Tool for searching across branches and through history in a Git repository

gitsearchsoftware-rec

I'm looking for a tool, or for suggestions towards a script, that would be able to search a Git repository for files based on both filenames and file contents (find/grep-like). It would need to be able to search not just in the currently checked out branch, but through that branch's history as well as in other branches.

The very specific example of a Git repository that I have is a checkout of dspinellis' unix-history-repo. I often look for historical implementations of things in there, but it's a pain to track down files as I often need to guess what branch I need to be looking at (there are 165 branches in that repository).

Example of thing I would like to do with this tool is to find the wow command, which may have been an external command or a built-in command in sh or csh at some point, if it existed as part of some early BSD Unix (if it existed at all). To do this, I would want to search for wow.* as a filename pattern, and also for files that may at some point have included a wow C function, across the 165 branches of dspinellis' Git repository.

Best Answer

Heh, guess what I’ve been doing too...

To look for a file name across all branches, I use

git log --all --name-only --pretty=format:%H -- wow\*

wow can be replaced by any glob. This runs quite quickly on the Unix history repository. The format shows the hash leading to the creation of the matching file, so you can then check the tree out at that point and explore further.

To search file contents across all branches, I use

git rev-list --all | xargs git grep "excited too"

which lists all commit objects and searches through them. This is very, very slow on the Unix history repository; listing all the branches and grepping there is quicker:

git grep "excited too" $(git branch -r | awk '{print $1}')

Related Solutions

Bash – How to pass output of one command as an argument to another

First, you need to massage the git branch output into a useable format

$ git branch
  experiment
* master
  new feature

$ git branch | awk '/^\* / { print $2 }'
master

Now, you want to use that as an argument:

$ git pull --rebase origin $(git branch | awk '/^\* / { print $2 }')

(or you can use backticks as in psusi's answer).

This should be ok, the awk command should always match exactly one line, and I'm assuming you can't have spaces in branch names. For anything much more complicated, I'd probably wrap it into a function (or a script) so you can do some sanity checking on the intermediate values.

How to recover broken/partially deleted git repository

All of the commits and the files that they reference would be stored as objects in the objects directory. Git creates those as read-only, so they should all still be present.

To recover, I'd advise creating a new, empty repository and copying the contents of your broken repository's objects directory into that of the new one. That should get you to a point where git will at least recognize that it's a repository and it will have all of your objects. Working with a copy will also help to avoid causing even more damage while trying to fix things.

Shell commands to create the temporary repository and copy over the objects:

git init /tmp/recovery
cd /tmp/recovery
cp -r /path/to/broken/repo/.git/objects .git

Once that is done, you could use git fsck to get a list of objects that aren't referenced by anything. This should include all of the branch heads, but it would also include any commits that were made obsolete by git commit --amend or by rebasing.

Since you still have the logs directory that is likely to be an even bigger help. There should be a logs/refs/heads/<branch> file for each branch that you had. The second column of the last line will contain the ID of the commit which was at the head of that branch when the deletion was done. There should also be logs/HEAD with the same information for where the HEAD was, but unless you'd been working with a detached HEAD it's probably better to just recover the branches and then checkout a branch normally.

For each branch that you want to restore you can run:

git branch <name> <commit_id>

Once you've restored the branches you can copy over the config file, and you should be fairly close to where you were as of the latest commit.

Best Answer

Related Solutions

Bash – How to pass output of one command as an argument to another

How to recover broken/partially deleted git repository

Related Question