├── ts.txt
├── README.rst
└── markov.py


/ts.txt:
--------------------------------------------------------------------------------
https://raw.githubusercontent.com/mattharrison/pycon-beg-markov-2017/master/ts.txt


--------------------------------------------------------------------------------
/README.rst:
--------------------------------------------------------------------------------
 1 | ========================================
 2 |  Hands-on Beginning Python 3.6 Bootcamp
 3 | ========================================
 4 | 
 5 | This repository contains the final output from the 2017 PyCon
 6 | tutorial `Beginning Python Bootcamp
 7 | <https://us.pycon.org/2017/schedule/presentation/160/>`_.
 8 | 
 9 | 
10 | Resources
11 | -----------
12 | 
13 | * https://github.com/mattharrison/Tiny-Python-3.6-Notebook
14 | * http://setosa.io/blog/2014/07/26/markov-chains/
15 | * http://www.gutenberg.org/ebooks/74
16 | * https://pymotw.com/3/argparse/
17 | 
18 | 


--------------------------------------------------------------------------------
/markov.py:
--------------------------------------------------------------------------------
  1 | """
  2 | This is a docstring. Here is an example
  3 | of running a Markov prediction:
  4 | 
  5 | >>> m = Markov('ab')
  6 | >>> m.predict('a')
  7 | 'b'
  8 | 
  9 | >>> m.predict('c')
 10 | Traceback (most recent call last):
 11 |   ...
 12 | KeyError
 13 | 
 14 | >>> get_table('ab')
 15 | {'a': {'b': 1}}
 16 | 
 17 | >>> random.seed(42)
 18 | >>> m = Markov('Find a city, find yourself a city to live in', 4)
 19 | >>> test_predict(m, 20, 'F', 4)
 20 | 'Find a city, find a c'
 21 | 
 22 | >>> with open('ts.txt', encoding='windows_1252') as fin:
 23 | ...     data = fin.read()
 24 | >>> m2 = Markov(data, 4)
 25 | >>> test_predict(m2, 100, 'T', 4)
 26 | """
 27 | 
 28 | 
 29 | import argparse
 30 | import random
 31 | import sys
 32 | 
 33 | 
 34 | class Markov:
 35 |    
 36 |     def __init__(self, data, size=1):
 37 |         self.tables = []
 38 |         for i in range(size):
 39 |             self.tables.append(get_table(data, i+1))
 40 |         #self.table = get_table(data)
 41 |         
 42 |     def predict(self, data_in):
 43 |         table = self.tables[len(data_in) - 1]
 44 |         options = table.get(data_in, {})
 45 |         if not options:
 46 |             raise KeyError()
 47 |         possible = ''
 48 |         for result, count in options.items():
 49 |             possible += result*count
 50 |         result = random.choice(possible)
 51 |         return result
 52 | 
 53 |     
 54 | def get_table(line, numchars=1):
 55 |     
 56 |     results = {}
 57 |     for i, char in enumerate(line):
 58 |         #print(i, char)
 59 |         chars = line[i:i+numchars]
 60 |         try:  # if i == len(line): # Look before you leap
 61 |             out = line[i+numchars]
 62 |         except IndexError:
 63 |             # easier to ask for forgiveness than permission
 64 |             break
 65 |         char_dict = results.setdefault(chars, {})
 66 |         char_dict.setdefault(out, 0)
 67 |         char_dict[out] += 1
 68 |         results[char] = char_dict
 69 |     return results
 70 | 
 71 | def test_predict(m, num_chars, start, size=1):
 72 |     res = [start]
 73 |     for i in range(num_chars):
 74 |         let = m.predict(start)
 75 |         res.append(let)
 76 |         start = ''.join(res)[-size:]
 77 |     return ''.join(res)
 78 | 
 79 | def repl(m, size=1):
 80 |     """
 81 |     This starts a repl, provide a Markov and
 82 |     optional size
 83 |     """
 84 |     while 1:
 85 |         txt = input(">")
 86 |         try:
 87 |             res = m.predict(txt[-size:])
 88 |         except KeyError:
 89 |             print("Try again...")
 90 |         print(res)
 91 | 
 92 | def main(args):
 93 |     p = argparse.ArgumentParser()
 94 |     p.add_argument('-f', '--file', help='Input file')
 95 |     p.add_argument('--encoding', help='File encoding default(utf8)',
 96 |                    default='utf8')
 97 |     p.add_argument('-s', '--size', help='Size of input default(1)',
 98 |                    default=1, type=int)
 99 |     p.add_argument('-t', '--test', help='run tests', action='store_true')
100 | 
101 |     opts = p.parse_args(args)
102 |     if opts.file:
103 |         with open(opts.file, encoding=opts.encoding) as fin:
104 |             data = fin.read()
105 |         m = Markov(data, opts.size)
106 |         repl(m)
107 |     if opts.test:
108 |         import doctest
109 |         doctest.testmod()
110 |     
111 | 
112 | if __name__ == '__main__':
113 |     print("EXECUTED")
114 |     #import doctest
115 |     #doctest.testmod()
116 |     main(sys.argv[1:])
117 | else:
118 |     print("IMPORTED")
119 | 
120 | 
121 | 


--------------------------------------------------------------------------------