Making git auto-commit

Question

On Linux you could use inotifywait to automatically execute a command every time a file's content is changed.

Edit: the following command commits file.txt as soon as it is saved:

inotifywait -q -m -e CLOSE_WRITE --format="git commit -m 'autocommit on change' %w" file.txt | sh

The earlier inotifywait answer is great, but it isn't quite a complete solution. As written, it is a one shot commit for a one time change in a file. It does not work for the common case where editing a file creates a new inode with the original name. inotifywait -m apparently follows files by inode, not by name. Also, after the file has changed, it is not staged for git commit without git add or git commit -a. Making some adjustments, here is what I am using on Debian to track all changes to my calendar file:

/etc/rc.local:

su -c /home/<username>/bin/gitwait -l <username>

/home/<username>/bin/gitwait:

#!/bin/bash
#
# gitwait - watch file and git commit all changes as they happen
#

while true; do

  inotifywait -qq -e CLOSE_WRITE ~/.calendar/calendar

  cd ~/.calendar; git commit -a -m 'autocommit on change'

done

This could be generalized to wait on a list of files and/or directories, and the corresponding inotifywait processes, and restart each inotifywait as a file is changed.

The previous answers recommending inotifywait for this job tipped me off in the right direction when I had this problem myself, so I wrote a little script. First this could only watch whole folders recursively (the opposite of Lester Buck's example), but then I also wanted to watch a file somewhere else, so I expanded it.

The result is a script currently called gitwatch, as that is what it does: it watches a file or folder for changes (using inotifywait), and commits them to a git repository.

You can find the script, more info and instructions over on github: https://github.com/nevik/gitwatch

git-wip is a great solution that works well for me. "WIP" stands for "work in progress". Every time you run 'git wip', the changes are commited to a separate branch. It can be run on the command line, but there are extensions for vim and emacs to automatically run git-wip each time a file is written.

I wanted to do this in windows, and found the best way was to use Directory Monitor to check for changes then when it detected a change have it run:

Program: cmd.exe

Params: /C C:\pathToBatchFile.bat

That batch file contained:

c:
cd c:\gitRepoDirectory\
(if exist "%PROGRAMFILES(X86)%" (
"%PROGRAMFILES(X86)%\git\bin\sh.exe" --login -i -c "git commit -am AutoCommitMessage"
) else (
"%PROGRAMFILES%\git\bin\sh.exe" --login -i -c "git commit -am AutoCommitMessage"
))

I also tried having another command in there to add files ("%PROGRAMFILES(X86)%\git\bin\sh.exe" --login -i -c "git add *.*"), but I don't think I got that working properly.

I also made a post-commit hook containing:

#!/bin/sh
git.exe pull -v --progress  "origin"
git.exe push    --progress  "origin" master:master
curl.exe -s https://webserverdomain.com/updateFromGitHook.x?r=repoName

(If there were any conflicts then it would abort the pull and abort the push, but there wasn't any clear way to tell that had happened - in the end we abandoned the whole idea because of this one flaw.)

That curl command told my server that it needed to do a pull on the code. All that was needed to handle it in php was:

<?
$r = $_GET['r'];
if (!empty($c)) {
    //use system instead of exec if you want the output to go back to the git client
    exec("cd /path/to/repo/parent/$r; sudo git reset --hard HEAD; sudo git pull;");
    echo "\n\nServer: Updated\n\n";
} else {
    echo "\n\nServer: UPDATE FAILED\n\n";
}
?>

The only problem with that was it needed to be run by the root user instead of the apache user, so I also had to make a file in /etc/sudoers.d/ containing:

www-data ALL = NOPASSWD: /usr/bin/git

For me, I think that worked pretty solidly. Directory Monitor can be configured to run on startup and start minimized, and it can watch several different folders