Wednesday, May 30, 2012

Run multiple python versions on your system

I'm a software development consultant.  I write python (as well as other languages) code for many clients and I don't get to dictate what their environment looks like.  I've got clients running python as old as 2.4 while others are on the bleeding edge.  Additionally, each client may have their own packages installed as well as differing lists of third party packages.

This post is a description of how I went about getting multiple versions of python installed in my ubuntu development machine and how I go about managing different package sets for different clients.

Get multiple pythons installed

Ubuntu typically only supports one python 2.x version and one 3.x version at a time.  There's a popular ppa (personal package archive) called deadsnakes that contains older versions of python.  You can find it:

To install it (per the instructions in the link above), you do the following:

steve@ubuntu64 ~ $ sudo add-apt-repository ppa:fkrull/deadsnakes

Then you need to update your cache:

steve@ubuntu64 ~ $ sudo apt-get update

Finally, simply install the other versions (I'm running on ubuntu 12.04 LTS, so I have python 2.7 already):

steve@ubuntu64 ~ $ sudo apt-get install python2.4 python2.5 python2.6

If you're following along, we now have python versions 2.4 through 2.7 installed the computer.  If you run 'python', you'll see the default version is still 2.7.

steve@ubuntu64 ~ $ python
Python 2.7.3 (default, Apr 20 2012, 22:39:59) 
[GCC 4.6.3] on linux2
Type "help", "copyright", "credits" or "license" for more information.

Here's why:  Each python version is stored in /usr/bin as python2.X where X is the version.  There is a symbolic link named python that points to the version you want to be the default.  Instead of typing python from the bash prompt, you could just as easily type python2.5:

steve@ubuntu64 ~ $ ls -l /usr/bin/python*
lrwxrwxrwx 1 root root       9 Apr 17 13:20 /usr/bin/python -> python2.7
lrwxrwxrwx 1 root root       9 Apr 17 13:20 /usr/bin/python2 -> python2.7
-rwxr-xr-x 1 root root 1216520 May 21 12:13 /usr/bin/python2.4
-rwxr-xr-x 1 root root 1403624 May  3 00:17 /usr/bin/python2.5
-rwxr-xr-x 1 root root 2652056 May 12 08:43 /usr/bin/python2.6
-rwxr-xr-x 1 root root 2993560 Apr 20 19:37 /usr/bin/python2.7

steve@ubuntu64 ~ $ python2.5
Python 2.5.6 (r256:88840, May  3 2012, 04:16:14) 
[GCC 4.6.3] on linux2
Type "help", "copyright", "credits" or "license" for more information.

This is a point of interest.  I would not mess with the symbolic link.  Ubuntu runs python for many internal maintenance scripts and those scripts are expecting the python version that shipped with ubuntu.

Use virtualenv to manage your python installations and package sets

So now that you have multiple versions of python on your system, how to manage them?  How do you keep packages installed for one version separate from packages installed for another.  What if you want to run one version of django for client X and a different version for client Y?

That's where virtualenv comes in.  If your a ruby programmer, this is analogous to rvm.  Virtualenv lets you manage the python versions and package installations separately for different projects or clients.

Installing virtualenv is simple

As always, you're just a single apt-get command away from having virtualenv ready to go:

steve@ubuntu64 ~ $ sudo apt-get install python-virtualenv

That it.  Virtualenv is ready to go now.

Quick example

Say your starting a new project for a client.  They are running python2.5 and want to use the mocks, nose and coverage packages for testing.  Here a walkthrough of how to use virtualenv to manage the project.

First, let's create a directory for the project:

steve@ubuntu64 ~ $ mkdir -p ~/dev/project1
steve@ubuntu64 ~ $ cd ~/dev/project1

Next, run virtualenv to create the environment for the project:

steve@ubuntu64 ~/dev/project1 $ virtualenv -p /usr/bin/python2.5 .env
Running virtualenv with interpreter /usr/bin/python2.5
New python executable in .env/bin/python2.5
Also creating executable in .env/bin/python
Installing distribute.............................................................................................................................................................................................done.
Installing pip...............done.
steve@ubuntu64 ~/dev/project1 $ 

This command tells virtualenv to create a .env directory and to place a copy of the 2.5 version of python in it.  This copy of the 2.5 python is brand-spankin' new.  It doesn't have any packages (beyond the standard library) installed.  You will need to install them yourself.  Any packages that you install in this instance of python will not be available to main python installation or other virtualenv instances.

Before your can use this new copy, you need to activate it:

steve@ubuntu64 ~/dev/project1 $ source .env/bin/activate
(.env)steve@ubuntu64 ~/dev/project1 $

The activate script manipulates your path environment variable, placing the new python instance first in your path.  This makes is so that when you run python, it will use the version from your instance:

(.env)steve@ubuntu64 ~/dev/project1 $ which python
(.env)steve@ubuntu64 ~/dev/project1 $ python
Python 2.5.6 (r256:88840, May  3 2012, 04:16:14) 
[GCC 4.6.3] on linux2
Type "help", "copyright", "credits" or "license" for more information.

Also, notice that your prompt starts with (.env).  This tells you that you're running with the virtualenv instance activated.  To install packages in your instance, use the pip command:

(.env)steve@ubuntu64 ~/dev/project1 $ pip install mock nose coverage
Downloading/unpacking mock
  Downloading mock-0.8.0.tar.gz (749Kb): 749Kb downloaded
  Running egg_info for package mock
    warning: no files found matching '*.png' under directory 'docs'
    warning: no files found matching '*.css' under directory 'docs'
    warning: no files found matching '*.html' under directory 'docs'
    warning: no files found matching '*.js' under directory 'docs'
Downloading/unpacking nose
  Downloading nose-1.1.2.tar.gz (729Kb): 729Kb downloaded
  In the tar file /tmp/pip-dH_WYa-unpack/nose-1.1.2.tar.gz the member nose-1.1.2/doc/doc_tests/test_selector_plugin/support/tests/mymodule/my_function$py.class is invalid: 'filename None not found'
  In the tar file /tmp/pip-dH_WYa-unpack/nose-1.1.2.tar.gz the member nose-1.1.2/doc/doc_tests/test_restricted_plugin_options/restricted_plugin_options.rst.py3.patch is invalid: 'filename None not found'
  Running egg_info for package nose
Downloading/unpacking coverage  Downloading coverage-3.5.2.tar.gz (115Kb): 115Kb downloaded
  Running egg_info for package coverage
    no previously-included directories found matching 'test'Installing collected packages: mock, nose, coverage
  Running install for mock
    warning: no files found matching '*.png' under directory 'docs'
    warning: no files found matching '*.css' under directory 'docs'
    warning: no files found matching '*.html' under directory 'docs'
    warning: no files found matching '*.js' under directory 'docs'
  Running install for nose
    Installing nosetests script to /home/steve/dev/project1/.env/bin
    Installing nosetests-2.5 script to /home/steve/dev/project1/.env/bin
  Running install for coverage
    building 'coverage.tracer' extension
    gcc -pthread -fno-strict-aliasing -DNDEBUG -g -fwrapv -O2 -Wall -Wstrict-prototypes -fPIC -I/usr/include/python2.5 -c coverage/tracer.c -o build/temp.linux-x86_64-2.5/coverage/tracer.o
    coverage/tracer.c:3:20: fatal error: Python.h: No such file or directory
    compilation terminated.
    ** Couldn't install with extension module, trying without it...
    ** SystemExit: error: command 'gcc' failed with exit status 1
    no previously-included directories found matching 'test'
    Installing coverage script to /home/steve/dev/project1/.env/bin
Successfully installed mock nose coverage
Cleaning up...
(.env)steve@ubuntu64 ~/dev/project1 $

To see that the packages have been installed, simply use them:

(.env)steve@ubuntu64 ~/dev/project1 $ python
Python 2.5.6 (r256:88840, May  3 2012, 04:16:14) 
[GCC 4.6.3] on linux2
Type "help", "copyright", "credits" or "license" for more information.
>>> import mock
>>> import nose
>>> import coverage

When you're done working on the project, deactivate it.  You can always come back later and activate it again.

(.env)steve@ubuntu64 ~/dev/project1 $ deactivate
steve@ubuntu64 ~/dev/project1 $

Notice that when you deactivate the environment, your packages are no longer available:

steve@ubuntu64 ~/dev/project1 $ python
Python 2.7.3 (default, Apr 20 2012, 22:39:59) 
[GCC 4.6.3] on linux2
Type "help", "copyright", "credits" or "license" for more information.
>>> import mock
Traceback (most recent call last):
  File "<stdin>", line 1, in <module>
ImportError: No module named mock
>>> import nose
Traceback (most recent call last):
  File "<stdin>", line 1, in <module>
ImportError: No module named nose
>>> import coverage
Traceback (most recent call last):
  File "<stdin>", line 1, in <module>
ImportError: No module named coverage

You can create as many virtualenv environments as you like.  I create one for each project that I work on.

Bonus material

To make life even easier, here's a couple additional things I do that you might find helpful!

First, once you've started to use virtualenv with some frequency, you start to get tired of downloading and installing the same packages over and over.  Pip has the ability to cache your downloaded packages for reuse.  To do that, you'll need to create a directory to store the download packages in:

steve@ubuntu64 ~ $ mkdir ~/.pip_download_cache

Then you'll need to set a variable to inform pip of the new directory.  Add the following to your .bashrc file:

export PIP_DOWNLOAD_CACHE=/home/steve/.pip_download_cache

Now when you do a pip install, it will keep the downloaded files in the ~/.pip_download_cache directory.  The next time you do a pip install of the same package, it will just use the copy from the directory instead of downloading it again.

Second, it can be tedious to always have to type 'source .env/bin/activate' every time you want to activate an environment.  Since I always put my virtual environments in a .env directory I can count on the command to activate always being the same.  So I create an alias for it.  I added the the following to my ~/.bash_aliases file:

alias activate='source .env/bin/activate'

Now once I cd into the projects directory, I simply type activate to activate my virtual environment.

Friday, May 25, 2012

File encryption with vim

You may not know it, but you can encrypt your files using vim... and it's pretty easy to do.

Turning encryption on for a file

If you're editing a file, to encrypt it, simply enter X at the : prompt.

Vim will then prompt you for a key.  A key is the password for decrypting the file in the future.  You'll have to enter it twice to ensure you typed it correctly.

Once you've entered your key, the next time you save/write, the file should be encrypted, see the status text at the bottom to be sure.

Once the file is encrypted, it will be unreadable to others.
Vim auto-detects whether files are encrypted.  If you try to open an encrypted file, vim will prompt you for the key.
Vim uses whatever you enter for the key to decrypt the file.  If you enter the correct things will look great.  If you enter an invalid key, vim will present you with garbage to edit.
The encryption will stay on for the file until you tell vim you'd like to remove it.

Be mindful of your key.  If you forget it, you will not be able to retrieve the contents of your file.

Turning off encryption for a file

Turning encryption off is equally as easy.  Simply set the key setting to nothing.  With the command:

:set key=

Be sure to write the file.  Note that it is no longer encrypted (status text)

Wait!  Before you go

As of vim 7.3, vim uses one of two encryption methods: zip and blowfish.  Zip is the same encryption that PkZip uses and is somewhat weak (can be cracked).  Conversely, blowfish is more contemporary and is much harder to crack.  Zip, unfortunately, is the default encryption method in vim.  I would strongly suggest that you use blowfish when encrypting your files.  To do this is simple enough.  Simply enter the command:

:set cm=blowfish

I put it in my .vimrc file so it's always set.

Thursday, May 24, 2012

Some light string manipulation in bash

This post is more for me than you.  I had to do a little bash work this morning and thought I'd keep a record of some of the syntax and concepts that I used in case I have to do more in the future.

There are two files here: a function library and a script that consumes it.  The function library is used to do different kinds of string justification (left justify, right justify and centering).  Kind of like 'echo' on steroids.  This could have easily been done in python, perl or ruby, but I wanted it to stay in bash this time to keep the calling scripts uniform.

Here's the sample file that consumes the library:


. echo_helper

x='Steve is here!'
left_justify "$x" 40 .
right_justify "$x" 40 .
center_justify "$x" 40 .

echo $(pad_string 40 -)

x='Steve was here.'
left_justify "$x" 40 .
right_justify "$x" 40 .
echo $(center_justify "$x" 40 .)

And here's what the output should look like:

Steve is here!..........................
..........................Steve is here!
.............Steve is here!.............
Steve was here..........................
.........................Steve was here.
............Steve was here..............

Finally, here's the echo_helper library code:


# Syntaxes to remember:
# String length of a variable:  ${#varname}
# String offsetting of a variable: ${varname:offset} or ${varname:offset:length}

# Usage: pad_string LENGTH {PADDING_CHARACTER=' '}
function pad_string {
local line=''
local length=$1
local padding_character=' '
if [[ "$2" != "" ]]; then
while((${#line} < $length)); do
echo -n "$line"

# Usage: left_justify MESSAGE LENGTH {PADDING_CHARACTER=' '}
function left_justify {
local message=$1
local length=$2
local padding_character=' '
if [[ "$3" != "" ]]; then
echo "$message$(pad_string $length-${#message} $padding_character)"

# Usage: right_justify MESSAGE LENGTH {PADDING_CHARACTER=' '}
function right_justify {
local message=$1
local length=$2
local padding_character=' '
if [[ "$3" != "" ]]; then
echo "$(pad_string $length-${#message} $padding_character)$message"

# Usage: center_justify MESSAGE LENGTH {PADDING_CHARACTER=' '}
function center_justify {
local message=$1
local length=$2
local padding_character=' '
if [[ "$3" != "" ]]; then
local half_length=$(expr $length / 2)
local half_message_length=$(expr ${#message} / 2)
local padding_length=$(expr $half_length - $half_message_length)
padding="$(pad_string $padding_length $padding_character)"
if [[ ${#message} > $length ]]; then # handle even/odd length issue
echo "${message:1}"
echo "$message"

Wednesday, May 23, 2012

QuickLook from vim

Here's another nifty trick I came up with.  I'm probably not the first to think of it but I'll put it out there anyway.

Quicklook is a neat feature of OSX (versions since Leopard) that allows you to preview the contents of a file without having to open it.  The way I've typically used it is to select a file in finder and hit the spacebar.  OSX then opens a nicely formated window with the contents of my selected file displayed.

I keep a lot of my documentation, notes, etc in markdown format.  Back in January, I attended CodeMash in Sandusky, Ohio and someone turned me onto a markdown rendering Quicklook plugin.  It's a great little tool.  I can keep writing my docs and notes in markdown, but when I Quicklook them, they are rendered for display nicely.  If you have interest in this free plugin, you can download it here.

Anyway, I settled into this habit of editing a file in vim, saving it, then pressing spacebar on it from finder to preview it.  It got me thinking, wouldn't it be great if I could just open Quicklook directly from vim?

After some quick googling around, I found a command-line tool in OSX that does just that.  It's called qlmanage.  From there it was easy enough to come up with a vim key mapping.

Put this (all one line) in your .vimrc file and you should be able to do Quicklooks of your files right from vim simply by hitting \v:

map <Leader>v :write<cr>:sil !/usr/bin/qlmanage -p % > /dev/null &<cr>:redraw!<cr>

It works from both MacVim as well as the terminal version.  You don't have to use it solely with markdown files.  In fact you can use it with any file that Quicklook understands (html for example).

Tuesday, May 22, 2012

Fun with gnu screen

In my day job, I spend a fair amount of time working on numerous tasks on the command-line in Linux and OSX.  Way back in my college days, I used a program called gnu screen (aka 'screen') but recently I've "refound" it.  I have a new appreciation for screen and now use it every day.

What is screen?

Screen let's you have multiple shell sessions running simultaneously in the same terminal window.  Here's some of the features that I personally find appealing:
  • With a little muscle memory, you can switch rapidly between the sessions.
  • It has the ability to split up your terminal window into multiple "window-splits" so you can see two or more shells sessions at the same time.  The splits are even resizable.
  • You can 'detach' and 'reattach' to a screen session, allowing you to sign-out, leave and come back later, all the while keeping your terminal sessions active and running.
  • It's configurable and scriptable.
  • You can copy and paste between sessions with 'vi' like keystrokes.

How do you get screen?

Well, if you're running ubuntu like me, it's pretty darn easy.  You're one command away from having it:

steve@ubuntu64 ~ $ sudo apt-get install screen

Pretty simple.

Getting started

Once you have screen installed, simply run it.

steve@ubuntu64 ~ $ screen

It will bring you to a welcome screen (I'll show you how to disable the welcome screen later).

At the bottom of the welcome screen, it will say

[Press Space or Return to end.]

Go ahead and do that.  You'll be returned to a bash prompt.  It's important to note that screen hasn't terminated, but rather it's spawned another shell prompt within itself and you're sitting at that new prompt. You can now begin using screen via keyboard command sequences.  To prove it, press Ctrl-a and then press " (double quote).  You should get a screen that looks like this:

This is the terminal session list.  If you have more than one terminal session running, you'd seem them all listed here.  Let's create another terminal session.  Press Ctrl-a and then press c.  This should take you back to a bash prompt.  Your now in a new bash prompt.  See see the updated list, press Ctrl-a and " again.

Now there's two bash sessions.  You can select the one you'd like to see by using your arrow keys (or vi cursor movement keys) to highlight the one you want and then press enter/return.

When you want to quit out, simply exit out of all your bash prompts.  Once the last bash session ends, screen will automatically terminate, leaving where you were when you initially started it.

My most used key sequences

As you've probably realized by now, screen uses Ctrl-a plus a key to drive it's functionality.  Here's a list of key sequences that I use most often.

Ctrl-a Ctrl-a - Toggles between two sessions.  It's kind of like the "last" button on your TV remote.

Ctrl-a c - Creates a new terminal session.

Ctrl-a A - Let's you set the title of a terminal session.  The title is what appears in the session list.

Ctrl-a " - Lists the terminal sessions.  You can then use arrow keys or vi keys to select which session you wan to go to.

Ctrl-a d - Detach screen so you can log off and come back later.  Later when you sign back on, just type screen -r to reattach to your running screen session.

Ctrl-a : - Bring up screen's colon prompt (similar to vi).  From the colon prompt you can issue commands directly to screen.  (You'll probably use it most resizing window splits).

Ctrl-a [ - Start copy mode (for copy/paste).

Ctrl-a ] - Paste what you copied from copy mode.

Ctrl-a S - Split the screen horizontally.

Ctrl-a | (pipe) - Split the screen vertically.

Ctrl-a {tab} - Move focus from window split to window split

Ctrl-a X - Close the current window split.

Split example

Here a quick example of doing a couple window splits.  First, start screen and a few terminal sessions:

steve@ubuntu64 ~ $ screen
Ctrl-a c
Ctrl-a c
Ctrl-a "

Your terminal window should look like this:

Now press Ctrl-a S to split the screen.  You should now have an upper split and a lower split.  Press {return} now to select terminal session 2 in the upper split.  In that terminal session run the top command.

Now press Ctrl-a {tab}.  This will move the cursor down to the bottom split.  Then press Ctrl-a " to bring up the session list.

Select session 0 and then issue the following command in the lower split: watch ls.

Next, let's split the lower window vertically.  Press Ctrl-a | (pipe), Ctrl-a {tab}, Ctrl-a ".  Select session 1.

As you can see, you can keep splitting, sub-splitting, etc.  Remember, Ctrl-a {tab} moves the cursor to the next window split.  That's how you move around.

When you're done, you can use Ctrl-a X to close the window splits.  This does not quit/kill the process running in that split.  The process stays in the session list until you exit it normally.


Screen reads and processes the ~/.screenrc file at startup.  Here's mine:

vbell off
startup_message off
defscrollback 1000
bindkey -k k1 colon
bindkey -k k2 screen
bindkey -k F2 windowlist -b
hardstatus off
caption always '%{Yb}[%=%t%=]'
altscreen on
shell -${SHELL}

You can go through the screen man page for what each line does.  You can customize screen however you'd like.

There are two features that combined make startup processing especially interesting:
  1. You can use the -c command line argument to give screen an alternate rc file to process on startup.
  2. Using commands, you can do anything in the rc file that you could do interactively.
Using this notion, I create different screen rc files for different contexts.  For example, when I'm working at a certain client, I would like to have screen set up a certain way, with specific terminal sessions setup.  I create a bash aliases for each client.

In my ~/.bash_aliases file I have:

alias clientX='screen -c ~/screen_profiles/clientX'
alias clientY='screen -c ~/screen_profiles/clientY'

My ~/screen_profiles/clientX file looks like this:

altscreen on
vbell off
attrcolor b ".I"
termcapinfo xterm 'Co#256:AB=\E[48;5;%dm:AF=\E[38;5;%dm'
defbce "on"
defscrollback 1500
bindkey -k k1 colon
bindkey -k k2 screen
bindkey -k F2 windowlist -b
screen -t "Home" bash
screen -t "Build" ssh my_username@build_server
screen -t "Dev2" ssh my_username@dev_server
screen -t "WebLogic" my_username@weblogic_server
screen -t "Todo/Notes" vi ~/notes/todo.markdown
hardstatus off
caption always '%{Yb}[%=%t%=]'
select 0
split -v
resize +20
select 5

This rc file does the following:
  • Does some normal setup that I do anyway (key bindings, colors, etc.)
  • Creates a terminal session titled, "Home" and runs bash in it.
  • Creates a terminal session titled, "Build" and ssh's me into the build server.
  • Creates a terminal session titled, "Dev2" and ssh's me into the dev 2 server.
  • Creates a terminal session titled, "WebLogic" and ssh's me into the weblogic server.
  • Creates a terminal session titled, "Todo/Notes" and edits my ~/notes/todo.markdown file using vi in it.
  • Splits the window vertically, resizing the left window to be 20 columns bigger.
  • Selects the "Home" session for the left window.
  • Selects the "Todo/Notes" session for the right window.
  • Puts the cursor focus on the left window.
So when I come into clientX's office for the day, I simply type clientX at the bash prompt and within a second, I have screen all set up with my splits and automatically logged into all the servers I commonly use.  Pretty cool.

Thursday, May 17, 2012

Gitolite... my new shiny penny

At my current client engagement, I'm serving as the build-guy and all around code librarian.  I'm managing 35 or so git repositories... keeping them tidy and organized.

Since the client seems to heavily utilize consultant help, they have a veritable revolving door of staff.  Before my arrival they decided to purchase and use a tool called gitorious to aid in managing the repositories and access to them.  It's pretty much like a private github.  It has all kinds of features including web-based management of users/keys, browse-able git logs, etc.  They have it running as an appliance on a virtual server here at the office.

It works pretty well and I've embraced it because it's the "officially-blessed-by-enterprise-architecture" tool.

It's not without it's drawbacks though:

  • In it's default configuration, EVERYONE/ANONYMOUS has READ access to ALL repositories.
  • Gitorious is heavily integrated with git's hooks, so if you want to add your own hooks to a repository, you're pretty much out of luck... unless you like tampering with their code - good luck with that.
  • If you don't want use their appliance and want to install it yourself, it's not all that straight forward.
  • The UI isn't really that intuitive.  For example if you want to see a list of tags for a repository, you have to drill into a branch rather than seeing them at the repository level.
  • It costs money if you want to have your own on-site copy.
All that said, I started looking around for alternatives for managing medium to large numbers of git repositories (more than 5'ish) both for my own stuff as well as for future clients.

In looking around, I found a tool called gitolite.  Gitolite seem to provide all the features that I want/need and has the benefits of being simple/easy to implement as well as being opensource/free.

Gitolite doesn't have a user interface.  You mange it with... what else... git!

Installation is easy

If you're running on Ubuntu, you're only a command away from installation:

steve@ubuntu64 ~ $ sudo apt-get install gitolite

That get's app installed and ready to configure.

Configuration is easy too

Instead of housing gitolite on my personal account (which you could easily do... I just choose not to), I created an account for it called git.

steve@ubuntu64 ~ $ sudo adduser git
Adding user `git' ...
Adding new group `git' (1002) ...
Adding new user `git' (1002) with group `git' ...
Creating home directory `/home/git' ...
Copying files from `/etc/skel' ...
Enter new UNIX password:
Retype new UNIX password:
passwd: password updated successfully
Changing the user information for git
Enter the new value, or press ENTER for the default
        Full Name []: Git Repository Home
        Room Number []:
        Work Phone []:
        Home Phone []:
        Other []:
Is the information correct? [Y/n] y
steve@ubuntu64 ~ $

After creating the account, you'll want to copy your public key to the git account's home directory for the one and only step of initial gitolite configuration.

steve@ubuntu64 ~ $ sudo cp .ssh/ /home/git/
steve@ubuntu64 ~ $ sudo chown git.git /home/git/
steve@ubuntu64 ~ $

Now we're ready to do the initial configuration.  You'll need to become the git account to do it.

steve@ubuntu64 ~ $ su - git

Then you'll want to issue the command gl-setup from the bash prompt, passing in the key you just copied.

git@ubuntu64:~$ gl-setup

It will prompt you to edit the default .gitolite.rc file, but don't change anything, it's fine as it is for now.  You may have a need to change it in the future but I haven't so far.  Just save and quit out of your editor without changing anything.

The default settings in the rc file (/home/git/.gitolite.rc) are fine for most
people but if you wish to make any changes, you can do so now.

hit enter...
creating gitolite-admin...
Initialized empty Git repository in /home/git/repositories/gitolite-admin.git/
creating testing...
Initialized empty Git repository in /home/git/repositories/testing.git/
[master (root-commit) 09aad52] start
 2 files changed, 6 insertions(+)
 create mode 100644 conf/gitolite.conf
 create mode 100644 keydir/

That's it.  Gitolite is all set up and ready to use!

git@ubuntu64:~$ exit
steve@ubuntu64 ~ $

Using gitolite is a piece-o-cake

Managing your git repositories with gitolite is done entirely by manipulating a git repository called gitolite-admin.  First let's use gitolite to create a new repository.  Step one:  clone the gitolite-admin repository from your system:

steve@ubuntu64 ~ $ cd dev
steve@ubuntu64 ~/dev $ git clone git@localhost:gitolite-admin.git
Cloning into 'gitolite-admin'...
remote: Counting objects: 6, done.
remote: Compressing objects: 100% (4/4), done.
remote: Total 6 (delta 0), reused 0 (delta 0)
Receiving objects: 100% (6/6), done.
steve@ubuntu64 ~/dev $

You'll notice that the gitolite-admin clone has a pretty simple directory structure.

steve@ubuntu64 ~/dev $ cd gitolite-admin
steve@ubuntu64 ~/dev/gitolite-admin(master) $ ls -la
total 20
drwxr-xr-x  5 steve steve 4096 May 17 09:23 .
drwxrwxr-x 13 steve steve 4096 May 17 09:23 ..
drwxrwxr-x  2 steve steve 4096 May 17 09:23 conf
drwxrwxr-x  8 steve steve 4096 May 17 09:24 .git
drwxrwxr-x  2 steve steve 4096 May 17 09:23 keydir
steve@ubuntu64 ~/dev/gitolite-admin(master) $

It has two directories: conf and keydir.  The conf directory contains a file called gitolite.conf.  You edit the gitolite.conf file to create repositories and manage permissions to those repositories.

steve@ubuntu64 ~/dev/gitolite-admin(master) $ ls -l conf
total 4
-rw-rw-r-- 1 steve steve 92 May 17 09:23 gitolite.conf
steve@ubuntu64 ~/dev/gitolite-admin(master) $

The keydir directory is where you should place public key files of the users you want to grant access to repositories.  You'll notice that the public key you copied for your self is already there!

steve@ubuntu64 ~/dev/gitolite-admin(master) $ ls -l keydir
total 4
-rw-rw-r-- 1 steve steve 391 May 17 09:23
steve@ubuntu64 ~/dev/gitolite-admin(master) $

Let's add that new repository.  To do so, simply edit the conf/gitolite.conf file with your favorite text editor.

steve@ubuntu64 ~/dev/gitolite-admin(master) $ vi conf/gitolite.conf

Add the following lines to it:

repo    gitolite-admin
        RW+     =   steveb
repo    testing
        RW+     =   @all
repo my_cool_new_repository
RW+ =   steveb

The two new lines in the file establish that you want to add a new repository named my_cool_new_repository and that you want to grant the user with the public key file full READ/WRITE/REFEDIT permissions.

Save the file. Commit it.  Then push it back to the repository origin.

steve@ubuntu64 ~/dev/gitolite-admin(master) $ git add .
steve@ubuntu64 ~/dev/gitolite-admin(master) $ git commit -m"Added new repository my_cool_new_repository"
[master 393ef15] Added new repository my_cool_new_repository
 1 file changed, 3 insertions(+)
steve@ubuntu64 ~/dev/gitolite-admin(master) $ git push
Counting objects: 7, done.
Delta compression using up to 2 threads.
Compressing objects: 100% (3/3), done.
Writing objects: 100% (4/4), 433 bytes, done.
Total 4 (delta 0), reused 0 (delta 0)
remote: creating my_cool_new_repository...
remote: Initialized empty Git repository in /home/git/repositories/my_cool_new_repository.git/
To git@localhost:gitolite-admin.git
   09aad52..393ef15  master -> master
steve@ubuntu64 ~/dev/gitolite-admin(master) $

Gitolite sees your change to the conf/gitolite.conf file.  It then creates a new blank, bare git repository called my_cool_new_repository.git.  It also ensures that the file has the appropriate access to the git account (check out the git account's ~/.ssh/authorized_keys).

That's it.  Your new repository is ready to clone!

steve@ubuntu64 ~/dev/gitolite-admin(master) $ cd ..
steve@ubuntu64 ~/dev $ git clone git@localhost:my_cool_new_repository.git
Cloning into 'my_cool_new_repository'...
warning: You appear to have cloned an empty repository.
steve@ubuntu64 ~/dev $ cd my_cool_new_repository/
steve@ubuntu64 ~/dev/my_cool_new_repository(master) $ date > file1.txt
steve@ubuntu64 ~/dev/my_cool_new_repository(master) $ git add .
steve@ubuntu64 ~/dev/my_cool_new_repository(master) $ git commit -m"Added file1.txt"
[master (root-commit) 90e1635] Added file1.txt
 1 file changed, 1 insertion(+)
 create mode 100644 file1.txt
steve@ubuntu64 ~/dev/my_cool_new_repository(master) $ git push origin master
Counting objects: 3, done.
Writing objects: 100% (3/3), 257 bytes, done.
Total 3 (delta 0), reused 0 (delta 0)
To git@localhost:my_cool_new_repository.git
 * [new branch]      master -> master
steve@ubuntu64 ~/dev/my_cool_new_repository(master) $

Adding/removing user access is as simple as adding/removing key files from the keydir directory, updating the conf/gitolite.conf file and pushing.

The gitolife.conf file, while very simple, is incredibly powerful.  It supports groups.  Also, it will not only allow you to control access to git repositories, but will also allow you to assign permissions to branches and tags with regular expressions.  Take a look at the documentation for much more info.

Tuesday, May 1, 2012

Use automator to create X11 shortcuts in OSX

I have a Mac and yet I do a fair amount of work in X11apps on Linux systems.  I really like the ssh's -Y flag that let's me run my Linux X11 apps and have their display output (and input) sent back to my Mac.

I just stumbled onto this in Automator and thought it might be worth sharing.

You can create Automator "Application" workflows that look/feel/behave like local OSX applications, but are actually running on remote Linux servers.  In this posting, I will walk through creating an Automator workflow for Synaptic, an application package manager for apt based Linux distributions.


You will need to ensure you have an environment that meets the following requirements in order to follow along:

  • A Mac running OSX (I'm using Lion.  I've not tested this with other versions so if you're using another version, your milage may vary.)
  • You will need or installed on your Mac. comes with OSX up to Lion.  After Lion, you will need to install yourself.
  • Access to a Linux system that is running SSH with X11 tunneling enabled (a default I believe) and gksudo and synaptic installed.
  • You need to have key-based authentication set up with your account so you are not prompted for a password when logging into the Linux machine via ssh.
gksudo is normally not needed.  In this case we're running synaptic, a package manager for Linux, which needs elevated privileges in order to add/remove software packages on the Linux machine.  If the X11 app you are wanting to run doesn't require elevated privileges you won't need gksudo.

Verify that you can log in without as password prompt

Before we create the Automator workflow, let's verify that you can log into the Linux system from your Mac without a password prompt. Start a terminal window and issue a command in the following format:

ssh username@linux_hostname

If you are prompted for a password you will still need to set yourself up for key-based authentication.  Check out any of these links and then come back.  You should be let in without a password as seen below:

Start a new Automator Workflow

Once you have confirmed that you can ssh from your Mac into your Linux system, you're ready to get started.  Run Automator and start a new "Application' workflow.
Then we're going to add a single action to the workflow.  In the Actions library locate the Utilities group and select it.  Then drag the Run Shell Script action to your workflow.
Once you've done that, click on the Options link at the bottom of the action.  It will reveal some options.  Check the Ignore this action's input checkbox.
Finally, replace the cat command with the following ssh command:

ssh -Y username@linux_hostname gksudo -g synaptic &

Where username is your username and linux_hostname is your Linux host's hostname.  Don't forget the & at the end.  Then save your workflow as Synaptic.  You can click the run button if you'd like to test it out.
Once you're satisfied with it, quit out of automator.  Now you can run the workflow directly from OSX.  Locate it in Finder and double-click on it.


That's it.  You can create any number of workflows for individual Linux apps.  I have one for LibreOffice, one for synaptic, one for Eclipse and one for gnome-system-monitor.

Optional - Set icon for your workflow

If you're like me, the generic Automator icon on my Synaptic workflow is a little annoying.  You can set the icon to whatever you'd like.  Here's one way to do it.

First, obtain a copy of the synaptic.png file from your Linux machine and place it on your Mac.  I use the following command from a terminal:

scp steve@ubuntu32.local:/usr/share/pixmaps/synaptic.png ~/Desktop

This puts the synaptic.png file on my Mac's desktop.  Then I double-click the synaptic.png file.  It should open up in Preview.  In preview, I select Edit->Select All, then Edit->Copy.
Next, I control-click on the Synaptic automator workflow and select Get Info.  This brings up an informational dialog.
In the top-lefthand corner of the dialog, there is a small Automator icon.  I click on it, causing it to glow blue-ish.  While it's glowing, I press Command-V
 Finally I close the dialog.  I also delete the synaptic.png file as I no longer need it.  Now  My synaptic Automator workflow has he official synaptic icon.

I hope you found this posting helpful.