gracinet added a comment.
@yuja: do you mean one of those Differential Revisions of this system for each commit, sure I can do. With respect to rust-cpython bindings, I'm currently waiting for feedback on https://github.com/dgrunwald/rust-cpython/issues/164 Perhaps you'd have an idea ? Short summary: sporadic segfaults that I can for now reproduce only by running the whole test suite. That being said, I do have rust-cpython bindings for AncestorsIterator and MissingAncestors (plus the full lazyancestors class). Unless I'm mistaken, that means that together with the existing C implementations, there's a potential for a full native version of `mercurial/ancestor.py`. I can send them to the mailing-list or here. What would be the correct flag for that ? RFC ? I'm also working on a perfmissingancestors (started a few hours ago). For now it confirms measurements I'd observed earlier: about x5 performance boost. REPOSITORY rHG Mercurial REVISION DETAIL https://phab.mercurial-scm.org/D5370 To: gracinet, #hg-reviewers Cc: yuja, durin42, kevincox, mjpieters, mercurial-devel _______________________________________________ Mercurial-devel mailing list Mercurial-devel@mercurial-scm.org https://www.mercurial-scm.org/mailman/listinfo/mercurial-devel
