From 1ec08892882e2d4dc72a858fcc965926024cd615 Mon Sep 17 00:00:00 2001
From: foldsters <37962412+foldsters@users.noreply.github.com>
Date: Sat, 28 Apr 2018 20:07:49 -0400
Subject: [PATCH 1/3] Huffman encoding python implementation

Compatible Python 2.7+
---
 .../data_compression/huffman/code/python.py   | 116 ++++++++++++++++++
 1 file changed, 116 insertions(+)
 create mode 100644 chapters/data_compression/huffman/code/python.py

diff --git a/chapters/data_compression/huffman/code/python.py b/chapters/data_compression/huffman/code/python.py
new file mode 100644
index 00000000..246a39f9
--- /dev/null
+++ b/chapters/data_compression/huffman/code/python.py
@@ -0,0 +1,116 @@
+# Huffman Encoding
+# Python 2.7+
+# Submitted by Matthew Giallourakis 
+
+from collections import Counter
+
+# constructs the tree
+def build_tree(message):
+   
+    # get sorted list of character,frequency pairs
+    frequencies = Counter(message)
+    trees = frequencies.most_common()
+
+    # while there is more than one tree
+    while len(trees) > 1:
+        
+        # pop off the two trees of least weight from the trees list
+        tree_left,weight_left = trees.pop()
+        tree_right,weight_right = trees.pop()
+        
+        # combine the nodes and add back to the nodes list
+        new_tree = [tree_left,tree_right]
+        new_weight = weight_left+weight_right
+        trees.append((new_tree,new_weight))
+
+        # sort the trees list by weight
+        trees = sorted(trees, key=lambda n: n[1], reverse=True)
+
+    tree = trees[0][0]
+    return tree
+
+# constructs the mapping with recursion
+def build_mapping(tree,code=''):
+
+    results = []
+
+    # split the tree
+    left_tree,right_tree = tree
+
+    # if the left node has children, find the mapping of those children
+    # else pair the character with the current code + 0 
+    if type(left_tree) is list:
+        results += build_mapping(left_tree,code+'0')
+    else:
+        results.append((left_tree,code+'0'))
+
+    # if the right node has children, find the mapping of those children
+    # else pair the character with the current code + 1 
+    if type(right_tree) is list:
+        results += build_mapping(right_tree,code+'1')
+    else:
+        results.append((right_tree,code+'1'))
+
+    return results
+
+# encodes the message
+def encode(mapping,message):
+    
+    encoding = ""
+
+    # build a char -> code dictionary
+    forward_dict = dict(mapping)
+
+    # replace each character with its code
+    for char in message:
+        encoding += forward_dict[char]
+    
+    return encoding
+
+# decodes a message
+def decode(mapping,encoding):
+    
+    message = ""
+    key = ""
+
+    # build a code -> char dictionary
+    inverse_dict = dict([(v,k) for k,v in mapping])
+
+    # for each bit in the encoding
+    # if the bit is in the dictionary, replace the bit with the paired character
+    # else look at the bit and the following bits together until a match occurs
+    # move to the next bit not yet looked at
+    for index,bit in enumerate(encoding):
+        key += bit
+        if key in inverse_dict:
+            message += inverse_dict[key]
+            key = ""
+    
+    return message
+
+def main():
+
+    # test example
+    message = "bibbity_bobbity"
+    tree = build_tree(message)
+    mapping = build_mapping(tree)
+    encoding = encode(mapping,message)
+    decoding = decode(mapping,encoding)
+
+    print('message: '+message)
+    print('tree: '+str(tree))
+    print('mapping: '+str(mapping))
+    print('encoding: '+encoding)
+    print('decoding: '+decoding)
+
+    # prints the following:
+    #
+    #  message: bibbity_bobbity
+    #  tree: ['b', [[['_', 'o'], 'y'], ['t', 'i']]]
+    #  mapping: [('b', '0'), ('_', '1000'), ('o', '1001'),
+    #            ('y', '101'), ('t', '110'), ('i', '111')]
+    #  encoding: 01110011111010110000100100111110101
+    #  decoding: bibbity_bobbity
+    
+if __name__ == '__main__':
+    main()

From 7d74ce2f419d0483bd348286a642213225045e8a Mon Sep 17 00:00:00 2001
From: foldsters <37962412+foldsters@users.noreply.github.com>
Date: Sat, 28 Apr 2018 20:11:56 -0400
Subject: [PATCH 2/3] Rename chapters/data_compression/huffman/code/python.py
 to chapters/data_compression/huffman/code/python/python.py

---
 chapters/data_compression/huffman/code/{ => python}/python.py | 0
 1 file changed, 0 insertions(+), 0 deletions(-)
 rename chapters/data_compression/huffman/code/{ => python}/python.py (100%)

diff --git a/chapters/data_compression/huffman/code/python.py b/chapters/data_compression/huffman/code/python/python.py
similarity index 100%
rename from chapters/data_compression/huffman/code/python.py
rename to chapters/data_compression/huffman/code/python/python.py

From 43a56e1fc81adbb81d38cb142dc154ceeeb82e95 Mon Sep 17 00:00:00 2001
From: foldsters <37962412+foldsters@users.noreply.github.com>
Date: Sat, 28 Apr 2018 20:13:00 -0400
Subject: [PATCH 3/3] Rename python.py to huffman.py

---
 .../huffman/code/python/{python.py => huffman.py}                 | 0
 1 file changed, 0 insertions(+), 0 deletions(-)
 rename chapters/data_compression/huffman/code/python/{python.py => huffman.py} (100%)

diff --git a/chapters/data_compression/huffman/code/python/python.py b/chapters/data_compression/huffman/code/python/huffman.py
similarity index 100%
rename from chapters/data_compression/huffman/code/python/python.py
rename to chapters/data_compression/huffman/code/python/huffman.py