<div dir="ltr"><div class="gmail_extra">John, I think your question (which I'm not sure you've answered yet), boils down to "What happens in the file system when I checkout a branch"? Am I understanding correctly?</div>
<div class="gmail_extra"><br></div><div class="gmail_extra">I think your set up is to have Apache in your XAMPP setup pointed to some directory in your git working copy. When you issue `git checkout rewind_xxxx`, git updates the files in your working copy so they match the snapshot of the files a the time HEAD of that branch was committed.</div>
<div class="gmail_extra"><br></div><div class="gmail_extra">So when you browse <a href="https://localhost/mysite" target="_blank">https://localhost/mysite</a>, XAMPP goes and fetches the files/runs the scripts as the were at commit X.</div>
<div class="gmail_extra"><br></div><div class="gmail_extra">To know what this means in terms of which features are present and what behaviour to expect, you need to have a good idea of what the commit history. This is why everyone harps about good commit messages.</div>
<div class="gmail_extra"><br></div><div class="gmail_extra">I'm confused by your message from midnight. If you `git checkout branch`, modify indexTest.php, then `git checkout otherbranch', git should warn you that you have uncommitted changes and refuse to change branches.</div>
<div class="gmail_extra"><br></div><div class="gmail_extra">If you commit your fix, the way to apply it to the other branches is to merge or rebase the branch with your fix on to the target branch. There is also the option to cherry-pick just that change, in case fix-branch has other changes that are not desired in target-branch.</div>
<div class="gmail_extra"><br></div><div class="gmail_extra">As for determining the impact of the fix on the target branch, git can only marginally help with that. As you are probably aware, git will tell you if there have been other changes made to the particular chunks of code in the target branch. However, if you change the behaviour of a function, git can't help you figure out how consumers of that function will adapt to the change. This is why good software engineering techniques from software design to automated unit and integration testing are still important.</div>
</div>