Greg Ercolano wrote: > On 12/29/11 01:22, Torsten Giebl wrote: >> GIT is great, no question about it, >> but i hate one thing about it and some other SCMs, it is the hash IDs. >>Are you using "6ff87c4664981e4397625791c8ea3bbb5f2279a3" ? >> I like the numbering of SVN. >> >> 1000 >> 1001 >> 1002 >> . >> . >> It is easier to remeber when you are filing a bug report, it is easier when >> you have to backport fixes. >> >> We need to backport the fixes from "4020-4025". >> Nice, easy to read, try that with hashes :-( > > Indeed, and here's an example of how those hashes find their way > into /public/ readable pages, non-programmers just trying to download > the latest binary builds of git snapshots: > > http://ffmpeg.zeranoe.com/builds/ > > Quoting that page: > > 32-bit Builds (Static) > FFmpeg git-5f0105b 32-bit Static (Latest) (2012-10-26) > FFmpeg git-04bf2e7 32-bit Static (2012-10-20) > FFmpeg git-1a104bf 32-bit Static (2012-10-10) > FFmpeg git-f3f35f7 32-bit Static (2012-10-09) > ^^^^^^^ > > 64-bit Builds (Static) > FFmpeg git-5f0105b 64-bit Static (Latest) (2012-10-26) > FFmpeg git-04bf2e7 64-bit Static (2012-10-20) > FFmpeg git-1a104bf 64-bit Static (2012-10-10) > FFmpeg git-f3f35f7 64-bit Static (2012-10-09) > ^^^^^^^ > > They're not even sequential, so you can't use them to determine > if one is newer or older than the other. > > Apparently this lack of sequential numbering has to do with the distributed, > decentralized nature of git's design, as there is no server to manage > the sequential number. > > Apparently there are hacks to do this though; not sure if they're easy to > administer and use in practice, though: > > http://stackoverflow.com/questions/4120001/what-is-the-git-equivalent-for- revision-number > http://stackoverflow.com/questions/677436/how-to-get-the-git-commit-count > http://gitready.com/beginner/2009/02/03/tagging.html
When I started using GIT on a serious project I was frustrated about not having something similar to SVN ID. Then, once people started cloning my repository, and working on their own forks I realised that having something like SVN ID is literally not possible. What would it mean - the ID of a commit in MY repository, or in someone else's repository? git-svn is great, I got entire FLTK repo with all branches and histories in my GIT FLTK repository here, and I automatically merge new code from FLTK (svn) into FLTK (git) "master". -- Dejan Lekic - http://dejan.lekic.org _______________________________________________ fltk-dev mailing list fltk-dev@easysw.com http://lists.easysw.com/mailman/listinfo/fltk-dev
