Fetching latest commit…
Cannot retrieve the latest commit at this time.
|Failed to load latest commit information.|
To use eg, simply put this directory in your path. If you are familiar with other version control systems, the syntax in eg is pretty similar ('eg commit', 'eg diff', 'eg info', etc.). To help get you started, you can run eg help which will provide a list of available subcommands. You can also get help on each subcommand; for example, running eg help commit will explain how to commit and provide many examples of doing so. To Msysgit users [From Dan Fabulich]: 1) Msysgit users aren't supplied an ordinary git.exe executable on the command-line, but a script called "git.cmd". Out of the box, "eg" can't find this script. It's easy to update eg 0.99 to support a non-standard git executable (for example: my $git_cmd = "git.cmd"), but that'll get you in trouble... 2) ... because Perl's backtick (`) operator won't correctly detect the return code of git in that case; it'll think that git.cmd always returns 0. This confuses eg's revision-detector code, etc. The only workaround I've found is to make a copy of git.cmd and change its behavior. git.cmd terminates the script using "@exit /b %ErrorLevel%"; I've made a copy called "gitx.cmd" that just uses "@exit %ErrorLevel%" and I refer to "gitx.cmd" in eg; that works perfectly. 3) eg assumes that you can set environment variables on the backtick command-line like this: `GIT_PAGER_IN_USE=1 git log -1` ... that works fine on UNIX but not on Windows. I had to go through and comment out references to GIT_PAGER_IN_USE to get "eg log" to work. This disables color, but color is in a wonky state on msysgit anyway, so I scarcely miss it. I think there's another clever use of inline environment variables in "eg revert"; it attempts to temporarily set GIT_INDEX_FILE when reverting just unstaged changes. I've never needed to do that, but I'm sure I will someday :-p I assume these inline variables are being used to make them temporarily apply only to the child process? Would it be harmful to just set them as %ENV variables instead? (We should soon remove #3 from this README; I agree with Dan that we should just modify %ENV instead.) Note: bash-completion-eg.sh is NOT needed to run eg; it's here just for the people that have bash completion set up and want to use it with eg (google for bash_completion if you want to know more or how to make use of bash-completion-eg.sh). All you need to run eg is to make sure the 'eg' script is in your path.