#!/usr/bin/env python
# coding: utf-8

# # Encoding
# 
# ## Goals
# 
#  - A string is more than a sequence of bytes
#  - A string is a couple (bytes, encoding)
#  - Use unicode_literals  in python2
#  - Manage differently encoded filenames
#  - A string is not a sequence of bytes

# ## Modules

# In[ ]:


import os
import os.path
import glob


# In[ ]:


from os.path import isdir
basedir = "/tmp/course"
if not isdir(basedir):
    os.makedirs(basedir) 


# # Song of Childhood
# 
# <table><tr>
# <td width=10%>
# Als das Kind Kind war,
# ging es mit hängenden Armen,
# wollte der Bach sei ein Fluß,
# der Fluß sei ein Strom,
# und diese Pfütze das Meer.
# 
# Als das Kind Kind war,
# wußte es nicht, daß es Kind war,
# alles war ihm beseelt,
# und alle Seelen waren eins.
# 
# </td>
# <td>
# <h1><center><i>
# "When the child was a child,<br>
# characters were bytes, and <br>
# strings list of bytes."</i>
# </center></h3>
# </td>
# <td width=10%>
# When the child was a child
# It walked with its arms swinging,
# wanted the brook to be a river,
# the river to be a torrent,
# and this puddle to be the sea.
# 
# When the child was a child,
# it didn’t know that it was a child,
# everything was soulful,
# and all souls were one.
# 
# </td>
# </tr>
# </table>

# # Encoding is a map
# 
# Encoding is a map between typographical characters and byte-sequences.
# 
# Decoding is its reverse map.
# 
# 
# |char -> | utf8 | cp1252 | ascii |
# |-------------|--------|---|
# |y -> | [121] | [121] | [121] |
# |z -> | [122] | [122] | [122] |
# |{ -> | [123] | [123] | [123] |
# |¢ -> | [194, 162] | [162] | - |
# |£ -> | [194, 163] | [163] | - |
# |¤ -> | [194, 164] | [164] | - |
# |¥ -> | [194, 165] | [165] | - |
# |Ɓ -> | [198, 129] | - | - |
# |Ƃ -> | [198, 130] | - | - |
# |ƃ -> | [198, 131] | - | - |
# 

# In[ ]:


# Py3 doesn't need the 'u' prefix before the string.
the_string = u"S\u00fcd" # Sued
print(the_string)


# In[ ]:


# the_string Sued can be encoded in different...
in_utf8 = the_string.encode('utf-8')
in_win = the_string.encode('cp1252')

# ...byte-sequences
assert type(in_utf8) == bytes 


# In[ ]:


# Now you can see the differences between 
print(repr(in_utf8))
# and
print(repr(in_win))


# In[ ]:


# Decoding bytes using the wrong map...
# ...gives Süd results
print(in_utf8.decode('cp1252'))


# In[ ]:


# Filenames are actually binary data
#  we should be careful when our scripts read
#  eg from a vfat filesystem.

# To make Py2 encoding-aware we must
from __future__ import unicode_literals, print_function

# Create 3 windows-encoded filenames in

# using the provided function
from course import create_espana
create_espana(basedir)


# In[ ]:


# Just list the newly created files
# and check that they are not showing correctly (unless we have windows :D)
get_ipython().system('dir {basedir}')


# In[ ]:


from glob import glob as ls 
#expands wildcards like ls

# To avoid encoding issue like the following...
files = ls("/tmp/course/*.txt")

#UnicodeDecodeError: 'ascii' codec can't decode
# byte 0xe9 in position 5: # remember ñ in cp1252
# ordinal not in range(128)


# In[ ]:


# We must explicitly use bytes prefixing with "b"
files = ls(b"/tmp/course/*.txt")
# And the file names are shown with bytes.
print(files)


# In[ ]:


# Exercise: don't run this cell! 
# Which outcome do you expect from the following instruction?
print('\n'.join(files))


# In[ ]: