import hashlib
sha1_hash_function = hashlib.sha1
sha1_hash_function

message = "git is a rude word in UK English"
hash_value = sha1_hash_function(message).hexdigest()
hash_value

sha1_hash_function("git is a rude word in UK English.").hexdigest() # Add a full stop

# Make the directory for the paper and change into the directory
from os import chdir, getcwd, mkdir, listdir
from os.path import split as psplit, abspath, isdir
import shutil
# If we are in the nobel_prize directory already, get out
current_path = abspath(getcwd())
if current_path.endswith('nobel_prize'):
    chdir('..') # Change down a directory
# If the directory exists, nuke it
if isdir('nobel_prize'):
    shutil.rmtree('nobel_prize')
# Make the new directory, and change to it
mkdir('nobel_prize')
chdir('nobel_prize')
# Check we are there
print(getcwd())
# The directory should be empty at this point
assert listdir('.') == []

%%file nobel_prize.txt
The brain is just a set of random numbers
=========================================

We have discovered that the brain is a set of random numbers.

We have charts and graphs to back us up.


%%file very_clever_analysis.py
# The brain analysis script
import numpy as np

# Make brain data
brain_size = (128, 128)
random_data = np.random.normal(size=brain_size)


# Get the figure from storage
!cp ../images/stunning_figure.png .

import os
os.listdir('.')

!mkdir .fancy_backups

!mkdir .fancy_backups/1
!mkdir .fancy_backups/1/files

!cp * .fancy_backups/1/files

""" A little utility to display the structure of directory trees

Don't worry about the detail here, you'll see below what it does.
The code is not complicated, but it's not relevant to the main points on git.

Inspired by:

http://lorenzod8n.wordpress.com/2007/12/13/creating-a-tree-utility-in-python-part-2/

with thanks.

This version is my own copyright (Matthew Brett) released under 2-clause BSD
"""
from os import getcwd, listdir
from os.path import isdir, join as pjoin

# Unicode constants for constructing the tree trunk and branches
ALONG = u'\u2500'
DOWN = u'\u2502'
DOWN_RIGHT = u'\u251c'
ELBOW_RIGHT = u'\u2514'
BLUE = u'\033[94m'
ENDC = u'\033[0m'
DOWN_RIGHT_ALONG = DOWN_RIGHT + ALONG * 2 + u" "
ELBOW_RIGHT_ALONG = ELBOW_RIGHT + ALONG * 2 + u" "
CONTINUE_INDENT = DOWN + u' ' * 3
FINISH_INDENT = u' ' * 4


def print_tree(root_path=None, indent_str=u''):
    """ Print tree structure starting from `root_path`

    Parameters
    ----------
    root_path : None or str, optional
        path from which to print directory tree structure.  If None, use current
        directory.
    indent_str : str, optional
        prefix to print for every entry in the tree.  Usually '', and then set
        by recursion into the function when printing subdirectories.

    Returns
    -------
    None
    """
    if root_path is None:
        root_path = getcwd()
    # ensure return of unicode paths from listdir
    root_path = unicode(root_path)
    paths = sorted(listdir(root_path))
    if len(paths) == 0:
        return
    for path in paths[:-1]:
        print_path(root_path, path, indent_str, False)
    print_path(root_path, paths[-1], indent_str, True)


def print_path(root_path, path, indent_str, last_entry=False):
    """ Print individual `path` (file or directory name) from `root_path`

    Parameters
    ----------
    root_path : str
        path containing `path`
    path : str
        file name or directory name
    indent_str : str
        string to prefix to entry for this `path`
    last_entry : bool, optional
        Whether this is the last entry in a list of paths.

    Returns
    -------
    None
    """
    full_path = pjoin(root_path, path)
    have_dir = isdir(full_path)
    leader = ELBOW_RIGHT_ALONG if last_entry else DOWN_RIGHT_ALONG
    path_colored = BLUE + path + ENDC if have_dir else path
    print(indent_str + leader + path_colored)
    if have_dir:
        new_indent = FINISH_INDENT if last_entry else CONTINUE_INDENT
        print_tree(full_path, indent_str + new_indent)

%%file .fancy_backups/1/info.txt
Date: April 1 2012, 14.30
Author: I. M. Awesome
Notes: First backup of my amazing idea

print_tree()

!echo "The charts are very impressive\n" >> nobel_prize.txt
!mkdir .fancy_backups/2
!mkdir .fancy_backups/2/files
!cp * .fancy_backups/2/files

%%file .fancy_backups/2/info.txt
Date: April 1 2012, 18.03
Author: I. M. Awesome
Notes: Fruit of enormous thought

!echo "The graphs are also compelling\n" >> nobel_prize.txt
!mkdir .fancy_backups/3
!mkdir .fancy_backups/3/files
!cp * .fancy_backups/3/files

%%file .fancy_backups/3/info.txt
Date: April 2 2012, 11.20
Author: I. M. Awesome
Notes: Now seeing things clearly

print_tree()

# Now we just cheat, and copy the commits
!cp -r .fancy_backups/3 .fancy_backups/4
!cp -r .fancy_backups/3 .fancy_backups/5
!cp -r .fancy_backups/3 .fancy_backups/6

!mkdir .fancy_backups/staging_area
!cp .fancy_backups/6/files/* .fancy_backups/staging_area
!ls .fancy_backups/staging_area

!cp nobel_prize.txt .fancy_backups/staging_area

!mkdir .fancy_backups/7
!mkdir .fancy_backups/7/files
!cp .fancy_backups/staging_area/* .fancy_backups/7/files

%%file .fancy_backups/7/info.txt
Date: April 10 2012, 14.30
Author: I. M. Awesome
Notes: Changes to introduction

!cp very_clever_analysis.py .fancy_backups/staging_area

# Our commit procedure; same as last time with "8" instead of "7"
!mkdir .fancy_backups/8
!mkdir .fancy_backups/8/files
!cp .fancy_backups/staging_area/* .fancy_backups/8/files

%%file .fancy_backups/8/info.txt
Date: April 10 2012, 14.35
Author: I. M. Awesome
Notes: Crazy new analysis

print_tree('.fancy_backups')

!mkdir .fancy_backups/objects

import hashlib
# read the bytes for the figure into a string
figure_data = open('.fancy_backups/1/files/stunning_figure.png', 'rb').read()
# Calculate the hash value
figure_hash = hashlib.sha1(figure_data).hexdigest()
figure_hash

!cp .fancy_backups/1/files/stunning_figure.png .fancy_backups/objects/$figure_hash

import os

def copy_to_hash(fname):
    # Read the data
    binary_data = open(fname, 'rb').read()
    # Find the hash
    hash_value = hashlib.sha1(binary_data).hexdigest()
    # Save with the hash filename
    hash_fname = '.fancy_backups/objects/' + hash_value
    if os.path.isfile(hash_fname):
        print "We already have a hash file for " + fname
    else:
        open(hash_fname, 'wb').write(binary_data)
    return hash_value

paper_hash = copy_to_hash('.fancy_backups/1/files/nobel_prize.txt')
script_hash = copy_to_hash('.fancy_backups/1/files/very_clever_analysis.py')

print 'Hash for stunning_figure.png:', figure_hash
print 'Hash for nobel_prize:', paper_hash
print 'Hash for very_clever_analysis.py:', script_hash

print_tree('.fancy_backups/objects')

%%file .fancy_backups/1/directory_list
Filename                Hash value
========                ==========
stunning_figure.png     aff88ecead2c7166770969a54dc855c8b91be864
nobel_prize.txt         3af8809ecb9c6dec33fc7e5ad330e384663f5a0d
very_clever_analysis.py e7f3ca9157fd7088b6a927a618e18d4bc4712fb6

# Nothing to for unchanged figure - already have hash value file
figure_hash = copy_to_hash('.fancy_backups/2/files/stunning_figure.png')
# Makes a new copy
paper_hash = copy_to_hash('.fancy_backups/2/files/nobel_prize.txt')
# Nothing to for unchanged script - already have hash value file
script_hash = copy_to_hash('.fancy_backups/2/files/very_clever_analysis.py')
print 'Hash for stunning_figure.png:', figure_hash
print 'Hash for nobel_prize:', paper_hash
print 'Hash for very_clever_analysis.py:', script_hash

%%file .fancy_backups/2/directory_list
Filename                Hash value
========                ==========
stunning_figure.png     aff88ecead2c7166770969a54dc855c8b91be864
nobel_prize.txt         bc330d8886bb1b36a49d8e7ebb07d3443190b0e6
very_clever_analysis.py e7f3ca9157fd7088b6a927a618e18d4bc4712fb6

tree_hash = copy_to_hash('.fancy_backups/1/directory_list')
tree_hash

copy_to_hash('.fancy_backups/2/directory_list')

%%file .fancy_backups/1/commit
Date: April 1 2012, 14.30
Author: I. M. Awesome
Notes: First backup of my amazing idea
Tree: 63d466086dd359c0b34d21e04b812781c7153b23

commit_hash = copy_to_hash('.fancy_backups/1/commit')
commit_hash

ls .fancy_backups/objects

!cat .fancy_backups/objects/f473e51a38d40772722205c92dddd4bdfd941ef8

%%file .fancy_backups/temporary_file
Date: April 1 2012, 18.03
Author: I. M. Awesome
Notes: Fruit of enormous thought
Tree: d45095659d1ea567b90aaf923e265bf41cbb126f
Parent: 63dab6e48ed2e9111624a3b5391bdf784c040b86

commit2_hash = copy_to_hash('.fancy_backups/temporary_file')

!git config --global user.name "Matthew Brett"
!git config --global user.email "matthew.brett@gmail.com"

# Put here your preferred editor. 
!git config --global core.editor gedit

!git config --global color.ui "auto"

!git

# First pull the contents of the first commit from .fancy_backups
!cp .fancy_backups/1/files/* .
# Then delete .fancy_backups - we're done with those guys
!rm -rf .fancy_backups

!ls -a

!git init

!ls .git

print_tree('.git/objects')

!git add nobel_prize.txt

print_tree('.git/objects')

!git status

import zlib # A compression / decompression library

def read_git_object(filename):
    """ Read an object in the ``.git/objects`` directory

    Yes, it's this simple - in this case.  But git reserves the right to use
    more complicated storage formats to save space when your repository is larger
    """
    compressed_contents = open(filename, 'rb').read()
    return zlib.decompress(compressed_contents)

filename = '.git/objects/d9/2d079af6a7f276cc8d63dcf2549c03e7deb553'
new_object_data = read_git_object(filename)
print new_object_data

!git add stunning_figure.png
!git add very_clever_analysis.py

!git status

print_tree('.git/objects')

print read_git_object('.git/objects/8a/dc8bf371d669242ea998f78f9916867cc6203c')

!git commit -m "First backup of my amazing idea"

print_tree('.git/objects')

print read_git_object('.git/objects/1e/322f4e782df4dfe963eda96517886f4e39a454')

!git ls-tree 1e322f4

!git log

!git log --parents

!git branch

!git branch -v

!ls .git/refs/heads

!cat .git/refs/heads/master

!echo "\nThe charts are very impressive\n" >> nobel_prize.txt

!git diff

!git status

!git commit

!git status

!git add nobel_prize.txt

!git status

!git commit -m "Fruit of enormous thought"

!cat .git/refs/heads/master

!git log

!git log --parents

!git log --oneline --topo-order --graph

# We create our alias (this saves it in git's permanent configuration file):
!git config --global alias.slog "log --oneline --topo-order --graph"

# And now we can use it
!git slog

!git mv very_clever_analysis.py slightly_dodgy_analysis.py
!git status

!git commit -m "I like this new name better"

!git slog

!cat .git/HEAD

!cat .git/refs/heads/master

!git branch -v

!git status
!ls

!git branch experiment

!ls .git/refs/heads

!git branch -v

!cat .git/refs/heads/experiment

!git checkout experiment

!git branch -v

!cat .git/HEAD

!echo "Some crazy idea" > experiment.txt
!git add experiment.txt
!git commit -m "Trying something new"
!git slog

!ls

!cat .git/refs/heads/experiment

!git checkout master
!cat .git/HEAD

!ls

!git slog

!echo "All the while, more work goes on in master..." >> nobel_prize.txt
!git add nobel_prize.txt
!git commit -m "The mainline keeps moving"
!git slog

!ls

!git merge experiment -m "Merge in the experiment"

!git slog

!git branch trouble
!git checkout trouble
!echo "This is going to be a problem..." >> experiment.txt
!git add experiment.txt
!git commit -m"Changes in the trouble branch"

!git checkout master
!echo "More work on the master branch..." >> experiment.txt
!git add experiment.txt
!git commit -m"Mainline work"

!git config merge.conflictstyle diff3

!git merge trouble -m "Unlikely this one will work"

!cat experiment.txt

%%file experiment.txt
Some crazy idea
More work on the master branch...
This is going to be a problem...

!git status

!git add experiment.txt
!git commit -m "Completed merge of trouble, fixing conflicts along the way"
!git slog