))) ;;; Indentationhttps://imgur.com/a/iifAmcd
That’s not what I wrote.
The variable matcher, indeed, is the one I was talking about…
I’m glad we’re on the same page.
You’re welcome.