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

# ## Practice - read last line from external file   
# Python Workbook, chp5  
# August 7, 2020  (past several days, internet lookup has better solutions)

# ### 0. Check python version, active conda env  

# In[1]:


get_ipython().system('python --version')


# In[2]:


get_ipython().system('conda env list')
# * is next to active conda env.  


# ### 1. Start practice

# In[3]:


f = open('demofile.txt', 'rt')
blob = f.read()  # read all 
print(blob)


# In[4]:


print(f.closed)  # check if f is closed at this point.


# In[5]:


f = open('demofile.txt')  # Open again to go back to file start point.  
line = f.readline()  # read first line, no S
print(line)


# In[6]:


f = open('demofile.txt')
lines = f.readlines()  # read all lines, iterable list. With S
print(lines[-1])  # print last line
f.close()


# ### 2. Binary format reads 'rb'  

# In[7]:


# Try binary format reading.  
with open('demofile.txt', 'rb') as f:  # binary format
    first = f.readline()
    f.seek(-5, 2)  # Go to pointer 5th bytes from end of file 
    #last = f.read()  # does readline work?  May not read all of the last line.  
    last = f.readline()  # readline does work.  Still not all of the line.  
    print(first, last)


# In[8]:


# Seek from end of file needs a loop.  
with open('demofile.txt', 'rb') as f:  # binary format
    first = f.readline()
    
    f.seek(-2, 2)  # Go to pointer 2 bytes relative to end   
    while f.read(1) != b"\n":   # while read one byte is not equal to binary '\n'
        f.seek(-2, 1)   # go back 1 byte.  
    # after while loop, this is beginning of last line    
    last = f.read()  # read all bytes from here.   
    # last = f.readline() gives same output.  
    
    print(str(first)+"\n", str(last))


# ### 3. Good for short files, readlines() all at once, save, index [ ] into each line

# In[9]:


# Good for short text files.  
with open('demofile.txt', 'rt') as f:  
    lines = f.readlines()
    first = lines[0]
    last = lines[-1]
    print(first, last)

print(f.closed)  # check file is cloased after "with" block end.  


# ### 4. seek(0) to go back to start, works with 'rt' 

# In[10]:


f2 = open('demofile.txt')  
lines = f2.readlines()
print("lines: ", lines)

f2.seek(0)  # Goto beginning of file
blob= f2.read()
print("blob: ", blob)

f2.seek(0)
one = f2.readlines(1)
print("one: ", one)

f2.seek(0)
two = f2.readlines()[1]  # Works!  prints 2nd item. nifty. 
print("two: ", two)


# ### 5. Only binary read 'rb' can use relative indexing  

# In[11]:


# seek can be used with text format?  
# Yes, but can't do relative location other than 0, current point. 
# If the file is opened in text mode (without ‘b’), only offsets returned by tell() are legal. 
# none of the relative indexing from back/current location works without 'b' binary format.  

with open('demofile.txt', 'rb') as f:  # need to be in binary format
    first = f.readline()
    print(first)
    
    f.seek(0, 0)  
    last = f.readlines()[2]
    print(last)
 
    #f.seek(-2, 2)  # not work, only 0 offset works with 2, 1 mode.  
    f.seek(0, 2)  # end of file work with text format file.  
    print(f.tell())
    f.seek(-11, 1)
    print(f.tell())
    last = f.read()
    print(last) 


# In[12]:


with open('demofile.txt') as f:  # text format again
    f.seek(0, 2)  # end of file  
    loc = f.tell(); print(loc)
    
    f.seek(0, 0)  # go back to beginning of tile 
    print(f.tell())
    first = f.readline()
    print(first, f.tell())
    
    last = f.readlines()[-1]
    print(last, f.tell())
    
    f.seek(31, 0)
    two = f.readlines()
    print(two, f.tell())
    
    f.seek(31, 0)
    second = f.readline()
    print(second, f.tell())
    
    third = f.readline()
    print(third)


# ### 6. Doen't solve large file last line problem, but another way to move around 'rt' format.

# In[14]:


with open('demofile.txt') as f:  # text format again
    
    # num_ln = len(f.readlines()); print(num_ln)
    
    num_ch_all = len(f.read())
    print(num_ch_all)
    
    f.seek(0)  # back to beginning
    num_ch_last = len(f.readlines()[-1]); 
    print(num_ch_last)
    
    offset = num_ch_all - num_ch_last  # offset from file beginning
    f.seek(offset, 0)
    print(f.tell())
    print(f.read())


# In[ ]: