git stash -> merge stashed change with current changes

tl;dr

Run git add first.

I just discovered that if your uncommitted changes are added to the index (i.e. "staged", using git add ...), then git stash apply (and, presumably, git stash pop) will actually do a proper merge. If there are no conflicts, you're golden. If not, resolve them as usual with git mergetool, or manually with an editor.

To be clear, this is the process I'm talking about:

mkdir test-repo && cd test-repo && git init
echo test > test.txt
git add test.txt && git commit -m "Initial version"

# here's the interesting part:

# make a local change and stash it:
echo test2 > test.txt
git stash

# make a different local change:
echo test3 > test.txt

# try to apply the previous changes:
git stash apply
# git complains "Cannot apply to a dirty working tree, please stage your changes"

# add "test3" changes to the index, then re-try the stash:
git add test.txt
git stash apply
# git says: "Auto-merging test.txt"
# git says: "CONFLICT (content): Merge conflict in test.txt"

... which is probably what you're looking for.

Running git stash pop or git stash apply is essentially a merge. You shouldn't have needed to commit your current changes unless the files changed in the stash are also changed in the working copy, in which case you would've seen this error message:

error: Your local changes to the following files would be overwritten by merge:
       file.txt
Please, commit your changes or stash them before you can merge.
Aborting

In that case, you can't apply the stash to your current changes in one step. You can commit the changes, apply the stash, commit again, and squash those two commits using git rebase if you really don't want two commits, but that may be more trouble that it's worth.

What I want is a way to merge my stashed changes with the current changes

Here is another option to do it:

git stash show -p|git apply
git stash drop

git stash show -p will show the patch of last saved stash. git apply will apply it. After the merge is done, merged stash can be dropped with git stash drop.

The way I do this is to git add this first then git stash apply <stash code>. It's the most simple way.

As suggested by @Brandan, here's what I needed to do to get around

error: Your local changes to the following files would be overwritten by merge:
       file.txt
Please, commit your changes or stash them before you can merge.
Aborting

Follow this process:

git status  # local changes to `file`
git stash list  # further changes to `file` we want to merge
git commit -m "WIP" file
git stash pop
git commit -m "WIP2" file
git rebase -i HEAD^^  # I always use interactive rebase -- I'm sure you could do this in a single command with the simplicity of this process -- basically squash HEAD into HEAD^
# mark the second commit to squash into the first using your EDITOR
git reset HEAD^

And you'll be left with fully merged local changes to file, ready to do further work/cleanup or make a single good commit. Or, if you know the merged contents of file will be correct, you could write a fitting message and skip git reset HEAD^.

May be, it is not the very worst idea to merge (via difftool) from ... yes ... a branch!

> current_branch=$(git status | head -n1 | cut -d' ' -f3)
> stash_branch="$current_branch-stash-$(date +%yy%mm%dd-%Hh%M)"
> git stash branch $stash_branch
> git checkout $current_branch
> git difftool $stash_branch

you can easily

Commit your current changes Unstash your stash and resolve conflicts Commit changes from stash Soft reset to commit you are comming from (last correct commit)

Another option is to do another "git stash" of the local uncommitted changes, then combine the two git stashes. Unfortunately git seems to not have a way to easily combine two stashes. So one option is to create two .diff files and apply them both--at lest its not an extra commit and doesn't involve a ten step process :|

how to: https://stackoverflow.com/a/9658688/32453