Felipe Contreras comments

Results 83 comments of


                                            Felipe Contreras

cloning pypy is some orders of magnitude slower than cloning directly with hg

@jedbrown I analyzed all this before, there's no way to improve performance because of mercurial's stupid design. In both git and bazaar a file with the same contents has the...

cloning pypy is some orders of magnitude slower than cloning directly with hg

Walking the chain is not what takes most time, and even then, it's per mercurial design; mercurial uses deltas, so that's what you have to do.

cloning pypy is some orders of magnitude slower than cloning directly with hg

FWIW. remote-hg is 45% faster while cloning pypy's repository.

Attempting to clone pypy repo failed for me.

Works fine in remote-hg: https://github.com/felipec/git/commit/a160978911801114c781efb15973c26e0b8b9822

Attempting to clone pypy repo failed for me.

This was pointed out by Junio http://article.gmane.org/gmane.comp.version-control.git/220034. Which shows why patch review by git developers is valuable.

Git pull failes, ERROR FOUND -> FIXED/WORK AROUND

I think the bug is in mercurial, this should work: `git clone gitifyhg::ssh//repoUserSSH/` (with a slash at the end).

Fix multipart parsing

@mikel Hello? This is an obviously correct fix.

"fixup" ignores resolved resolution

Can you try the following patch? ```diff --- a/git-reintegrate +++ b/git-reintegrate @@ -475,7 +475,14 @@ def cmd_fixup(message, fixup_commit, *args) system(*%w[git cherry-pick --no-commit], fixup_commit) && system({ 'EDITOR' => ':' }, *%w[git...

Usability enhancements

I added the reflog messages. Not sure about adding the highlights. All the git commands have similar output.