Vbench: run Python benchmarks throughout your repository history(wesmckinney.com)
wesmckinney.com
Vbench: run Python benchmarks throughout your repository history
http://wesmckinney.com/blog/?p=373
6 comments
Since 1.5.1 git has had `git-bisect run` which can fully automate the whole thing.
http://lwn.net/Articles/317154/
Only used this a couple of times, but I highly recommend it. It's one of those tools that's so clever you feel a bit like a jedi just using it.
http://lwn.net/Articles/317154/
Only used this a couple of times, but I highly recommend it. It's one of those tools that's so clever you feel a bit like a jedi just using it.
Inspired by that, someone else has made "svn-bisect" http://search.cpan.org/dist/App-SVN-Bisect/bin/svn-bisect Does the same thing, and is a godsend when trying to track down bugs in a svn project.
Also in darcs since 2.5 as an option for the (way older) linear "trackdown" command http://wiki.darcs.net/Using/Trackdown
That's definitely on my todo list.
Very interesting. I'd be interested in potential uses for other languages. I might do some research.
Seems like a really nice addition to this would be to demonstrate a quick integration with git bisect (http://book.git-scm.com/5_finding_issues_-_git_bisect.html). That could help you quickly identify troublesome commits where the speed dropped significantly.