Skip to content

Conversation

savetheclocktower
Copy link

@savetheclocktower savetheclocktower commented Sep 3, 2025

Now that we have a context-aware version of git-utils published, this isn't a high priority, but I'm opening it anyway so it doesn't get forgotten on my hard drive.

Almost a year ago, when I was going through all of our native modules and assessing the context awareness of each one, I rewrote this module using N-API, mainly just to get practice with N-API. It was a straightforward process and it gave me an opportunity to update some other dependencies.

Separately: @mauricioszabo indicated that his original reason for vendorizing libgit2 had something to do with yarn. It occurs to me that he was likely running into problems that were specific to how yarn handled dependencies that were git repositories instead of NPM packages. But now that we have a package-publishing strategy, there should be no need for vendorizing, and we can return to the submodule strategy. (If I'm wrong, we'd have a perfectly good 7.0.1 release to fall back on in PulsarNext.)

Anyway, to reiterate: this isn't a necessary part of the drive toward PulsarNext, but it solves a couple of long-term action items.

@savetheclocktower
Copy link
Author

(Because the diff removes a bunch of vendored files from source control, the meaningful N-API changes can't be seen in the web view. But you can view binding.cc and binding.hh directly; they are total rewrites from the old files (repository.cc and repository.hh).

Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment

Labels

None yet

Projects

None yet

Development

Successfully merging this pull request may close these issues.

1 participant