For Part 2 of our 60-second Git tutorial, we’ll look at some super simple branching and merging.

git branch tells you which branch you’re on.

git branch -a lists all branches, even remote ones.

git checkout mybranch checks out the branch you want.

git checkout -b newbranch creates a new branch from the one you’re on, and checks it out for you.

git checkout master && git merge mybranch merges changes from one branch to your master branch. If you don’t yet want to commit the changes, you can do git merge --no-commit mybranch instead.

So far, we’ve only talked about local git usage. Isn’t it cool that you can do so much without even having a server set up with our repo? I think so.

Part 3 will include pulling and pushing your changes to remote repos.

Update: Yeah, ok, so I never got around to writing part 3. Back when this was written, super simple Git how-tos were less plentiful. Now, we have git-scm.com, Why Git is Better than SVN, and of course GitHub’s help site.

First, the mental hurdles:You don’t need a server to have a git repository. All your commit history is available locally and quickly, with simple commands. Same with branches and tags. Git “checkout” and “commit” are commands you run locally. Svn “checkout” and “commit” are really more like “pull” and “push” with git.

Now, the commands:

rails myapp && cd myapp && git init && git add . && git commit -a creates a repo and makes the first commit (but still nothing is sent to any server; all work is done locally in your own repository)

echo "my app rocks" > README && git add README && git commit -a makes another commit

git rm path/to/remove && git commit -a

git reset --hard HEAD returns your repo to last committed state (warning: cancels working changes) I’m still fighting with this one. It seems the best way might be to do a git checkout -f instead.

That’s it for the first part. Part 2 will be next.

OneBody now has a real, bona fide blog of its own. Thanks to WordPress.com, creating a new blog is waaaay too easy, evidenced by the fact that I now have 37 (just kidding).

Our little project continues to grow and improve. I’m still waiting on that influx of talented, eager volunteers to make it awesome, though :-) For now, it will continue to be Tim’s pet project, though I refuse to stop referring to the development “team” as “we” and “us.”

Just saw this in my referrer logs… You can search on Google for all occurrences except in a certain site.

Many know about adding site:timmorgan.org or similar to only search within a certain domain. Well, maybe it was obvious, but now I know you can do -site:timmorgan.org to search everywhere but that domain. Nice.

Long title, I know. Here’s how I did it…

1
2
3

sudo apt-get install fetchmail postfix
cd ~
vim .fetchmailrc

Add the following to the file (replacing everything in CAPS appropriately):

1
2
3

poll pop.gmail.com with proto POP3 and options no dns
  user 'YOUR_GMAIL_ADDRESS' there with password 'YOUR_PASSWORD' is 'YOUR_UBUNTU_USERNAME' here
    options ssl

Now:

1
2

chmod 710 .fetchmailrc
sudo vim /etc/crontab

Add the following to your crontab file:

1

10 7 * * 1 YOUR_UBUNTU_USERNAME fetchmail -k

This will run fetchmail every Sunday and grab your email. It seems either fetchmail or gmail limits each connect to something around 400-500 messages, so to get started, you can run fetchmail -k a few dozen times to download all your mail, then let crontab do the rest on a weekly basis.

Your mail is downloaded to a single mbox file in /var/spool/mail/YOUR_UBUNTU_USERNAME. Being a single file makes it nice for backing up, moving, etc. I hear Thunderbird can easily read the file, and I think Apple’s Mail.app can as well.

Just makes you feel good having all your mail backed up, now doesn’t it?