Re: State of the overlay tree branch?

[Sebastien Chapuis <sebastien@chapu.is>]

Eli Zaretskii writes:

>> From: Sebastian Sturm <s.sturm@arkona-technologies.de>
>> Date: Sun, 18 Mar 2018 21:14:53 +0100
>>
>> [1] I'm using cquery for my C++ editing needs, which comes with an
>> overlay-based semantic highlighting mechanism. With my emacs
>> configuration, lsp-mode/lsp-ui emit 6 calls to line-number-at-pos per
>> character insertion, which consume ~20 to 25 ms each when performing
>> edits close to the bottom of a 66KB C++ file (measured using
>> (benchmark-run 1000 (line-number-at-pos (point))) on a release build of
>> emacs-27/git commit #9942734...). Using the noverlay branch, this figure
>> drops to ~160us per call.
>
> If lsp-mode/lsp-ui needs a fast line counter, one can easily be
> provided by exposing find_newline to Lisp.  IME, it's lightning-fast,
> and should run circles around count-lines (used by line-number-at-pos).
>
> (I'm not sure I even understand how overlays come into play here,
> btw.)

The language server protocol defines a position in file with zero-indexed
line and column offsets [1]:

```
interface Position {
	line: number;
	character: number;
}
```

lsp-mode uses heavily line-number-at-pos to convert an Emacs buffer
point to a LSP position, and vice-versa [2].
This can happen thousands times on each keystroke.

If Emacs could provide a function to do the conversion very fast (or at
least faster than with line-number-at-pos), it would be great.


[1] https://github.com/Microsoft/language-server-protocol/blob/gh-pages/specification.md#position
[2] The conversion from a LSP position to point doesn't use
line-number-at-pos, but forward-line and forward-char. It's still very slow.

--
Sebastien Chapuis