It is funny, one of the things that has bothered me for quite a while is that diffs don’t understand code.
For example if I add an additional usage to a global variable to a file, while someone else renames it in other files, git happily merges with no conflict.
This tight integration between language and general purpose tool is a nice idea with a terrible price. It would make tools a lot harder to develop and they would be a lot slower as they now have to perform significantly more work.
That strikes me as outside the scope of a feasible version control tool (and is probably undecidable in most cases). Seems like you’d be much better off handling that with a test suite and CI.
I kind of agree, but if you have a very tight coupling, you could version control objects of the language and not text files. For example, a function could be versioned as a function, not a chunk of text.
For example, I have seen cases with diffing where based on whatever algorithm, a very small logical change, (like moving a function up or down in the file along with a minor edit), creates a very confusing side by side diff based on whatever similar lines it think match up.
It can, but like with other “rich” structures (type systems, data schemas, IDLs...) there is usually no single good solution, and when you need different ones to interoperate (either because you hit the limits of one of them or because you need to bridge two preexisting systems) you need to write a lot of tedious boilerplate at best and are completely stuck at worst. AFAICS that’s why amorphous sequences of bytes won as the fundamental (absence of an) abstraction in many places (files, networks, RAM, etc.).
I would very much like for this not to be true, but there doesn’t seem to be a good solution.
With syntax trees in particular, it’s also frequently difficult to find even a single good option for what you would reasonably want a diff to be: last I checked rich-text diffs usable over arbitrary HTML (even sans CSS) still sucked in half the cases you’d want. Line- and word-based diffs over the source also suck, naturally, but at least they do so uniformly. It’s not an accident Stack Overflow will usually try to show you a rich-text diff when editing but will occasionally fall back to word-diffing the Markdown.
And a version control system needs to interoperate above all. Unlike, say, an IDE, and you do see attempts at semantic and even structured editing in IDEs.
/me continues to fantasize about interbreeding Paredit and Pijul one day
Git allows you to use a custom diff tool for a specific file format. The problem is that these centralized source hosting/review tools only let you do about half of what Git can do.
For example if I add an additional usage to a global variable to a file, while someone else renames it in other files, git happily merges with no conflict.
Tighter language integration can help with that.