Posts tagged with modernization
You might’ve noticed that the releases around here have been getting sparser and sparser. We haven’t even had a release since May. Is this the dreaded death spiral?
No, it isn’t; we’ve just entered the cocoon and are preparing to emerge as a more modern and performant butterfly. And you can help us test that butterfly!
I’ve spent a number of posts talking about Tree-sitter as though it makes sliced bread look mundane. But it also drives me nuts on a regular basis. Let’s wrap up the series by talking about what makes Tree-sitter hard to work with — with an optimistic look toward the near future.
In the beginning, Atom appeared. It created an API to make packages, but together with this API, it also allowed authors to use web APIs together with node.js packages, modules (including "nat…
We’ve been telling a series of stories about all the different ways that Tree-sitter can improve the editing experience in Pulsar. Today’s story about symbols-view starts a bit slowly, but it’s got a great ending: the addition of a major new feature to Pulsar 1.113.
One annoying thing that software developers do is insist on writing in more than one language at once. Web developers are espeically obnoxious about this — routinely, for instance, putting CSS inside their HTML, or HTML inside their JavaScript, or CSS inside their HTML inside their JavaScript.
Code editors like Pulsar need to roll with this, so today we’ll talk about how the modern Tree-sitter system handles what we call injections.
Last time we looked at Tree-sitter’s query system and showed how it can be used to make a syntax highlighting engine in Pulsar. But syntax highlighting is simply the most visible of the various tasks that a language package performs.
Today we’ll look at two other systems — indentation hinting and code folding — and I’ll explain how queries can be used to support each one.
Last time I laid out the case for why we chose to embrace TextMate-style scope names, even in newer Tree-sitter grammars. I set a difficult challenge for Pulsar: make it so that a Tree-sitter grammar can do anything a TextMate grammar can do.
Today I’d like to show you the specific problems that we had to solve in order to pull that off.
In the last post, I tried to explain why the new Tree-sitter integration was worth writing about in the first place: because we needed to integrate it into a system defined by TextMate grammars, and we had to solve some challenging problems along the way.
Today I’ll try to illustrate what that system looks like and why it’s important.
The last few releases of Pulsar have been bragging about a feature that arguably isn’t even new: our experimental “modern” Tree-sitter implementation. You might’ve read that phrase a few times now without fully understanding what it means, and an explanation is long overdue.
How did I decide to start working on tree-sitter, and all preparations to modernize it on Pulsar